THESIS
3
zoo;

5%?30157

 

This is to certify that the

dissertation entitled

CHARACTERIZATION OF THE SPATIAL DISTRIBUTION
OF Heterodera glycines Ichinohe 1955 (NEMATODA),
SOYBEAN CYST NEMATODE IN TWO MICHIGAN FIELDS

 

presented by

Maria Felicitas Avendano

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Entomology

 

 

 

Date Z/z/a}

MS U is an Affirmative Action/Equal Opportunity Institution

 

0-12771

 

 

 

LIBRARY

Michigan State

University

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65«p.15

 

 

 

CHARACTERIZATION OF THE SPATIAL DISTRIBUTION OF Heterodera glycines
Ichinohe 1955 (N EMATODA), SOYBEAN CYST NEMATODE IN TWO MICHIGAN
FIELDS
By

Maria Felicitas Avendano

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Entomology

2003

ABSTRACT
CHARACTERIZATION OF THE SPATIAL DISTRIBUTION IN TWO MICHIGAN
FIELDS OF Heterodera glycines Ichinohe 1955 (N EMATODA), SOYBEAN CYST
NEMATODE
By

Maria Felicitas Avendafio

Heterodera glycines Ichinohe 1955 (N EMATODA) (soybean cyst nematode,
SCN) is recognized as the major pest limiting soybean [Glycine max (L.) Merr.]
production, accounting for approximately 1.67 x 109 US. dollars of soybean yield loss
annually in the United States. Despite current management efforts, SCN continues to
spread throughout soybean producing areas worldwide. The goal of this project was to
understand SCN spatial distribution in soybean fields as the first step towards developing
site-specific management (SSM) strategies for SCN. If SCN is to be managed site-
specifically its spatial distribution should be structured and relatively time invariant, and
it has to be related to yield-limiting factors easier to monitor and manage. The literature
suggests that SCN may meet these requirements for SSM. Geostatistical tools and
classical statistics were applied to test the hypotheses that SCN’s spatial distribution
within a field is sufficiently structured and time invariant; that SCN spatial distribution
and population densities are related to soil properties; and that the relations among SCN
population density, soil properties and soybean yield are sufficient in magnitude to aid in
the management decision-making process. A nested survey sampling design was applied
on two SCN-infested fields in MI and soil and soybean root samples were collected at

monthly intervals during the growing seasons of 1999 and 2000. Soil samples were

analyzed for SCN population density, soil fertility and soil texture. The SCN population
in the roots was also quantified. To assess host response, soybean leaf samples were
collected twice in 2000 for tissue analysis, and soybean yield was recorded in 1999 and in
2000.

The within field variability in cysts, eggs per cyst, and eggs was large in both
fields. The spatial structure in SCN population varied with sampling times, but a periodic
pattern in semivariograms appeared consistently fi'om planting to harvest in both fields.
The difficulty in adequately fitting wave models to the empirical semivariograms
underestimated in some cases the spatial structure in SCN population. Soil texture, pH,
and calcium concentration in the soil were strongly correlated and cross- correlated with
SCN density in the soil, and to a lesser extent in the roots. Correlations were maintained
consistently over time. The nutritional status of the crop reflected the interactions of soil
fertility, soil texture, and SCN population density. Bean yield was also strongly
correlated with soil texture, soil pH and calcium concentration, and SCN population
density in the soil.

The results contribute significantly towards the understanding of SCN spatial
distribution in soybean fields, provide evidence of the underlying factors involved in
determining grain yield, and lay the base for further research on cause and effect relations
to advance understanding SCN biology, ecology, and management opportunities. The
spatial variability in yield correlated to the combined effect of SCN density and
unfavorable soil conditions observed in this work provided support for the notion of

management zone delineation. Thus, the hypotheses tested were demonstrated true.

To Gustavo and Lupecita

iv

ACKNOWLEDGMENTS

I am most grateful to Dr. Haddish Melakeberhan for giving me the opportunity
and the means to pursue a Doctoral Degree and to Drs. Fran J. Pierce, Oliver
Schabenberger, James Miller, George W. Bird, and Stuart Gage, my Advisor Committee
for their guidance, support and encouragement through these years. Within their own
expertise they have become my models, setting high academic standards for me to
follow. They fostered my critical thinking skills and taught me the way to be the best
scientist I can be through constructive critics and productive discussions.

I am thankful to the Department of Entomology for providing the framework for
my program and for giving me the opportunity to teach while pursuing my degree. I
greatly appreciate the assistance of the faculty and staff, and the numerous opportunities
provided to discuss my project with highly recognized scientists through seminars and
invited guests.

The work presented here would not have been possible without the help and
support of others. Mr. Charles and Mr. Dave Eickholt kindly made the research fields
available, with the assistance of Mr. Neil Miller. Data collection was possible thanks to
Chad Holovach, and Nathan Nye, Megan Kierzek, Carolyn Stein, Nathan Cottrell,
Bindiya Shah, Dan Armstrong, Marisol Soto, Carolina Holteuer, Katrina Blakely and
Teresa Raslich, who stood by me in the field during long, hot summer days.

My thanks to Dr. Melakeberhan, the Department of Entomology, The
Graduate School, USDA Project through the Michigan Agricultural Experimental

Station, and Michigan Soybean Promotion Committee, North Central Soybean Research

Program and United Soybean Board for providing the financial support for my program
and to attend scientific meetings.

And many, many thanks to my office mates and friends Nathan Cottrell, Kirsten
F ondren, Nate Siegert, Laura Lazarus, Marcelo J ans, and all those who came and left
during my stay in room 42. Thank you for the good times, and thank you for your support
during the tough phases.

Words are not enough to thank my husband Gustavo and my daughter Lupe for
their endless loving patience as I made my way through the program. I could not have

done it without their love, support and encouragement.

vi

 

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES
KEY TO SYMBOLS AND ABBREVIATIONS

CHAPTER 1
Introduction and Review of Literature

CHAPTER 2
General Materials and Methods
Selection of Experimental Sites
Soil and soybean root sampling for SCN and soil characterization
Soybean tissue analysis and grain yield
Statistical Analysis
Software Used

CHAPTER 3
Geostatistical Analysis of Field Spatial Distribution Patterns of
Soybean Cyst Nematode
Abstract
Introduction
Materials and Methods
Experimental Sites and Site Management
Soil Sampling Design
Soil Sampling for SCN
Cyst Extraction
Statistical Analysis
Geostatistical Analysis
Results
Population Densities of SCN
Geostatistical Analysis
Kriging
Discussion
CHAPTER 4
The Spatial Distribution of Soybean Cyst Nematode in Relation to Soil
Texture and Soil Map Unit
Abstract
Introduction

Materials and Methods
Research Site, Sampling Design, and Soil Sample Collection

vii

xiii

xix

14
14
15
15
15
16

17
19
20
22
22
23
25
25
26
27
30
30
34
39

50
52
53
57
57

SCN Analysis
Soil texture Analysis
Statistical Analysis
Results
Soil Particle Size Distribution and Spatial Analysis
SCN Population Density
Relationship between Soil Map Units and SCN Population
Density
Relationship between Soil Particle Size and SCN Population
Density
Discussion

CHAPTER 5
Relationship between Soil Texture and Soybean Cyst Nematode
population dynamics in soil and in soybean roots
Abstract
Introduction
Materials and Methods
Research Site and Sampling Design
Soil and Soybean Root Sampling for SCN
Soil Texture
Statistical Analysis
Results
SCN Population in the Soil
SCN Population in Soybean Roots
SCN Population Dynamics and Soil texture
Spatial Analysis of Cyst Population
Discussion
Conclusion

CHAPTER 6
Spatio-temporal Analysis of Soybean Cyst Nematode, soil Fertility,
and Plant Nutrient Status
Abstract
Introduction
Materials and Methods
Experimental Sites and Site Management
Sampling Design, Collection of Samples and Soil Analysis
SCN Analysis
Soybean Leaf Tissue Nutrient Analysis
Statistical Analysis
Results
Soil Analysis
Tissue Analysis
SCN Population and Soil Fertility
SCN Population and Tissue Analysis

viii

 

58
59
60

66

67

7O
78

82
84
85
87
87
88
90
91
94
94
94
97
100
1 13
I 17

119
121
122
124
124
125
125
126
126
128
128
133
140
144

Discussion
Conclusion

CHAPTER 7
Spatial Analysis of Soybean Yield in Relation to Soil Texture, Soil
fertility, and Soybean Cyst Nematode Population Density
Abstract
Introduction
Materials and Methods
Statistical Analysis
Results
Soybean Yield
Soybean Yield and Soil Attributes
Soybean Yield and SCN Population
Discussion

CHAPTER 8
General Discussion and Conclusions

APPENDIX A
Elutriation Procedure Efficiency to Extract SCN Cysts
from Three Soil Types
Materials and Methods
Soil Types, SCN Inoculum, and Experimental
Design
Elutriator and Extraction Methods
Statistical Analysis
Results
Method Efficiency by Soil Type
Contamination
Method Efficiency by Inoculum Doses, by Soil Type
Discussion

APPENDIX B
Software for Geostatistics and Spatial Analysis

APPENDIX C
Record of Deposition of Voucher Specimens

LITERATURE CITED

ix

146
152

153
155
156
156
158
159
159
165
167
170

173

181
182

182
182
184
185
185
186
186
188

191

197

200

LIST OF TABLES

3.1. SCN (soybean cyst nematode) cyst and egg population densities, and eggs per cyst
from two fields in Shiawassee Co., MI before planting in 1999 and 2000. Summary
statistics. Page 32.

3.2. Parameters of the theoretical semivariograrn models of SCN (soybean cyst
nematode) population density in two fields in Shiawassee Co., MI before planting in
1999 and 2000. Page 36.

4.1. Soil texture in Field A and Field B as determined using a modified Bouyoucos
hydrometer method (Gee and Bauder, 1986). Page 64.

4.2. Parameters of the theoretical semivariograrn models of sand, silt, and clay fractions
in Field A and Field B before planting in 1999. Page 65.

4.3. Mean SCN eggs per cyst by soil type in Field A and Field B in 1999 and in 2000.
Page 69.

4.4. Regression models and coefficients of determination of the relationship between cyst
population density and sand, clay, or silt proportion in the soil in Field A and Field B
in 1999 and in 2000. Page 71.

4.5. Regression models and coefficients of determination of the relationship between
eggs per cyst and sand, clay, or silt proportion in the soil in Field A and Field B in
1999 and in 2000. Page 73.

5.1. Sampling dates for soil and soybean root samples collected from Fields A and B in
1999 and in 2000. Page 89.

5.2. SCN mean population density in soybean roots and in the soil in Field A in 1999 and
in 2000. Page 95.

5.3. SCN mean population density in soybean roots and in the soil in Field B in 1999 and
in 2000. Page 96.

5.4. Simple linear regression parameters for the effect of soil texture on SCN cyst
population density in the soil in Fields A and B in 1999 and in 2000. Page 98.

5.5. Simple linear regression parameters for the effect of soil texture on the number of
SCN eggs per cyst in Fields A and B in 1999 and in 2000. Page 99.

 

5.6. Simple linear regression parameters for the effect of soil texture on SCN population
density in soybean roots in Fields A and B in 1999 and in 2000 shown in Figs. 5.1-3.
Page 104.

5. 7. Parameters of the theoretical semivariograrn models of SCN cyst population density
fi'om samples collected at monthly intervals in Fields A and B in 1999 and in 2000.
Pages 108-109.

6.1. Parameters of the theoretical sernivariogram models of soil pH, P, K, Ca, and Mg in
Fields A and B before planting in 1999. Page 130.

6.2. Pearson’s correlation coefficients for percent sand, clay and silt, soil pH, and soil
macronutrients (Kg ha") with tissue nutrients at V9 and R3 in 2000 in Field A.
Page 137.

6.3. Pearson’s correlation coefficients for percent sand, clay, and silt, soil pH, and soil
macronutrients (Kg ha") with tissue nutrients at R2 in 1999, and at V9 and R3 in
2000 in Field B. Page 138-139.

6.4. Pearson’s linear correlation coefficients for SCN population density in the soil or in
soybean roots in relation to soil pH, Ca, and Mg concentration. Page 140.

6.5. Pearson’s linear correlation coefficients for cyst population density in the soil at
planting and at 19 and 44 DAP in 1999 and 200, respectively, in relation to macro and
micronutrients concentration in soybean leaves at R2 in 1999, and V9 and R3 stages
of growth in Field B in 2000. Page 145.

7.1. Soybean yield within the areas sampled in Fields A and B in 1999 and in 2000.
Page 160.

7.2. Parameters of the theoretical semivariogram models of soybean yield in Fields A and
B in 1999 and in 2000. Page 163.

7.3. Pearson’s correlation coefficients for soybean seed yield with sand, clay, silt, and soil
pH and concentrations of P, K, Ca and Mg (Kg ha") in Fields A and B in 1999 and in
2000. Page 165.

7.4. Pearson’s correlation coefficients for soybean seed yield and SCN population density
sampled at monthly intervals in Fields A and B in 1999 and in 2000. Page 168.

A. 1. Efficiency (%) of four elutriation methods to extract SCN cysts from artificially
inoculated soil. Page 186.

A2. Average number of cysts recovered from the control sandy loam, muck and clay
loam samples, using different elutriation methods. Page 186.

xi

3.1. Summary of some of the benefits and drawbacks of the software package Surfer 7.
Page 193.

B2. Summary of some of the benefits and drawbacks of the software package Variowin
2.21. Page 194.

3.3. Summary of some of the benefits and drawbacks of the software package
SAS/STAT. Page 194.

8.4. Summary of some of the benefits and drawbacks of the software package GSLIB.
Page 195.

xii

LIST OF FIGURES

3.1. Soil series maps and location of the soil samples collected from a- Field A and b-
Field B in Shiawasee Co., MI in 1999 and 2000. c- Nested sampling design with
geometric progression reduced distances. Distances between pairs of samples are
indicated; the direction in which two consecutive samples were oriented was
randomly selected. This design was applied within 50 x 50 m cells of a grid centered
in each field. Black circles in figures a and b indicate sets of sample locations in
selected cells for both 1999 and 2000. The crosses in figures a and b indicate the
locations of the additional samples collected in 2000 from alternate nodes of a 25 x
25 m grid. Soil series maps were digitized from Threlkeld and Feenstra (1974).

Page 24.

3.2. Frequency distribution and cumulative distribution function (cdt) of a- cysts 100 cm-
3 of soil, b- eggs per cyst, and c- eggs 100 cm-3 of soil from Field A; and d- cysts 100
cm-3 of soil, e- eggs per cyst, and f- eggs 100 cm-3 of soil from Field B. Samples
were collected before planting in 1999 and 2000. The number of eggs per cyst was
estimated as the average number of eggs found after crushing three randomly selected
cysts in each sample. Page 33.

3.3. Semivariograms of a, b- cysts [loglo (cysts 100 cm-3 of soil +1)], 0, d- eggs per cyst
[logm (eggs per cyst +1)] and e, f- eggs [loglo (eggs 100 cm-3 of soil +1)] from Field
A in a, c, e- 1999 and b, d, f- 2000. Black circles indicate omnidirectional empirical
semivariogram, the solid line indicates the theoretical model fitted by means of least
squares and the dashed line is the sample variance. Page 37.

3.4. Semivariograms of a, b- cysts [logm (cysts 100 cm-3 of soil +1)], c, d- eggs per cyst
[loglo (eggs per cyst +1)] and e, f- eggs [logm (eggs 100 cm-3 of soil +1)] from Field
B in a, c, e- 1999 and b, d, f- 2000. Black circles indicate omnidirectional empirical
sernivariogram, the solid line indicates the theoretical model fitted by means of least
squares and the dashed line is the sample variance. Page 38.

3.5. Log-kriged maps represent the distribution of cysts in the area sampled within field
A in a— 1999 and b- 2000; the distribution of eggs per cyst in c- 1999 and d- 2000, and
the distribution of eggs in e- 1999. The shading scale indicates levels of cysts [loglo
(cysts 100 cm; 3of soil +1)], eggs per cyst [loglo (eggs per cyst +1)], and eggs [log 10
(eggs 100 cm 3of soil +1)]. A solid line delineates the field boundaries. Page 40.

3.6. Cross-correlograms between a, d- cysts [loglo (cysts 100 cm 3of soil +31)], b, e- eggs
per cyst [loglo (eggs per cyst +1)], and c, f- eggs [log 10 (eggs 100 cm 3of soil +1)] in
1999 and 2000 m Field A (a, b, c) and Field B (d, e, f). r— = linear correlation
coefficient. Page 41.

xiii

3. 7. Cross-correlograms between a, c- eggs per cyst [loglo (eggs per cyst +1)] in 1999 and
cysts [loglo (cysts 100 cm 3of soil +1)] m 2000; and b, d- eggs [log l0 (eggs 100 cm'3
of soil +1)] m 1999 and cysts in 2000, in Field A (a, b) and Field B (c, d). r— = linear
correlation coefficient. Page 42.

3.8. Log-kriged maps represent the distribution of cysts in the area sampled within field B
in a- 1999 and b- 2000; the distribution of eggs per cyst in c- 1999; and the
distribution of eggs in d- 1999 and e- 2000. The shading scale indicates levels of cysts
[logm(cysts 100 cm :of soil +1)], eggs per cyst [loglo (eggs per cyst +1)], and eggs
[log 10 (eggs 100 cm 3of soil +1)]. A solid line delineates the field boundaries.

Page 43.

4.1. Field A. a- Nested design for the collection of soil samples. Samples were collected
at the indicated distances in random directions within grid cells of 50 x 50 m. b-
Location of sampled sites and soil series present in Field A. Crosses indicate the
location of samples collected for the determination of SCN population density; black
circles indicate which of those samples were also used for soil particle size analysis.
Soil series are l- Newaygo sandy loam, 2- Conover loam, 3- Brookston loam, 4-
Breckenridge sandy loam, and 5- Belding sandy loam. Soil series map was digitized
from Threlkeld and F eenstra (1974). c-Spatial distribution of sand, d- clay, and e- silt
percentages in the soil as interpolated by ordinary kriging within the area sampled.
The shadings in c, d, and e represent percentage ranges. f- Soil type delineation
determined based on the proportion of sand, silt, and clay. Soil types are 1- Loamy
sand, 2- Sandy loam, 3- loam, and 4- sandy clay loam. Page 62.

4.2. Field B. a- Location of sampled sites and soil series present in Field B. Crosses
indicate the location of samples collected for the determination of SCN population
density (see Fig. la); black circles indicate which of those samples were also used for
soil particle size analysis. Soil series are l- Newaygo sandy loam, 2- Brookston loam,
and 3- Berville loam. Soil series map was digitized from Threlkeld and Feenstra
(1974). b—Spatial distribution of sand, c- clay, and d- silt proportion in the soil as
interpolated by universal kriging within the area sampled. The shadings in c, d, and e
represent percentage ranges. e- Soil type delineation determined based on the
proportion of sand, silt, and clay. Soil types are 1- Loamy sand, 2- Sandy loam, and
3- Sandy clay loam. Page 63.

4.3. Boxplots of SCN (a) cyst population density and (c) eggs per cyst in Field A; and (b)
cyst population density and (d) eggs per cyst in Field B. The y-axis indicates the
sampling time: P’99: Planting 1999, H’99: Harvest 1999, P’OO: Planting 2000 and
H’OO: Harvest 2000. The sample mean is indicated with a dashed vertical line and the
median with a solid vertical line in each box. Significant differences among the
means at different sampling times within each field are indicated with lower case
letters. Mean comparisons were performed on logarithmic transformed data; original
data are shown for clarity. Means with the same letter were not significantly different
(protected LSD, 5 % significance level). Page 67.

xiv

4.4. SCN mean cyst population density [loglo (cysts 100 cm'3 of soil + 1)] a: standard
deviation by soil type in Field A (open circles) and in Field B (black circles) at (a)
planting 1999, (b) harvest 1999, (c) planting 2000 and (d) harvest 2000. Soil types are
(SCL) sandy clay loam, (SL) sandy loam, (LS) loamy sand and (L) loam. Means with
the same letter are not significantly different (protected LSD, 5 % significance level).
Page 68.

4.5. Relationship between SCN cyst population density [loglo (cysts 100 cm'3 of soil +
1)] and the proportion of each soil fi'action in the sample. Means 3: standard deviation
for Field A (open circles) and for Field B (black circles) are indicated. The left
column (a, b, c, (1) corresponds to the percentage of sand, the central column (e, f, g,
h) corresponds to clay percentage and the right column (i, j, k, 1) corresponds to silt
percentage; at (a, e, i) planting 1999, (b, f, j) harvest 1999, (c, g, k) planting 2000, and
(d, h, l) harvest 2000. Regression curves fitted are indicated as solid lines (Field B)
and dashed lines (Field A) where significant (5 % significance level). Regression
coefficients are shown in Table 4.4. Page 72.

4.6. Relationship between SCN eggs per cyst [logm (eggs per cyst 100 cm'3 of soil + 1)]
and the proportion of each soil fraction in the sample. Means 3: standard deviation in
Field A (open circles) and in Field B (black circles) are indicated. The left column (a,
b, c, (1) corresponds to sand percentage, the central column (e, f, g, h) corresponds to
clay percentage and the right column (i, j, k, 1) corresponds to silt percentage; at (a, e,
i) planting 1999, (b, f, j) harvest 1999, (c, g, k) planting 2000, and (d, h, l) harvest
2000. Regression curves fitted are indicated as solid lines (Field B) and dashed lines
(Field A) where significant (5 % significance level). Regression coefficients are
shown in Table 4.5. Page 74.

4.7. Cross-correlograms of SCN cyst population density and percent sand (solid line) or
clay (dashed line) in the soil in Field A (a-d) and Field B (e-h). a, e- Planting 1999; b,
f- Harvest 1999; c, g- Planting 2000; d, h- Harvest 2000. Linear correlation
coefficients for cyst density with sand and clay are indicated. Page 76.

4.8. Cross-correlograms of SCN eggs per cyst and percent sand (solid line) or clay
(dashed line) in the soil in Field A (a—d) and Field B (e-h). a, e- Planting 1999; b, f-
Harvest 1999; c, g- Planting 2000; d, h- Harvest 2000. Linear correlation coefficients
for eggs per cyst with sand and clay are indicated. Page 77.

5.1 . Relationship between sand percentage in the soil and SCN second- (J 2), third/fourth-
juveniles (J 3/4), and immature females in soybean roots in Field A (A-C) and Field B
(D-F). In Field A, root samples were collected at 31 and 17 days after planting
(DAP), and in Field B at 44 and 19 DAP in 1999 and in 2000, respectively.
Parameters of the regression lines are shown in Table 5.6. Page 101 .

5.2. Relationship between clay percentage in the soil and SCN second- (J 2), third/fourth-

juveniles (J 3/4), and immature females in soybean roots in Field A (A-C) and Field B
(D-F). In Field A, root samples were collected at 31 and 17 days after planting

XV

(DAP), and in Field B at 44 and 19 DAP in 1999 and in 2000, respectively.
Parameters of the regression lines are shown in Table 5.6. Page 102.

5.3. Relationship between silt percentage in the soil and SCN second- (J2), third/fourth-
juveniles (J 3/4), and immature females in soybean roots in Field A (A-C) and Field B
(D-F). In Field A, root samples were collected at 31 and 17 days after planting
(DAP), and in Field B at 44 and 19 DAP in 1999 and in 2000, respectively.
Parameters of the regression lines are shown in Table 5.6. Page 103.

5.4. Semivariograms of cyst population density [Loglo(cysts 100 cm'3 of soil +1)] in Field
A at (A) 31, (B) 59, (C) 86, and (D) 126 DAP in 1999; and at (E) 17, (F) 45, (G) 73,
and (H) 147 DAP in 2000. Black circles indicate omnidirectional empirical
semivariograrn, the solid line is the theoretical model fitted by means of least squares,
and the dashed line is the sample variance. Page 106.

5.5. Semivariograms of cyst population density [Logm(cysts 100 cm'3 of soil +1)] in Field
B at (A)44, (B) 73, (C) 100, and (D) 125 DAP in 1999; and at (E) 19, (F) 46, (G) 77,
and (H) 155 DAP in 2000. Black circles indicate omnidirectional empirical
sernivariogram, the solid line is the theoretical model fitted by means of least squares,
and the dashed line is the sample variance. Page 107.

5.6. SCN cyst spatial distribution in Field A at planting (Pl) (Avendano et al., 2003-
Chapter Three), and at the days after planting (DAP) indicated, in 1999 and in 2000.
Shadings scale indicate levels of loglo (cysts 100 cm'3 soil + 1). Parameters of the
empirical semivariograms are shown in Table 5.7. Page 111.

5.7 . SCN cyst spatial distribution in Field B at planting (Avendano et al., 2003- Chapter
Three), and at the days after planting (DAP) indicated, in 1999 and 2000. Shadings
scale indicates levels of loglo (cysts 100 cm'3 soil + 1). Parameters of the empirical
semivariograms are shown in Table 5.7 . Page 112.

6.1. Field A and Field B, location of sampling sites. White irregular areas indicate where
soybean leaf samples were collected in 1999 from Field B, and in 2000 from Fields A
and B. Each circle includes fi'om one to four sites from the nested survey sampling
design where soil samples were previously collected (black circles)(Chapter Three).
Page 128.

6.2. Box-plots of pH and P, K, Ca, and Mg quantified in soil samples collected at
planting in 1999 from Field A and Field B. Means (dotted line) with the same lower
case letter were not significantly different between fields (protected LSD, 5 %
significance level). Medians are indicated with a full line. Whiskers represent the 5th
and 95th percentiles; outliers are indicated with black circles. Page 129.

6.3. Spatial distribution of pH, and concentration of P, K, Ca, and Mg in the soil as

interpolated by kriging from samples collected before planting in 1999 in Field A at
the locations indicated with black circles. Page 131.

xvi

6.4. Spatial distribution of pH, and concentration of P, K, Ca, and Mg in the soil as
interpolated by kriging from samples collected before planting in 1999 in Field B at
the locations indicated with black circles. Page 132.

6.5. Box-plots of N, P, K, Ca, Mg, and S concentration (%) in soybean leaf tissue
samples collected at R2 in 1999 from Field B, and at V9 and R3 in 2000 fi'om fields
A and B. Significant differences among the means (dotted lines) at different sampling
times within each field are indicated with lower case letters. Means with the same
letter were not significantly different (protected LSD, 5 % significance level).
Medians are indicated with a full line. Whiskers represent the 5th and 95th percentiles,
outliers are indicated with black circles. Page 134.

6.6. Box-plots of Fe, Zn, Mn, Cu, and B concentrations (ppm) in soybean leaf tissue
samples collected at R2 in 1999 from Field B, and at V9 and R3 in 2000 from fields
A and B. Significant differences among the means (dotted lines) at different sampling
times within each field are indicated with lower case letters. Means with the same
letter were not significantly different (protected LSD, 5 % significance level).
Medians are indicated with a full line. Whiskers represent the 5th and 95th percentiles,
outliers are indicated with black circles. Page 135.

6.7. Cross-correlograms of soil pH and SCN population density in Field A and Field B.
Lag distance is the separation distance between samples. SCN population density was
determined in soil samples, or in root samples where indicated. Page 142.

6.8. Figure 6.8. Cross-correlograms of soil Ca and SCN population density in Field A and
Field B. Lag distance is the separation distance between samples. SCN population
density was determined in soil samples, or in root samples where indicated. Page 143.

6.9. Cross-correlograms of Mg concentration in the soil and SCN population density in
Field A in the cases where correlation analysis was significant. Cyst population
density was quantified in the soil at 31 DAP in 1999, and at 17 DAP and harvest in
2000. SCN population in the roots was quantified at 17 DAP in 2000. Lag distance is
the separation distance between samples. Page 144.

7.1. Soybean seed yield in Field A in 1999 and in 2000. Seed yield was recorded by a
yield monitor system connected to a GPS unit mounted on the combine. Black circles
indicate the locations where soil samples were collected for SCN, soil texture and soil
fertility analyses. This image in color in the dissertation. Page 161.

7.2. Soybean seed yield in Field B in 1999 and in 2000. Seed yield was recorded by a
yield monitor system connected to a GPS unit mounted on the combine. Black circles
indicate the locations where soil samples were collected for SCN, soil texture and soil
fertility analyses. This Image in color in the dissertation. Page 162.

xvii

7.3. Anisotropic semivariograms of soybean yield (Kg/ha) in Fields A and B in 1999 and
in 2000 in different directions in degrees from East. The dashed line is the sample
variance. Parameters of the sernivariogram models are shown in Table 7.2. Page 164.

7.4. Cross-correlograms of yield and soil texture (% sand, % clay, and % silt), and soil
fertility (pH, and Mg, Ca, K) in Field A in 1999 and in 2000. Cross-correlation at zero
lag distance equals the linear correlation coefficient. Page 166.

7.5. Cross-correlograms of yield and soil texture (% sand, % clay, and % silt), and soil
fertility (pH, and P, K, Ca in Kg ha'l) in Field B in 1999 and 2000. Cross-correlation
at zero lag distance equals the linear correlation coefficient. Page 167.

7.6. Cross-correlograms of yield and SCN population density in Field A in 1999 and in
2000. SCN population density was quantified in soil (cysts) and in soybean root
samples (all developmental stages) collected at the days after planting (DAP)
indicated. Cross-correlation at zero lag distance equals the linear correlation
coefficient. Page 169.

7.7. Cross-correlograms of yield and SCN cyst population density in soil samples in Field
B in 1999 and in 2000. Soil samples were collected at the days after planting (DAP)
indicated. Cross-correlation at zero lag distance equals the linear correlation
coefficient. Page 169.

7.8. Cross-correlograms of yield and SCN population density in soybean roots in Field B
in 1999. Root samples were collected at the days after planting (DAP) indicated.
Cross-correlation at zero lag distance equals the linear correlation coefficient.

Page 170.

8.1. SCN-soybean system. The diagram represents some of the factors and elements
involved in this system in a commercial soybean field. Page 178.

A]. Extraction efficiency (means i sd) of four elutriation methods to recover SCN cysts

from sandy loam, muck and clay loam artificially inoculated with 50, 100 and 150
cysts 100 cm'3 of soil. Page 187.

xviii

Co
cdf
DAP

GPS

SCN
sd

3,, s,-

SSM
Z(s)

70!)

KEY TO SYMBOLS AND ABBREVIATIONS

Nugget effect

Cumulative distribution function

Days after planting

Global positioning system

Lag distance

Pearson’s correlation coefficient

Soybean cyst nematode

Standard deviation

Locations of observations separated by the lag h
Site-Specific Management

Realization of a mean-constant spatial process

The sernivariance at a given lag distance

xix

CHAPTER ONE

INTRODUCTION AND REVIEW OF LITERATURE

The phylum Nematoda (N emata) Cobb, 1919 (position reviewed by Chitwood
and Chitwood, 1950 and Maggenti, 1991) is comprised of organisms defined as
roundworm invertebrates with a body cavity and complete digestive tract, which are
nonsegrnented, appendageless, and have a bilateral symmetry (Hirschrnann, 1971; Poinar
1983). The digestive tract of nematodes includes a stoma (mouth), alimentary canal
(esophagus and intestine), and anus. Nematodes have a complex nervous system, a
secretory-excretory system, a reproductive system, and musculature system (longitudinal
muscles), but have no specialized respiratory or circulatory systems. The body is
completely covered by a flexible and permeable cuticle. Traditionally, nematodes have
been placed in two classes within the phylum: the Adenophorea and the Secementea
based largely on morphological and physiological differences (Maggenti, 1991). More
recently, embryology and molecular techniques such as the use of small subunit
ribosomal DNA sequences are being used to redefine the phylogenetic tree and taxonomy
of the phylum Nematoda (Thomas et al., 1997, Blaxter et al., 1998, Voronov et al., 1998).

Nematodes are found in all habitats and ecosystems of the biosphere. They occur
in soil, decaying organic matter, all forms of plant life and most animals, including
domesticated and wild species (Norton, 1978). To date, some 20,000 species of
nematodes have been described, and estimates of the actual number range from 40,000 to
10 million (Blaxter, 1998). If the known nematode species are grouped according to

simple habitat categories, then of the total, 50% are marine nematodes, 25% are soil

inhabiting nematodes (microbivorous and predators), 15% are animal-parasitic
nematodes and 10% are plant-parasitic nematodes (V iglierchio, 1991). Although many
nematodes parasitize plants or animals, others are beneficial, as they contribute to
nutrient cycling or serve as biological control agents of plant pests (Stirling, 1991;
Yeates, 1996; Niles and Freckman, 1998).

Nematode diversity is often high in the soil environment and it is important for
the long-term stability of soil structure and function (Ettema, 1998). The use of feeding
groups classifications has provided significant advances for investigating the role of
nematodes in soil ecosystem processes (Wardle et al., 1995). Major trophic groups of
nematodes include plant-feeding nematodes (parasites), plant-associated nematodes
(herbivores), hyphal-feeding nematodes, bacterial-feeding nematodes, predators, animal
parasites, and omnivores (Y eates, et al., 1993). Relationships between nematode
functional groups and ecological processes have been found in field experiments (Y eates,
1999). The coexistence of species depends largely on the stability of the environment and
the relationships in the abundance of species and trophic groups shifts with different
levels of disturbance (Bongers, 1990). For example, agroecosystems, particularly those
under monoculture management, are less diverse in nematode species than less disturbed
environments such as grasslands (Y eates and Bongers, 1999). The proportion of plant-
parasitic nematodes in the soil ecosystem is relatively small when compared to bacterial
feeders, for example. Nonetheless, populations of plant parasites can increase because of
many environmental stressors (W asilewska, 1995). Fertilization, and in turn increased

nutrient uptake by higher plants, seems to shift the composition of plant-feeding

nematodes, so that plant-associated nematodes are replaced by plant-parasitic nematodes
(Bongers et al., 1997).

Agronomists recognize that plant-parasitic nematodes often constrain crop
growth. The annual worldwide losses caused by nematodes on life-sustaining crops are
estimated to be about 11 percent, adding to an estimate of 100 x 109 US. dollars yield
loss annually (Sasser and Freckman, 1987). Severe infestations of fields with nematodes
such as Meloidogyne spp. or H. glycines Ichinohe 1952 (soybean cyst nematode, SCN)
often result in annual yield losses of 10 to 50 % (McSorley, 1987; Sasser and Freckman,
1987; Wrather et al., 2001a).

Plant-parasitic nematodes are usually invisible to the naked eye (0.5 to 5 mm
long, and 50 to 250 u wide). All known plant-parasitic nematodes possess a buccal stylet
with which they puncture and feed upon the cells of their hosts (Hirschmann, 1971).
Based on their feeding behavior, nematodes can be classified into three groups:
destructive (host cells killed), adaptive (cells modified), and neoplastic (cells modify and
undergo new growth). Root-lesion (Pratylenchus spp.), cyst (Heterodera and Globodera
spp.), and root-knot (Meloidogyne spp.) nematodes are representatives of the three
feeding behaviors respectively (Dropkin, 1980). All three genera are the most wide
spread nematodes in almost all of the life sustaining crops (Sasser and Freckman, 1987).

Plant-parasitic nematodes can traverse limited distances by their own active
movements (Wallace, 1959; Prot and Netscher, 1979). Although distances are short (less
than a meter) active movement can be important if it enables nematodes to be

disseminated by other agents such as wind, water, vehicles and animals. Nematodes may

be moved on agricultural equipment fi'om farm to farm, or with soil, seeds, plant material,
animals and humans to different locations around the world (Lehman, 1994).

Although plant-parasitic nematodes occur in most cultivated soils, associated
damage usually results from high population densities, rather than from mere occurrence.
Expected economic losses for annual crops generally are inversely correlated with the
level of infestation at the time the crop is planted. In any cultivated field, estimates of
current population density and future increase are, therefore, critical in anticipating crop
losses, and fundamental in making nematode management decisions (Duncan and Noling,
1998). The fact that initial numbers of nematodes can be related to the yield of annual
crops has enabled nematologists to develop fimctional advisory programs (Barker and
Nusbaum, 1971) focusing on maintaining phytopathogenic nematode populations
densities below economic thresholds (Ferris, 1978; Heald, 1987). If the predicted crop
loss is less than the economic decision interval, management is unnecessary; if it is above
the economic decision interval, management is justified. If the predicted crop loss is
within the decision interval, a subjective decision must be made based upon the grower’s
economic status or risk aversion/risk acceptance level (Ferris, 1984). Yet, the
development of reliable estimates for economic losses caused by nematodes has proven
to be very difficult (Ferris, 1993; Roberts, 1993). This situation results from a number of
factors, including the impact of environment on the activity of detrimental and beneficial
nematodes (Y eates et al., 1993), as well as crop plants and the frequent involvement of
nematode and disease complexes, and concomitant plant nematode species.

Variability in nematode population density is a serious problem in the analysis

and interpretation of experimental data (Noe and Campbell, 1985). Plant-parasitic

nematodes are not unifome distributed through cultivated soils, but occur in clusters
(Goodell and Penis, 1981; Alby et al., 1983; McSorley et al., 1985; Webster and Boag,
1992; Robertson and Freckman, 1995), with frequency distributions typically describing
negative binomial functions (Taylor et al., 1979; Seinhorst, 1982). This aggregation adds
a substantial degree of uncertainty to most estimates of population size and adds
significantly to the effort required for comprehensive measurement (McSorley and
Parrado, 1982).

Estimates of nematode field populations require accurate and affordable sampling
procedures. The recommended practice for estimating population density is to sample an
area of two hectares or less, taking composite sample units (Barker, 1978). Soil cores in
each sample unit are combined and mixed thoroughly so that a representative portion
(usually 100 cm3) can be processed to extract the nematodes (Dropkin, 1980). The degree
of precision necessary in sampling and the structure of the sampling plan itself, depend
on the purpose for which the sample is taken (Barker and Campbell, 1981).

Studies of nematode population dynamics seek to understand and predict
nematode population growth, and to use this knowledge to improve nematode
management (Seinhorst, 1970; Nusbaum and Barker, 1971; Duncan and McSorley, 1987;
Ferris and Noling, 1987; McSorley and Philips, 1993; McSorley, 1998; Donald et a1,
1999). Many agricultural practices are implemented with the intent of lowering nematode
population densities and improving plant growth.

Management practices designed to limit crop losses to plant-parasitic nematodes
involve one or more tactics that focus on the strategies to reduce the initial inoculum,

and/or limit the rate of nematode population density increase (Roberts, 1993). The

prevention of spread of nematodes, general land management, and cultural practices, such
as crop rotation, physical treatment of infested plant material or media, and the use of
chemical treatments are among the traditional management tactics used in different crop-
nematode systems. Other promising treatments are biological controls (Stirling, 1991)
and the development of host resistance through genetic engineering (Atkinson et al.,
1994; Opperman et al., 1994; Cai et al., 1997).

Crop rotation is a management practice of primary importance for limiting losses
due to plant-parasitic nematodes (Noe, 1998). The emphasis in crop rotation studies has
been to reduce the population levels of plant-parasitic nematodes to below-threshold
levels. The effectiveness of crop rotations to suppress nematode populations is variable
depending on the nematode species, the host range of the nematode of interest and the
interactions between nematode species, and among pathogenic races of the same species
(Noe, 1998; Hirunsalee et al., 1995). The main limitation to the use of crop rotations is
the lack of a suitable or non-profitable nonhost crop for some plant-parasitic nematode
systems.

Planting resistant cultivars and using rotations are practical methods of
suppressing nematode damage to crops of low economic value (Young, 1998). Resistant
cultivars without nematicide treatment often yield as much as high-yielding susceptible
cultivars treated with nematicides (Epps et al., 1981). Use of resistant cultivars has the
following advantages: suppresses nematode reproduction, reduces need for toxic
chemicals, shortens length of rotations, does not require use of specialized equipment,
maintains cost of seed generally equal to that of susceptible cultivars (Boerrna and

Hussey, 1992), and may limit disease complexes associated with nematodes (Mai and

Abawi, 1987). There are some limitations to the use of resistant cultivars, however. The
quality, yield, and economic return of some resistant cultivars may be lower than the use
of higher quality susceptible cultivars with nematicide treatments (Johnson, 1990).
Infestations with multiple nematode species can pose difficulties for effective use of
resistant cultivars, since resistance is usually effective against a single nematode race,
species or genus (Young, 1998). Also, planting highly resistant cultivars places selection
pressure on the nematode population for biotypes that can reproduce on the resistant
cultivar (Hartwig, 1981; Young, 1990). Shifts in pest species also limit the effectiveness
of nematode-resistant cultivars (Barker, 1989; Johnson, 1989).

The soybean cyst nematode is recognized today as the major pest limiting
soybean production, accounting for approximately 54% (approximately 1.67 x 109 US.
dollars) of the soybean yield loss annually attributed to disease causing agents in the
United States (Wrather et al., 2001a, 2001 b). Soybean [Glycine max (L.) Merr.] is one of
the four major crops in the world, along with maize, wheat, and rice (F A0, 2002), and is
probably the leading source of protein and vegetable oil (Riggs and Niblack, 1993). Over
100 species of nematodes other than SCN parasitize soybeans (Schmitt and Noel, 1984).

The soybean cyst nematode has a narrow host range, primarily in the
Leguminosae, soybean being the most economically important host (Riggs and Wrather,
1992). The life cycle of SCN can be described as follows (Ichinohe, 1955). The first-
stage juvenile develops within the egg where it molts, emerging as the infective-stage
juvenile (J2). The 12 moves through soil, invades a root, and establishes a feeding site
disrupting vascular tissue. Root penetration occurs by slitting of plant cell walls from

thrusts of the robust stylet (Endo, 1978). In susceptible plants, cells in contact with the

initially stimulated cell coalesce into a multinucleated syncytium via the dissolution of
adjacent cell walls. The induction of a syncytium is essential for SCN development and
survival (Endo, 1992). After feeding begins, the 12 becomes sedentary, and molts three
more times, as it enlarges. The fourth-stage male juvenile ceases feeding about 9 days
after infection (Endo, 1992) and reverts to an elongate form. Unlike the sedentary female,
the adult male is mobile and leaves the root after mating. The lemon-shaped adult female
changes color fiom white, to yellow, to brown as it matures, ceases feeding about 21 days
after infection (Endo, 1992), and becomes a cyst. The female may produce 200-600 eggs
(Young, 1992); some are deposited in a gelatinous matrix outside the vulva, but most of
them are retained within the body. Duration of the life cycle varies from 3 to 4 weeks
(Ichinohe, 1955). The most common dispersal mechanism of SCN is movement of soil
(Lehman, 1994). Strong winds, farming equipment, animals and flooding may disperse
eggs and cysts in a field or move them to new locations.

The feeding stages 12, 13, J4, and adults destroy plant roots, interfere with nutrient
uptake, and serve as vehicles for other pests like fungi and bacteria (Dropkin, 1980;
Melakeberhan et al., 1985, 1987, 1988, 1990; Blevins et al., 1995). Producers associate
nematode damage with severe stunting and chlorosis of plants. However, the
aboveground disease symptoms are non-specific and thus non-diagnostic. Noel (1992)
measured 20 % to 30 % yield losses in fields infested with the nematode when there was
an absence of severe stunting and chlorosis. The universal symptom of SCN infection of
a susceptible soybean cultivar is reduction of yield (Riggs and Niblack, 1993). Detection
of the nematode involves examining roots for presence of cysts or white females

followed by soil sampling.

The three major steps in managing SCN according to Young (1998) are: periodic
sampling for nematode infestations, identification of the race present, and selection of
appropriate control measures. Planting resistant cultivars is the most widely used
management practice. Triantaphyllou (1975) correctly pointed out field populations are
not pure races but have virulence genes in different frequencies. The term ‘ ce” or H-G
type is used to distinguish intra-specific forms that show physiological variation, such as
differences in host preference. SCN races are distinguished by reproduction on a set of
soybean genotypes called differentials (Riggs and Schmitt, 1988; Niblack, 2002). The
selection pressure of growing a resistant cultivar may change the race classification of a
population (Anand et al., 1994). Since resistance genes fail to remain effecfive for many
years, practices that extend the durability of the genes have been proposed. Rotation of
resistant and susceptible cultivars, often in combination with non-host crops, and rotation
of cultivars with different sources of resistance (Young, 1982; Leudders and Dropkin,
1983) have been suggested to extend the time that resistance genes are effective. Other
cultural practices like late date of planting (Hussey and Boerma, 1983; Koenning and
Anand, 1991); tillage management (Wrather et al., 1992; Gavassoni et al., 2001);
irrigation (Barker and Koenning, 1989); herbicide application (Kraus et al., 1982; Bostian
et al., 1984), and flooding (Stover, 1979) are also used to control SCN population and
reduce soybean yield losses to SCN. Melakeberhan (1997) discussed the role of plant-
nematode-nutrient interactions in nematode management. Increased nutrition generally
increases plant growth and nutrient accumulations in plant tissue with or without an
effect on nematode population densities (Oteifa, 1952; Oteifa and Elgindi, 1976; Trudgill,

1980 and 1987; Melakeberhan et al., 1987; Melakeberhan, 1997b; Melakeberhan et al.,

1997). If nutrition benefits the host in the presence of nematodes, it is logical to suggest
that healthy plant-based nutrient recommendations may not be adequate under nematode
infested conditions, and that nutrition can be used as a tool for nematode management
(Melakeberhan, 1997).

Despite current management efforts, SCN continues to spread through out
soybean producing areas worldwide (Y okoo, 1936; Hung, 1958; Norton, et al., 1983;
Nishizawa, 1984; Anderson et al., 1988; Noel, 1992; Doucet, 1999). In the United States,
it is currently found in 26 of 28 soybean-producing states (Schmitt and Riggs, 1989), a
rapid spread since it was first reported in 1954 in North Carolina (W instead et al., 1955).
In Michigan, SCN first report was in 1987 in Gratiot County (Bird et al., 1988), and in
1994, SCN was found in 12 of 16 counties surveyed (Warner et al., 1994).

The potential for site-specific management (SSM) of plant-parasitic nematodes
has been explored for a variety of crop-plant-parasitic nematode systems. In potato
production for example, SSM strategies are being explored to reduce use of pesticides
and increase economic return to farmers in nematode-infested fields (Evans et al., 2002;
Morgan et al., 2002). The works of Workneh et a1. (1999), and Donald et a1. (2001) in
soybean systems, and of Wyse-Pester et a1. (2002) in irrigated comfields are further
examples of exploratory analysis towards the development of SSM for plant-parasitic
nematodes.

The goal of SSM is to manage each parcel of agricultural land for maximum crop
production, while protecting or improving natural resources. In practice, SSM involves
applying the right management, at the right time, at the right place, in the right way

(Pierce et al., 1994). In order to apply SSM strategies for the control of nematode

10

populations, it is necessary to know its spatial distribution, and how it changes through
time.

While it is difficult to compare over a range of experimental conditions,
variability in the spatial distribution of the cyst nematodes has been clearly demonstrated.
A study on local variance of SCN using geostatistics showed that the scale of
heterogeneity in the distribution of eggs between-rows was similar to that within-row
(Francl, 1986a, 1986b). Levels of H. avenae and Globodera rostochiensis in soil are
strongly autocorrelated at a normal working scale (Webster and Boag, 1992). In a 4-year
field study the spatial dependence in SCN egg population varied both within season and
from season to season (Donald et al., 1999). A better understanding of the spatial and
temporal dynamics of the incidence of SCN would enable its more effective management
by farmers. While previous studies clearly show the spatial dependence of SCN
population density, more details are needed on the nature of the spatial dependence and
how it varies in adjacent fields and over time.

The goal of this dissertation was to understand SCN spatial distribution in
soybean fields as the first step towards the long-term goal of developing SSM strategies
for SCN.

For SCN to be managed site-specifically, its spatial distribution has to be
sufficiently structured and time invariant, and it has to be related to yield limiting factors
that are easier to monitor and manage. The literature reviewed suggests that SCN may
meet the requirements for SSM. SCN can cause severe yield loss, and there is evidence
that it has an aggregated spatial distribution. In addition, correlations between SCN

population density and soil attributes (soil physical and chemical properties) have been

11

found from experimental studies. The spatial relations between SCN population
dynamics and soil properties within a given field, and the combined effect of the two on
soybean yield are key elements to accurate assess the potential for SSM of SCN,
however, a within field spatial analysis of all the variables combined has not been carried
out yet.

The working hypotheses of this Dissertation were that SCN’s spatial distribution
within a field is sufficiently structured and time invariant; that SCN spatial distribution
and population densities are related to soil properties; and that the relations among SCN
population density, soil properties and soybean yield are sufficient in magnitude to aid in
the management decision-making process.

In Chapter Three, the objective was to assess the magnitude, structure, and
persistence in time of the spatial distribution patterns of SCN cysts, eggs, and eggs per
cyst under field conditions using geostatistical tools. In Chapter Four, the relationship
between soil texture and SCN population density variability were characterized
addressing the following objectives: i- to assess the spatial structure of texture within
fields of known SCN population density in Michigan and its relationship to published soil
survey maps; ii- to determine the extent to which the spatial variability in SCN cyst
population density relate to soil texture; iii- to quantify the relationship between soil
separates (sand, silt, and clay) to SCN population density; and iv- to assess the extent to
which this relationship holds between fields with similar soil types but different SCN
populations. Population dynamics of SCN were studied in Chapter Five. Specific
objectives were: i- to characterize SCN population dynamics in soybean roots and the

surrounding soil in two fields in Michigan over two growing seasons; ii- to investigate

12

the extent of the correlation of soil texture with SCN population in the roots and in the
soil, and with eggs per cyst; and iii- to analyze SCN population dynamics spatially in
relation to soil texture. The purpose of Chapter Six was to answer the question of whether
soil fertility affects SCN distribution and root infection, and is this reflected in tissue
analysis? The following objectives were addressed: i- to characterize soil fertility
variability and its relationship with soil texture and tissue analysis in two fields of known
SCN infection in Michigan; and ii- to analyze the relation of SCN in the soil and in
soybean roots with soil fertility and tissue analysis. Finally, in Chapter Seven soybean
yield was analyzed as an integrator of the effect of the variables analyzed in previous
chapters. The extent to which yield was related to soil texture, soil fertility, and SCN
population densities was investigated by correlating spatial information on these

variables.

13

CHAPTER TWO

GENERAL MATERIALS AND METHODS

Selection of Experimental Sites

In fall of 1998, an exploratory soil sampling for SCN was performed in nine
commercial farms in Shiawassee and Saginaw counties, M1, to locate two fields with a
range of SCN infection levels fi'om high to undetectable. In each field, soil samples were
collected every 30 to 40 m following a Z pattern. The location of each sample was
marked with a flag and the geographic position was determined with a Global Positioning
System (GPS). A soil sample consisted of ten soil cores collected with a cone auger at a
depth of 20 cm within a 30 cm radius of each flag. The number of samples collected per
field varied with the dimensions of the fields. Cysts were extracted fiom the soil by
passing a suspension of 100 cm3 of soil in water through an 850-um pore sieve (mesh
#20) nested on a 75-11 pore sieve (mesh #200). Cysts retained in the #200 sieve were
further separated from soil particles following the sugar flotation-centrifugation method
(Dunn, 1969), and counted under a stereoscopic microscope. Georeferenced cyst counts
were plotted within maps of the corresponding field sampled to identify general patterns
in SCN spatial distribution. Fields with very low or undetected SCN population densities
(less than 10 cysts 100 cm'3 of soil), as well as fields where all samples collected had
high SCN levels (more than 30 cysts 100 cm‘3 of soil), were not considered firrther.

Two fields (Field A and Field B) located 3.2 km apart in Shiawassee County, M1,
were chosen based on the range in SCN population levels, management history, and

accessibility, and the research was conducted from spring 1999 until fall 2000. Field

14

information such as management, soil series, soybean cultivars planted, and planting and
harvesting dates are provided in Materials and Methods section in Chapter Three.
Soil and soybean root sampling for SCN and soil characterization

Soil samples were collected following a geostatistical sampling design applied in
each field to study the spatial distributions of SCN population and soil attributes. The
sampling design and SCN extraction procedures are explained in Chapter Three. The
efficiency of the extraction procedure was tested for different soil types and SCN cyst
densities, and alongside other extraction methods (Appendix A). Soil texture analysis is
presented in Chapter Four, and soil fertility details are presented in Chapter Six.

To study spatial dynamics in SCN spatial distribution, soybean root samples were
collected at approximately monthly intervals in 1999 and 2000. The procedures followed,
sampling dates, and nematode staining techniques are described in Chapter Five.
Soybean tissue analysis and seed yield

Soybean leaf samples were collected from Field A and B to evaluate the
nutritional status of the plants. Related procedures are described in Chapter Six. The
cooperating farmer using a yield monitoring system mounted on the combine and
connected to a GPS receiver recorded soybean yield. More information is provided in
Chapter Seven.

Statistical Analysis

Classical statistical methods applied are described in each chapter. The spatial

analysis of SCN, soil attributes, grain yield, and spatial relations among these variables

were analyzed with geostatistical tools. Geostatistics have been used in nematology to

15

study the spatial distribution of a variety of plant-parasitic nematodes (Wallace and
Hawkins, 1994; Webster and Boag, 1992; Donald et al., 1999).
Software Used

Classical statistics analyses were performed with SAS® System Release 8 (SAS
Institute, Cary, NC). Geostatistical analyses such as variography and mapping were
performed with the Surfer 7.02 software package (Golden Software, 1999), and cross-
correlograms with the Variowin 2.21 software package (Panatier, 1998). A review of the

software selected for geostatistical analyses is presented in Appendix B.

16

CHAPTER THREE

GEOSTATISTICAL ANALYSIS OF FIELD SPATIAL DISTRIBUTION PATTERNS
OF SOYBEAN CYST NEMATODE.

Avendar‘io, Felicitas, Oliver Schabenberger, Francis 1. Pierce and Haddish Melakeberhan.

2003. Agronomy Journal (July-August issue- in Press) Manuscript A02-106T

l7

GEOSTATISTICAL ANALYSIS OF FIELD SPATIAL DISTRIBUTION PATTERNS
OF SOYBEAN CYST NEMATODE.

Felicitas Avendafio, Oliver Schabenberger, Francis J . Pierce and Haddish

Melakeberhan".

F. Avendafio, H. Melakeberhan, Dep. of Entomology, 243 Natural Science Building,
Michigan State University, East Lansing, MI 48824; O. Schabenberger, SAS Institute
Inc., Cary, NC 27513; and F. 1. Pierce, Center for Precision Agricultural Systems,
Washington State University, 24106 North Bunn Road, Prosser, WA 99350. Financial
support for the project was provided by USDA Hatch Project through the Michigan
Agricultural Experimental Station and grants from the Michigan Soybean Promotion
Committee, North Central Soybean Research Program and United Soybean Board to the
last author. * Corresponding author (melakebe@msu.edu)

AKNOWLEDGEMENTS

The project is part of a Ph.D. Dissertation of the first author. We gratefully acknowledge
anonymous reviewers for their helpful comments and criticisms, which improved the
quality of the manuscript. We thank Chad Holovach, Nathan Nye, Megan Kierzek,
Carolyn Stein, Nathan Cottrell, Bindiya Shah, Dan Armstrong, Marisol Soto, Carolina
Holteuer, Katrina Blakely and Teresa Raslich for assistance in collecting and processing
samples; and Dave Eickholt, Charles Eickholt, (farmers) and Neil R. Miller (ABC-

Agribusiness Consultant) for providing and maintaining the sites.

18

GEOSTATISTICAL ANALYSIS OF FIELD SPATIAL DISTRIBUTION
PATTERNS OF SOYBEAN CYST NEMATODE.

ABSTRACT

Site-specific management of SCN is plausible if its spatial and temporal dynamics
are adequately known and structured. The hypothesis that variation in the spatial
distribution of SCN is sufficient in magnitude and structure and sufficiently time-
invariant to support the use of site-specific management in SCN infested fields was
tested. A nested survey sampling design with distances reduced by geometric progression
was applied on two fields in Michigan. Cysts were extracted from single-core soil
samples collected before planting in 1999 and 2000, the number of eggs per cyst was
estimated and the number of eggs per sample was obtained by multiplying eggs per cyst
by the number of cysts. The distribution of the three variables was characterized using
geostatistical tools including semivariograms, kriging, and cross-correlograms on log-
transformed values of the original data. Mean cyst population density ranged from 6 to 33
cysts 100 cm'3 of soil in the two fields. Although the spatial structure of SCN population
was insufficient for SSM and varied between fields, and SCN population density varied
between years, the location of areas of high or low cyst density could be identified
repeatedly. The reasons why nematodes exhibited an aggregated distribution are not well
understood. The evaluation of factors associated with the determination of SCN spatial
distribution is part of an ongoing project towards the development of SCN site-specific
management.
Key words: Heterodera glycines, nested design, kriging, semivariogram, site-specific

management.

19

The soybean cyst nematode (SCN), Heterodera glycines Ichinohe, is a major pest
of soybean [Glycine max (L.) Merr.] worldwide, accounting for approximately 54%
(Approximately 1.67 x 109 US. dollars) of the soybean yield loss armually attributed to
disease-causing agents in the United States (W rather et al., 2001a, 2001 b). Conducive
cropping systems, highly adaptive behavior, and limited sources of resistance (Young,
1992; Young and Hartwig, 1992; Diers et al., 1999; Wang et al., 2000) are among the
reasons SCN continues to be an economic threat. Eradication has been unsuccessful and
the repeated planting of resistant cultivars in the field results in the selection of a
nematode population that can overcome plant resistance, reducing the longevity of the
cultivar (Young, 1998). Thus, a realistic strategy for managing SCN appears to be
through cultural-based nematode population suppression practices (Bridge, 1996). An
important consideration in managing SCN is that its aggregated distribution varies in
space and time (Francl, 1986a, 1986b; Donald etal., 1999). The spatial-temporal
variability in SCN is often overlooked in commonly used nematode sampling strategies
that may miss field population clusters and in existing management thresholds based on
whole-field sampling procedures. Thus, current SCN management practices involve
treating whole fields rather than nematode-infected areas only, and with uniform rather
than condition-specific inputs. Advances in precision agriculture suggest that site—specific
management of SCN is plausible, but only if the spatial and temporal dynamics of SCN
are adequately known and structured (Pierce and Nowak, 1999).

It is known that plant-parasitic nematodes have aggregated spatial distribution
with fi'equency distributions generally described by the negative binomial function

(Anscombe, 1950; Seinhorst, 1982). Taylor’s Power Law (Taylor, 1984) has been used to

20

describe the distribution and to devise sampling strategies for nematodes (Ferris et al.,
1990; McSorley and Dickson, 1991). Unfortunately, while recommendations call for
systematic soil sampling to obtain samples that represent the entire area sampled (Ferris
et al., 1990; McSorley and Dickson, 1991), samples are composited to form a single
sample representative of the site average. Such spatially non-explicit sampling is not
conducive to examination of the possibility of site-specific management practices. The
high cost of obtaining and analyzing nematode samples makes it imperative that methods
of assessing spatial variability of SCN are highly effective and efficient. Efficiencies can
be achieved through robust sampling designs or by relating SCN populations to more
easily measured properties such as soil pH or soil texture, that are known to have a
somewhat structured spatial variation (Pierce and Nowak, 1999).

Geostatistics provides tools for describing spatial variation of soil properties and
for local interpolation (kriging) to predict and map values at unsarnpled locations.
Geostatistical methods can be applied to describe spatial autocorrelation of nematode
distribution, soil properties, and host response to nematode infestation (Boag, 1998).
While it is difficult to compare over a range of experimental conditions, variability in the
spatial distribution of the cyst nematodes has been clearly demonstrated. A study on local
variance of SCN using geostatistics showed that the scale of heterogeneity in the
distribution of eggs between-rows was similar to that within-row (Francl, 1986a, 1986b).
Webster and Boag (1992) showed that the levels of Heterodera avenae and Globodera
rostochiensis in soil are strongly autocorrelated at a normal working scale. Donald et a1.
(1999) found in a 4-year field study that the spatial dependence in SCN egg population

varied both within season and from season to season. In the field studied, the direction of

21

spatial variation was related to the direction of tillage but not to other factors such as soil
type or weed distribution (Donald et al., 1999). A good understanding of the spatial and
temporal dynamics of the incidence of SCN would enable its more effective management
by farmers. While previous studies clearly show the spatial dependence of SCN
population density, more details are needed on the nature of spatial dependence and how
it varies in adjacent fields and over time.

The objective of this work was to assess the magnitude, structure, and persistence
in time of the spatial distribution patterns of SCN cysts, eggs, and eggs per cyst under
field conditions using geostatistical tools. We postulate that the spatial distribution of
SCN is sufficiently structured and time-invariant to support the use of site-specific

management in SCN infested fields.

MATERIALS AND METHODS
Experimental Sites and Site Management

The study was conducted in Shiawassee County, MI in 1999 and 2000 on two
fields (Field A and Field B) located 3.2 km apart and maintained by the cooperating
farmer. Field A was 24 ha, managed under no-tillage since 1996, and planted to soybean
in 1996 and 1997, and to corn in 1995 and 1998. Field B was 13 ha, conventionally tilled
in 1995 and in 1998, and managed under no-tillage in between and thereafter. Field B
was planted to corn in 1995, soybean in 1996 and in 1997, and wheat in 1998. In 1999, a
SCN -susceptible soybean variety (Asgrow 1901), Roundup-Ready, was grown in both
fields. Soybean was planted in 19.1-cm rows at a rate of 519 000 viable seeds ha". Row

orientation was North-South in Field A and East-West in Field B. Fields A and B were

22

planted within the same week in mid-May in 1999 and in early June in 2000; planting
delayed by wet soil conditions. Weed control was maintained using Roundup at the
recommended rate with one preplant application in Field A in 1999 and 2000, and one
application postemergence in 1999. In Field B, there was one preplant application in
2000, and one postemergence application in 1999.

Soil series in Field A were Belding sandy loam (Coarse-loamy, mixed, fiigid
Argic Endoaquods), Breckenridge sandy loam (Coarse-loamy, mixed, nonacid, fiigid
Mollic Endoaquepts), Brookston loam (Fine-loamy, mixed, superactive, mesic Typic
Argiaquolls), Conover loam (Fine-loamy, mixed, active, mesic Udollic Endoaqualfs), and
Newaygo sandy loam (Fine-loamy over sandy or sandy-skeletal, mixed, fiigid Alfie
Haplorthods) (Figure 3.1 .a). Soil series in Field B were Brookston loam, Newaygo sandy
loam, and Berville loam (Fine-loamy, mixed, mesic Typic Argiaquolls) (Soil Survey
Division-NRCS-USDA, 2001). Soil series maps were digitized from Threlkeld and
Feenstra (1974) (Figure 3.1 .b).

Soil Sampling Design

1999:

A geostatistical sampling design was established in an 8 ha area and a 5.25 ha area in
the center of Fields A and B, respectively (Figure 3.1.a and b). A grid of 50 x 50 m cells
was marked on the area sampled. A nested survey sampling design with geometric
progression reduced distances (adapted from Webster and Boag, 1992) was applied to

both fields within alternate cells of the grid. In each selected cell, a first pair of single-

23

a- Field A

   

Belding Sandy Loam
{54%; Breckenbridge Sandy Loam
T I Brookston Loam

300‘

Conover Loam
E] Newaygo Sandy Loam

North (m)

 

c 20.0 m

 

7-90'“ o 7 160’ 260’ 560

a
2.70 m
m 0.90 m Ea“ (m)

.3 m , »
> 0 0 'Bervllle Loam

 

 

 

Brookston Loam
E] Newaygo Sandy Loam

 

460 7 0 100 200 300
East (m)

Figure 3.1. Soil series maps and location of the soil samples collected from a- Field A
and b- Field B in Shiawasee Co., MI in 1999 and 2000. c- Nested sampling design with
geometric progression reduced distances. Distances between pairs of samples are
indicated; the direction in which two consecutive samples were oriented was randomly
selected. This design was applied within 50 x 50 m cells of a grid centered in each field.
Black circles in figures a and b indicate sets of sample locations in selected cells for both
1999 and 2000. The crosses in figures a and b indicate the locations of the additional
samples collected in 2000 from alternate nodes of a 25 x 25 m grid. Soil series maps were
digitized fi'om Threlkeld and Feenstra (1974).

core samples was collected 20 m apart so that the segment connecting them passed
through the center of the cell. A new point 7.9 m away from each sampled location was
chosen in a random direction and also sampled. The procedure was repeated at 2.7, 0.9,
and 0.3 m (Figure 3.1.c). The angles for each new location were randomly selected. Each
angle was expressed in north and east coordinates to facilitate the location of the
sampling sites in the field. This arrangement produced ten sampling locations per cell that

were flagged and geo-referenced using GPS (Figure 3.1.a-b, black circles). The

advantage of the nested survey design is to provide pairs of observations at various short-

24

and long-range separation distances. This design guards against the potential pitfalls of
regular grid sampling designs when the shortest separation distance exceeds the range of
the spatial process, in which case description of the spatial dependency structure of the
spatial process would not be possible. Sampling locations were staked before harvest in
1999 to mark them for sampling in 2000.

2000:

In addition to the 1999 sampling desigr, a gid with 25 x 25 m cells was
superimposed on each field in 2000. The additional gid was added to expand the level of
sampling detail because the nested desigr applied in 1999 provided very few pairs of
samples for separation distances between 20 and 60 m. Soil samples were obtained from
alternate nodes of the new gid, providing 119 and 77 additional samples from Fields A
and B, respectively (Figure 3.1.a, b, crosses).

Soil Sampling for SCN

At each flag location, single-core soil samples were obtained within a week
before planting using an 8-cm diameter by 23-cm depth bucket auger (Riverside Angers,
Eijkelkamp, Giesbeek, The Netherlands). A total of 160 and 279 samples was collected
fi'om Field A and 110 and 187 samples were collected from Field B in 1999 and 2000,
respectively. Soil cores were placed in individual plastic bags and, upon arrival at the lab,
were stored in lO-gal Rubbermaid containers at 4°C until they were processed (within 30
days).

Cyst Extraction
Cysts were extracted by adding 100 ml of soil in a beaker containing 400 ml of

tap water. A semi-automatic elutriator (Research Services Instrument Shop, The

25

University of Georgia, Athens, GA) was used for the extraction following standard
procedures (Byrd, etal. 1976) with 60% extraction efficiency (Appendix A). The last step
in the elutriation process consists of the collection of nematodes and small soil particles
suspended in water in a bowl. The bowl drains through 15 T ygon tubes into a 7 S-u
aperture sieve (#200 mesh) where cysts are retained. The volume of water and the
amount of soil particles flowing from 15 tubes overflowed the sieve, so cysts were
collected from seven tubes instead. Cysts collected in the sieve were further separated
from soil particles following the sugar flotation-centrifugation method (Dunn, 1969).
Cysts were then counted under a stereomicroscope and counts were adjusted to estimate
the total number of cysts per sample if all 15 tubes had been used. Three cysts were
randomly selected fiom each sample and crushed. All eggs and second-stage juveniles
contained in a cyst were counted. The average was used to determine the eggs per cyst
for each sample containing at least one cyst. Egg numbers were estimated by multiplying
the average number of eggs per cyst by the total number of cysts in each sample.
Statistical Analysis

SCN population in Fields A and B were analyzed separately, and no statistical
comparison was made between fields. For reasons beyond experimental control, a few
samples were not collected fiom Fields A and B in 1999 or 2000. In order to compare
populations between years, incomplete pairs (1999 or 2000 sample missing for a specific
location) and additional data collected in 2000 were omitted only for the descriptive
statistical analysis. Descriptive statistics for cysts, eggs per cyst, and eggs were calculated
with the SAS® System Release 8 (SAS Institute, Cary, NC). Histograms and cumulative

distribution functions (cdf) were plotted to compare the fi'equency distribution of cysts,

26

eggs per cyst, and egg counts between years in each field. Logarithmic transformation of
the data was performed whenever frequency distributions were highly skewed.
Cumulative distribution functions for cysts [loglo (cysts+l)], eggs [loglo (eggs+l)], and
eggs per cyst [logo (eggs per cyst +1 )] were calculated and tested for log-normality with
the Kolmogorov-Smirnoff test, used to compare probability distributions to a specific
function. Logarithmic transformed means and sample variances were compared between
years within fields with a paired t-test and an F-test, respectively (a=0.05).

We used Taylor’s Power Law to provide a measure of aggegation in SCN
population density (Taylor, 1961; 1984; Ferris etal., 1990). Means and variances were
computed from pairs of samples of cysts, eggs, and eggs per cyst, the samples being
combined in different ways to give a range of spatial separations. The slope of the
regession of log variance on log mean (parameter b’) was estimated by simple linear
regession. Parameter b’ is an index of aggegation, varying continuously from zero for a
regular distribution, through one for a random distribution, to 00 for a highly contagious
distribution.

Geostatistical Analysis

The semivariogam is a structural tool for depicting the spatial dependency in a
realization of a mean-constant spatial process Z(s). Attributes of interest for which
serrrivariogarn analysis was performed were the numbers of cysts, eggs per cyst, and egg
population densities. Various estimators of the semivariogam are used in practice;
Schabenberger and Pierce (2002) summarize several of these. Here, the classical

Matheron estimator

Z{Z(s.)— Z(s )1

)=2—1—|N(h)lr. -. n: r.

27

was used (Matheron 1963). The semivariance tax) at a given lag distance h is estimated

as one half the average squared difference between all observations at locations 3;, Sj that
are separated by the lag h. The senrivariogam for a given direction is displayed as a plot

of ?(h) versus distance. Depending on the data and sampling interval used, the shape of

the experimental semivariogam may take many forms. In general, the semivariance
increases with increasing distance between sample locations, rising to a more or less
constant value (the sill) at a given separation distance called the range of spatial
dependence. Samples separated by distances closer than the range are spatially related.
Those separated by distances geater than the range are no longer spatially autocorrelated.
Semivariances may also increase continuously without showing a defined range and sill,
thus preventing definition of a spatial variance, indicating that the range is geater than
the largest lag (h), or the presence of a trend effect and/or nonstationarity (Webster and
Burguess, 1980). Stationarity means that the random field sampled looks similar
everywhere. A random field is second-order stationary if the mean of the random field is
constant and does not depend on locations, and the covariance between two observations
is only a function of their spatial separation (Schabenberger and Pierce, 2002). Whenever
serrrivariogams showed nonstationarity, the data were detrended by carrying out a
polynomial least squares regession and senrivariogarn analysis was performed on the
residuals. Other semivariogams show complete absence of spatial structure, implying
that the value observed at one location canies no information about values at other
locations. The nugget effect (Co) is a discontinuity of the semivariance near the origin. It
consists of measurement error variability and/or the sill of a micro-scale spatial process.

The error variance is a measure of repeatability of the data measurements whereas micro-

28

scale variance is a measure of variation that occurs at separation distances less than the
smallest sample spacing (Cressie, 1993).

To reduce the influence of extreme values and to achieve geater symmetry, log-
transformed data were used for the geostatistical analysis. Experimental omnidirectional
semivariogams of cysts [logo (cysts + 1)], eggs per cyst [logo (eggs per cyst + 1)], and
eggs [loglo (eggs + 1)] were calculated for each field and year for lags ranging from 1 to
30 m, with a lag tolerance of one half of the lag used (h/2). The minimum number of
pairs required for each lag was 30. The reduced number of samples in the E-W direction
in Field A, and the predominant SW-NE arrangement of the samples in Field B prevented
the calculation of reliable directional semivariogams. Variogaphy was carried out with
the Surfer 7 .02 software package (Golden Software, 1999).

Kriging is the best linear unbiased prediction of regionalized variables at
unsampled locations using the structural properties of the serrrivariogam and the sampled
values at observed locations. Soil properties often exhibit logrorrnal probability
distributions, in which case log-Gaussian kriging is employed. It involves computation of
semivariogams and kriging on log-transformed values of the original data using the same
procedures as for simple linear kriging (Cressie, 1993). When a drift or trend (non-
stationarity of the mean) existed within the area of interest, universal kriging was used;
otherwise, ordinary kriging was the method of choice. Universal kriging takes the drift
into account provided the form of the drift and the semivariogam are known (J oumel and
Huijbregts 197 8). The distributions of cysts, eggs, and eggs per cyst were mapped

separately for each field and year with ordinary or universal log-kriging predicting values

29

at the nodes of a 1x1 m cell gid using all the data points in each sample and the
parameters item the models fitted to the empirical semivariogams.

The cross-correlogam is used to describe the spatial continuity between
measurements of different attributes or of the same attribute measured at different times.
The cross-correlation function given by Goovaerts (1997) was used here to calculate
cross-correlogams for logarithmic transformed cysts, eggs per cyst, and eggs between
years; and for logarithmic transformed eggs per cyst and eggs in 1999 with cysts in 2000.
Only the data points from locations sampled in both 1999 and 2000 were used for this
analysis.

RESULTS
Population Densities of SCN

Cyst and egg population densities of SCN, as well as eggs per cyst, varied
sigrificantly between fields and years. Cyst population density for entire fields ranged
from as low as 6 cysts 100 cm’3 of soil in Field A in 1999 to as high as 33 cysts 100 cm'3
of soil in Field B in 2000. The lowest and the highest egg population densities were also
found in Field A in 1999 (87 eggs 100 cm'3 of soil) and in Field B in 2000 (4939 eggs per
100 cm'3 of soil), respectively (Table 3.1).

Field A

While mean cyst density in Field A remained similar and low in 1999 and in
2000, with positively skewed frequency distribution (Table 3.1, Figure 3.2.a), the sample
variance of cyst density increased sigrificantly in 2000 (Table 3.1). Taylor’s Power Law
index of aggegation b’ indicated a moderately aggregated distribution of cysts in 1999

and 2000. Mean egg density and sample variance were sigrificantly higher in 2000 than

30

in 1999 as a result of the increased production of eggs per cyst in 2000 (Table 3.1, Figure
3.2.b, c). The degee of aggegation in the population of eggs was also geater in 2000
than in 1999, whereas the opposite was true for the number of eggs per cyst (Table 3.1).
The proportion of cysts without eggs decreased from 36% in 1999 to 23% in 2000,
indicating that more cysts had eggs at planting and, therefore, that there was a geater
infection potential in 2000 (Figure 3.2.b). The presence of empty cysts was proof that
SCN was present in the field.
Field B

SCN population density in Field B was moderate in 1999 and high in 2000. The
mean number of cysts found in 2000 was 2.3 times geater than in 1999, with geater
variability as well (Table 3.1). The distribution of relative frequencies was positively
skewed in both years (Figure 3.2.d). In 1999, cysts were not detected in 19.3% of the
samples whereas in 2000 this proportion decreased to 13.6% (Figure 3.2. d). The same
moderate degee of aggegation in cyst population was observed in 1999 and 2000 (Table
3.1). The number of eggs found in 2000 was sigrificantly geater and more variable than
in 1999 (Table 3.1, Figure 3.2.f). In contrast to Field A, the production of eggs per cyst
was reduced by 25% from 1999 to 2000 (Table 3.1), and 13.7% of the cysts were without
eggs in 2000 compared to only 2.3% in 1999 (Figure 3.2.e). Therefore, the increase in
egg density in 2000 was due to the geater number of cysts and not to increased egg
production. Taylor’s index of aggegation indicated aggegation in egg population, and
insufficient evidence of aggegation in the number of eggs per cyst in 1999 and 2000
(Table 3 .1).

3l

:32 Essex: 85 2e a 858888 ..

.Aomv 8.8 Eavgm :5 b 888%? me 58:5 33 830: «.835. t

8.8 c:o 88— 8 :23 8383 hie 8:205 woman—«:8 8.6 .8: mwwo 28 mwwo 8% one 838m #

.358 8m :30fi 8.8 83

namwto me @2838: @8985 .83-: 85 $8: 88» 8838: 38:88 8.83 See 38.—8.0.8.: 3:2:ng— mo “80:35, 298m F
Eta—o :8 :32? 88 Se: BEES

.«o 38:: 38855. amok-“ :28: 85 wfim: v.88» 88382 38:88 883 8% 3.8.8853 2:33.88— .«o :82: owofifig m
2:88 :28 :m $98 “8828 388:8 885 328.8 8%“ ©2580 ammo mo 89:5: omm8>e 85 mm “who .8: ammo mo 83:5: 2: “n
838% 85

E 958 398 .8 89:3: :38 85 ,3 “who 8: ammo .«o 8:85: owa8>e 85 mfibfizsfi 3 38838 83 29:3 :28 :m 388: wwm...

 

 

 

dog 2 59 2 A38 3 A39 3 :38 S :89 5 E83 8.? b
a. E E E 2: a: a a: 283m
33 mag an 8-0 8m :5 «are 55 82-32
3 cm a: a; e: «2 a. .>o
m2 5% m2 3% _. mam _.. 2m: .. Sm . RN F 8838 e383
.. ms .. we: .. Ema. .. 33a .. 3m .. 2: @308 unease...
. I: Eula
:38 I 33: w: 5.8 3 $2: 3 :83 2 $.98 3 t. an 83> b
m: m: m: m: 5 E Saginaw
we; 23 $35 885 $5 :10 82.5:
3 w: a: a: am: 8: .x. .5
.. Se .. 0.8 .. .15 ._. «a: .. M2: .. one F 883% eaequ
.. 3:. .. or: .. 2mm .. new mz aw m2 :3 a =88 08855
=8 .688 2: 3550 |< e. an:
82.. 82 88 82 88 82
H ammo 8m wwwm ._. mwmm "ammo

 

.8838 $585 .88 ea 82 a wees: song
=2 :00 comme3eEm 5 88% O3: 80¢ 8.3 8: ammo :8 £388: gum—«ac: wwc 98 “who 638:8: «who S8983 20m 4.». 838:.

32

2:83 :58 8 33° 3828 380:5: 8:: M838

8:: :85: mwmo .«o 8:88 8585 2: 5 858:8 53 «who 8: ammo mo 898:: BE. doom :5 3o: 8 m::5_: 8.8.8: 88280
883 8.85m .m 22: 88.: :8 mo :80 cc: wwwo .m :5 Ammo 8: ammo .8 do» we M.8: 8: $90 -: :5 m< 23: 88: :8 :0 M88 9:
wwwo -o :5 J93 8: ammo -9 .28 .8 n8: co: $98 .5 mo Q:ov 8:88: 85:98:: 85883 :5 8:358: 8828:: .N.m 8:3:

:8 .o 9:8 8: $500 8.6 8: 3.58 30 0583.. :8 6 9:8 09 8:30

oooow 88F oooww 800 80v 0 com ova 00— ON: on c 8.. on on 9. ON 0

 

 

 

 

 

y
m
9
U
m H
m o m somaaom -. - m 22896.8 33 -m m
U
m a -1 ._ I
U
C

0: II I

8 LI II

3+ -

00—. 41 T

 

 

 

 

< Eomaaum -o n < 22:85 8: $3 .3 < :_o_n_.am>o .a

 

33

Eggs per cyst relative frequency distributions appeared normally distributed in both years
with medians (109 eggs cyst'l in 1999 and 75 eggs cyst'l in 2000) close to the means (113
eggs cyst" in 1999 and 86 eggs cyst" in 2000) (Figure 3.2.e).
Geostatistical Analysis

Although the spatial structure of SCN population density varied between fields
and years, the spatial autocorrelation in cyst and egg population densities, as well as eggs
per cyst, was structured to a lesser degree in 2000 than in 1999. Empirical
semivariograms were calculated and graphed for lags ranging from 3 to 30 m, but only
those calculated for a lag of 10 m were selected for modeling because they represented
more clearly the structure of the spatial variability in these fields. The models fitted, and
their corresponding parameter estimates are given in Table 3.2.
Field A

The data exhibited non-stationarity in mean for logarithmic transformed cyst data
in 1999 (data not shown). Therefore, the semivariogram was calculated with the residuals
after a polynomial trend was removed. The experimental semivariograms of cyst density
in 2000, and egg density and eggs per cyst in both years showed stationarity in the
distribution (Figure 3.3. b-f). Empirical semivariograms clearly showed periodicity in the
spatial structure of the distribution of cysts, eggs, and eggs per cyst (Figure 3.3. a-f).
However, because of the large nugget effect in some cases, a wave or hole-effect model
could not be fitted (Figure 3.3.b, c, f). The empirical semivariogram of cysts in 1999
revealed two scales of spatial structure. One described by a wave effect model with a
short range indicating small clusters of cysts, and the second described by a spherical

model with a much greater range describing the spatial autocorrelation of those

34

Table 3.2.

1' Models were fitted by least squares based on empirical semivariograms calculated for
lags (h) ranging fiom 1 to 25 m, with a lag tolerance of NZ. The minimum number of
pairs required for each lag was 30.

1 Whenever semivariograms showed nonstationarity, the data were detrended carrying
out a simple polynomial least squares regression and semivariogram analysis was
performed on the residuals. The polynomial order of the trend is indicated when a drift
or trend was removed.

§ Co is the nugget effect or a discontinuity in semivariance at the origin due to rnicroscale
variability or sampling error.

1[ C is the partial sill defined for spherical models.

# Observations that are spatially separated by more than the range are uncorrelated.

1‘1“ Co/(C+Co) is an indicator of the degree of spatial structure, the lower the number the
stronger the spatial autocorrelation.

11 Semivariograms for cysts were calculated for loglo (cysts 100 cm'3 of soil + 1). Cysts
were extracted from a 100 cm3 soil subsample with a semiautomatic elutriator and
counted.

§§ Semivariograms of eggs per cyst were calculated for logo (eggs per cyst + 1). The
number of eggs per cyst was estimated as the average number of eggs counted after
crushing three randomly selected cysts in each sample.

1H1 Semivariograms for eggs were calculated for loglo (eggs 100 cm'3 of soil + 1). Egg
density was determined for each sample by multiplying the average number of eggs per
cyst by the number of cysts in the sample.

35

Table 3.2. Parameters of the theoretical semivariogram models of SCN (soybean cyst
nematode) population density in two fields in Shiawassee Co., MI before planting in
1999 and 2000. T

 

Drifi 1 Model function C0§ C 1| Range # CO H

 

C + Co

Field A 1999 m

Cysts 11 Linear Nugget 0.08 0.50
Wave 0.04 3.6
Spherical 0.04 819.3

Eggs per cyst §§ No Nugget 0.34 0.77
Spherical 0.1 168.3

Eggs W No Nugget 0.8 0.77
Wave 0.24 4.5

Field A 2000

Cyst No Nugget 0.1 8 0.69
Spherical 0.08 122

Eggs per cyst No Nugget 0.33 0.44
Wave 0.41 7. l 3

Eggs No Nugget 1.32

Field B 1999

Cysts Linear Nugget 0.07 0.35
Wave 0.13 12.16

Eggs per cyst No Nugget 0.08 0.44
Wave 0.1 1 1.93

Eggs No Nugget 0.25 0.43
Wave 0.33 10.22

Field B 2000

Cyst Linear Nugget 0.09 0.37
Wave 0.08 2.62
Spherical 0.07 268

Eggs per cyst No Nugget 0.44

Eggs No Nugget 1.1 0.86
Wave 0.18 4.24

 

36

smaller clusters (Figure 3.3.a). A similar pattern was observed in 2000, although the
periodicity could not be modeled in this case (Figure 3.3.b). The semivariance of eggs per
cyst increased with increasing lag distance up to a range of almost 170 m in 1999 (Figure
3.3. c). In 2000, a wave effect model with a range of 7 m described the periodicity in the
distribution of eggs per cyst (Figure 3.3.d). The semivariance for eggs fluctuated about
the sample variance with increasing separation distance between data pairs in 1999 and
2000, indicating a clustered distribution of eggs within a very short range, even though a
wave effect model could not be fitted in 2000 because of the large nugget (first lag
represented by 131 pairs) (Figure 3.3.e-t).

Cysts Eggs per cyst Eggs
0-3‘: a-1999 ‘1 c-1999 31 e-1999

 

 

 

 

 

c 'l
.w l
c 0 "-"~"" v-- '—'-1'--“~'— —r W" 'r—*-' ' ~ O‘P__“'_"fi.' ‘- ‘—‘— v‘ ----.—— 7—~*—r‘"—'"“
g 020406080100020406080100
s ..
2%
F .
1...
[..__-_ t__ L L
I] .~
!
l
I l
— .m. r . t n. Dis—w aw..-_-n.—-r...flrv..._q
020406080100020406080100020406030100

 

Lag distance (m)

Figure 3.3. Semivariograms of a, b- cysts [loglo (cysts 100 cm’3 of soil +1)], c, d- eggs per
cyst [loglo (eggs per cyst +1)] and e, f- eggs [logm (eggs 100 cm'3 of soil +1)] from Field
A in a, c, e- 1999 and b, d, f- 2000. Black circles indicate omnidirectional empirical
semivariogram, the solid line indicates the theoretical model fitted by means of least
squares and the dashed line is the sample variance.

37

Field B

The data exhibited non-stationarity in mean for logarithmic transformed cyst data
in both years (data not shown). Therefore, semivariograms were calculated fi'om residuals
after removal of a polynomial trend. In 1999, the distribution of cysts showed strong
spatial autocorrelation described by a wave efiect model with a range of 12 m (Figure
3.4.a). In 2000, the structure of the empirical semivariogram was similar to the one
observed for cysts in Field A in 1999. A wave effect model and a spherical model

described the short and the long-range structures, respectively (Figure 3.4. b).

Eggs per cyst Eggs
“3 c-1999 3‘» e-1999

 

 

Semivariance

   

Lag distance (m)

Figure 3.4. Semivariograms of a, b- cysts [logm (cysts 100 cm”3 of soil +1)], c, d- eggs per
cyst [loglo (eggs per cyst +1)] and e, f- eggs [log10(eggs 100 cm'3 of soil +1)] fiom Field
B in a, c, e- 1999 and b, d, f- 2000. Black circles indicate omnidirectional empirical
semivariogram, the solid line indicates the theoretical model fitted by means of least
squares and the dashed line is the sample variance.

38

The periodicity observed in the empirical semivariograms of eggs per cyst and
eggs could be described by wave effect models, except for eggs per cyst in 2000 where
the nugget effect was too large (174 pairs in the first lag) (Figure 3.4.c- f). These
semivariograms indicate clustered distribution of eggs and eggs per cyst.

Kriging

The distributions of cysts, eggs, and eggs per cyst were mapped separately for
each field and year with ordinary or universal log-kriging using the parameters from the
models fitted to the experimental semivariograms (Table 3.2).

The distribution of cysts in Field A varied slightly from 1999 to 2000. In 1999,
small clusters of cysts were aggregated in larger patches. Cyst density was lower in the
center of the area sampled and in the north end. The highest cyst density was found in the
south portion of the field and in a band of approximately 100 m wide located between
550 and 650 m north (Figure 3.5. a). In 2000, cyst density increased throughout the field,
and the distribution pattern changed in some portions of the field when compared to the
pattern observed in 1999 (Figure 3.5. b). Even though the grouping in small clusters
disappeared, cyst density remained low in the center and the northwest corner of the field.
The most significant change in the distribution was observed in the southeast corner of
the field where cyst density decreased markedly in 2000. Despite of the similarities in
cyst distribution between years, the cross-correlogram indicated weak correlation
between cysts in 1999 and 2000 for very short lags, and no cross—correlation beyond a
separation distance of 20 m (Figure 3.6.a). The poor spatial structure and the low number
of eggs per cyst in 1999 generated a distribution of kriged values rather uniform

throughout the field (Figure 3.5.c).

39

.8353 20¢

05 83253 on: 2.8 < .22. :8 mo «-80 2: wwwov 2 we: ammo 98 .5+ “menu be wwwov Ewe: “96 Ba wwwo .5+ :8 mo .80 2:
39on ewe: 393 .«o £96— 88085 03% E5323 0.5. .33 -o E ammo mo noun—93m? 2: can doom 6 v5 33 .o 5 “who 5% ammo mo
833.5% 2: seem a e5 82 .e a < 22» 55? Bass 83 es s was .8 8.535% 2: sewage ass pauses .3 am

 

 

 

 
 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

:5 mom
9: o 8, o 2: o
_ . . r _ b
.. \xlfl... a v... u . .. \l r . \kt . . 1 . o
n Weofi r .. is . ., r .Sw
.. o QR .- .. r . d‘ 1 saw
a . . L f . V
. a... saw” . . - . . . 58 N
I. WM. om I J l n T ‘8' W
when... m.
.. Warn... fl. . - . 1 .8m
. 1.90 a r . . T THE
W sum 8“
r r so 3... . . 1 . , . .
Ll my ._ .05. \.\ I “.11. 1 . . . l\.\ .. r80
mom 70 ooom.“U moo—nu

 

33 “m3 .8 33

40

Field A Field 8

 

 

 

 

 

 

a- r=0.09 d- r=0.59
o.4<
O
‘<
3
oz
02
b- r=0.14
§ 0.1
g» In"
E» ‘8
g E
g or 3
0

 

 

 

b
to
4
l
1

 

.0
A

 

 

 

 

 

 

“0 go go go so teeters. as to so as asses...

Lag distance (m)
Fig. 3.6. Cross-correlograms between a, d- cysts [loglo (cysts 100 cm'3 of soil +1)], b, e-
eggs per cyst [loglo (eggs per cyst +1)], and c, f- eggs [log 10 (eggs 100 cm'3 of soil +1)]
in 1999 and 2000 in Field A (a, b, c) and Field B (d, e, f). r = linear correlation
coefficient.
The increased number of eggs per cyst and cysts with eggs in 2000 may have contributed
to a better-defined spatial structure. Kriged values for eggs per cyst in 2000 showed small
clusters of high and low values distributed randomly throughout the field (Figure 3.5.d).
The distribution of high and low egg density areas resembled that of cysts in 1999
(Figure 3.5.e). Because of the nature of the semivariogram model (pure nugget effect),
the map of egg distribution in 2000 represented the egg mean density throughout the field

(not shown). The distribution of eggs per cyst and eggs were uncorrelated between 1999

41

and 2000 (Figure 3.6.b, c). Cyst population density at planting in 2000 was poorly

correlated with eggs per cyst or eggs at planting the previous year (Figure 3.7. a, b).

 

 

 

 

 

 

 

 

 

 

Field A Field 3
1.0
a-r=0.13 c-r=0.19
0.5 r 1
0.0 .W
-O.5 ‘ r
'- .1.0 a - a
g
E b- r=0.14 d- r=0.42
g 0.5 .
5 00 \‘ J ; W ‘R
-O.5
-'l.0

 

osoroorsozooasoosoroorsozoo
Lagdistance(m)

Fig. 3.7. Cross-correlograms between a, c- eggs per cyst [loglo (eggs per cyst +1)] in
1999 and cysts [loglo (cysts 100 cm'3 of soil +1)] in 2000; and b, d- eggs [log 10 (eggs 100
cm'3 of soil +1)] in 1999 and cysts in 2000, in Field A (a, b) and Field B (c, d). r = linear
correlation coefficient.

A well defined area with high cyst density was found in the northeast corner and
an area with low cyst density in the southwest corner of the sampled site in Field B in
1999. Clusters of high and low cyst density were mixed in between these extreme
locations (Figure 3.8. a). The pattern was maintained in 2000 with a relatively high linear
correlation coefficient (Figure 3.6.d), but the infected area was larger, extended towards
the south and without the inclusion of low-density clusters (Figure 3.8. b). The cross-
correlogram for cysts between years showed a decrease in correlation as the separation

distance between samples increased (Figure 3.6.d). Patches of eggs per cyst density

slightly higher than the mean were distributed throughout the field in 1999, with the

42

05 82353. on: Ben < .2? :8 we ran. 9.: mwwuv E we: wwwo we“ .2? «who Hem wwon Ewe: ammo be mwwo 42+ :8
3&8 Ewe: was? we £05— mnxaomcfi 28m wfiefim 2F doom .0 can 33 -w E wwwo we =32:me 05 93 ”mag -o E “who

.8558 see

me .80 E:
Em

mwwo me

8:355 2: seem -e Ea 82 .e a m Eon 555 3&5 as es a see as 8:235 2: “sees as: confine; a; a;

 

.4

L

  

 

 

 

Tl].

o8m -o

 

. ~ _

‘_.,

 

 

82 .c
83

E 55

com cop o

 

 

 

 

 

82 .0
“96 Log mmmm

 

 

 

 

 

 

 

 

 

82 a
296

(w) ”MON

43

lowest numbers located on the southeast corner of the site (Figure 3.8. c). In 2000, the
distribution of eggs produced per cyst was represented by the mean throughout the field
because of the nature of the semivariogram model (not shown). The distribution of eggs
matched well that of cysts in 1999 and 2000 as expected, since the number of eggs is a
linear function of cysts and eggs per cyst and the CV of cysts was much greater than the
CV of eggs per cyst (Figure 3.8. d, e, Table 3.1). In 1999, it was possible to identify
clusters of spatial autocorrelation in the distribution of eggs, whereas in 2000 the clusters
appear to merge into more homogeneous bands. The correlation in the distribution of
eggs per cyst between years was poor at short lags, becoming negative with increasing
separation distance, and non-existent beyond 60 m (Figure 3.6.e). The distribution of
eggs was moderately correlated between years at very short lags and uncorrelated beyond
20 m (Figure 3.6.f). Cyst population density in 2000 was poorly correlated with the
number of eggs per cyst in 1999, but relatively well correlated with egg density at

planting in 1999 (Figure 3.7.0, d).

DISCUSSION
The objective of this work was to assess the magnitude, structure, and persistence
in time of the spatial distribution patterns of SCN cysts, eggs and eggs per cyst under
field conditions using geostatistical tools. Geostatistics has been used to study the spatial
distribution of soil inhabiting plant pathogens such as fungus-vectored viruses (W orkneh
et al., 2001), soil nematode community structure (Robertson and Freckman, 1995), and
cyst nematodes under semi-controlled and field conditions (Francl, 1986a, 1986b;

Webster and Boag, 1992; Evans et al., 1998; Donald et al., 1999; Donald et al., 2001). In

almost all cases, spatial patterns have been demonstrated. However, the precision of
defining the spatial structure seems to vary with the type of sampling design used. The
sampling design selected for this study allowed for the construction and analysis of
detailed semivariograms for short separation distances, thus contributing to the
understanding of SCN spatial distribution patterns. A regular grid may provide more
uniform coverage of the area to be sampled, but the scale of spatial autocorrelation may
be missed if the distance between nodes is larger than or equal to the range of the
semivariogram. For exploratory spatial analysis, the nested sampling design of Webster
and Boag (1992) has an advantage over regular grids in that it provides information for a
variety of short separation distances (lags), thus enabling the analysis of the structure of
the semivariogram when the scale of spatial variability is unknown. Taylor’s Power Law
has been used to determine the degree of aggregation of many nematode species (Boag
and Topham, 1984; Duncan et al., 1989; Ferris et al., 1990; McSorley and Dickson, 1991;
Webster and Boag, 1992). While the index of aggregation b’ was not affected by sample
size (Boag and Topham, 1984), it differed by species and by the separation distance
between samples (Boag and Topham, 1984; Webster and Boag, 1992). In Fields A and B,
Taylor’s index of aggregation did not correspond with the information obtained from the
semivariograms on the spatial structure in SCN population. This might be due to the
periodicity encountered in SCN population, with positive and negative correlation
between samples as the separation distance increased. A plot of Taylor’s index b’ versus
separation distance between samples would probably have reflected the fluctuations

observed in the empirical semivariograms.

45

The spatial distribution of SCN cysts, eggs, and eggs per cyst exhibited varying
degrees of aggregation and structure, and the spatial structure varied between years.
While the exact causes of varied SCN spatial distribution are poorly understood,
nematode multiplication and population density equilibrium are influenced by host-
nernatode interactions and the prevailing environmental conditions (Seinhorst, 1967). For
example, SCN reproduction and subsequent survival could be influenced by non-host
plants or the presence of a resistant cultivar (Riggs, 1987). These, in turn, may lead to a
less structured and more random distribution of the surviving cysts in soils. Hence, the
low number of SCN cysts observed in Field A before planting in 1999 appears to be the
outcome of com, a non-host crop, grown the previous year and thereby possibly
explaining the poorly structured spatial distribution observed for cysts in Field A in 1999.
When infective juveniles hatch in the presence of a susceptible soybean variety, SCN can
complete several generations during a growing season and population density can
increase sharply if conditions are adequate (Lauritis et al., 1983; Bonner and Schmitt,
1985). SCN population densities that were below the detection limit at planting in 1999
may have increased to detectable levels in 2000 after susceptible soybean, generating
changes in the spatial structure poorly correlated between years. The cysts and eggs
distribution maps in Field B showed that the areas with more eggs in 1999 resulted in
increased cyst density in 2000 and areas with fewer eggs in 1999 resulted in relatively
fewer cysts in 2000 (Figure 3.8), indicating a positive correlation based on spatial
location between eggs at planting and cyst density the following spring. This correlation

was corroborated by the cross-correlogram (Figure 3.7.d). Over time, and under adequate

46

conditions, the spatial distribution of cysts in this field, and possibly in Field A too, may
become more structured.

Plant-parasitic nematodes act as sinks of photoassimilates and nutrients and
their ability to function as such will vary considerably depending on their physiological
age (Bird and Loveys, 1975). SCN females are more likely to reach maturity in primary
rather than in secondary or tertiary roots, thus, maturity may be related to the distance
between the infection site (the sink) and the shoot (Melton et al., 1986). Plants with large
root biomass offer more feeding sites to nematodes, favoring infection but not necessarily
improved egg production. In Field B, we observed that the soybean plants located where
cyst and egg densities were high grew less well than plants in areas with lower nematode
densities. At the same time, cysts located in the northeast comer of the field contained
similar numbers of eggs in 2000 as cysts found in areas with healthier plants (pure nugget
effect semivariogram). Soil properties are very likely also involved in this relationship
(Koenning et al., 1988; Todd and Pearson, 1988; Koenning and Barker, 1995). Among
other factors, poor growth could be a function of soil moisture, where generally more,
deeper, and less evenly distributed roots develop under drought stress, offering more
feeding sites to nematodes (Huck et al., 1986). The portion of the field where plants grew
poorly was well drained (N ewaygo sandy learn), suggesting that drought stress may have
been an adverse factor for plant growth in addition to the abundance of SCN. These
observations correspond to those of Koenning and Barker (1995) where plant growth was
adversely affected in coarse soils with poor water holding capacity even when there was a

continuous water supply. A combination of causal factors could result in patches of cyst

47

density and variation of egg production, thereby generating spatial autocorrelation over
time and in the presence of a suitable host.

Although strong winds, farming equipment, animals and flooding may disperse
eggs and cysts in a field or move them to new locations, the most cormnon dispersal
mechanism of SCN is movement of soil (Lehman, 1994). Even though both fields were in
no-tillage management between 1999 and 2000 with a residue coverage protecting the
soil, some SCN movement is possible. For example, soil peds containing eggs and cysts
adhered to the residue may have been moved by the wind and by planting and harvesting
operations. In addition, the surface run-off in the spring of 2000 may have altered
previously existent spatial structure towards a more uniform pattern.

Earlier reports indicated a strong relationship between the spatial distribution of
SCN and soil texture (W orkneh et al., 1999; Donald et al., 2001). In this study, variations
in soil properties are likely to have had the most influence on SCN spatial structures and
trends observed, especially in Field B. The influences may have been direct on the
nematode population density, indirectly mediated through the host, or a combination of
both.

The site-specific management of SCN is plausible because SCN studies have
shown that it is not unifome distributed within fields and because the cost and
performance of eradication practices suggest that management offers the most viable
control option for SCN. To make site-specific SCN management successful, the spatial
distribution of SCN must be highly structured and temporally stable within a given field.
In the fields evaluated in this study, the within-field variability in cysts and eggs per cyst

was large (CV > 100%), but spatial variability was weakly structured as evidenced by

48

high nugget variances and poor fit of semivariogram models. Poor spatial structure and
the high cost of sampling nematodes make the success of site-specific management in
these fields unlikely. However, although there was poor correlation in cyst density
between years in Field A, the temporal variability in SCN distribution was small. There
were areas within each field in which SCN was not detected or occurred at only low
density, and the areas of high or low SCN population density remained approximately at
the same locations between years. We also know from remote sensing and yield maps
(not presented here) that soybean performance was correlated with SCN infection.
Therefore, the notion of SCN management zones, areas in which different SCN
management practices would be applied, could be a viable option if appropriate criteria
for the delineation of effective management zones were determined. Deriving and
evaluating SCN management zone delineation criteria is a continuing goal of this

research effort.

49

CHAPTER FOUR

THE SPATIAL DISTRIBUTION OF SOYBEAN CYST NEMATODE IN RELATION
TO SOIL TEXTURE AND SOIL MAP UNIT

Felicitas Avendano, Francis J. Pierce, Oliver Schabenberger and Haddish Melakeberhan.
2003. Agronomy Journal (submitted). Manuscript A02-0289.

50

THE SPATIAL DISTRIBUTION OF SOYBEAN CYST NEMATODE IN RELATION
TO SOIL TEXTURE AND SOIL MAP UNIT

Felicitas Avendano, Francis J. Pierce, Oliver Schabenberger and Haddish Melakeberhan.m

F. Avendafio, H. Melakeberhan, Dep. of Entomology, 243 Natural Science Building,
Michigan State University, East Lansing, MI 48824; F. J. Pierce, Center for Precision
Agricultural Systems, Washington State University, 24106 North Bunn Road, Prosser,
WA 99350; and O. Schabenberger, SAS Institute Inc., Cary, NC 27513. Financial support
for the project was provided by USDA Hatch Project (No MICL01813 of the last author)
through the Michigan Agricultural Experimental Station. * Corresponding author
(melakebe@msu.edu)

ACKNOWLEDGEMENTS

.w‘

/

The project is part of a Ph.D. Dissertation of the first author. We thank Jeannette Makries
and Mohamed Elwadie for technical assistance with soil particle size analysis; and Chad
Holovach, Nathan Nye, Megan Kierzek, Carolyn Stein, Nathan Cottrell, Bindiya Shah,
Dan Armstrong, Marisol Soto, Carolina Holteuer, Katrina Blakely and Teresa Raslich for
assistance in collecting and processing samples. We also thank Dave Eickholt and
Charles Eickholt (farmers), and Neil R. Miller from ABC-Agribusiness Consultant for

providing and maintaining the sites.

51

THE SPATIAL DISTRIBUTION OF SOYBEAN CYST NEMATODE IN
RELATION TO SOIL TEXTURE AND SOIL MAP UNIT

ABSTRACT

Evidence suggests that variation in soil texture may be a key variable to explain
the variability of soybean cyst nematode (SCN), Heterodera glycines Ichinohe,
population density within infested fields and be important to the delineation of SCN
management zones. The purpose of this work was to assess the spatial structure of soil
texture in two fields of known SCN population density and its relationship to published
soil survey maps; and to quantify the relationship between sand, silt, and clay with SCN
population density variability across fields and over time. Cysts were extracted by
elutriation from single-core soil samples collected in a geostatistical sampling design.
Soil texture analysis was performed for a subset of samples from each field using a
modified hydrometer method. Classical and geostatistical analyses were employed to
characterize and map soil texture for each field and analyze the effect of sand, silt, and
clay on SCN population. Cyst population density was consistently higher in loamy sand
than in sandy clay loam. The proportions of sand, clay, and silt in the soil were spatially
structured and affected SCN population density strongly and consistently over time. The
number of eggs per cysts was not related to soil type or texture (or = 0.05). This study
demonstrates the value of soil survey maps as indicators of where SCN can be expected
in an infested field and how the addition of site-specific texture data can improve the
spatial prediction of SCN. In addition to providing the basis for future experimentation to
define soil texture tolerance limits for SCN, this study lays a foundation for new and

integrated approaches to site-specific management (SSM) of SCN.

52

Soybean cyst nematode (SCN), Heterodera glycines Ichinohe, is a major
economic pest in soybean with wide geographic distribution in the major soybean
growing areas of the US. (Wrather et al., 2001b). The resilience of SCN makes
management and not eradication the most viable option for minimizing its impacts on
soybean production (Bridge, 1996). Site-specific management of SCN is of interest
because population densities vary spatially within fields (Donald et al., 1999; Avendar'io
et al., 2003- Chapter Three), and with variable management in response to SCN
population density soybean growers might increase the efficacy and reduce the costs of
SCN management practices. To be of value, SSM requires that the spatial variability be
highly structured to ensure that spatial prediction and corresponding management maps
are accurate (Pierce and Nowak, 1999). However, based on geostatistical sampling of
SCN populations, Avendano et a1. (2003, Chapter Three) found that the spatial variability
of SCN in two Michigan fields was poorly structured leading them to conclude that the
success of SSM of SCN in these fields is unlikely, particularly given the high cost of
sampling nematodes. In areas within these same fields, however, there were repeated
occurrences of non-detectable or low densities of SCN and the authors reported
correspondence between SCN infection and remote sensed imagery and yield maps. This
observation suggested that other criteria might be available to delineate management
zones for SSM of SCN in these fields (Avendafio et al., 2003, Chapter Three). This
notion is supported by evidence that environmental conditions created by the interaction
of weather, soil, landscape, and plant factors assist in the dispersion of eggs, determine

SCN survivability, or limit its growth potential and thereby regulate the spatial dynamics

53

of SCN (Lehman, 1994; Koenning and Sipes, 1998; Donald et al., 1999; Workneh et al.,
1999; Donald et al., 2001).

The determination of spatio-temporal dynamics of yield-limiting factors and the
identification of cause-and-effect relationships among limiting factors are also critical
components of SSM of nematodes (Evans et al., 2002; Melakeberhan, 2002; Wyse-Pester
et al., 2002). Within the two fields studied by Avendaiio et al. (2003, Chapter Three),
SCN population densities appeared to vary by soil mapping unit, which were
differentiated primarily on soil texture differences. This would suggest that soil texture or
some combination of individual soil separates (sand, silt, and clay), are related directly or
through the soil properties and processes they influence to SCN population densities and
may be useful in delineating SCN management zones. The literature supports two
important points in this regard: that nematode population dynamics are related to soil
texture (particle size composition), structure (spatial arrangement and continuity of the
soil pores between and within the particles), and related soil hydraulic properties and that
these soil properties vary spatially and often with strong spatial structure.

Soil texture and structure strongly affect crop production and ecosystem health,
including the nematode community (McKeague and Wand, 1982; Heal and Dighton,
1985; Gupta, 1994; Topp et al., 1997; Workneh et al., 1999). While there seems to be
some variation among nematodes, generally coarse sandy soils favor nematode population
growth by providing more space for nematode movement than do poorly structured soils
containing finer particles which cannot form stable compound aggregates and so pack
closer and diminish total porosity (Jones et al., 1969). Population densities of

Pratilenchus penetrans, Aglenchus agricola, T ylenchorhynchus spp., H. trifolii, and

54

Paratylenchus spp. were significantly correlated with sand or silt particle size classes
(Wallace et al., 1993). However, this relationship may be reversed for other nematodes.
For example, higher densities of Meloidogyne incognita were associated with higher
levels of clay in a loamy sand soil (Noe and Barker, 1985). Tillage influences nematode
prevalence and population density by increasing the amount of space available for
nematode movement even in fine soils rich in clay (Jones et al., 1969; Workneh et al.,
1999; Donald et al., 2001). In fields under tillage management, repeated soil disruption
during land preparation and cultivation may have alleviated oxygen deficiencies arising
from water saturation due to high clay content, thus favoring the nematode population
(Young, 1987; Workneh et al., 1999). In a field study on silt loam, the response of field-
grown soybean to SCN varied depending upon the water status of the soil and SCN level
(Johnson et al., 1993). In this case, the increase in SCN penetration of the soybean root
system corresponded positively to the increase in soil oxygen diffusion rate and
corresponding decrease in water potential. Water holding capacity was found to be the
most important soil factor affecting the success of the oat crop at various levels of H.
avenae (F idler and Bevan, 1963). A positive correlation was also found between soil
moisture and SCN survival (Slack et al., 1972), although Barker and Koenning (1989)
noticed that numbers of SCN eggs, infective juveniles, and cysts were affected by soil
texture, but not by soil moisture. Soybean plants respond to moisture stress by increasing
root biomass, which would favor reproduction of SCN thus increasing population density
under drought conditions (Koenning et al., 1988; Barker and Koenning, 1989; Koenning
and Barker, 1995). Thus, the correlation found by Slack et al. (1972) could have been an

indirect effect on the nematode population of soil moisture on plants. Koenning and

55

Barker (1995) also found that although SCN can increase to damaging levels in fine
textured soils, the low rate of increase in these soils limits the damage potential of this
nematode to soybean, as does the fact that damage is less severe per unit increase in
population density. The low reproductive rate in soils with high clay content results in a
longer time being necessary for the nematodes to attain damaging levels in fine textured
soils. When the crop is damaged, nematode population equilibrium will decline to levels
at which soybean yield suppression in subsequent years may not be perceptible.

Given that SCN population dynamics are regulated at least in part by soil
properties and associated processes, the delineation of site-specific SCN management
zones would appear feasible if soil properties are spatially structured and if quantitative
relationships between SCN and these soil properties occur and are known. Considerable
evidence supports the spatial variability of soil properties and that this variability can
have spatial structure adequate for SSM (Robertson et al., 1993; Pierce and Nowak, 1999;
Kravchenko and Bullock, 2000 and 2002, Cassel et al., 2000, Basso et al., 2001; Gaston
et al., 2001; Mueller et al., 2001). We conclude that there is sufficient evidence to suggest
a significant relationship between SCN populations and soil texture and that soil texture
varies spatially. However, quantitative relationships on the spatial co-variance of SCN
with soil texture and/or predictive relationships between soil properties and SCN needed
for management zone delineation are not available. We therefore hypothesize that the
variation of SCN population density between soil mapping units observed in the fields
studied by Avendafio et a1. (2003, Chapter Three) suggest that soil texture may be a key
variable to explain the variability of SCN population density within these fields and be

important to the delineation of SCN management zones.

56

The purpose of this work was to characterize the relationship between soil texture
and the variability observed in SCN population density for the MI fields reported by
Avendar‘io et al. (2003, Chapter Three). Specific objectives were to: assess the spatial
structure of texture within fields of known SCN population density in Michigan and its
relationship to published soil survey maps; determine the extent to which the spatial
variability in SCN cyst population density relate to soil texture; quantify the relationship
between soil separates (sand, silt, and clay) to SCN population density; and assess the
extent to which this relationship holds between fields with similar soil types but different

SCN populations.

MATERIALS AND METHODS
Research Site, Sampling Design, and Soil Sample Collection
The experimental design of the initial phase of this study to assess the spatial
variability of SCN population was reported by Avendafio et a1. (2003, Chapter Three).
Briefly, the study was conducted in Shiawassee County, MI in 1999 and 2000 on two
fields, (Field A and Field B) maintained by the cooperating farmer. Field A was 24 ha,
managed under no-tillage since 1996, and planted to corn in 1998. Field B was 13 ha,
conventionally tilled after wheat in 1998, and managed under no-tillage thereafter. In
1999, an SCN-susceptible soybean variety (Asgrow 1901), and in 2000, an SCN-resistant
variety (Asgrow 2201), both Roundup-ready, were grown in both fields. Soybean was
planted in l9-cm rows at a rate of 519 000 viable seeds ha'1 in 1999, and 494 000 viable

seeds ha'1 in 2000. Rows orientation was north-south in Field A and east-west in Field B.

57

Soil series in Field A were Belding sandy loam (Coarse-loamy, mixed, fiigid
Argic Endoaquods), Breckenridge sandy loam (Coarse-loamy, mixed, nonacid, fiigid
Mollic Endoaquepts), Brookston loam (Fine-loamy, mixed, superactive, mesic Typic
Argiaquolls), Conover loam (F inc-loamy, mixed, active, mesic Udollic Endoaqualfs), and
Newaygo sandy loam (Fine-loamy over sandy or sandy-skeletal, mixed, fiigid Alfie
Haplorthods). Soil series in Field B were Brookston loam, Newaygo sandy loam, and
Berville loam (Fine-loamy, mixed, mesic Typic Argiaquolls) (Soil Survey Division-
NRCS-USDA, 2001). Soil series maps were digitized fi'om Threlkeld and Feenstra (1974)
(Figure 4.1.b and 4.2.b).

The spatial sampling for SCN population density consisted of a geostatistical
sampling design applied within 8 and 5.25 ha in the center of both fields, as shown in
Figure 4.1 .a and described by Avendafio et al. (2003, Chapter Three). A bucket auger (8
cm diameter by 23 cm depth; Riverside Augers, Eijkelkamp, Giesbeek, The Netherlands)
was used to collect 160 single-core soil samples fi'om Field A and 110 from Field B,
within one week before planting and within three days after harvest in 1999 and in 2000.
Soil cores were placed in individual plastic bags and, upon arrival at the lab, were stored
in 10-gal Rubbermaid containers at 4°C until they were processed (within 30 days).

SCN analysis

Cysts were extracted from 100 cm3 sub-samples using a semi-automatic elutriator
(Research Services Instrument Shop, The University of Georgia, Athens, GA; Byrd et al.,
1976). The system had an extraction efficiency of 60%. Cysts were further separated fiom
soil particles following the sugar flotation-centrifugation method (Dunn, 1969) and then

counted under a stereo-microscope. Three cysts per sample were randomly selected,

58

crushed, and eggs and second-stage juveniles were counted, with the average used to
determine the eggs per cyst for each sample containing at least one cyst.
Soil texture analysis

A subset of the 1999 sampling locations was selected to evaluate the relationship
between soil texture and SCN population densities..Sarnple sites were chosen to include
all soil series described for each field (Threlkeld and Feenstra, 1974) and to include areas
of high as well as undetectable cyst density. Particle size analysis was conducted in
duplicate on a sub-sample of soil from each of the 25 and 24 selected samples from Field
A and Field B, respectively, using a modified hydrometer method (Gee and Bauder,
1986). Oven-dried soil was lightly crushed on a tray using a rolling pin to break up soil
structure until the sample passed through a 2-mm aperture sieve (mesh #10). Forty gams
of the < 2mm sieved soil was pre-treated in 100 m1 deionized water (DI-water) with 30%
hydrogen peroxide to oxidize the organic matter. Samples were then soaked in 100 ml of
calgon solution (sodium metaphosphate 50 gl") added to 200 ml of soil-DI-water
suspension for a minimum of 7 hours. Suspensions were then mixed for 5 minutes at
medium speed with an electric mixer to complete the dispersion of soil particles, and
transferred to a 1000 ml sedimentation cylinder. Volume was adjusted to 1000 ml with
DI-water. After thorough mixing of the soil suspension, the cylinder was left undisturbed
for exactly eight hours; the suspension density was then measured with a hydrometer
(ASTM 152H Bouyoucos style) reading the upper edge of the meniscus. Next, the
contents of the cylinder were poured through a 45-p. aperture sieve (mesh #325) to retain
sand particles. The sand retained in the sieve was carefully rinsed with DI-water,

transferred to previously tarred and labeled aluminum weighing dishes, oven-dried

59

overnight at 105°C, and weighed upon cooling immediately after removal from the oven.
The hydrometer reading was corrected for a blank cylinder containing 100 ml calgon
solution in 900 ml DI-water. The proportion of sand in each sample was calculated as net
sand weight divided by the initial weight of the sample (40 g) multiplied by 100. The clay
fraction was calculated dividing the difference between the hydrometer reading of a
sample and the blank reading by the initial sample weight, and multiplying this number
by 100. Silt fiaction was calculated subtracting the percentages of sand and clay from the
initial sample weight. Soil type was determined for each sample based on the percentage
contributed by each soil fraction as defined in the texture triangle recommended by
USDA (Soil Survey Division Staff, 1993).
Statistical Analysis

Descriptive statistics were applied to characterize soil particle size distribution for
each field. The soil separates sand, and clay were subjected to geostatistical analysis to
determine empirical omnidirectional semivariogams (Matheron, 1963). The parameters
of theoretical semivariogam models fit to the empirical semivariogams were estimated
by (nonlinear) least squares. The spatial distributions of sand, and clay were mapped by
predicting values at the nodes of a 1 x 1 m gid with universal or ordinary kriging using
the structural properties of the estimated theoretical semivariogam and the sampled
values at observed locations. The predicted values for the spatial distribution of silt were
determined as loo-(predicted sand value + predicted clay value) at each node of the 1 x 1
m gid.

Cyst population densities and numbers of eggs per cyst at planting and harvest are

shown as box-plots in Figure 3. Cyst density and number of eggs per cyst means were

60

compared across sampling times within fields with Fisher’s (protected) LSD test (a=
0,05) using logarithmic transformed data [loglo (cysts 100cm-3 soil + 1), logo (eggs per
cyst + 1)] to increase symmetry and to stabilize the variance.

Each SCN observation was associated with a kriging-predicted value for the
proportion of sand, clay, and silt, matched by location. A soil type classification was
assigred to each SCN observation based on the predicted proportion contributed by each
soil separate as defined by the texture triangle recommended by USDA (Soil Survey
Division Staff, 1993). Cysts and eggs per cyst were then compared among soil types by
sampling time within fields using Fisher’s (protected) LSD test for means (a= 0.05). The
effects of each of the soil texture fractions on transformed cysts and eggs per cyst were
analyzed with analysis of variance (ANOVA). Regession coefficients were determined
on means by simple linear regession analysis. Regession models were compared
between fields, and between sampling times within fields for parallelism when
appropriate ((1 = 0.05) (SAS®, Release 8; SAS Institute, Cary, NC).

The cross-correlogam is a geostatistical tool used to describe the spatial
continuity between measurements of different attributes or of the same attribute measured
at different times. The cross-correlation function given by Goovaerts (1997) was used
here to calculate cross-correlogams for logarithmic transformed cysts and eggs per cyst
with sand and clay percentage in the soil. Only the data points from locations sampled for

SCN and soil texture in both 1999 and 2000 were used in this analysis.

61

.82 :20 .380 -v :8 .88— -m .88: 38w -N .080 :83 A 08 89$ :ew 5:0

:8 J8 680 we 8:888 2: no :08: 88800: 838:8: 0&3 :ew .: .mewefi 0w8808: “8098.. 0 :8 i .0 E mwfieanm
2:. 60380 00.8 05 55:5 wEwE 8:88 .3 wean—e988 ma :8 05 E 8wfi88: am .0 :8 S20 -0 680 we seesaw—E:
8%? :8: 828a Ea 23:2: ace 880:. as as: are :8 82 has 05200 -m Ea .85 €50 008880
-v .82 8:00:85 -m .82 8880 -N .58: >98 ew>030z A 08 00:00 :em .8308 050 20:8: :8 .5: :00: em? 083
00383 085 we :05? 0.8%.: 8:08 :83 5:80: 8:038: 20m we 8:05:88: 2: new 880:8 8388 me 8:82 0.:
28%.: 009.80 .< 20:: E 80005 008m :8 :8 8% 02:58 we 8:83 A: .8 cm x on me 0:00 :tw 55:3 0880:: E88
E 8086:. 8:00:05 05 an 880:8 0.53 00388 .8380 :8 we 880:8 05 He: cwmme: 83.02 -0 .< 205 .3. 05me

E: “mam
2.:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

62

     

Soil
Series

 

 

 

North (m)
Q
&’

 

 

 

 

 

 

 

 

 

0100200300

East (m)

63

fl Berville Loam
'_' Brookston Loam

E] Newaygo Sandy Loam

Figure 4.2. Field B. a- Location
of sampled sites and soil series
present in Field B. Crosses
indicate the location of samples
collected for the determination
of SCN population density (see
Fig. la); black circles indicate
which of those samples were
also used for soil particle size
analysis. Soil series are l-
Newaygo sandy loam, 2-
Brookston loam, and 3- Berville
loam. Soil series map was
digitized from Threlkeld and
Feenstra (1974). b-Spatial
distribution of sand, c- clay, and
d- silt proportion in the soil as
interpolated by universal kriging
within the area sampled. The
shadings in c, d, and e represent
percentage ranges. e- Soil type
delineation determined based on
the proportion of sand, silt, and
clay. Soil types are l- Loamy
sand, 2- Sandy loam, and 3-
Sandy clay loam.

RESULTS
Soil Particle Size Distribution and Spatial Analysis
Soil particle size distribution of the surface layer in Field A and Field B
correspond to an overall surface texture of sandy loam (Table 4.1). Even though on
average soil particle size composition was similar, there were differences between Field
A and Field B in the range of variability observed for each fraction, and considerable
differences were observed when the data was analyzed spatially. Variation in the sand
separate was similar in both fields but clay varied more in Field B while Field A varied
more in the silt fraction.

Semivariogram models showed considerable spatial structure in the distribution of
sand and clay, with similar ranges of autocorrelation for each separate in Field A and
Field B (Table 4.2). The values predicted by kriging were used to generate contour maps
of the levels of sand, clay, and silt (Table 4.2, Figures 4.1.c-e and 4.2.c-e). In Field A, the
proportion of sand was the highest in the areas delineated for Belding sandy loam, and

lowest in the area of Brookston loam (Figure 4.1 .b, c).

Table 4.1. Soil texture in Field A and Field B as determined using a modified Bouyoucos
hydrometer method (Gee and Bauder, 1986).

Fraction Sample size Mean Std. DevT. Mini Max.§ CV

 

 

Fiel A % % %
Sand 25 63.9 8.81 44.7 74.1 13.78
Clay 25 15.6 3.00 10.0 20.0 19.17
Silt 25 20.5 8.47 10.6 42.8 41.50
Fiel B
Sand 24 67.4 10.23 49.9 85.3 15.15
Clay 24 14.0 5.33 6.2 22.5 38.34
Silt 24 18.6 5.29 8.4 28.8 28.48

 

1' Standard deviation, 1 Minimum, § Maximum.

Table 4.2. Parameters of the theoretical semivario gram models of sand, silt, and chy
fractions in Field A and Field B before planting in 1999. T

 

 

DrifiI MOdCl funct'nn CO§ C1] Range# Co 1.1.
C+Co
FieldA m
SandII None Nugget ~0 0
Exponential 87.0 70
Clayfl None Nugget 3.9
Gaussian 6.9 130 0.4
Field B
SandII Linear Nugget ~0
Spherical 28.8 66 0
Clayfl Linear Nugget ~0
Spherical 11.1 129 0

 

1' Modeb were fitted by least squares based on empirical semivariogram calculated for
lags ranging fi'om 30 to 90 m (h), with a lag tolerance of N2. The minimum number of
pairs required for each lag was 30
1 Whenever semivario grams showed nonstationarity, the data were detrended carrying
out a simple polynomial least squares regression and semivariogram analysis was
performed on the residuals. The polynomial order of the trend is indicated when a drift or

trend was removed.

§ Co is the nugget effect or a discontinuity in semivariance at the origin due to microscale
variability or sampling error.
1] C is the partial sill defined for spherical, exponential and gauss'nn models.

# Observations that are spatially separated by more tlnn the range are uncorrehted. The
range is indicated {or spherical models, and the practical range is indicated for
exponential and gaussian rmdels.
‘H' Co/(C+Co) is an indicator of the degree of spatial structure, the bwer the number the
stronger the spatial autocorrelation.
II Semivariograms were calculated for the percentage of sand and clay presem in soil
sanples, quantified using a modified Bouyoucos hydrometer method (Gee and Bauder,

1986)

The proportion of clay in the soil in Field A was mostly between 10- 20 % except

for a few small patches where the proportion reached values slightly higher than 20 % in

the area of Breckenridge sandy loam (Figure 4.1 .b, d). The highest level of silt was

located within the area of Conover loam (Figure 4.1.b, e). In Field B, the lowest level of

sand, and the highest of clay and silt corresponded with the delineation for Brookston

loam, whereas the highest level of sand and the lowest of clay and silt were located in the

area of Newaygo sandy loam and Berville loam. Intermediate levels were found in the

65

transition zone between Brookston loam and the other two soil types (Figure 4.2.a-d).
Overlaying the soil separate maps, we mapped the spatial distribution of the resulting soil
map units in each field based on the proportion contributed by each separate (Figures
4.1 .fand 4.2.e). Sand content seemed to dictate soil map units delineation in Field A
(Figure 4.1 .b, t), while clay content dictated soil map units delineation in Field B (Figure
4.2.c, e). Field B graded from sandy clay loam in the south to loamy sand in the north,
while Field A was predominately sandy loam and loamy sand with patches of sandy clay
loam and loam (Figures 4.1.f and 4.2.e). The soil maps obtained this way were useful to
locate areas of very distinct soil map units located within the delineations reported by
Threlkeld and Feenstra (1974) in each field.
SCN Population Density

The within fields and between years variability in preplant cyst and eggs per cyst
for 1999 and 2000 have been reported in Avendafio et al. (2003, Chapter Three). Briefly,
the number of cysts 100 cm'3 soil at planting and at harvest was similar in 1999 and in
2000 in Field A (Figure 4.3.a); whereas in Field B, there were more cysts in 2000 than in
1999 and more so at harvest than at planting (Figure 4.3.b). A section of Field B
containing 26% of the samples was ponded with water at harvest in 2000. The localized
flooding reduced the number of samples characterized by low cyst density observed in
previous samplings influencing the results obtained for this particular sampling time as
evidenced by a sample mean greater than expected (Figure 4.3.b). Generally, there were
more eggs per cyst in Field B than in Field A in both years and sampling times, with the

greatest number of eggs per cyst observed at harvest in 1999 (Figure 4.3.c, d).

66

 

 

 

A C
P'99‘bh ~cI4->° °
H‘99‘al"° ”bl—”m ° g!
P‘OO‘bl‘” ”al.—l" E
€H‘OO‘C'“° *bl-i-M
g . . . . . . . . . . .
ps99~ci—poo "8+0 0
H'gg-bfi—POOO‘D Lad-[ho g!
P‘ooibfih OO OrbI—l—+0- a
m. In» ~b+I—~ ..

 

 

 

 

 

01W200W4W50001N2MM4W5006M

Cysts 100 cm3 of soil Eggs per cyst

Figure 4.3. Boxplots of SCN (a) cyst population density and (c) eggs per cyst in Field A;
and (b) cyst population density and ((1) eggs per cyst in Field B. The y-axis indicates the
sampling time: P’99: Planting 1999, H’99: Harvest 1999, P’OO: Planting 2000 and H’OO:
Harvest 2000. The sample mean is indicated with a dashed vertical line and the median
with a solid vertical line in each box. Significant differences among the means at different
sampling times within each field are indicated with lower case letters. Mean comparisons
were performed on logarithmic transformed data; original data are shown for clarity.
Means with the same letter were not significantly different (protected LSD, 5 %
significance level).

Relationship between Soil Map Units and SCN Population Density
Cyst population density varied by soil map unit. In Field A, cyst density was
consistently higher in loamy sand and sandy loam than in the other two soil map units,
although differences in cyst density among soil map units were statistically significant
only at harvest in 1999 and planting in 2000 (Figure 4.4.a—d). In Field B, the number of

cysts per 100 cm3 of soil was the highest in loamy sand and lowest in sandy clay loam in

67

both years and sampling times (Figure 4.4.a-d). The number of eggs per cyst was not

significantly different among soil map units in either field or sampling time (Table 4.3).

 

 

 

 

 

 

 

 

 

J A 0 Field A
2 5 0 Field 8
c . 13
g 2.0 g
as :r.
05 1.5 4 3
0 to
'O 1_o J .{ .s
“O
(U .
U 0.5 a; a } a
c
,3 0.0 -
+1 2.5 < B
A
I 2.01 0:;
_ 2
’5 1.5 < o
a) $9.
'5 1.0 1 cf i "
e a a g
. ab
g 0.5
8 0.0 . b . - -
I; 2.5 l C
‘0'
5 2.0« :9
v m
2 1.5 1 2*.
3
g, 1.0 < b ‘9
c b a g
(6 , . a
o 05 b
2 0.0 . b f a .
2.5 1 D
2.0 - a} I
1.5 1 3
b 0
1.0 4 53
N
0.5 « a
a 3 §
0.0 . n - . .
SCL SL LS L
Soil type

Figure 4.4. SCN mean cyst population density [loglo (cysts 100 cm'3 of soil + 1)] :1:
standard deviation by soil type in Field A (open circles) and in Field B (black circles) at
a- planting 1999, b harvest 1999, c planting 2000 and d harvest 2000. Soil types are
(SCL) sandy clay loam, (SL) sandy loam, (LS) loamy sand and (L) loam. Means with the
same letter within field are not significantly different (protected LSD, 5 % significance
level).

68

55—0 .8: 850% 0.8 80: 88308.83 ._0>0_ 5:338: nod 05 8 ~00. qu 308883 3.0%: m new? 890
:8 mecca 88:88 083 008: mam—980 55:3 5+ :8 mo «-8 :2 “who 8: wwwov ewe: 0808 88.8853 £822.8on H

.282 0% Essa stem meme <8: 3 3808882
288.: 0.538: 05 .3 8::0: 00 Emo— 3: :8 693. >82 6.: .82 30:00 Gm: .82 >30 .380 30m: 083 89»: :em ._.

 

 

 

 

 

 

 

m2 m2 m2 m2 02 02 m2 m2 ”3o 03
N6 3% mdfimmm in We: Ndfiem A
me 130 Qwflvw edfimwc: e6 3mm: odflom Gwamm Shem NNHQ m1:
:4 13m $203 ext. 362 wfifim: Ndfimm ewaev Ewawv Q33 Am
edefiofi 0.3an wd—Hmo— m.mmace o o.w ENS. méwfivw 0632 40m
00 H “who 8: 030 802 00 a 8:0 .5: $0 802
508: was—BE 8020: wares—m “0083 @585 8023: mates—m +095 :om
coon 3e: ooom aeo—
m 20:: < 20::

 

.88 5 Ba 32 a 0 22m as < 22.1. a 25 =8 3 sea 8 am... 2% as: .3. 2.5

69

Relationship between Soil Particle Size and SCN Population Density

SCN population density was strongly affected by the proportion of sand, clay, and
silt in the soil (Figure 4.5, Table 4.4). The effect of sand was characterized by a positive
slope, not significantly different across sampling times in Field B (Table 4.4, Figure 45.a—
d). The regression model at harvest 1999 in Field A had the same slope and intercept (a =
0.05) as the model for Field B at harvest 1999 and at planting in 2000 (Figure 4.5.b). Cyst
density decreased with increasing clay proportion in the soil at all sampling times in Field
B and at harvest both years in Field A, with the same slope (a = 0.05) across sampling
times and between fields (Table 4.4, Figure 4.5.e-h). Silt proportion in the soil also had a
negative effect on SCN; cyst population density decreased linearly with increasing
percentage of silt at all sampling times in Field B (Table 4.4, Figure 4.5.e-h). Regression
models were also parallel for silt across sampling times. The relationship between silt and
cysts was not significant in Field A.

Unlike the cyst population, the relationship between eggs per cyst and soil particle
size was highly variable (Table 4.5, Figure 4.6). The proportion of sand in the soil had a
negative effect on eggs per cyst at planting in Field A, but a positive effect at harvest the
same year (Figure 4.6.a, b). Clay percentage had a negative effect on eggs per cyst at
harvest in Field A in 2000 (Figure 4.6.h), and silt percentage had a negative effect on eggs
per cyst at planting in 1999 in Field B (Figure 4.6.i). Otherwise, sand, clay, or silt had no

significant effect on eggs per cyst.

70

5:5 5\5 E 52555 6:55 5\5 u x 5+ :83 5-80 9: $953 Swo— u a H
.855 us 55.5% 5558525...
0.53 52525 2: mo 282 .88: $5.588 550.55 8520:5555 .85 3558 0.53 3% £25 .983 85.395 :8 :55... 5353 £322 ._.

 

 

 

 

 

555 5.5555555"5 555 5555.555"5 555 x555.5+55.5.n5 5555 585:
555 5555.555".A $5 52555.3?5 55.5 5.555.555. Tu» 5555 5:55:
55.5 555.555.?» 55.5 655.545?» 55.5 5555.55.5fiA 552 5255
$5 $555-55.?» 55.5 5555555.?» R5 U85.5.5..m5.~.n5. 552 5:55:
I'm so:
52 55.5 55555.85"5 52 5555. 585:
52 52 52 5555 555%:
52 55.5 5.855.555”.A £5 5.255355L 552 5832.5
52 52 52 5555 55.525
III< 225
m :38: m 5.5252 x E5552 2:5 535855
:5 .35 555m

 

doom 5 can 33 E m 2on 98 < 20E 5 =8 2: E :Btomoa 5:5
S $5.5 6555 can bin—ow cog—smog ammo 5953 5535:0556.— ofi no 535585205 mo 356508 98 £259: commmohwom .vé 035,—.

71

 

 

 

 

2.5 < A E I 0 Field A '2
20« 0 Field B m
3.
1.5‘ 3
D
1.0 J _.
0.0
2.6< I
n)
2.0 . 2
(D
1.5- g
1.0 1 —L

 

 

Log1o (cysts 100 cm'3 of soil + 1) i standard deviation

oooz ISONBH oooz Bunueld

 

 

 

 

 

 

% Sand % Clay

Figure 4.5. Relationship between SCN cyst population density [loglo (cysts 100 cm'3 of
soil + 1)] and the proportion of each soil fiaction in the sample. Means i standard
deviation for Field A (open circles) and for Field B (black circles) are indicated. The left
column (a, b, c, (1) corresponds to the percentage of sand, the central column (e, f, g, h)
corresponds to clay percentage and the right column (i, j, k, 1) corresponds to silt
percentage; at (a, e, i) planting 1999, (b, f, j) harvest 1999, (c, g, k) planting 2000, and (d,
h, 1) harvest 2000. Regression curves fitted are indicated as solid lines (Field B) and
dashed lines (Field A) where significant (5 % significance level). Regression coefficients
are shown in Table 4.4.

72

5mm 5\5 .8 acow—5.x. €555 5\5 u x 5+ memmo 5-89 2: “who 5m wwwov ewe— " a H

536 NS Engage
>358me6 8: 203 memo—m 52:: mam—9:55 550.85 85:05:83 com @952 0.53 26 £30 .983 38535 :8 some Eng? £832 ...

 

 

 

52 52 52 555m 58$:
52 52 52 5555 555.25
52 52 52 552 58%:
~55 £55555“.A 52 52 552 5555:
lm 22m
52 R5 £555-55.an 52 5555 58$:
52 52 52 5555 55:5:
52 mz 555 $85.55;..I.s 552 535:
52 52 R5 58.5-3.1 5555 wean:
Ill< 235
x 50552 No 50552 x 23552 2:5 méefim
555 Eu 2%

 

doom 5 EB 33 E m 23m use < 20E 5 =8 2: E eoanoE
E5 .8 $20 €555 93 596 be ammo 5053 9:20:38 05 we 503258.583 mo 356508 98 fleece commmoewom .mé 035.5

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3:21“ a 1,.- 5.: '- 2:23;; a
E éiflifl l i “ii WM} 5;
3;: that - Ethnic} fig} ({1})? g
t E l 5* "r ‘1 pm” “r *‘l
E... . i ii} {i «t fig} } 1. ::
1.22:? :fiitl t . - i 9% i. ii}. g
EZQ {1* g G . if {KK- . ° g
3112. fifi‘ifl Fiiljii {Witihlii {} Mill} 2 g
géofwifi'; :5}? H! i§+iiiili.i Lgfifiiiiéhoi g
5:933 Hi i. . W; W '3’

Figure 4.6. Relationship between SCN eggs per cyst [loglo (eggs per cyst 100 cm'3 of soil
+ 1)] and the proportion of each soil fraction in the sample. Means a: standard deviation in
Field A (open circles) and in Field B (black circles) are indicated. The left column (a, b,

c, (1) corresponds to sand percentage, the central column (e, f, g, h) corresponds to clay
percentage and the right column (i, j, k, 1) corresponds to silt percentage; at (a, e, i)
planting 1999, (b, f, j) harvest 1999, (c, g, k) planting 2000, and (d, h, l) harvest 2000.
Regression curves fitted are indicated as solid lines (Field B) and dashed lines (Field A)
where significant (5 % significance level). Regression coefficients are shown in Table
4.5.

74

Linear correlation coefficients for cysts with sand were very low in Field A
(Figure 4.7.a-d). The low correlation decreased even more with increasing separation
distance between samples. Cysts were better correlated with clay at harvest than at
planting, and more so in 2000 (Figure 4.7.b, d). The separation distance at which the
cross-correlation cysts-sand or cysts-clay reached zero was the same (approximately 110
m) at harvest in 2000 (Figure 4.7 .d).

Cyst population density was highly correlated with sand and clay at all times in
Field B, and the correlation decreased in absolute value at the same rate for both
separates, reaching zero at the same separation distance between samples (Figure 4.7.e-h).
The distance at which cross-correlations were zero increased from approximately 115 m
to 130 m from planting in 1999 to harvest in 2000. The symmetry in the cross-
correlograms indicated that the effect of sand on cyst population was equal in magnitude
to the effect of clay, but with opposite sign.

The number of eggs per cyst was only correlated with clay at harvest in 2000 in
Field A; otherwise, correlations were very low as indicated by the linear correlation
coefficients (Fig 4.8.d). Cross correlograms showed fluctuations in correlation between
eggs per cyst and sand or clay as separation distance between samples increased. In Field
B, the symmetry between sand and clay cross correlograms observed for cysts was also

evident for eggs per cyst (Figure 4.8.e—h).

75

Field A

Field B

 

 

0.75 4

0.50 4

—0.50 4

-0.75 4

 

 

 

r sand = 0.20
r clay = -0.05

 

 

 

 

0.75 ~ Bi

-0.50 4
-0.75 4

 

 

r sand = 0.26
r clay = -0.26

 

 

 

 

 

 

 

0.75 4
0.50 4

Cross correlogram

0.25 (
0.00 «
0.25 ‘
-0.50 -

-0.75 4

I,

 

 

 

rund=0.62

 

 

 

 

0.75 J D
0.50 « ‘
0.25 1
0.00 «
0.25 <
050 ~

 

0.75 ‘ i
0 20

V v

4060

 

 

r sand = 0.28
r clay = -0.41

00100120140100100

  

 

 

‘0’.
rcIay=-0.04 fclay='0.58
4 reand=0.77 I-I
rday=-0.70 )0

a /

' ' 1r

 

\.

 

 

 

 

Lagdistance(m)

0 20400000100120140100

0003 Bunueud 666L isatueH 666l Bunueld

0008 ISGNBH

Figure 4.7 . Cross-correlograms of SCN cyst population density and percent sand (solid

line) or clay (dashed line) in the soil in Field A (a-d) and Field B (e-h). a, e- Planting
1999; b, f- Harvest 1999; c, g- Planting 2000; d, h- Harvest 2000. Linear correlation

coefficients for cyst density with sand and clay are indicated.

76

Field A Field B

 

 

04A E

0.2 4

 

 

 

 

 

 

 

 

 

 

666i fiuuueld

.02 .. 5 E '1' . it

0.4 1

 

 

 

 

0.2 4

 

 

 

 

-
00 q ——q_—
0 .

 

0.21 /

 

 

666i lSGNBH

 

 

0.4 4

 

 

 

 

 

 

 

 

 

 

0.2 4 E E 4 I 1
- i '\
:5 /’_'0 r/‘\ \/ \x s__ s

Cross correlogram

 

 

 

0.0 ~ -—.

f
\

‘/

T

i

II
\

-0.2 « - ‘b

 

 

 

 

 

 

f
l
f
X
\
0003 6“9“?ch

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i
l

0.2 J .
d.- ' I

-_,. _.._i/ l ; JD 4 _ ' V’\
0.04 f / / K / k 7‘7’4

 

 

 

 

. ‘1’ ‘1f .
0.2 J .r . %

/.‘ ‘11.. r sand = 0.04 l' and = 0.10
4144" ‘J rcIay=-0.33J ‘ ' rday=-0-11

0 20406080100120140160180 o 20 40 60 80100120140160
Lagdistance(m)

 

 

 

 

 

 

0003 lseNeH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4.8. Cross-correlograms of SCN eggs per cyst and percent sand (solid line) or clay
(dashed line) in the soil in Field A (a-d) and Field B (e-h). a, e- Planting 1999; b, f-
Harvest 1999; c, g— Planting 2000; d, h- Harvest 2000. Linear correlation coefficients for
eggs per cyst with sand and clay are indicated.

77

DISCUSSION

This study was conducted as part of a project designed to investigate the potential
application of SSM of SCN in Michigan soybean production systems (Avendano etal.,
2003, Chapter Three). The work presented here provides further insights into the potential
of SSM for SCN by describing detailed relationships between SCN population density
and soil texture.

While it is not known when SCN was introduced into the fields and what
undetermined factors may have contributed to the difference in SCN population density,
Field A and Field B presented us with the opportunity to study the relationship between
SCN population and soil texture under two different conditions frequently encountered by
soybean growers. Cyst population density was high in Field B and it increased during the
study whereas in Field A, cyst density was much lower. While soil survey maps suggested
that Field A and Field B had very similar soil types, geostatistical sampling and analysis
provided a different perspective into the relationship between soil texture and SCN
population dynamics. The spatial distribution of the soil separates was highly structured
in both fields, allowing the construction of reliable maps for the distribution of sand, silt,
and clay in each field. The arrangement of soil map units obtained from superimposing
these maps corresponded in general terms with the soil survey maps reported by
Threlkeld and Feenstra (1974).

The relationship between soil map units and SCN population density was
consistent between the two fields, although differences were attenuated in Fields A,

where cyst population remained low at all times. Cyst population density was consistently

78

higher in loamy sand than in the other soils; it was lowest in sandy clay loam (Figure 4.4).
While this observation is consistent with previous studies, where SCN was found more
abundantly in coarser soils than in finer soils (Dropkin et al., 1976; Koenning and Barker,
1995; Donald et al., 1999), the relationship between soil types and ranges of textural
composition needs careful consideration. Koenning and Barker (1995) reported highest
SCN egg density in Fuquay sand (91% sand: 6%silt: 3% clay), Norfolk sandy loam
(84:12:4) and Portsmouth loamy sand (72:18:10) when compared with Cecil sandy clay
loam (53:18:29) and Cecil sandy clay (48:13:39). The lowest nematode density was found
in soils with more than 25% clay. We propose that SCN can sustain high population
density levels (above 20 cysts per 100 cm3 of soil) only in soils composed of more than
60% sand, less than 20% silt and less than 20% clay. However, the soil texture defined
for this composition corresponds to sandy loam, loamy sand and sand. The difference in
texture between sandy loam and sandy clay loam in our research fields was the result of
reduced sand content (~60%) combined with increased clay (>20%). It is important to
note that only a portion of the area defined for sandy loam in the texture triangle is
favorable for SCN. Sandy loam with more than 20% silt was associated with low levels
of cysts in our study. The particle size composition of sandy clay loam and loam are
beyond the 60:20:20 limits, accordingly with our proposition, these soil types had
significantly less cysts than loamy sand (Figure 4.4). Therefore, it might be best that
references to any relationship between SCN population and soil include soil texture in
addition to soil classification.

The number of eggs per cysts was not related to soil map units or texture in our

study, with a few exceptions. The data suggest that soil texture affects SCN population at

79

the mobile stages during root finding and penetration, and perhaps development in the
roots, rather than the reproductive potential, fecundity, or hatching. This phenomenon
was previously reported by Todd and Pearson (1988) when they recovered more SCN
females and cysts from newly infested roots in sandy loam (60:30:10) than in silty loams
(30:46:24 and 14:60:26). Nevertheless, Young and Heatherly (1990) attributed the lower
rate in SCN reproduction to soil type in a study using Sharkey clay (85:34:57.5) and
Dubbs silt loam (23:60.5: 16.5). This indicates that a high proportion of clay and very low
sand content are necessary to interfere with SCN reproductive potential. However, the
effect of this kind of soil on root development should also be taken into consideration
(Russell, 1977), since the effect of soil texture on cyst fecundity may be strongly
influenced by plant conditions (Koenning and Barker, 1995).

It is difficult to discriminate which of the separates: sand, silt or clay, has the
greatest influence on cyst density. The analysis of cyst density by soil separate indicated
that sand had the opposite effect of clay or silt. These observations indicate that the
combination of the separates, that is, the resulting soil texture has a greater influence on
SCN population than a specific separate by itself. This is the first report to document
consistency in the relationship between SCN and soil texture across fields and over time.
Wyse-Pester et al., (2002) explored the possibility of using correlation between soil
attributes (soil texture) and nematode density to reduce the cost of sampling in an effort to
map nematode distribution for SSM. Although the spatial dependency indicated a
potential for mapping Helicotylenchus spp., T ylenchorhynchus capitatus, and
Pratylenchus neglectus infestations, the small variation in soil texture in their research

fields resulted in inconsistent and weak correlations with nematode density.

80

Site-specific management of soybean yield-limiting factors requires an
understanding of the spatio-temporal dynamics of the prevailing conditions. In addition to
providing the basis for future experimentation to define soil texture tolerance limits for
SCN, this study lays a foundation for new and integrated approaches to SSM of SCN and
other yield-limiting factors. The co-variation of SCN with other factors in soils is helpful
to assess spatial variability of SCN populations within fields if SSM is to be a plausible
strategy to manage SCN in soybean production. We conclude that soil survey maps can
be useful in predicting expected levels of SCN in an already infested field but they should
be used with caution. Soil map unit delineations based on the texture triangle may not be
sufficient when referring to SCN population and the proportion of each soil fiaction
should be used in addition. We have shown in this study the benefit of texture-based
analysis over soil type-based analysis therefore, a soil survey map can be used to identify
high-risk sectors, and then a texture analysis of soil fiom these zones may help

delineating soil map units of expected high cyst density.

81

CHAPTER FIVE

RELATIONSHIP BETWEEN SOIL TEXTURE AND SOYBEAN CYST NEMATODE
POPULATION DYNAMICS IN SOIL AND IN SOYBEAN ROOTS

Felicitas Avendafio, Francis J. Pierce and Haddish Melakeberhan. Manuscript to be
submitted to Nematology.

82

RELATIONSHIP BETWEEN SOIL TEXTURE AND SOYBEAN CYST NEMATODE
POPULATION DYNAMICS IN SOIL AND IN SOYBEAN ROOTS

F elicitas Avendafiol, Francis J. Pierce2 and Haddish Melakeberhanl.

1- Dep. of Entomology, 243 Natural Science Building, Michigan State University, East
Lansing, MI 48824 (avendan1@msu.edu); 2- Center for Precision Agricultural Systems,
Washington State University, 24106 North Bunn Road, Prosser, WA 99350
(fjpierce@wsu.edu); 3- Dep. of Entomology, 243 Natural Science Building, Michigan State
University, East Lansing, MI 48824 (melakebe@msu.edu).

Correspondence should be sent to Haddish Melakeberhan

ACKNOWLEDGEMENTS

The project is part of a Ph.D. Dissertation of the first author. We gratefully
acknowledge anonymous reviewers for their helpful comments and criticisms of the
manuscript. We thank Chad Holovach, Nathan Nye, Megan Kierzek, Carolyn Stein,
Nathan Cottrell, Bindiya Shah, Dan Armstrong, Marisol Soto, Carolina Holteuer, Katrina
Blakely and Teresa Raslich for assistance in collecting and processing samples; and Dave
Eickholt, Charles Eickholt, (farmers) and Neil R. Miller (ABC-Agribusiness Consultant)
for providing and maintaining the sites. Financial support for the project was provided by
the Michigan Agricultural Experimental Station (Hatch Project) and grants from the
Michigan Soybean Promotion Committee, North Central Soybean Research Program and

United Soybean Board to the last author.

83

RELATIONSHIP BETWEEN SOIL TEXTURE AND SOYBEAN CYST
NEMATODE POPULATION DYNAMICS IN SOIL AND IN SOYBEAN ROOTS

ABSTRACT

The purpose of this work was to identify i) the extent to which soil texture affects
Heterodera glycines, soybean cyst nematode (SCN) population dynamics and ii) to assess
the potential for management zone delineation based on soil map units for designing an
alternative SCN management practice. Traditional statistics and geostatistics were
applied to analyze SCN population dynamics and spatial distribution in the soil and in
soybean roots in relation to soil texture in two Michigan fields. Soil texture had a strong
and consistent effect on SCN cyst population density and spatial distribution in one field,
but not in the other. Coarser soils supported higher SCN population density than finer
soils delineating areas of high or low population density within the field. The effect of
soil texture was weaker on the third and fourth larval stages and immature females, and
not significant on infective larvae in the host roots. It is not clear from this work whether
the effect of soil texture on the number of eggs per cyst was on fecundity (egg
production) or on hatching. The stability in the relationship between SCN spatial
population dynamics and soil properties indicates the potential for delineation of

management zones to reduce the economic loss due to SCN in infested fields.

84

Soybean cyst nematode’s (SCN, Heterodera glycines Ichinohe) ability to adapt to
a wide range of environmental conditions seems to be one of the reasons for its
widespread distribution and economic significance in soybean [Glycine max (L.) Merr.]
production worldwide (Winstead et al., 1955; Wrather et al., 2001 a, b). The broad
distribution over a range of production systems and soil types in the USA demonstrates
the limitations of current SCN management tactics (Bradley et al., 1996; Workneh et al.,
1999; CTIC, 2000). Avendafio et al. (2003, Chapter Three) investigated SCN spatial
distribution within fields to assess the potential for SCN site-specific management
(S SM). Even though the spatial structure of SCN population density was poor, and varied
between fields and years, it was possible to identify areas within a field where SCN
population densities remained high or low, over time. Wyse-Pester et al., (2002) explored
the possibility of using correlation between soil attributes (soil texture) and nematode
density to reduce the cost of sampling in an effort to map nematode distribution for SSM.
Although the spatial dependency indicated a potential for mapping Helicotylenchus spp.,
T ylenchorhynchus capitatus, and Pratylenchus neglectus infestations, small variation in
soil texture resulted in inconsistent and weak correlations with nematode density.
However, SCN has been found correlated with soil texture consistently over time
(Chapter Four), with higher equilibrium density in coarse soil than in finer soils (Dropkin
et al., 1976; Koenning and Barker, 1995; Workneh et al., 1999). In addition, SCN
population density was spatially structured to a greater degree in fields with diverse soil
types than in more homogeneous fields (Avendafio et al., 2003-Chapter Three; Donald et
al., 2001). The temporally stable location of hot spots, or areas of high SCN population

density reported in Chapter Four, consistently correlated with soil texture makes the

85

delineation of SCN management zones based on soil map units an attractive alternative
(Doerge, 1998).

The relationship between SCN development and the soil environment is poorly
understood. SCN development is certainly affected by the genetic characteristics of the
host plant. There is disagreement on whether exudates from susceptible or resistant
soybean plants stimulate more SCN hatching (Sikora and Noel, 1996, Schmitt and Riggs,
1991; Teffi and Bone, 1985). Nevertheless, it is clear that development is most successful
on SCN-susceptible soybean varieties than on SCN-resistant varieties (Wallace et al.,
1995), especially during the vegetative phase (Sikora and Noel, 1996; Hill and Schmitt,
1989; Tefit and Bone, 1985). The average number of eggs per cyst, however, was not
correlated with soil texture (Chapter Four), and infective juveniles were equally
successful infecting roots in a clay soil (8.5% sand, 34% silt, 57.5% clay) and a silt loam
(23% sand, 60.5% silt, 16.5% clay) in a greenhouse test (Young and Heatherly, 1990).
Hence, we postulate that soil texture affects SCN survival in the soil and development
rather than reproduction and root invasion.

Measuring the extent to which soil texture affects SCN population dynamics and
particular life stages is a necessary further consideration to the possibility of management
zone delineation based on soil map units as a plausible SCN management alternative. The
objectives of this work were to: i- characterize SCN population dynamics in soybean
roots and the surrounding soil in two fields in Michigan over two growing seasons; ii-
investigate the extent of the correlation of soil texture with SCN population in the roots
and in the soil, and with eggs per cyst; and iii- analyze SCN population dynamics

spatially in relation to soil texture.

86

MATERIALS AND METHODS
Research Site and Sampling Design

The study was conducted in Shiawassee County, MI in 1999 and 2000 on two
fields (Field A and Field B) maintained by the cooperating farmer. Field A was 24 ha,
managed under no-tillage since 1996, and was planted to corn before this study in 1998.
Field B was 13 ha, conventionally tilled after wheat in 1998, and was managed under no-
tillage thereafter. In 1999, a SCN -susceptible soybean variety (Asgrow 1901), and in
2000, an SCN-resistant variety (Asgrow 2201 ), both Roundup-ready, were grown in both
fields. Soybean was planted in l9-cm rows at a rate of 519 000 viable seeds ha'1 in 1999,
and 494 000 viable seeds ha’1 in 2000. Fields A and B were planted 5/22/99 and 5/16/99,
respectively, and 6/9/00.

The experimental design was developed at the initial phases of this study to assess
the spatial variability of SCN population and the effect of soil texture on SCN cyst
population density at harvest and at planting reported by Avendafio et a1. (2003, Chapter
Three and Chapter Four). The spatial sampling for SCN population density consisted of a
nested survey sampling design with distances reduced by geometric progression (adapted
from Webster and Boag, 1992) applied within 8 and 5.25 ha in the center of Fields A and
B, respectively (Figure 3.1.a,b). Pairs of single-core samples were collected 20, 7.9, 2.7,
0.9, and 0.3 m apart (Figure 3.1 .c). Each sampled location was chosen in a random
direction. The angles for each new location were expressed in north and east coordinates
to facilitate the location of the sampling sites in the field. This arrangement produced 160
and 110 sampling locations that were flagged and geo-rcferenced using GPS in Field A

and Field B, respectively.

87

Soil and Soybean Root Sampling for SCN.

Soil samples for SCN cyst quantification and soybean root samples for SCN
developmental stages were collected from Field A and Field B at about 30 day intervals
from planting to harvest in 1999 and 2000 (Table 5.1). Adverse weather conditions
prevented sample collection after 76 DAP in 2000, and delayed harvest until November,
when roots were at an advanced state of decomposition so only soil samples were
collected at this time. Sampling dates were selected based on the length of SCN life cycle
(3-4 weeks). In 2000, the first set of samples was collected before 20 days after planting
to detect the early stages of the infection.

A single core of soil was obtained at each flag location using an 8 cm diameter by
23 cm deep bucket auger (Riverside Augers, Eijkelkamp, Giesbeek, The Netherlands).
The position of the auger was rotated around the flag in successive samplings. Soil cores
were placed in individual plastic bags and stored at 4°C upon arrival at the lab until
processed. Cysts were extracted from a 100 ml subsample of soil measured by water
displacement from each sampled location. A semi-automatic elutriator (Research
Services Instrument Shop, The University of Georgia, Athens, GA) was used for cyst
extraction following standard procedures (Byrd, et al. 1976) with 60% extraction
efficiency. Cysts collected in 75-}; aperture sieves (#200 mesh) were further separated
from soil particles following the sugar flotation-centrifugation method (Dunn, 1969).
Cysts were then counted under a stereo-microscope. Three cysts per sample were
randomly selected, crushed, and eggs and second-stage juveniles were counted, with the

average used to determine the eggs per cyst for each sample containing at least one cyst.

88

Table 5.1. Sampling dates for soil and soybean root samples collected from Fields A and
B in 1999 and in 2000.

 

 

 

 

 

Field A Field B
Sample Date DAPT Date DAPT
collected
1999
Soil May 13 -8 May 20 4
Soil and roots June 22 31 June 29 44
Soil and roots July 20 59 July 28 73
Soil and roots August 16 86 August 24 100
Soil and roots September 25 126 @arvest)1 September 18 125(harvest)1
2000
Soil June 5 -4 June 12 3
Soil and roots June 26 17 June 28 19
Soil and roots July 24 45 July 25 46
Soil and roots August 21 73 August 25 77
Soil November 3 147 (harvest); November 11 155 (harvest):

 

1' Days after planting. A negative sign indicates days before sampling.
1 Samples were collected within a week before or after soybean was harvested.

Three soybean plants were dug out at each flag location using a small shovel, and
the soil adhered to the roots was gently shaken off before cutting off the stem of the plant
at its base and placing the roots in a plastic bag. At the lab, roots were gently washed with
tap water to remove soil, cut into approximately 4-cm sections, and mixed. Two grams of
root fragments were stored at 4°C until stained within 24 hours and the rest was
discarded. SCN inside roots were stained with a NaOCl-acid fuchsin technique (Byrd et
al., 1983) modified as follows. The treatment time and the proportion of chlorine bleach
(5.25% NaOCl) used to partially break down plant tissue and facilitate the penetration of
the dye were adjusted depending on the age and thickness of the roots. Roots were lefi in
1:3 bleach: water solution (1.31% NaOCl) for 3 minutes at 17 - 19 DAP, and for 5
minutes at 44 - 46 DAP; older roots were left in 1:1 bleach: water solution (2.62%

NaOCl) for 5 minutes at 73 DAP, for 6 minutes at 100 DAP, and for 7-10 minutes at 125

89

DAP. Nematodes were then stained with acid-fuchsin following the procedure described
by Byrd et al. (1983). Stained samples were kept at 4° C until counted.

Roots preserved in acidified glycerol were spread forming a single layer and
pressed between clear plastic plates to facilitate visualization of nematodes under a
stereo-microscope. SCN developmental stages were determined as illustrated in Agrios
(1997) and counted in four categories as follows. Vermiform infective juveniles (J2) were
counted separate from J3 and J4 males and females (J 3/J 4) identified as short, stout, well-
stained nematodes; females without eggs were classified as immature females; and
whenever eggs were visible inside the females or inside gelatinous matrix, they were
counted as mature (gravid) females.

Soil Texture

Maps of sand, clay, and silt proportion in the soil in Field A and Field B from the
previous chapter (Chapter Four) (Figure 4.1 .c-e, and 4.2.b-d) were used to associate
percentage values of sand, clay, and silt predicted by kriging to each SCN observation
based on location in the field. In Field A, percent sand in the soil ranged from 45% to
74%, percent clay ranged from 10% to 21%, and percent silt ranged fi'om 8% to 43%
(Chapter Four). Soil series in Field A were Belding sandy loam (Coarse-loamy, mixed,
frigid Argic Endoaquods), Breckenridge sandy loam (Coarse-loamy, mixed, nonacid,
fiigid Mollic Endoaquepts), Brookston loam (Fine-loamy, mixed, superactive, mesic
Typic Argiaquolls), Conover loam (Fine-loamy, mixed, active, mesic Udollic
Endoaqualfs), and Newaygo sandy loam (Fine-loamy over sandy or sandy-skeletal,
mixed, fiigid Alfic Haplorthods) (Threlkeld and Feenstra, 1974). In Field B, percent sand

ranged from 50% to 80%, percent clay ranged from 8% to 23%, and percent silt ranged

9O

from 11% to 29% (Chapter Four), and soil series were Brookston loam, Newaygo sandy
loam, and Berville loam (Fine-loamy, mixed, mesic Typic Argiaquolls) (Threlkeld and
Feenstra, 1974).
Statistical Analysis

Descriptive statistics were applied to characterize the population of
developmental stages in the roots, cysts in the soil, and eggs per cyst at each sampling
date for each field. The data was logarithmic transformed to increase symmetry and to
stabilize the variance. Cysts, developmental stages in roots, and eggs per cyst means were
compared between sampling dates within fields, with Fisher’s (protected) LSD test for
means (01= 0.05). The effects of each of the soil texture fractions (sand, clay, silt) on
transformed cysts, eggs per cyst, and developmental stages in the roots were analyzed
with analysis of variance (ANOVA). Only data from samples collected at 31 and 44 DAP
in 1999, and 17 and 19 DAP in 2000 were used for analysis of the relationship between
SCN and soil texture because SCN counts at later samplings were too low for the
analyses to be significant. The regression coefficients were determined on means by
simple linear regression analysis. Significant regression models were tested for
parallelism by soil fi'action across developmental stages, sampling times, and fields.
Models were compared by pairs with a sum of squares reduction test on dummy variables
(a = 0.05) (SAS®, Release 8; SAS Institute, Cary, NC ).

The spatial variability in cyst population was quantified by describing the spatial
dependence in the distribution at each sampling time with empirical omnidirectional

semivariograms. The semivariogram is a structural tool for depicting the spatial

91

dependency in a realization of a mean-constant spatial process Z(s). Here, the classical

Matheron estimator

. __ __l__ _ 2

 

was used (Matheron 1963). The semivariance ’Kh) at a given lag distance 11 is estimated

as one half the average squared difference between all observations at locations st, Sj that
are separated by the lag h. Depending on the data and sampling interval used, the shape
of the experimental semivariogram may take many forms. In general, the semivariance
increases with increasing distance between sample locations, rising to a more or less
constant value (the sill) at a given separation distance called the range of spatial
dependence. Samples separated by distances closer than the range are spatially related.
Those separated by distances greater than the range are no longer spatially autocorrelated.
Semivariances may also increase continuously without showing a defined range and sill,
thus preventing definition of a spatial variance, indicating that the range is greater than
the largest lag (h), or the presence of a trend effect and/or nonstationarity (Webster and
Burguess, 1980). Stationarity means that the random field sampled looks similar
everywhere. A random field is second-order stationary if the mean of the random field is
constant and does not depend on locations, and the covariance between two observations
is only a function of their spatial separation (Schabenberger and Pierce, 2002). Whenever
semivariograms showed nonstationarity, the data were detrended by carrying out a
polynomial least squares regression and semivariogram analysis was performed on the
residuals. Other semivario grams show complete absence of spatial structure, implying
that the value observed at one location carries no information about values at other

locations. Nugget effect (Co) is a discontinuity of the semivariance near the origin (lag

92

h=0). It consists of measurement error variability and/or the sill of a micro-scale spatial
process. The error variance is a measure of repeatability of the data measurements,
whereas micro-scale variance is a measure of variation that occurs at separation distances
less than the smallest sample spacing (Cressie, 1993).

Empirical omnidirectional semivariograms of cysts [logto (cysts 100 cm'3 of soi1+
1)] at planting for both fields were obtained from Avendano et al. (2003, Chapter Three),
and for all other sampling times were calculated for lags ranging from 1 to 30 m, with a
lag tolerance of one half of the lag used (h/2). The minimum number of pairs required for
each lag was 30. The reduced number of samples in the E-W direction in Field A, and the
predominant SW-NE arrangement of the samples in Field B prevented the calculation of
reliable directional semivariograms. The parameters of the semivariograms were
estimated by least squares fitting of theoretical semivariogram models with the Surfer
7.02 software package (Golden Software, 1999).

Kriging is the best linear unbiased prediction of regionalized variables at
unsampled locations using the structural properties of the semivariogram and the sampled
values at observed locations. When a drift or trend (non-stationarity of the mean) existed
within the area of interest, universal kriging was used; otherwise, ordinary kriging was
the method of choice. Universal kriging takes the drift into account provided the form of
the drift and the semivariogram are known (Journel and Huijbregts 1978). The
distributions of cysts, were mapped separately for each sampling time and field with
ordinary or universal kriging predicting values at the nodes of a 1x1 m cell grid using all

the data points in each sample and the parameters fi'om the models fitted to the empirical

93

semivariograms. Maps of cyst distribution at planting in both fields were obtained from

Avendafio et al. (2003, Chapter Three).

RESULTS
SCN Population in the Soil

Cyst density in the soil was relatively low in Field A and Field B at the beginning
of the study (Tables 5.2 and 5.3). In Field A, cyst population density remained low
throughout 1999, decreased slightly overwinter and over the following season, reaching
an average of 7 cysts 100 cm’3 of soil at harvest in 2000 (Table 5.2). In Field B, cyst
population tripled its density 44 DAP, then decreased significantly at 73 DAP and
remained stable until harvest in 1999. Cyst densities declined over the 2000-growing
season but reached almost 47 cysts 100 cm'3 of soil at harvest, density comparable to that
observed at 44 DAP in 1999 (Table 5.3).

The number of eggs per cyst varied between fields, sampling times and years
(Tables 5.2 and 5.3). In Field A, the number of eggs per cyst was lower in 1999 than in
2000 with the lowest number found at 59 DAP in 1999; and the highest at 17 DAP in
2000 (Table 5.2). In Field B, the number of eggs per cyst varied greatly in 1999, but
stayed constant at a moderate level in 2000 (Table 5.3).

SCN Population in Soybean Roots

SCN population in the roots behaved similarly in Fields A and B in 1999, but
differed greatly in 2000 (Tables 5.2 and 5.3). All stages, including a few mature females,
were detected in the roots collected at 31 and 44 DAP from Fields A and B, respectively.

Very few nematodes were found in the roots collected later in the 1999 season.

94

.55er £050 E 530 Mo 00556 05 mesa—50 5:5 03:85 5 S 35.3 005 $2800 0055 ammo 005 030 .«o 0055:. 0w80>< F

.< 20E 80.5 360:8 503855 0.50 0355 805 A18 mo 80 8: 00385335 58m secures—0 .3 5050055 0.03 3&9 m

Age 5055058 wean—«E .055 535 .«o 5055.: 05 50 538:8 0.03 50.95% N

.Am0—580m 003—25 5508 3050—0w E .5 0205 $0 53 50350.5 55 A8380.“

0.555883 mww0 03mm; 8053 50380.5 50935-5on .Avac 00530800 50555 >5:th .535 55 505502 055.8— 033 5.55

55 E5 2585 5:5 258 .AND 0.0—503:. 03500.55 .Eobeug 003 308 50935 50555 5 505550 50mm; 50083—055 20m A
5:30 05 :32? 0.8 See 30%—.8 .850 50805555.: 02:55?—

ee Ame. 085:3 5505 qu 30500583 5 10%—m 5:25.55 banana—ma Hoe 005 0050. 0:85 05 .3 330:5 55:00 0 E53 9802

 

 

 

- - - 55:3.“th 059.2550 n3

condumo undeno .5 _.oao e o obiwmamwA 50 530 Q.

n 3 fled 0 Adamo e mda Ad 5 o prmafimm 0 Neaoe me

e o a menuwd em.m:am._m “30.5555 59515.3 wood— 52. S
- - - - 9 05¢ H v.3 0 2: 0“ Nw 95:53 ocom

e o 0 o c o e o nedhamsv e 5.2 at: cm—

eomdfi _.o 0 mdflo e mdamd e mdamd film“ mum 5 w.m_ «5.: mm

0 Would o Noao 0 ALan 0 egawd 0995.5ng 0 0.2..an em

5 Qaawd 50.35 QM: 5.5.—m ”Sewn 5 w.wma wdA 05.5mav.e_ a Q? «ed— 2.“
- - - - e 555 a 52 e 55 0 e5 555555 555.

55 a :85
55 0” ~08 50935 no _.w m 3550 55 a $550 060 0A: 3550 .35
thMWw/w 52560.5 0558:: 8%.. NA #593 00a mwwm oflmAU mMMHNMm

 

+588 eschew

.88 E 5:5 592 E < 20E E :8 05 5 5:5 588 500.85 E 56005 50523qu508 20m .m.m 035,—.

new

95

.5550 £050 5 mwm0 .8 85:5: 05 55:80 5:5 08:85 5 5 $550 005 85558 855 55.50 8: mww0 .8 85:5: 0w58>< F
.m 585 :85 5080—80 88:55 0.80 2m:_m :85 A505 .8... :8 8: 50885555 :85 855520 .3 508558 083 35.50 a

Age 505505:— 55:58 855 5.555 .8 855: 05 55 5080—80 083 8355 H

.m 035... 5 5058505 55 50585—0 588 809.58 50:55 8 50880 50w58 3888—085 20m 5.

5:5—0 :5 :32? 08 S55 5:858 .5555 5088.88: 055%:—

:o -.Amo ©5558 5505 Om: 5080853 5 .8555 4:08.55 55:555me 8: 08 850— 0855 05 .3 503555 5:200 5 5:3 5:502

 

 

 

 

- .. 0955.33 5 5.355 m3

5 o.mHm._ 5 2.35.: s 5.85.55 9 m.m_Hm.m 505 o. mm H _. mm 50 5. mm H 5. ON 2.

0: fime; 9 EVNHoé 5 EmmHmS 50 m.mH5.o 5 ath 9:. 05 wfim H 93 55

5 deo 5 5.53:; 5 ”$35.52 9 5.55.5 5 5.5m H 03 505 5.3 H ES 3
- - - - 5 5.55 H 5.55 85 5.55 H 5.55 5555.5 5555

50 de50 0 _.OHo 0 _.oHo 5 o 5 5.3 H 552 on .529. H mfim m2

0 wded 0 5.oH~.o 0 NAHmd 50 mAHmd n5 5%: H 052 a m.mm H w.wm o3

a ugHN; a _.~Hm._ 5 m.mHm.m 0 «.1556 5 54m H mdo 0: 5:55 H Twm m5

5 5.5.35.5 5 fimmHS: 0 YNNHENN 5 waHndv 0 5.5m H Wmm 5 Nmm H Qwv 5.5
- - - - 5o .25 H 5.55 o 555 H 5.: 5555.: 555_

555 H :05 .8
555 H 80: :50985 50 _.mm 5880 555 H 9:80 M.50 2: 58:00 nan
MHMM 85:5 0555 529. NH F558 8: mwwm @3550 wmwmmwwm
.780.— :50€Aom 50m

 

.5555 a 555 555_ a m 2...: a =8 25 a 55 38. .5358 a 5885 8525555 :85 205 .55 2.5.5

96

In Field A there was a large number of J 2 and J 3/J4 in the roots 17 DAP in 2000,
but after this sampling nematode counts were practically zero (Table 5.2).

In Field B there was a very large number of 13/14, as well as a relatively high
number of immature females at 19 DAP in 2000 (Table 5.3). Although all stages except
mature females were well represented in the roots, numbers fell at 46 DAP. At 77 DAP, a
new root invasion was detected, with 12 counts comparable to those observed at 19 DAP.
A large number of necrotic tissue marks were observed in the roots fi'om both fields in
2000 (not quantified), even where not many nematodes were found.

SCN Population Dynamics and Soil Texture

The two fields were distinct in regards to the effect of soil texture on SCN
population in the soil and in the roots. Sand, clay and silt proportion in the soil had little
effect on SCN population density or eggs per cyst in Field A (Tables 5.4-6, Figures 5.1-
3). The number of cysts in the soil increased with increasing percentage of sand only at
59 DAP (Table 5.4), and decreased with increasing percentage of clay at the same rate at
59, 86, and 126 DAP in 1999 and at 17 and 147 DAP in 2000 (Table 5.4). Silt had no
effect on cysts in Field A (Table 5.4). The number of eggs per cyst increased at 59 and
126 DAP, and decreased at planting in 1999 with increasing sand percentage (Table 5.5).
Although coefficients of determination were higher for clay, the effect was also only
significant for three sampled times. The rate of decrease in eggs per cyst numbers as clay
percentage in the soil increased was similar at 86 DAP in 1999, and at 76 and 150 DAP
in 2000 in Field A (Table 5.5). The proportion of silt had no effect on eggs per cyst in
Field A (Table 5.5). Only sand and silt proportion affected SCN population in soybean

roots in Field A, and this effect was only significant on immature females in the roots at

97

31 DAP in 1999 (Figures 5.1.c, 5.3.c, Table 5.6). Although the number of immature
females decreased with increasing clay percentage (Figure 5.2.c), the effect was not

statistically significant (Table 5.6).

Table 5.4. Simple linear regression parameters for the effect of soil texture on SCN cyst
population density in the soil in Fields A and B in 1999 and in 2000.

 

 

 

 

 

 

Sampling Soil separate
eventT Sand Clay Silt
DAP Yell 8§ 7 YoI a§ 12 3’01 a§ r2
FieldA1999
Planting NS NS NS
31 NS NS NS
59 -0.56 0.02b 0.21 1.42 -0.04a 0.57 NS
86 NS 1.83 -0.05a 0.69 NS
126 NS 1.89 -0.06a 0.64 NS
FieldA2000
Planting NS NS NS
17 NS 1.35 -0.04a 0.40 NS
45 NS NS NS
73 NS NS NS
147 NS 2 -0.09a 0.75 NS
FieldB 1999
Planting -2.03 0.04a 0.77 2.21 -0.09a 0.76 2.53 -0.08a 0.81
44 -1.23 0.0% 0.61 2.55 -0.06a 0.71 2.98 -0.07a 0.77
73 -1.35 0.04a 0.75 2.35 -0.07a 0.60 2.49 -0.06ab 0.73
100 -1.5 0.04a 0.75 2.42 -0.08a 0.80 2.61 -0.06ab 0.62
125 -O.89 0.03a 0.54 2.14 -0.06a 0.89 2.59 -0.07a 0.66
FieldB2000
Planting -l.58 0.04a 0.76 2.53 -0.09a 0.63 2.76 -0.08a 0.76
19 -0.85 0.03a 0.65 1.77 -0.05a 0.54 2.00 -0.04b 0.50
46 -1.07 0.03a 0.52 1.82 -0.05a 0.40 1.94 -0.04b 0.41
77 NS NS NS
155 -1.69 0.05a 0.67 2.69 -0.09a 0.79 2.83 -0.07a 0.70

 

Regression models significant at the 5% level.

1' Soil samples were collected at the days after planting (DAP) indicated.

1 Yo is the intercept of the regression model fitted [Loglo (cysts 100cm”3 of soil +1)]

§ a is the slope of the regression model fitted. Slopes followed by the same letter within
each soil separate were not significantly different (a = 0.05).

98

Table 5.5. Simple linear regression parameters for the effect of soil texture on the number
of SCN eggs per cyst in Fields A and B in 1999 and in 2000.

 

 

 

 

Sampling Soil separate
event'l’ Sand Clay Silt
DAP YoI a§ F YoI a§ 1'2 YOI a§ Fr
Field A 1999
Planting 1.57 -0.01b 0.18 NS NS
31 NS NS NS
59 -0.91 0.02a 0.17 NS NS
86 NS 2.86 -0.12b 0.48 NS
126 -0.92 0.03a 0.17 NS NS
Field A 2000
Planting NS NS NS
17 NS NS NS
45 NS NS NS
73 NS 2.43 -0.1 1b 0.49 NS
147 NS 2.53 -0.09b 0.53 NS
Field B 1999
Planting NS NS 2.86 -0.05b 0.32
44 2.09 -0.01b 0.26 0.55 0.0% 0.46 0.46 0.03a 0.34
73 NS NS NS
100 0.43 0.023 0.22 NS 2.69 -0.04b 0.41
125 NS NS NS
Field B 2000
Planting -0.l4 0.02a 0.23 2.62 -0.08b 0.32 2.45 -0.05b 0.33
19 NS NS NS
46 0.03 0.02a 0.28 2.2 -0.06b 0.34 NS
77 NS NS NS
155 NS NS NS

 

Regression models significant at the 5% level.
T Soil samples were collected at the days after planting (DAP) indicated.
1 Yo is the intercept of the regression model fitted [Logm (eggs per cyst 100cm'3 of soil

+1)]
§ a is the slope of the regression model fitted. Slopes followed by the same letter within
each soil separate were not significantly different (a = 0.05).

In Field B, the proportion of sand, clay, and silt had a strong effect on SCN cyst
population in the soil. Cyst densities increased linearly with increasing sand percentage,

with a slope similar across sampling times (a = 0.05), and different from the model fitted

99

for Field A (Table 5.4). The effect of clay was also consistent across sampling times, cyst
densities decreased with increasing clay percent, with a slope not significantly different
from the trend fitted for Field A (Table 5.4). Cyst densities were also negatively
correlated with silt, with a slope consistent across sampling times (Table 5.4). The
relationship between soil texture and eggs per cyst varied greatly (Table 5.5). At 44 DAP
eggs per cyst decreased with increasing sand, and increased with increasing clay and silt
percentages, but the slope of the regression lines had the opposite sign of the trends fitted
for the other significant sampling times and of those observed for cysts (Table 5.5). The
number of J 3/J4 stages and immature females in the roots was affected by sand, clay, and
silt (Table 5.6, Figures 5.1.e,f— 3.e, f). The population density of 12s in the root was not
affected by soil texture (Table 5.6, Figures 5.1.d — 3.d). In Field B, The number of J3/J4
and immature females increased with increasing sand percentage at the same rate in 1999
and 2000, and decreased with increasing clay percent in 1999, and increasing silt percent
in 1999 and 2000 (Table 5.6, Figures 5.1.e-f— 3.e-f).
Spatial Analysis of Cyst Population

The structure in the spatial distribution of cysts was highly variable between fields
and in time. Semivariogram models revealed variable degree of spatial structure, ranging
from highly structured (Field A at 31 DAP in 1999) (Table 5.7, Figure 5.6.a) to complete
absence of spatial dependence (Field B at 100 DAP in 1999 and 19 DAP in 2000) (Table
5.7, Figure 5.7 .c, e)). Overall, the degree of spatial structure in cyst distribution was
higher in Field A than in Field B (lower nugget variance to total variance ratio) (Table
5.7). A wave model, also known as hole-effect model, is applicable when the attribute of

interest shows some degree of periodicity in the spatial distribution.

100

Field A Field B

 

 

 

 

2.5 " d 0
§ 2m

 

2.0 i

1.5 *

Zf‘

1.04

0.5 ‘

 

 

0.0 .
2.5 <

 

2.0 ‘
1.5 < o

I
. g

0.5 <

VP/El‘

 

 

0.0 . . . r
2.5 ‘

Log ,0 (Nematodes 29" of root + 1)

samurai emmwuu

 

 

 

 

 

4O 50 60 70 8040 50 60 70 80

Sand %

Figure 5.1. Relationship between sand percentage in the soil and SCN second- (12),
third/fourth-juveniles (J 3/4), and immature females in soybean roots in Field A (A-C) and
Field B (D-F). In Field A, root samples were collected at 31 and 17 days after planting
(DAP), and in Field B at 44 and 19 DAP in 1999 and in 2000, respectively. Parameters
of the regression lines are shown in Table 5.6.

101

Field A Field B

9 1999
2.5‘ A . D O

 

 

 

2.0 .

1.5- ‘ i f

. i i J i 5
‘1’: Jlgéiiiii . lg N glilé

3P

 

1.5. i§§§§ H 1

1.0‘ § §§

0.5 .

VP/Sl‘

 

 

0.0 V f f f ‘ i
2.5 ' c ‘ F

Log 10 (Nematodes 29" of root + 1)

2.0 4

1.5. ,I i
i

0.5—wL—k
0.04 § § 6§§c§

seleule; elniewwl

 

 

 

 

 

Clay %

Figure 5.2. Relationship between clay percentage in the soil and SCN second- (J 2),
third/fourth-juveniles (J 3/4), and immature females in soybean roots in Field A (A-C) and
Field B (D-F). In Field A, root samples were collected at 31 and 17 days after planting
(DAP), and in Field B at 44 and 19 DAP in 1999 and in 2000, respectively. Parameters
of the regression lines are shown in Table 5.6.

102

Field A Field 8

01999
2.5-A ,D 02000

 

 

 

 

 

2.0 <

.o ‘ f
1.51 o . 4 i
1.0+ '{ fig .. . Q}; J
' ssi blot l, ‘ . t. .l

25‘ B

l-O—l
F-O-I—O—l
i——e—i
Zf'

 

2.0 <

1.5‘

214 l—O-l
Wt,
*‘814
.fi 3
171791"

1.0 - .

t-O-l .
O 0
1'01
0
H-1
0 /
' 9—6—1

0.5 i

 

0.0 . . . . . . f

2.5. C . F

Log ,0 (Nematodes 29‘ of root + 1)
g an“ EP‘HH

SGIBUJG} 91111911111"

 

 

 

 

 

Figure 5.3. Relationship between silt percentage in the soil and SCN second- (J 2),
third/fourth-juveniles (J 3/4), and immature females in soybean roots in Field A (A-C) and
Field B (D-F). In Field A, root samples were collected at 31 and 17 days after planting
(DAP), and in Field B at 44 and 19 DAP in 1999 and in 2000, respectively. Parameters
of the regression lines are shown in Table 5.6.

103

.m «ea see 32% .88. «a 82

a «2d a «a a. a «8 .< 25 age boéooea .88 «8 82 a 55 S «a z a «28:8 23 Base. 82 __.
.62 u 3 agate esoeae

8: 0.83 Seaman =8 :08 555 5:2 ofium 05 3 332—3 mono—m 62E E58 comma“: 05 mo one—m 05 mm a a
5+ 88 mo rwm 838805 «$0.: 33m 358 commuekwoc 05 mo 38.—35 05 mm c> a

.N «38. a «2&3 «« «owe 3888.38 28 a.

._o>o_ «\«m 05 3 885:3». £0.58 cams—mom

 

 

 

 

 

Rd «3.? R; mz «No «86 2.? «case «Bass:
33 £8? «3 m2 «.3 «3o :3 «:2
m2 m2 m2 .2
ESIIIIN mega
who «8.? m3” :3 «8.? «3 3o «85 2- «case «53::
«we n.8? c: «3 «8.? m3 «3 «3o 83. :5
m2 m2 m2 2
Follllaa 3%:
m2 m2 «2 «case «saga:
m2 m2 m2 :5
m2 m2 m2 2
Fooollllm < gala
m3 «5.? «E «z 85 £3 86 83:3 usage:
m2 m2 m2 :5
m2 m2 m2 .2
flea < «. Elm
m am Hen m 2 E a. m« us we“... 28
5m 30 25
285mg mom

 

«-3 «mm a 55% 88 a «a 82 a m «a < «25
E mace .8238 E «Ewan—c noun—«ace 20m :0 23x8 :8 mo Soho 05 8% E80883 :ommmopwoc 58:: 295m 66 2%...

104

Certain periodicity was detected in all empirical semivariograms of cyst
distributions in our study. However, because of the large variance at short lags (nugget
variance) wave models could only be fitted at 31 and 45 DAP in Field A (Table 5.7,
Figures 5.4.a,f), and at 125, 46, 77, and 155 DAP in Field B (Table 5.7, Figures 5.5.d, f,
g, h) and other models had to be used instead with poor fit. In some cases, the spatial
dependence in cyst distribution was best described by nesting several models to describe
a short range variability nested within a longer range variability. This was the case at
planting and at 31 DAP in 1999 and at 147 DAP in 2000 in Field A, and at planting in
2000 in Field B (Table 5.7), when a wave model (range < 10 m) was nested within a
spherical or an exponential model (range > 100 m), in addition to the nugget effect. The
range of spatial autocorrelation in Field A decreased from 819 m (planting) to 125 m at
harvest in 1999 and 122m at planting in 2000. However, at 17 DAP in 2000 the range
increased and remained constant at approximately 290 m through out the 2000 growing
season until harvest (Table 5.7). In Field B, there was great variability in the range of
spatial dependence. The range for spherical semivariogram models decreased from 395 m
(44 DAP) to 153 m (73 DAP) in 1999, and increased at planting in 2000 (268 m) (Table
5.7). The range for wave models also decreased in 1999 from 12 m at planting to 8 m at

harvest, but varied greatly in 2000 (Table 5.7).

105

1999

   

 

 

0.151
0.11 0.1
0.05 0.055!
o. - , ,, 310A? 0L .21.! DAP
020406080100120 020406080100120
0.3] 03‘,
!F

025» _- 0.251

   

 

    

 

 

0.21
0.15
0.11
g 0.051 0051'
g 01 59 DAP 01 , “W, 45 DAP
E5 020408080100120 020MM.6000100120
>
E 0.31 0.3
a) C G
(D 0251, 025
I
0.2'_ 02<~—-—..'--— -__.- --’
0.15 0.151
0.11 0.11
0.05_ 0.05,
«L ,A 3 960» 01 73DAP
o 20 40 60 00100120 0 20 40 60 90100120
0.3

0.25
0.21”" " “I“

0.151

 

0.1 1

 

0.05 1 0.051

 

0; _ 126 DAP .._.-.-- p 147 DAP
0 20 40 60 eofl'i'ono 0 20% 1037*?100 120
Lag distance (m)

Figure 5.4. Semivariograms of cyst population density [Log10(cysts 100 cm”3 of soil +1)]
in Field A at (A) 31, (B) 59, (C) 86, and (D) 126 DAP in 1999; and at (E) 17, (F) 45, (G)
73, and (H) 147 DAP in 2000. Black circles indicate omnidirectional empirical
semivariogram, the solid line is the theoretical model fitted by means of least squares,
and the dashed line is the sample variance.

106

 

 

0.35« F

 

 

 

 

 

0.11 01
a) 0.05 0051

3 43.059- ._._. 3.- - .49 DAP.
.59 0 20 40 00 so 100 0 20 40 00 90 100
b

g 0.35%0 03516

E ' .l

0)

(D

 

 

 

 

 

 

 

0L... , -, .1250”, «l _, . . 1550.1)?
0 20 4O 60 80 100 O 20 40 60 00 100
Lag distance (m)

Figure 5.5. Semivariograms of cyst population density [Loglo(cysts 100 cm'3 of soil +1)]
in Field B at (A)44, (B) 73, (C) 100, and (D) 125 DAP in 1999; and at (E) 19, (F) 46, (G)
77 , and (H) 155 DAP in 2000. Black circles indicate omnidirectional empirical
semivariogram, the solid line is the theoretical model fitted by means of least squares,
and the dashed line is the sample variance.

107

 

 

 

m8 «2. 30:23
«.2 8.: 38:80::
«no :5 33:2 30::
«mm 8.: 30:23
«9o «3 03:2 30::
«.w mod «>55
wmd :.o «03:2 30:5
3% 85 EE:&
«3 :6 «09:2 50::
NS 8.: 30:25
«go «3 «.0me «z
«.«E «ed Steam
Rd «3 «03:2 90:5
E: :.o 3:33
We :.o «03:2 30::
5.2 «S 55.88
«to «3 03:2 «8:5
«3. So «>«B
mm: «3. 30:23
«No 85 «03:2 «.52
new :3 3E:&
9m «3. «>«3
omd mod 83:2 @054
E
«05
t «o .3QO =0 moo 58:3 3:22 .35

«:5 E
.29 m:
«:5 a.
.25 2

$352:

.29 8:
«Zn 3

«FE om

Erzn 3

$322:

ooom

coo _

H? 22:

 

38: «a «:« «:2 a m «:« < «22: a «3:35
«35:08 «a 380:8 838% Bob 36:30 core—smog “who 20m mo 22005 nauwotgubom 333.805 05 mo E06883 «an 2an

108

2:29 mats—:3 Ba: :8: 82:28:: :5 8 380:8 82:83:: mo: 88.: 3:238 083 3&0 Er

8:8: .8 8 8882 as: :8 8:8: 8

.3 i6: :0 M.5: 2: $38 Swo— :0m 3832:: 0.83 8882:5383 K

dog—2:839: 38% on: .8858 :5 89:5: 05 .532 on: .8303: 38% mo 88:8 05 :0 8888:: a: m: A:U+UV\:U .3
.83228038 38% mo :88: 2F 8

:8 8:8: 2: m: o F

.88 mam—9:8 :o 5232:? 8288me 9 02: Ewto 85 8 3:88:88: 8 b35888“: : :o «coho Emma: on: m: :0 a
.888me m: 33:8: 88:: :5 «o 88.8 38:53.: 2:. 2:528: 05 no 38.—8:3 83 838:: 8:809:38: 9:: 58882
8.8:? 8:2 38952: 285: 8 So warn—:8 8098.8: 8.83 85 on: 58:08:38 332:: mafiwowazcsm 85855 H
A5 E mm 9 2 88.: megs: 82 8.: 3833:: 8.882.838: 18598 no 8838 :82 «3 825 0.8? $252 +

 

Z 8.: 2:3
8.: 8.: 88:2 8:5 :5 8:
8 8.: 98>:
8.: 2.: 88:2 8:5 :<n R
m: 8.: 28>:
8.: 2.: 88:2 «8:5 :3 :v
: 8.: 88:2 8:5 :3 8
8: 8.: 88:8
8;: 8.: 283
8.: 8.: 88:2 8:5 88:52: 88
8 8.: 283
8.: 8.: 88:2 8:5 :«d 8:
: m _ .: 88:2 «8:5 :3 :2
:82 8.: 888::
8.: 2.: 88:2 8::2 :3 m:
:8 8.: 8:88:
5.: E .: 88:2 8:5 :<o 3
2.2 2.: 2:3
8.: 8.: 88:2 8:5 88:8: 8::

m 22:
8:8: :8 28:.

 

109

Cyst population densities were distributed in more or less defined clusters in Field
A (Figure 5.6) The size and shape of the clusters of higher cyst density in Field A,
changed over time, but the locations of hot spots or high cyst density areas remained the
same. Two relatively large areas located on the first 200 m and between 500 and 700 m
from the south boundary of the area sampled had consistently more cysts at each
sampling time than the rest of the field sampled (Figure 5.6). These areas corresponded
with areas of high sand percentage in the soil (Figure 4.1 .c). There was a trend towards a
better definition of the hot spots boundaries in Field A in 1999 (Table 5.7, Figure 5.6),
whereas in 2000 there was great variability in the spatial patterns, as reflected by the
structure of the semivariogram models (Table 5.7 , Figure 5.6). In the cases were a wave
model was nested within another model with a larger range, cyst distributions appeared in
very small clusters grouped into larger clusters showing two scales of spatial variability
(Figure 5.4.a, b, h, Figure 5.6).

The spatial distribution of cysts graded from high density in the northeast to low
density in the southwest side in Field B (Figure 5.7). As the 1999 season progressed the
area of high cyst density increased in size towards the south side of the field (Figure 5.7).
The lack of spatial structure in cyst population at 100 DAP (Figure 5.5.c) produced a map
showing uniform distribution of cysts throughout the field with a population density
equal to the mean (Figure 5.7). At harvest, the areas of highest cyst density were again
concentrated on the north side of the field (Figure 5.7). At planting in 2000, cyst densities
distribution was similar to that observed the year before, but with high cyst density in a
larger area of the field (Figure 5.7 ). Cyst density decreased somewhat as the season

progressed, but at harvest the soil in the north two thirds of the field had high levels of

110

5m 033. E 850% Pa anacwotazfiom 28395
05 co Egg.“ .2 + :8 was 2: £3 52% 22,2 38%.: 038 .385 .88 a e5 82 a .3865 p29 Managua
5% as 2: a E“ A85 page -88 .3 a OEBEV 9.: mega a < 22m 5 8:355 age 58 20m .3. 2am

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

W.
W7
v
W
W. . .
w. . W
W w
on W... . )
, . w
u fw . .
a. m , (.th
Mm W. a;
_ n W. W
nun“ h. .
. . .. WE... o8
PM

 

 

 

ooom .30 .E mam? u<o E

111

gum 03$. E 53% one mEEwotgwuom Row—Ens

2: .8 E2035 .2 + =8 we". 2: was 23% £32 sauces use... magma .88 e5 82 a .Bsses 3%: mega
5% as 2: 3 Ba A85 .8395 -88 ea 3 amnesia 9383 a m sea a nonsense 33% in. 28 «an 2&3

Erma
0 8m SF SF 8 o 8m 03 8. on o
4. at”. _ .n ..

..

 

o oomomwoow on

:4
W

 

cow 8.. 8w cm

L

  

   

82.6
88:2 05a om.

we 2
9:5...

(m) quN

ach n<n

Soto
8832 23m

.2 3.
2.2.2,.

 

mN —. oo —.
mmm _. n<n

112

cysts while levels remained low on the south side (Figure 5 .7). The spatial pattern of
cysts at 73 DAP in 2000 was somewhat different of that observed at the other sampling
times. At this time, there was more variability along the east-west axis than the north-

south axis of the field (Figure 5.7).

DISCUSSION

The relationship between soil texture and SCN population dynamics in the soil
and in soybean roots was studied as part of a project designed to investigate the potential
for SSM for SCN in Michigan. The two fields selected for this research provided an
interesting scenario since they differed in SCN population density and in soil physical
characteristics

SCN populations in Field A and Field B were active, and infections occurred
through out the season, as indicated by the presence of all developmental stages in the
roots at almost all sampling times since early in the season. The rapid increase in the
number of cysts in the soil and the presence of gravid females in the roots about a month
after planting followed by a decline in SCN population density in roots and soil indicate
that most population increase occurred in the first generation, as it seemed to be the case
in Illinois (Lawn and Noel 1986). The rate of increase in plant-parasitic nematode
populations is often density-dependent and driven largely by the amount of food available
(Seinhorst, 1966). The optimal activity of SCN and, therefore, highest population
increases normally occur during May and June when roots are primarily in or near the
plant row (Alston and Schmitt, 1987; Bonner and Schmitt 1985). When the density of a

population exceeds its food supply, starvation results, and the population decreases until a

113

new equilibrium with the food supply is reached (Seinhorst, 1966). Since root growth
deep into the soil and between rows occurs later, population increase of SCN lags behind
that in the row, even if population density does not exceed its food supply (Alston and
Schmitt, 1987; Bonner and Schmitt 1985). Yen et a1. (1995) attributed a decline in
hatching in August to early dormancy induced in July-August following planting in May-
June. They propose the existence of a primary generation that takes 20 to 40 days from
first hatching at planting (dependent on soil temperature) followed by dormancy as an
obligate condition for eggs retained within cysts of the first generation. Additional,
smaller generations may be produced by non-dormant 12s hatched from eggs produced in
the gelatinous matrix. Early damage to soybean roots caused by an abundance of 12s
early in the season, and consequently reduced root growth may have caused the drastic
population decline in the roots observed in Field A and Field B at mid season, similar to
the observations of Wallace et al. (1995). In addition, the reduction in number of larval
stages in the roots associated with the increase in eggs per cyst observed in August and at
harvest in Field B in 1999 and 2000, and in Field A in 1999 was probably the result of
reduced hatching due to induced early dormancy and a decline in the stimulatory effect of
roots with plant age (Sikora and Noel, 1996; Tefft and Bone, 1995, Yen et al., 1995; Hill
and Schmitt, 1989).

SCN overwinter survival was good based on the similarity in the number of cysts
at harvest in 1999 and at planting in 2000. Root penetration was reported equivalent in
susceptible and resistant soybeans (Acedo et al., 1984; Melton etal., 1986), but final egg
density and cyst production were significantly lower for the resistant cultivar than for the

susceptible cultivar (Wallace et al., 1995). Here, infective juveniles were successfully

114

invaded the SCN-resistant soybean variety planted in 2000, reaching the adult stage in
less than 20 days after planting. Six weeks after planting, however, the number of cysts in
the soil declined about 30% in Field B and 50% in Field A. At 77 DAP a second massive
infection of the roots was observed in Field B. This second infection was also successful
as evidenced by greater number of cysts at harvest than at planting. The numbers of eggs
per cyst at harvest in 2000 were also higher or similar to those recorded at planting the
same year or at harvest in 1999. Apparently, planting a resistant cultivar for one year was
not an effective tactic to reduce SCN levels, particularly in Field B.

Cyst population density was positively correlated with sand, and negatively
correlated with clay and silt proportion in the soil within the ranges of 45% to 80% sand,
8% to 23% clay, and 8% to 43% silt. Differences in soil type, topography and
management history may have contributed to the differences in population dynamics and
correlations observed between our research fields. Wallace et a1. (1995) observed that
SCN population dynamics differed between sites that differed in soil texture and organic
carbon, primarily. SCN can maintain higher equilibrium population density in soils with
higher sand content, and although it may increase to damaging levels in fine textured
soils, the low reproductive rate in soils with high clay content results in a longer time
being necessary for the nematode to attain damaging levels (Koenning and Barker, 1995).
In our study, the lowest cyst population density and eggs per cyst numbers were observed
in soils with low percent of sand; and roots of plants growing in soil with high clay
content (low sand) had more nematodes early in the season, but resulted in low number of
cysts later on. Soybean responds to moisture stress by increasing root biomass (Huck et

al., 1986), which would favor reproduction of SCN by offering more feeding sites

115

(Koenning and Barker, 1995). The stress posed to the plant by the large nematode
population in conjunction with possible water stress during the summer may have
promoted new root growth and a second massive infection of the roots in Field B in 2000.
Slow drainage (low hydraulic conductivity), associated with fine textured soils, often
results in anaerobic conditions persisting for relatively long periods of time (V rain,
1986). Soil oxygen levels may become the limiting factor for the aspects of the
nematode’s life cycle that require aerobic respiration, such as movement, hatch, and
development. The area of Field B characterized by Brookston loam, soil rich in clay and
poor drainage (Chapter Four), was also characterized by reduced number of cysts over
two years, although abundant number of infective juveniles were observed in the roots at
44 DAP in 1999, and 13/J4s at 19 DAP in 2000. This section of the field remained
flooded for several days after rainfalls, while sandier areas of the field dried faster,
therefore it is possible that oxygen availability may have been a factor in SCN population
dynamics.

A collection of semivariogram models can be used to describe a spatial process
(Schabenberger and Pierce, 2002). In nematology, the spherical model seems to be the
most frequently used (Donald et al., 1999, 2001; Workneh et al., 1999; Wyse-Pester et
al., 2002, Webster and Boag, 1992). The short separation distance between samples in the
nested design allowed us to detect a short-range spatial variability in the distribution of
cysts described by a wave or hole effect semivariogram model. The wave model permits
positive and negative autocorrelation as it fluctuates about the sill, with fluctuations
decreasing with increasing lag distance (Cressie, 1993, Schabenberger and Pierce, 2001).

Avendafio et al. (2003, Chapter Three) used the wave model to describe the spatial

116

structure in the distribution of cysts at planting in Field A and Field B, and we showed
here that the periodicity in the spatial structure was maintained over the season. Part of
the variability observed in the semivariogram models shown in Table 7 was because of
the difficulty in modeling the periodicity observed in empirical semivariograms. For
example, the peaks and valleys in the empirical semivariogram at 44 DAP in Field B in
1999 could not be modeled because of the large nugget. A spherical model was fitted
instead to describe the long-range spatial structure (Figure 5.5 .a), whereas a wave model
was adequate to describe the structure of the empirical semivariogram at 46 DAP in 2000
(Figure 5.5.f). The periodicity in the spatial structure was possibly generated by SCN’s
biology, low mobility and slow spread. Under favorable conditions of moisture,
temperature and host availability, infective juveniles hatched from eggs in a cyst infect
nearby roots creating a very small infestation. Over time these small patches increase in
size and merge onto nearby clusters creating the short scale spatial variability observed.
These clusters are in turn arranged in larger clusters, which creates the larger scale spatial
variability.
CONCLUSION

Soil texture had a strong and consistent effect on SCN cyst population density and
spatial distribution in one field, but not in the other. Soils with more than 60% sand
supported higher SCN population density and favor the nematode spread more than finer
soils. Soil texture also had an effect on SCN developmental stages that occur inside the
host, although this relationship seems to involve another factors not accounted for in this
study such as the physiological status of the host. It is not clear from this work whether

the effect of soil texture on the number of eggs per cyst was on fecundity (egg

117

production) or on hatching. This is an interesting question to address in the future under
controlled conditions to advance in the understanding of the effect of soil texture on SCN
population. The stability in the relationship between SCN spatial population dynamics
and soil properties indicates there is potential for delineation of management zones to

reduce the economic loss due to SCN in infested fields.

118

CHAPTER SIX

SPATIO-TEMPORAL ANALYSIS OF SOYBEAN CYST NEMATODE, SOIL
FERTILITY AND PLANT NUTRIENT STATUS.

F elicitas Avendafio, Francis J. Pierce and Haddish Melakeberhan. Manuscript to be
submitted to Nematology.

119

SPATIO-TEMPORAL ANALYSIS OF SOYBEAN CYST NEMATODE, SOIL
FERTILITY AND PLANT NUTRIENT STATUS.

Felicitas Avendafio', Francis J. Pierce2 and Haddish Melakeberhan3

1- Michigan State University. Current address: 1308 Sheldon Dr., Newark, DE 19711,
avendanl @msu.edu

2- Center for Precision Agricultural Systems, Washington State University, 24106 North
Bunn Road, Prosser, WA 99350, fipierce@wsu.edu

3- Michigan State University, Department of Entomology, 243 Natural Science Building,
East Lansing, MI 48824, melakebe@msu.edu

Correspondence should be sent to H. Melakeberhan.

AKNOWLEDGEMENTS

The project is part of a Ph.D. Dissertation of the first author. We thank Chad Holovach,
Nathan Nye, Megan Kierzek, Carolyn Stein, Nathan Cottrell, Bindiya Shah, Dan
Armstrong, Marisol Soto, Carolina Holteuer, Katrina Blakely and Teresa Raslich for
assistance in collecting and processing samples; and Dave Eickholt, Charles Eickholt,
(farmers) and Neil R. Miller (ABC-Agribusiness Consultant) for providing and
maintaining the sites. Financial support for the project was provided by USDA Hatch
Project through the Michigan Agricultural Experimental Station and grants fi'om the
Michigan Soybean Promotion Committee, North Central Soybean Research Program and

United Soybean Board to the last author.

120

SPATIO-TEMPORAL ANALYSIS OF SOYBEAN CYST NEMATODE, SOIL
FERTILITY AND PLANT NUTRIENT STATUS.

ABSTRACT

We studied the relationships between the spatial distribution of Heterodera
glycines Ichinohe, the soybean cyst nematode (SCN), and soil fertility under field
conditions, and between SCN and the nutritional status of the infected soybean. Soil
samples collected following a geostatistical sampling design were analyzed for SCN, soil
texture and soil fertility. Leaf-tissue samples were collected from selected locations for
complete nutrient analysis. Geostatistical analysis was applied in conjunction with
correlation and regression analysis. Soil fertility affected SCN spatial distribution,
especially soil pH. The spatial distribution of SCN was affected by a combination of soil
pH and soil texture consistently over time. SCN population density was also related to Ca
and Mg in the soil and the nutritional status of the infected soybean, with similarities and
differences between fields. The number of SCN cysts recovered at harvest was correlated
with the nutritional status of soybean at mid season in Field B and not in Field A, but the
number of eggs per cyst was not. This work laid the foundation for future research on the

interaction between plant nutrition and SCN population dynamics.

121

Soybean cyst nematode (SCN, Heterodera glycines Ichinohe) is distributed over a
wide range of soybean production zones in the USA (W instead et al., 1955; Wrather et
al., 2001a, b). Variability in soil structure and soil properties has direct and indirect
effects on SCN and other nematodes fitness. Soil texture has a strong effect on SCN cyst
population density and spatial distribution in the soil, a weaker effect on SCN
developmental stages inside soybean roots, and no effect on the number of infective
juveniles in roots (Chapter 5; Todd and Pearson, 1988). Following are some examples of
soil fertility effect on SCN. The relationships between soil fertility and soybean growth in
the presence of SCN have been studied mostly in relation to K levels in the soil with
diverse results depending on experimental conditions and SCN races used (Luedders et
al., 1979; Hanson et al., 1988; Blevins et al., 1995; Howard et al., 1998; Melakeberhan
1999; Smith et al., 2001). Most studies were done under controlled greenhouse conditions
for 30 days or less or in experimental plots by adding fertilizers. Even though K
concentrations varied in the different experiments, SCN population density in the soil
increased when moderate levels of K fertilizer were added, but not when none or high
levels were applied (Luedders et al., 1979; Hanson et al., 1988; Howard et al., 1998;
Smith et al., 2001). A positive linear correlation has been found between SCN population
density and soil pH within the range of 5.5 to 8.4, but it is not clear if there is a direct
relationship between soil pH and SCN population densities or if the effect is indirect and
plant mediated (Tylka et al., 1998; Grau et al., 1999). Melakeberhan et al. (1997) reported
that Pratylenchus penetrans population density was negatively affected while plant
growth rate was increased when cherry rootstocks were subjected to an optimum

fertilization regime as opposed to a nutrient deficient program, and Trudgill (1987) stated

122

that damage by potato cyst nematode (Globodera rostochiensis and G. pallida) was
highest at low fertilization levels. Increasing levels of SCN in the soil affected the
concentration and translocation of nutrients in plant tissue. For example, Mg and Ca
concentrations in roots were increased, whereas P remained unchanged, and Mg
translocation was increased with SCN treatment (Blevins et al., 1995). Melakeberhan
(1997, 1999) discussed a set of hypotheses about the possible roles of soil fertility on
plant growth in the presence of plant-parasitic nematodes.

Understanding the spatio-temporal variability of soil prevailing conditions might
help attempts to site-specific management strategies (Donald et al., 1999; Avendafio et
al., 2001) which, in turn, require that the spatial variability of the attribute of interest is
highly structured to ensure that spatial prediction and corresponding management maps
are accurate (Pierce and Nowak, 1999). Even though the spatial variability of SCN is
poorly structured in general (W orkneh et al., 1999; Donald et al., 1999; Avendafio et al.,
2003 -Chapter Three), the possibility of management zone delineation for SSM of SCN is
supported by evidence that environmental conditions created by the interaction of
weather, soil, landscape, and plant factors assist in the dispersion of eggs, determine SCN
survivability, or limit its growth potential and thereby regulate the spatial dynamics of
SCN (Lehman, 1994; Koenning and Sipes, 1998; Donald et al., 1999; Workneh et al.,
1999; Donald et al., 2001, Avendafio et al., 2003- Chapter Three and Chapter Four).
Moreover, Avendafio et al., (2003- Chapters Three, Four, and Five; Donald et al., 2001)
found areas within fields with repeated occurrences of non-detectable or low densities of
SCN and hot spots or areas of high density, consistently correlated with sand and clay

percent in the soil composition over time.

123

The purpose of this work was to answer the question does soil fertility affect SCN
distribution and root infection, and is this reflected in tissue analysis? Specific objectives
were: i) to characterize soil fertility variability and its relationship with soil texture and
tissue analysis in two fields of known SCN infection in Michigan; ii) to analyze the

relation of SCN in the soil and in soybean roots with soil fertility and tissue analysis.

MATERIALS AND METHODS
Experimental Sites and Site Management

The study was conducted in 1999 and 2000 on two fields (Field A and Field B) in
Shiawassee County, MI. Field A consisted of 24 ha, was managed under no-tillage since
1996, and was planted to corn prior to this study in 1998. Percent sand ranged from 45%
to 74%, percent clay ranged fiom 1% to 21%, and percent silt ranged from 8% to 43%
within Field A (Chapter Four). Field B was comprised of 13 ha, was conventionally tilled
after wheat in 1998, and was managed under no-tillage thereafter. Percent sand ranged
from 50% to 80%, percent clay ranged from 8% to 23%, and percent silt ranged fiom
11% to 29% within Field B (Chapter Four). An SCN -susceptible soybean variety
(Asgrow 1901), and an SCN-resistant variety (Asgrow 2201), both Roundup-Ready, were
grown in both fields in 1999 and in 2000, respectively. Soybean was planted in 19-cm
rows at a rate of 519 000 viable seeds ha'1 in 1999, and 494 000 viable seeds ha'1 in 2000.
Fields A and B were planted 5/22/99 and 5/16/99, respectively, and 6/9/00. Weed control
was maintained using Roundup at the recommended rate. There was one preplant

application in Field A in 1999 and 2000, one application postemergence in 1999, and a

124

midseason application in Aug. 2000. In Field B there was one preplant application in
2000, one postemergence in 1999 and a mid season application in Aug. 2000.
Sampling Design, Collection of Samples and Soil Analysis

The spatial sampling for soil samples consisted of a nested survey sampling
design with distances reduced by geometric progression (adapted from Webster and
Boag, 1992) applied within 8 and 5.25 ha in the center of Fields and B, respectively, as
described in Chapter Three. Single-core soil samples were collected using an 8 cm
diameter by 23 cm deep bucket auger (Riverside Augers, Eijkelkamp, Giesbeek, The
Netherlands) from 160 and 110 locations in Field A and Field B, respectively. Each
sample collected before planting in 1999 was thoroughly mixed and then divided in three
sub-samples. One sub-sample was used for SCN analysis (Chapters Three, Four, and
Five), the second was used for texture analysis (Chapter Four), and the third portion of
the sample was analyzed for pH, P, K, Ca, and Mg (Soil and Plant Analysis Laboratory,
Michigan State University).

SCN analysis

Cysts in the soil, eggs per cyst, and developmental stages in soybean roots were
quantified from soil and root samples collected at planting, at harvest, and at the dates
indicated in Table 5.1 (Chapters Three, Four, and Five). The maps of SCN spatial
distribution in Field A and Field B mapped in Chapter Four were used in this work to

relate SCN spatial distribution to soil fertility.

125

Soybean Leaf Tissue Nutrient Analysis.

A leaf sample consisted of the upper-most fully developed trifoliate from 20
plants randomly chosen from selected clusters including up to four neighboring locations
of the nested survey sampling design (Figure 6.1). In 1999, one set of 21 leaf samples
was collected from Field B 61 days after planting (DAP), approximately two weeks after
plants were in full bloom (R2 stage) (Ritchie and Benson, 1994). In 2000, two sets of 24
samples were collected from Field A and two sets of 20 samples from Field B. The first
set of samples was collected at 40 DAP, when only occasional flower buds were visible
(V9 stage), and the second set was collected at 81 DAP, when most plants had 2 cm-long
pods in the lower 2 or 3 nodes (R3 stage) (Ritchie and Benson, 1994). Sampling times
were designed to cover the vegetative and reproductive growth phases of the plant.
Leaves were collected in paper bags and dried at 80°C for 24 hours before the analyses.
Tissue concentrations of N, P, K, Ca, Mg, Fe, Zn, Mn, Cu, and B were analyzed by the
Plant and Soil Analysis Laboratory at Michigan State University in 1999, and by A & L
Great Lakes Laboratory, Fort Wayne, Indiana in 2000.

The one to four SCN and soil fertility observations comprised within each cluster
sampled for tissue analysis were averaged, and the mean was associated with each tissue
nutrient value for the corresponding cluster. Each tissue analysis observation was also
associated with a kriging-predicted value for the proportion of sand, clay, and silt
matched by location.

Statistical Analysis
Descriptive statistics were applied to characterize soil pH, soil fertility, and tissue

nutrients, the results are shown as boxplots. Pearson’s simple linear correlation

126

coefficients were calculated for soil fertility and soil texture; soil fertility and tissue
analysis; SCN population density in the soil and in the roots with soil pH and nutrients;
and SCN population density (The SAS System Release 8; SAS Institute, Cary, NC).

Geostatistical tools were applied to quantify the spatial variability in the
distribution of soil pH and soil nutrients. Omnidirectional empirical semivariograms
(Matheron, 1963) were calculated for lags ranging from 4 to 40 m (h), with a lag
tolerance of h/2. The minimum number of pairs required for each lag was 30. The
parameters of theoretical semivariogram models fit to the empirical semivariogram were
estimated by (nonlinear) least squares. The spatial distribution of soil pH and soil
nutrients were mapped by ordinary or universal kriging. Interpolated values were
predicted at the nodes of a 1 x 1 m grid using the structural properties of the estimated
theoretical semivariogram and the sampled values at observed locations. Details on the
geostatistical analysis were described by Avendafio et al., (2003- Chapter Three).

The cross-correlogram is another geostatistical tool used to describe the joint
variability or spatial continuity between measurements of different attributes or of the
same attribute measured at different times. The cross-correlation between two attributes
at the same location (zero lag distance) equals the linear correlation coefficient for those
two attributes. The cross-correlation function given by Goovaerts (1997) was used here to
calculate cross-correlograms for logarithmic transformed cysts and nematodes in roots in
1999 and 2000 with pH, Ca, and Mg in the soil. Only the data points from locations

sampled for both attributes were used for this analysis.

127

4779650 : 1,,
/

Northing (m)

' ”—fi

737700

Field A

__ ___..-_._._ ——_._. 1.— ”—\

 

 

 

. l
§
C9 5] {-‘fl’ ##fl)
1. " ‘1
5 l
'0' ‘
5.4) ll Field B
to _. l ’ I W— "fry” H71
, W ’\ 47801001. 1
. . 9
or? 1" t __ _M j :9
1 o
‘ " dJ / 4700000111 °° 0
q) \‘ l .o . Q s.
-' .3 j .1 ., “0 W
." . 4779900 1 ' ..
a I .- -’ 9
fifi / ”I CU '.
’ l
W. .‘x _T—-r'."_::::! _, ._.W_r1_ ___ 477m«_1\_::-:1:: :‘1'7—1‘1.'_‘L:T:..—T-~- '7" W 1 1
737900 739105 739268 739430
Easting (m)

Figure 6.1. Field A and Field B, location of sampling sites. White irregular areas indicate
where soybean leaf samples were collected in 1999 from Field B, and in 2000 from
Fields A and B. Each circle includes from one to four sites from the nested survey

sampling design where soil samples were previously collected (black circles)(Chapter

Three).

RESULTS
Soil Analysis

Soil pH in Field A was lower than in Field B (Figure 6.2), with a highly structure

spatial distribution within a range of 71 m (Table 6.1). Soil pH lower than 6.5 or higher

than 7.5 appeared in medium to small size clusters through out the area sampled in Field

A (Figure 6.3). The absence of nugget in the empirical semivariogram of pH in Field B

indicated a highly structured spatial distribution (Table 6.1). Soil pH was higher on the

north side of Field B than on the south side. The long range of spatial autocorrelation

(106 m) showed the variability in soil pH levels in large clusters in this field (Figure 6.4).

128

 

 

 

 

 

     

 

 

 

 

 

 

 

 

 

 

 

pH Phosphorous Calcium Magnesium
9 350 000 7000
b a a a a a
8‘ 3001 . 500 6000-
W 250 gm.
. . as w l ‘°°
7 I. a g 2m m<
~ 3: 300<
6, ‘ 150‘ 3000.
200«
100+ 2000‘
5-4
50« 100‘ 1000‘
O
4 v 0 o ' o
A B A a A B

   

Field
Figure 6.2. Box-plots of pH and P, K, Ca, and Mg quantified in soil samples collected at
planting in 1999 from Field A and Field B. Means (dotted line) with the same lower case
letter were not significantly different between fields (protected LSD, 5% significance
level). Medians are indicated with a full line. Whiskers represent the 5th and 95th
percentiles; outliers are indicated with black circles.

Phosphorus concentration in the soil in Field A was lower and with greater
variability than in Field B (Figure 6.2). The spatial distribution of P concentration was
moderately structured in field A, with a range of autocorrelation of 317 m (Table 6.1).
The contour map showed rather uniform distribution of P levels with higher
concentration in about one third of Field A (Figure 6.3). Phosphorus distribution in Field
B was highly structured within a range of 46 m (Table 6.1). Small patches of high P
concentration were identified scattered through out the field (Figure 6.4).

Potassium concentration in the soil was similar in Field A and Field B (Figure
6.2), and moderately structured spatially within a range of 81 m in Field A and 52 m in
Field B (Table 6.1 ). Patches of below average K concentration were distributed
throughout Field A in bands oriented more or less SW-NE (Figure 6.3). In Field B, two

rather large areas of low K concentration extended from the north side of the field to

almost the south boundary of the area sampled (Figure 6.4).

129

Table 6.1. Parameters of the theoretical semivariogram models of soil pH, P, K, Ca, and
Mg in Fields A and B before planting in 1999.:

 

 

DriftI Model function Co§ C fl Range # C0 Tl'
C+C0
Field A m

pH None Nugget 0.1
Spherical 0.34 71 0.23

P None Nugget 1494
Exponential 1213 317 0.55

K None Nugget 2380
Spherical 3420 81 0.41

Ca Linear Nugget 33700
Exponential 350000 106 0.09

Mg None Nugget 402
Exponential 12900 194 0.03

Field B

pH None Nugget ~ 0
Spherical 0.19 106 0

P None Nugget 13.7
Spherical 855 46 0.01

K None Nugget 2810
Exponential 1900 52 0.60

Ca Linear Nugget 31300
Spherical 405000 292 0.07

Mg None Nugget 7440
Spherical 44100 202 0.14

 

'1' Models were fitted by least squares based on empirical senrivariograms calculated for
lags ranging from 4 to 40 m (h), with a lag tolerance of h/2. The minimum number of
pairs required for each lag was 30.

I Whenever semivario grams showed nonstationarity, the data were detrended carrying
out a simple polynomial least squares regression and semivariogram analysis was
performed on the residuals. The polynomial order of the trend is indicated when a drift or
trend was removed.

§ Co is the nugget effect or a discontinuity in semivariance at the origin due to microscale
variability or sampling error.

11 C is the partial sill defined for spherical models.

# Observations that are spatially separated by more than the range are uncorrelated.

fl' CO/(C+Co) is an indicator of the degree of spatial structure, the lower the number the
stronger the spatial autocorrelation.

130

3900
3400

 

2400

19m

 

 

 

 

141]]

 

Figure 6.3. Spatial distribution of pH, and concentration of P, K, Ca, and Mg in the soil
as interpolated by kriging fi'om samples collected before planting in 1999 in Field A at
the locations indicated with black circles.

Calcium concentration in the soil was similar in Field A and Field B (Figure 6.2).
The spatial distribution of Ca was highly structured in both fields, but the range of spatial
autocorrelation was larger in Field B (Table 6.1). Calcium concentration in the soil was
higher in the south side of both fields, decreasing gradually towards the north (Figures
6.3 and 6.4). Magnesium concentration was lower in Field A than in Field B, with much
lower minimum and maximum values in Field A (Figure 6.3). The spatial distribution of
Mg was highly structured, with a range of about 200 m in both fields (Table 6.1).
Magnesium concentration was lower on the north side of Field A (Figure 6.3). High
concentration of Mg was found in a band about 100 m wide stretching from the east side

of the Field B towards the center (Figure 6.4).

131

 

East (m)

Figure 6.4. Spatial distribution of pH, and concentration of P, K, Ca, and Mg in the soil
as interpolated by kriging from samples collected before planting in 1999 in Field B at
the locations indicated with black circles.

The following soil fertility attributes were related to texture in Field A and in
Field B. Soil pH was correlated with sand (r = 0.58), clay (r = -0.48) and silt (r = -0.63)
percentages in the soil in Field B exclusively (a = 0.001). Phosphorus concentration was
not correlated with soil texture in either field. Potassium and soil texture were poorly
correlated in Field A [r (sand) = -0.25, r (clay) = 0.25, r (silt) = 0.23, a = 0.05], and only
significantly correlated with clay percentage in Field B (r = - 0.17, a = 0.05). Calcium
however, was strongly correlated with soil texture in both fields, especially with clay in
Field B [Field A: r (sand) = -0.48, r (clay) = 0.26, r (silt) = 0.42. Field B: r (sand) = -0.75,

r (clay) = 0.80, r (silt) = 0.63, a = 0.001]. Magnesium was more strongly correlated with

132

soil texture in Field A [r (sand) = -0.69, r (clay) = 0.51, r (silt) = 0.54, (F 0.001] than in
Field B [r (sand) = -0.22, r (clay) = 0.31, r (silt) = ns, (1 = 0.05].
Tissue analysis

The appearance of soybean plants in Field A was uniform throughout the field in
1999 and 2000. The canopy was completely closed at V9, and plants were homogeneous
in size and color at V9 and R3. Abundant number of large nodules (not quantified), pink
inside when dissected was observed on roots at V9. Soybean plants at V9 in Field B were
green and homogeneous in size in approximately the south half of the field. On the north
side, green, fully-grown plants were alternated with large patches of stunted plants with
fewer and chlorotic leaves, many of which had necrotic edges. Green plants had
abundant, large nodules pink inside; whereas, yellow plants had reduced number of
nodules, small in size and most of them brown or yellow inside when dissected.
Numerous cysts were easily observed with the naked eye on roots fiom yellow plants. At
R3, differences in plant appearance were accentuated in Field B.

Descriptive statistics of macronutrients and micronutrients concentration in leaf
tissue are shown as boxplots (Figures 6.5 and 6.6). In Field A, all nutrients were within
soybean sufficiency ranges (Vitosh et al., 1997). The maximum values of K, Fe, Zn, and
Cu concentration in leaf tissue at V9 were above normal levels, without reaching toxic
concentrations. At V9, N, Ca, Mg, Fe, and Cu concentrations were higher than at R3,
when the variability was greater. The concentration of B in leaf tissue was lower at V9
than at R3, whereas P, K, S, Zn, and Mn concentrations did not differ between samplings
(Figure 6.5). In Field B, Cu concentration in leaf tissue samples collected at R2 stage in

1999 was below the sufficiency limit and the minimum values of a few other elements

133

8288 :83 .23

Bass 2. 858 .8588 53 as an as .8838“ 8833 .8: a... s a? Bass 2.. 8.82 .225. 8588s. s.

m .qu 38883 inflame .m—ucaomfiwmm 8: 0.53 SE: 088 05 55 2:32 .828— 88 .532 53, 333?: 0.3 20¢ :80 £55
88a means... Beams s 85. 883 £88 05 525 868%. Eugen .m as. < 83 as... 88 a 2 as a> s e8 .m
2on use 82 a 2 s 88:8 saga... 25% 88. 8&8 a g 88988 m as a: so .2 .m .28 5288 .3 2%:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

39m 586w
mm a> mm mm m> mm mm a> am am a> mm mm m> «m
b p . mF.° p P p p L . b . p F F b o.”
o .3 . a . .
% . 8.0 m 0 H . N O. NO 0 O .. m5
0 . .9 . 0.? W m a .
B . mm o E o . no a . o e
a o . 8.0 . NF 0 a . .1 mm . mé
:1- m I o a a a . . w . _
. . v o i... H .
_. mm o .- ¢.—. . r o m
. a a A. . a.
a a . .
. 9.0 a a a . md a a a - m m
m; . 3 o
93 m . 9o
11 m3 . . No A a 1. ed
a. ad .
a a and W W to o m a a No a a an
o . o
. r o F . .
mmo A. a so D . m a no a 3.
M 86 F... m... m; 3 . m...
3 ad e WW 1W .1. .
H and . B. ._ .5 so on
O W. V; v L U
. o; a m .
. o... o . 3 . . no m m
. O
de Nw Illll. m 0.0 0%
SE5 828:3: E228 Eammfloa maouofimoza 8852

134

83.8 x83 55 @8865 8a 82:3

55:88 58 23 em 05 “Egg FEE? .8: =e F 53 8.865 as 23:52 .232 888% .x. m d8 383.3
Bogota. >=8miwmm “on 0.63 .832 088 05 55 882 .88_ 88 832 55 @2865 08 20m 88 55m? 88: mam—9:8
HESEE 3 $3: @883 888 05 macaw 888mg E8555 .m 98 < 3.0% 88m ooom 5 mm 98 a> «a 98 .m 23m

88m 33 5 NM 3 880:8 81:58 2.8: m8_ 5238 S 358 2538888 m 98 .30 .52 .:N an no muoakom 66 «Emma

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

owflm 8&5
mm a) mm mm m> NE mm a> NE mmgmm mm m> mm
a. a a S o a a a 2 IAIdldlé
o A a a o o .8 o ._ .
.8 .m m .8 .8m
0 m m . _. . L o n
B a a .8 W . f m w m a .8
d a! a 0 a... nu. or .. _ m .0? In .8?
m 8 a m _. .3
. F O
F H mw o . 80
g . . . r00 tom
.0m 0 Few o o
O \
o \8 .— a a 8 .8m
8 on
d
d
w
or F F F o .lmllmllrnle F F . or L F . o
0
cm m a 3. om a a om + .H 8a
a A a. a D «E. on o E
I m .... c on r
A .3. mum on .1... or _L L ow my. 0 00¢
M .. 3 _ . ....
.mw M“ -F 8 :4... HM... mw . L 0m 0 f.» O? A a
F F o o E. u a _ o8
r . r 8 . % om
. om \om . F. 2
. \om . . 8 08
.8

:98 L830 mflmcmacmz oEN , :2.

135

were also below sufficiency limits (P, Zn, and Mn). All other nutrients were within
normal ranges at this growth stage (V itosh et al., 1997). Iron concentration in the leaves
at the V9 stage in 2000 was high, whereas in a few samples, Zn, Mn, Cu, and B
concentrations were below sufficiency limits. At the R3 stage in 2000, N, P, K, and S, as
well as Zn, Mn, and Cu concentrations were below sufficiency limits in some of the
samples. Mean nutrient concentrations at R2 in 1999 were equal to or higher than at R3
in 2000, and equal to or lower than at V9, with the exception of B. Manganese was the
only element which concentration remained unchanged at different sampling times
(Figure 6.6).

Some of the nutrients in tissue from Field A were correlated with soil attributes
(Table 6.2). Potassium, Zn, and Mn were negatively correlated with pH, Ca, and Mg in
the soil; Mn was also negatively correlated with P. Calcium and Fe in tissue were not
correlated with soil fertility. Correlations for other nutrients in tissue varied with the
growth stage of the plant. Correlations between tissue nutrients and soil fertility varied by
year and by plant growth stage in Field B (Table 6.3). In 2000, Mg and pH were the soil
attributes correlated with the most tissue nutrients. High concentration of Mg in the soil
was related to high levels of P, Zn, and B in leaves at V9, and low levels of the same
elements at R3. Also, high pH was related to low levels of P, Zn, and B at V9, but high
levels at R3. The relation of S, Mn, and Cu to soil pH and Mg at V9 remained similar at
R3. Other nutrients were also related to soil fertility in Field B (Table 6.3).

Soil texture and tissue analysis were strongly correlated in Field B in 2000.
Correlations were weaker and only with a few tissue nutrients in Field A (T able6.3). Off

the tissue nutrients at R2 in 1999, only Mn was correlated with soil texture in Field B.

136

23:89.2 .35. £332; 8d Ba .85 a: a anamawa : ._..

 

 

 

 

 

 

137

m2 two- .93- m2 m2 mz m2 m2 3%.? m2 m2 .5
m2 m2 m2 m2 m2 mz m2 m2 m2 m2 m2 Eu
m2 m2 .1de m2 mz m2 m2 m2 2:2 m2 m2 2am
m2 and- :85- in? m2 m2 .23 m2 :Ed- .53. :35 32
m2 at? 2:2- 1.3.? m2 m2 m2 m2 :59 .85- $2 8
m2 m2 m2 m2 m2 m2 m2 m2 m2 m2 m2 2
m2 m2 ease- m2 m2 :«2 m2 m2 m2 m2 $3 a
:85- 1.3.? m2 rid- m2 m2 m2 m2 inc- m2 m2 :a
88 2 <25
m2 m2 m2 m2 m2 m2 m2 m2 m2 m2 m2 :5
m2 mz *3. Led- m2 mz .33 m2 m2 m2 m2 30
m2 m2 :3 m2 m2 m2 m2 m2 .36 m2 m2 gm
rag mz :Ed- :85- m2 m2 3de m2 1.5.? m2 m2 m2
.156 m2 rid- 2&3- m2 m2 :28 m2 1.2.6- m2 :2. 8
m2 m2 m2 m2 m2 E3- m2 m2 m2 m2 m2 m
m2 m2 12- m2 m2 tone m2 m2 1N3- m2 m2 a
m2 m2 38d- :03- m2 m2 :86 m2 :85- m2 m2 :a
Semis, < 25
Sam c\c
wt

m so =2 3 on m .2 8 M a 2 eggs as
03mm: buo— Swonxom E 355:2 bar—om mom

 

.< 205 E ooom E mm new o> B 355:: can: firs
3-2 3: $553888 :8 can in :8 Jam EB >20 .33 “583 com ficomommooo cog—oboe Maegan .N6 2an

 

 

 

 

 

 

 

138

......wh6 .iEuo 3.3.6- 1.656 .1. 36 m2 3.2.6- m2 m2 m2 1.2.6 Em
..:..hw6 3.006 51.06- .3656 1.3.6 m2 .Ibw6- m2 m2 m2 .1. :6 >20
:35- ..555- .355 $555- 1555. m2 $3.5 m7. m7. 52 .1555- 55m
.35 57. .555 .555 m2 :35- .555 m2 m2 .555 m2 53.
m2 52 57. m2 m2 m2 m2 .555 52 m2 m2 8
.555 :35- m2 m2 mz .555- m2 m2 m2 m7. .555 o.
.555- mz .555- m2 m2 mz m2 m2 m2 .2555- .555 5
.355- ..555 £555- ..555- mz :85 52 .35- mz £555- mz :5
5555 5> m 25...
57. 57. .555- 52 m7. m7. m2 m2 mz mz mz am
52 .52 .85- 52 m2 m7. m7. m7. m7. .52 m7. 556
m2 m2 .555 m2 m2 m7. m2 m2 52 m7. mz 555
2.5 m2 m7. m7. m7. m2 m7. m2 m7. 52 ms.
555 555 52 555 m2 5.7. .55. m7. m2 .52 m7. 8
m7. mz mz mz 555- 5.2 $5- 555- .55 m2 m2 5.
.52 57. m7. mz m2 m7. m2 m2 mz .555 52 5
555- 52 mz 85- m2 m7. .555 .555 .85- 52 .555- :5
555. .5. m 25...
Ema .x.
m
m .6 as. 5N 5.. m s. 5 v. 5 z 235. 5....
0:53 .58— :5358 5 355.. Z bat-Em mom

 

.m 20.... E ooom 5 mm 9.5 o> .5 2.5 .33 5 NM .5 3.55::
03mm: 55 A75: wva macofiscobma =8 98 .mm :8 5:5 2.5 $30 6:55 Eco-5m .8 8.56508 cog—oboe whom-80¢ .m6 035-.-

52585.2 .26. b.5535... .55 5.... .35 o... .5 5.85.5.5 .. ..

 

«$56
3&6
:Nw6-

3&66-
m2
m2
m2

1.656

.Ifim6-
1.606
the-6

m2
m2
m2
m2
3.36

.Ibm6-
mZ
*Su6

*mv6
m2
*av6
...6m6..
sit-6-

1.36
.I.:.6
1.656-

...6m6-
m2
m2
m2

*mv6

from-6.-
*ov6-
*mm6

1.696-
m2
m2
m2

*cv6

1.5.6-
1.596-
:66

flux-6-
m2

m2

m2
.3656

1.56-
1.56-
......aw6

m2
m2
m2
{3.6
m2

 

coon mm m 205

m2

m2
m2

1.606..
*Vm6
m2

m2

m2

m2
m2

m2

tux-06.-
m2

m2

m2
1.656

.3366.-
3.2.6..
3.3.6

1.3.6.-
m2
m2
m2

3.606

*mm6-
.3356.-
3.006

m2
m2
£1.06
*ov6
m2

558$ 55 2.3

Em
>30
69%

ms.
50
v.
nu
mm

139

SCN Population and Soil Fertility

Nematode population density was greater in Field B than in Field A. While cyst
density did not change much over the duration of the study in Field A, in Field B there
was great variability, with the number of cysts in the soil at harvest greater than at
planting in 1999 and 2000 (Figure 5.6 and 5.7). Correlation coefficients were low for cyst
density in relation to soil pH in both fields (Table 6.4). The spatial distribution of cysts at
planting, at harvest, and in June was cross-correlated with soil pH up to a distance of 60 -
70 m in Field A and up to 120-130 m in Field B (Figure 6.7). Areas of higher cyst density
were consistently observed at the locations were soil pH was also higher within the area

sampled in both fields (Figure 5.6 and Figure 6.3).

Table 6.4. Pearson’s linear correlation coefficients for SCN population density in the soil
or in soybean roots in relation to soil pH, Ca, and Mg concentration.

 

 

 

 

 

 

 

 

 

Field A 1999 Field B 1999
Sample pH Ca Mg pH Ca Mg
. Kg/ha Kg/ha
Planting SoilT 0.32" 0.24“ NS 0.42" -0.61** NS
June: Soil 0.49“ 0.46" 0.25" 0.55" -O.44** NS
Roots§ 0.22" 0.19* NS NS -0.24* NS
Harvesfi] Soil 0.45“ 0.27" NS 0.29“ -O.55** NS
Field A 2000 Field B 2000
Planting Soil 0.24M O. l 7* NS 0.48" -O.53 ** NS

June Soil 0.20” NS -O.18* 0.41" -O.32** NS
Roots 0.38“ 0.34" 0.27M 0.22* 0.19* NS

Harvest Soil 0.32" NS -O.17* 0.29" -O.66** -O.21*
*, ** Significant at the 0.05, and 0.01 probability level, respectively.
1' nglo (cysts lOOcm'3 of soil + 1). Cysts were extracted from the soil by elutriation and
sugar flotation.
1 June samples were collected at 31 and 44 DAP in 1999 and at 17 and 19 DAP in 2000
from Field A and Field B, respectively.
§ Logm (nematodes in 2g of root + 1). SCN developmental stages counted in 2 g of
stained soybean roots.
1[ Harvest samples were collected at 125 and 126 DAP in 1999 and at 147 and 155 DAP
in 2000, from Field A and Field B, respectively.

140

Cysts in the soil had a much larger range of cross-correlation than SCN in the
roots (Figure 6.7). SCN population in the root in June was poorly correlated with pH in
Field A and practically uncorrelated with pH in field B (Table 6.4, Figure 6.7). Cyst
population density in the soil was correlated with P only at planting in 1999 in Field A
with a very low correlation coefficient (r = 0.16, a = 0.05). SCN population density was
not correlated with K concentration in the soil. Cyst population was positively correlated
with Ca in Field A, with low correlation coefficients and cross-correlations over a very
short separation distance between samples (Table 6.4, Figure 6.4). Cysts were more
strongly correlated with Ca in Field B, but correlation coefficients were negative in this
field (Table 6.4). Calcium and cysts were negatively cross-correlated up to a separation
distance of about 110 m, consistently over time. Beyond that range, samples became
positively cross-correlated (Figure 6.8). Cross correlation between SCN in the roots and
Ca was poor in both fields (Figure 6.8). SCN population density was correlated with Mg
in Field A only. Cyst population density at 31DAP in 1999 and nematodes in the roots at
17 DAP in 2000 were positively correlated with Mg (r = 0.25 and r = 0.27, respectively,
a = 0.01), whereas cyst population density at 17 DAP and at harvest in 2000 were
negatively correlated with Mg (r = -0.18 and r = -0.l7, respectively, a = 0.05). Cross-
correlograms of cysts and Mg showed low correlation up to about 100 m before

becoming zero (Figure 6.9).

141

Field A Field B

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.0
0 Planting 99 0 Planting 99
o 5‘ 0 Planting 00 o. o Planting 00
0.0 1 .
\tst:
-o.5 +
1.0 - - . . -
0310AP99 0440AP99
017DAP00 019DAP00
0.5 r 0‘
N
\V ~
0.0 1 13‘
c -0.5 4
.9
‘6
g . 5 . 5 . v
8 031DAP99 O44DAP99
g 05 017DAP00 0190AP00
5 ‘OR‘K °'-'°\
0'0 1 W
-05 ,
Boots Boots
1.0 . . ' r 1 '
0 Harvest 99 0 Harvest 99
05, 0 Harvest 00 54.0,-.. 0 Harvest 00
O.\ \
' ‘ A- _ ‘o— “
0.0 ~ *V— r .r
V V W
0.5 J .
-1.0 . - - . . .
0 50 100 150 O 50 100 150
Lag distance (m)

Figure 6.7. Cross-correlograms of soil pH and SCN population density in Field A and
Field B. Lag distance is the separation distance between samples. SCN population
density was determined in soil samples, or in root samples where indicated.

142

 

1.0

0.5 t

0.0 . ............

-o.s J

    

Field A Field 3
. Planti 99 ' Planting 99
o Planti% 00 ° Plenum 0°

4

 

1.0

'7

 

 

 

0440AP99

 

 

 

 

 

 

 

 

 

O
,5 $3223 No...-
. j 1
o- o ’p“ I»
00 , \x‘ ’0' rm , r ‘0
V I
M
-0.5 . -
C
.9
E 1'0 T f T 44 DAP 99
‘9 . 31 DAP 99 '
§ 0.5, 017DAP00 °19°AP°°
g O--0 ’p“~°~
.05 <
m . , . .
0 Harvest 99 ' Harvest 99
0 Harvest 00 ° Harvest 00

0.5 r

0.0 1

-0.5 .

 

 

 

 

’3‘

e’ ‘0
I

 

 

50 100 150

Lag distance (m)

Figure 6.8. Cross-correlograms of soil Ca and SCN population density in Field A and
Field B. Lag distance is the separation distance between samples. SCN population
density was determined in soil samples, or in root samples where indicated.

143

 

 

 

 

0 Soil, 3lDAP 1999
A Soil, l7DAP 2000
A Roots, l7DAP 2000

Cross-correlation

 

 

 

 

_,_o - Soil, harvest 2000
0 so 100 150
Lag distance (m)

Figure 6.9. Cross-correlograms of Mg concentration in the soil and SCN population
density in Field A in the cases where correlation analysis was significant. Cyst population
density was quantified in the soil at 31 DAP in 1999, and at 17 DAP and harvest in 2000.
SCN population in the roots was quantified at 17 DAP in 2000. Lag distance is the
separation distance between samples.
SCN Population and Tissue Analysis

SCN cyst population densities at planting and at 17 DAP were not related to
nutrient levels in tissue at V9 or R3 in Field A (a = 0.05). However, SCN population in
roots in samples collected at 17 DAP were negatively correlated with K and Mn at V9 [r
(K) = -0.46, r (Mn) = -0.49, a = 0.05] and R3 [r (K) = -0.45, r (Mn) = -0.40, a = 0.05] and
with Zn at R3 [r (Zn) = -0.43, a = 0.05]; and positively correlated with Mg at V9 [r =
0.69, a = 0.01]. Tissue nutrients at R2, V9, or R3 were related to SCN population density
at planting and at 44 or 19 DAP in 1999 and 2000, respectively in Field B (Table 6.5).
The sign of the correlations varied by element and with plant phenology. The effect of
SCN on leaf tissue nutrients was mostly negative at V9 and mostly positive at R3 in
2000. Magnesium was the only element that was always positively correlated with cyst

density, whereas the correlation with Zn was always negative. Cysts population density in

soil and nematodes in root at 44 DAP were negatively correlated with B in 1999.

144

.2 + =8 .«o «-808—

flmmov 2w2 so Bang—om 33 max—28 noufioboo doufioa Ewan 98 833520 an :8 05 89¢ @8358 0.83 396 H
.852? 8: 3% £88 .3 E unnumEmmm 8: 0.83 8328 53, 332280 +

reggae 432 £5395 :3 as .86 2: a .fiomwamm r. ._.

 

 

3mg- .93 is: .53. m2 m2 :3 m2 3.3 m2 BE 2
«15.0- *smbd m2 send- m2 .3..de fraud swvd ssowd .3..de @557—
mm
:NS- :9? .53 .86- and- m2 m2 m2 m2 :39 $52
:Nwd- saved- send aamd- .3506. m2 .3de m2 m2 3.55.0- mandar—
.9.
1.8.? m2 m2 :35- m2 m2 :35 .34.? m2 mz $53.
$3. m2 m2 :86- m2 m2 5.3 m2 m2 m2 mama:
mlm.
Ema ck.
m 5 :2 E on m w: M a 2 Vans... 28

 

+253 .32 533m 5 $5E=Z

 

.88 a m 25 a 53on do swam 2 Ba
o> Ba .33 5 NM «a mogu— Swonzom 5 8338088 3:033:98? can 822: 8 douse E 533838 doom EB 33

E m<m So was 2 E can mega—m 3 =8 05 5 bacon cons—smog “who 8m $56508 coca—oboe .82.: $8853 «.0 2an

145

The concentration of K in leaf tissue was affected at the reproductive stages, with
opposite signs in 1999 and 2000. Manganese, N, Cu, Fe, P, and S concentrations in leaf
tissue were inconsistently correlated with cysts or SCN population in roots, positively in
some cases and negatively in other (Table 6). Calcium was the only element in leaf tissue

that was not correlated with SCN at any stage of the plant in either field.

DISCUSSION

This work showed that SCN population density and spatial distribution were
correlated with soil fertility, and that a combination of soil fertility and SCN population
density affected the nutritional status of soybean. The results presented here emphasize
the relevance of the interactions between SCN population and edaphic factors on the
nutritional status of the host as reflected in tissue analysis.

The spatial distributions of SCN and soil pH were positively correlated
consistently over time in the two fields studied, as it was observed from the experiments
of Tylka et al., (1998) and Grau et al., (1999). Even though the degree of spatial structure
(nugget variance/total variance ratio) and the maximum distance to which pH was auto-
correlated and cross-correlated with SCN density varied by field, correlation parameters
within field were consistent over time indicating that soil pH was an important factor
acting directly or indirectly (plant mediated) in shaping SCN population distribution in a
given field. It was reported in Chapter Four that SCN population in Field B was
negatively correlated with clay percentage in the soil. This study showed clay percentage
and soil pH were negatively correlated, indicating that the spatial distribution of SCN was

affected by the combination of soil pH and soil texture.

146

SCN population density was also related to Ca and Mg concentration in the soil,
with similarities and differences between fields. Calcium and SCN population density
were correlated consistently over time, but with great variability between fields. In Field
A high levels of Ca in the soil were associated with high levels of SCN. Conversely, in
Field B the correlation was negative and the effect of Ca on the spatial distribution of
SCN was more important than in Field A. Even though Ca concentrations in the soil were
similar between fields, the strong association of high levels of Ca with fine textured soils
generated the negative correlation with SCN population observed in Field B. Calcium
and Mg uptake and translocation mechanisms are similar under normal conditions
(Reinbott and Blevins, 1991), but they behaved differently in response to SCN treatment,
with levels in leaves increased with high SCN density (Blevins et al., 1995; Franc],
1993). Variability in Ca and Mg concentration in tissue between fields and sampling
times was observed in this study. Calcium variability however, was not correlated with
SCN population density, whereas Mg in tissue was positively correlated with soil SCN
population, pH, and soil Mg in both fields. The spatial distributions of Mg and SCN were
inconsistently cross-correlated over time. The inconsistent crosscorrelation of Mg with
SCN over time indicated that the two variables were only correlated at the same
locations, and samples at neighboring locations were no longer correlated.

The effect of SCN population density on K concentration in leaves is not clear
and there is some controversy on whether K fertilization is beneficial to soybean under
SCN infection. Part of the controversy results from differences in soil K levels in the
different studies (Luedders et al., 1979; Hanson et al., 1988; Blevins et al., 1995; Howard

et al., 1998). Smith et al., (2001) have recently shown that SCN population density after

147

the first 30 days of infection did not affect K levels in leaves, but it increased K in petiole
and stem tissue, and reduced K in root tissue. The results presented here were consistent
with a three-year field research (Hanson et al., 1988) in which SCN population density
was not correlated with K in the soil. Under field conditions, K in tissue at the end of the
vegetative and beginning of the reproductive phases was negatively correlated with levels
of SCN density early in the season in this study. Potassium concentration in the soil in
Field A and Field B was similar to the highest level tested by Luedders et al. (1979), at
which they observed a decrease in SCN numbers. Potassium concentration in tissue was
higher in Field A than in Field B, even though there were no differences in K
concentration in the soil. We also observed that K concentration in leaves was negatively
correlated with pH, Ca and Mg in the soil. The higher levels of pH and Mg in the soil in
Field B, and the strong positive correlation between pH and SCN population may have
contributed all together to reduce K concentration in leaves.

Nitrogen in leaves was sufficient in both fields, with concentrations at R3
somewhat lower than at V9.The relation between SCN population and N concentration in
leaves varied with the plant developmental stage. The maximum rate of N2 fixation is
reached at the beginning of flowering, followed by a rapid decline. The decline is most
likely an expression of sink competition for photosynthates between the developing pods
and the root nodules (Haystead and Sprent, 1981). Nitrogen at R3 was lower than at V9
probably because of nitrogen mobilization to the seeds and reduction in N2 fixation rate
after flowering. Nitrogen mobilization fiom leaves with reduced nitrogen concentration
may have been comparatively less; therefore, the negative relationship between SCN and

N at V9 reversed at R3. The pink coloration observed in nodules from healthy looking

148

plants was due to leghemoglobin, a red-colored enzyme with a central Fe atom in the
porphyrin ring. The concentration of leghemoglobin is closely correlated with the N2-
fixing capacity of root nodules (Werner, et al., 1981). The grey coloration and reduced-
size observed in nodule fi'om chlorotic plants could be a sign of reduced Nz-fixation
activity. Although not quantified, the number of cysts observed on roots was negatively
related to the size, number and quality of the nodules in our study. SCN competes with
nodules for photosynthates affecting Nz-fixation rates (Poskuta et al., 1986; Sinclair,
1994). In another study, nodules from SCN infected soybean had lower fresh weights per
plant and lower specific nitrogenase activity than nodules from uninfected plants (Huang
and Barker, 1983).

Phosphorous in tissue was within sufficiency ranges in Field A and in Field B
except at R3. Increasing SCN infection level did not vary P concentration in leaves, but
decreased the concentration in nodules (Blevins et al., 1995). In our study, P in leaves at
R3 was positively correlated with nematodes in the soil and soil pH, and was below
sufficiency limits in Field B. At this stage P was also negatively correlated with Mg in
the soil in both fields.

Micronutrients in tissue were also affected by SCN population density at planting
and early in the season. In Field B, high SCN population density at planting was
associated with lower Fe concentration in leaves at V9, but noticeably, Fe concentration
at this time was well above the normal sufficiency limit for soybean. In Fe-deficient
leaves, contents of chlorophyll decline (Morales et al., 1990) and leaves look chlorotic.
Chlorosis symptoms may also appear when plants are not using available Fe effectively,

and large amounts of Fe accumulate in the leaves (Mengel and Geurtzen, 1988). This

149

phenomenon is usually found in plants grown in sandy, calcareous soils, when there is a
large supply of P, or when different forms of nitrogen are supplied. It is possible that high
levels of SCN interfered in some way with Fe utilization causing high concentration of
this element in leaf tissue and the chlorotic appearance of plants in portions of Field B.
Iron concentration in tissue samples from Field A was not correlated with SCN
population density or soil fertility.

Manganese concentration in tissue was affected negatively by pH and P in the
soil, and by SCN population density in Field A, whereas SCN population had a positive
effect on Mn in tissue in Field B. Manganese was negatively correlated with Mg in the
soil, and since Mg in the soil was much lower in Field A than in field B, Mn in tissue was
not surprisingly lower in Field A. On the other hand, Mg in the soil was positively
correlated with Mg in tissue and SCN population, and Mg and Mn in tissue were
negatively correlated at V9 in Field A. High levels of Mg and SCN in the soil, correlated
with high levels of Mg in tissue, resulted in lower Mn in tissue. At higher levels of soil
Mg, Mg in the soil and Mn in tissue were positively correlated, and there was no relation
with nematodes as observed in Field B, and no correlation between Mg and Mn in tissue
either. Manganese in tissue was negatively correlated with Ca in the soil, and Ca was
positively correlated with nematodes. Calcium in the soil was similar in Field A and B,
but Ca had no effect on Mn in tissue in Field B, perhaps related to differences in pH and
P, or differences in soil texture between fields. In its chemical behavior, Mn shows
properties of alkali cations such as Mg and Ca and the heavy metals Zn and Fe. It is
therefore not surprising that these ion species affect uptake and translocation of Mn in the

plant although the mechanism of these effects needs still to be elucidated (Fox and

150

Guerinot, 1998). Nematodes can affect Mn uptake. Barley plants grown in Mn deficient
soil with and without Mn supply had similar number of nematode infections, but plant
growth was suppressed without Mn and not affected with Mn supplement (Wilhelm et al.
1985). Manganese concentration was lower in Field A than in Field B, and so were P in
the soil and pH, therefore the levels of Mn in tissue were the result of the interaction
between nematodes, pH, and P in the soil.

Zinc concentration in tissue was negatively correlated with SCN population
density. Zinc in leaves was also correlated with pH, Ca, and Mg in the soil with
variability by field and sampling time, similar to the observations for Mn. Zinc is
involved in the same enzymatic frmctions as Mn and Mg (Jones et al., 1998) therefore it
was probably affected by the same interactions as Mn.

Copper tissue concentration was within normal range. Calcium, Mg, and pH were
negatively correlated with Cu in tissue in Field A, and positively in Field B. The
movement of Cu is strongly dependent on the Cu status of the plant (Loneragan, 1981).
The relation of SCN with Cu in tissue varied by field and by sampling time. In Field B,
Cu in tissue at V9 was negatively related with SCN, but the relation was positive at R3.
Copper at R3 was lower than at V9, probably because of translocation from vegetative
parts to seeds (Caballero et al., 1996), but the more nematodes the more Cu at this stage,
so nematodes may be interfering with mobilization. Francl (1993) found negative
correlations between Cu and SCN.

Boron in tissue was within the sufficiency range, and while it was not affected by

SCN in Field A, it was consistently and negatively correlated with SCN in Field B. The

151

correlation of B with other elements such as Ca, Mg and K, or soil pH varied by field and

sampling time.

CONCLUSION

This study shed light on the complexity of the interactions among soil fertility and
texture, SCN population density, and the nutritional status of the host. The information
presented here based on observations on naturally infected fields provides the basis for
designing experiments to test cause and effect relationships and advance in the
understanding of the relationship between SCN and soybean for management purposes.
For example, our results indicate the recommendation to growers to increase K
fertilization of SCN infested fields to increase soybean yield should be carefully revised,
taking into consideration not only K concentration in the soil, but also soil pH and Ca
levels. In addition, this work assists with the delineation of management zones for SCN

control based on soil fertility in addition to soil texture.

152

CHAPTER SEVEN

SPATIAL ANALYSIS OF SOYBEAN YIELD IN RELATION TO SOIL TEXTURE,
SOIL FERTILITY AND SCN POPULATION DENSITY.

Felicitas Avendafio, Francis J. Pierce and Haddish Melakeberhan. Manuscript to be
submitted to Journal of Nematology.

153

SPATIAL ANALYSIS OF SOYBEAN YIELD IN RELATION TO SOIL TEXTURE,
SOIL FERTILITY AND SCN POPULATION DENSITY.

F elicitas Avendano], Francis J. Pierce2 and Haddish Melakeberhan3

1- Michigan State University. Current address: 1308 Sheldon Dr., Newark, DE 19711,
avendanl @msu.edu

2- Center for Precision Agricultural Systems, Washington State University, 24106 North
Bunn Road, Prosser, WA 99350. fipierce@wsu.edu

3- Michigan State University, Department of Entomology, 243 Natural Science Building,
East Lansing, M148824, melakebe@msu.edu

Correspondence should be sent to H. Melakeberhan.

AKNOWLEDGEMENTS

The project is part of a Ph.D. Dissertation of the first author. We thank Dave Eickholt,
Charles Eickholt, (farmers) and Neil R. Miller (ABC-Agribusiness Consultant) for
providing and maintaining the sites, and for supplying yield information. Financial
support for the project was provided by USDA Hatch Project through the Michigan
Agricultural Experimental Station and grants from the Michigan Soybean Promotion
Committee, North Central Soybean Research Program and United Soybean Board to the

last author.

154

SPATIAL ANALYSIS OF SOYBEAN YIELD IN RELATION TO SOIL
TEXTURE, SOIL FERTILITY AND SCN POPULATION DENSITY.

ABSTRACT

The purpose of this work was to identify the extent to which soybean yield was
spatially correlated with soil texture, soil fertility, and SCN population densities in two
fields in Michigan (Field A and Field B). Soybean yield was measured with a commercial
yield monitoring system mounted on the combine connected to a GPS receiver. Yield of
the susceptible soybean variety planted in 1999 was 20% and 30% greater than the
resistant variety planted in 2000 in Field A and Field B, respectively. Spatial variability
in yield was highly structured in Field B, and moderately in Field A. Correlation analysis
was performed on yield and sand, clay and silt percentage in the soil, soil pH, P, K, Ca
and Mg concentration in the soil, and cysts and SCN in roots population densities data
available from previous work. Yield was correlated with sand (r = -0.89), clay (r = 0.86),
and silt percentage (r = 0.84), and with soil pH (r= -0.60) and Ca concentration (r = 0.60)
in Field B in 1999. In 2000, Pearson’s correlation coefficients were slightly lower. These
variables were strongly spatially correlated with yield, as indicated by the cross-
correlograms. Yield was also negatively correlated and spatially cross-correlated with
SCN population density in the soil. In Field B, correlation coefficients between yield and
Pi (SCN population density at planting) were —0.48 in 1999, and —O.45 in 2000, in Field
A were -0.16 in 1999, and —0.20 in 2000. Thus, providing basis for future work on
delineating management zones for SCN. In fields where soil properties and SCN
densities appear spatially structured, and where there is a history of yield spatial
variability due to the combined effect of SCN and unfavorable soil conditions,

management zone delineation could be an appropriate strategy to overcome yield losses.

155

The geostatistical analysis of soybean cyst nematode (SCN, Heterodera glycines
Ichinohe) distribution patterns (Avendafio et al., 2003, Chapter Three), the relationship
between SCN and soil texture and soil map units (Chapter Four), soil texture and SCN
life cycle (Chapter Five), and spatio-temporal dynamics of SCN and soil and plant
nutrition (Chapter Six) were documented. From the preceding chapters, it is clear that the
two fields exhibited differences and similarities. For example, the fields differed in soil
type, texture, fertility, and SCN population density. SCN was correlated with pH, texture
and Ca in both fields, but more in Field B. Nutrients were related with SCN but not
consistently between fields or time.

Seed yield is the ultimate interest of a soybean grower, and knowing any
relationship between any of the correlations described above and crop yield will be
advantageous to the grower’s decision-making ability. Therefore, the purpose of this
work was to identify the extent to which yield was related to soil texture, soil fertility,

and SCN population densities by correlating spatial information on these variables.

MATERIALS AND METHODS
The study was conducted in 1999 and 2000 on two fields (Field A and Field B) in
Shiawassee County, MI. Field A consisted of 24 ha, was managed under no-tillage since
1996, and was planted to corn prior to this study in 1998. Percent sand ranged from 45%
to 74%, percent clay ranged from 10% to 21%, and percent silt ranged fi'om 8% to 43%
within Field A (Chapter Four). Field B was comprised of 13 ha, was conventionally tilled
after wheat in 1998, and was managed under no-tillage thereafter. Percent sand ranged

fiom 50% to 80%, percent clay ranged fi'om 8% to 23%, and percent silt ranged from

156

11% to 29% within Field B (Chapter Four). The soybean varieties planted were Asgrow
1901 in 1999 (SCN-susceptible) and Asgrow 2201 in 2000 (SCN-resistant). Both
varieties were Roundup-Ready. Soybean was planted in 19.1-cm rows at a rate of 519
000 viable seeds ha'1 in 1999, and 494 000 viable seeds ha'1 in 2000. Fields A and B were
planted 5/22/99 and 5/16/99, respectively, and 6/9/00. Weed control was maintained
using Roundup at the recommended rate. There was one preplant application in Field A
in 1999 and 2000, one application postemergence in 1999, and a midseason application in
Aug. 2000. In Field B there was one preplant application in 2000, one postemergence in
1999 and a mid season application in Aug. 2000. Rows orientation was north-south in
Field A and east-west in Field B.

SCN, soil texture and soil fertility data were obtained fiom Avendafio et al. (2003,
Chapter Three), and Chapters Four, Five and Six. The spatial sampling for soil samples
consisted of a nested survey sampling design with distances reduced by geometric
progression (adapted from Webster and Boag, 1992) applied within 8 and 5.25 ha in the
center of Fields A and B, respectively, as described in Avendano et al. (2003, Chapter
Three). Single-core soil samples were collected using an 8 cm diameter by 23 cm deep
bucket auger (Riverside Augers, Eijkelkamp, Giesbeek, The Netherlands) from 160 and
110 locations in Field A and Field B, respectively. Each sample collected before planting
in 1999 was thoroughly mixed and then divided in three sub-samples. One sub-sample
was used for SCN analysis, the second was used for texture analysis, and the third portion

of the sample was analyzed for pH, phosphorous, K, Ca, and Mg.

157

The collaborating farmer measured soybean yield in both fields in 1999 and 2000
with a commercial yield monitoring system mounted on the combine connected to a GPS
receiver.

Statistical Analysis

Sample mean and variance were calculated for yield in each field and year. Means
between years within fields were compared with the z-test for means (a= 0.05), and
variances between years within fields were compared with the F-test ((r= 0.05).

A subsample was extracted from each yield data set matched by the locations
were soil sampled had been collected for texture, fertility, and SCN analysis to perform
correlation analyses. Pearson’s simple linear correlation coefficients were calculated for
yield and soil texture, yield and soil fertility, and yield and SCN population density (The
SAS System Release 8; SAS Institute, Cary, NC).

Geostatistical analysis was used to quantify the spatial variability in yield in Field
A and Field B in 1999 and in 2000. The spatial analysis of yield was performed on a
subsample consisting on the data points collected within the area sampled for soil and
SCN analyses. General geostatistical methods were described in Avendafio et al. (2003,
Chapter Three). Directional semivariograms were calculated to explore anisotropy in the
spatial variability in yield. Geometric anisotropy can be visualized as elliptic iso-
correlation contours, defined by the ratio of two orthogonal axes (radius 1 and radius 2)
and an orientation angle. We define the orientation angle as the counterclockwise rotation
between the positive X-axis (east direction) and radius 1. The ratio of major and minor
axis is defined as the ratio of the largest and shortest range in the empirical directional

semivariograms. Directional semivariograms were calculated for lags of 12 111 so that at

158

least 30 data pairs were available for each lag in each direction with an angle tolerance of
22.5 angular degrees. The parameters of the semivariograms and the degree of anisotropy
were estimated by least squares fitting of theoretical senrivariograrn models.

The cross-correlogram is a geostatistical tool used to define the joint spatial
dependence or continuity between measurements of different attributes or of the same
attribute measured at different times. In the cross-correlation function, the cross-
correlation coefficient at lag equal to zero is the same as the classical correlation
coefficient between two variables. The cross-correlation function given by Goovaerts
(1997) was used here to calculate cross-correlograms for yield and the following
attributes: sand, clay and silt percentage in the soil, soil pH, P, K, Ca and Mg
concentration in the soil, and logarithmic transformed cysts and nematodes in roots
population densities. Cross-correlations were calculated for each field and year in the
cases were linear correlations were significant ((r= 0.05). Only the data points from

locations sampled for both attributes were used for these analyses.

RESULTS
Soybean yield
The SCN-susceptible soybean planted in 1999 yielded more kilograms per hectare
of seeds than the SCN-resistant variety planted in 2000 in both fields (T able 7.1). Yield
variance in Field B was significantly higher in 1999 than in 2000, and two times greater
than in Field A in both years. The georeferenced data collected by the yield monitor was

plotted and a map of seed yield in 1999 and 2000 were obtained for each field.

159

Table 7.1. Soybean yield within the areas sampled in Fields A and B in 1999 and in 2000.
Soybean seed yield (Kg ha")

 

 

 

Field A Field B
Year MeanT Std. Devi Mean]L Std. Dev:
1999 3104.5 a 239.5 a 2940.7 a 688.2 a
2000 2498.1 b 252.3 a 1997.1 b 543.9 b

 

1' Means within columns followed by the same letter were not significantly different at
the 5% level (z-test for means).

I Standard deviations followed by the same letter indicate variances within columns were
not significantly different at the 5%leve1 (F-test for variance).

In Field A, seed yield was relatively uniform through out the field in both years
(Figure 7.1). The lighter coloration along the south edge of the field indicated lower
yields in this area slightly elevated over the rest of the field. In Field B yield maps, two
distinct zones could be identified. Seed yield on a large section on the north side of the
field was low and appeared patchy, whereas on the rest of the field seed yield was much
higher and uniform (Figure 7.2). In 2000, the lower yield zone occupied a greater
proportion of the field. On the west side of the south zone there was a sector of lower
yield, corresponding to an area that was ponded for long periods fiom September until
harvest in 2000.

The directional semivariograms of yield in Fields A and B are shown in Figure
7 .3. The parameters of the corresponding semivariogram models fitted are shown in
Table 7.2. The spatial structure in yield distribution within the area sampled for soil was
moderate in Field A and high in Field B. In Field A, the shape of the semivariograms in
1999 and 2000 were similar, but the variability in yield distribution in 1999 was greater
than in 2000 as evidenced by greater sill and nugget variances in 1999. The longest and
shortest ranges of spatial autocorrelation extended beyond the longest and shortest

dimensions of the field, respectively.

160

decaf—83.. a... E 3.8 E own—E nib .mcmbafi tam—n.8,“ :8 was SE8 :8 .20m 8m 380:8 0.83
838% :8 80:3 28:82 05 33%": 83:0 gum—m .8388 2: no @8958 ES mew a 8 3858 82%?
Ease so: a 3 8282 as 2...; Row .88 a as 82 a < 20E 3 22» 83 533m .3 2am

       

 

 

 

 

 

 

 

AEV 95me
82.9. SEQ. cactus 8+5»
1 -80on
com o. o O A 7
89 2 eon A9 r
comp 2 89 A6 7
88 2 come as
comm o. 88 O m , 1089.:
88 0. comm O A ;
comm 0. coon O A Z
ooov 0. comm O n A W
-oomonnv
gave 7 H w
2% 5398 J n" m.
7 A \II
1, g -09.on W
L
7 , 2+
1 A I 08on
.

 

ooow 29> mmmr 22>

161

325.3% 2.. a 3.8 a. $2... a “8%: Efia :8 2a 2&3
=8 .ZUm he.“ 380:8 803 3388 :8 82.3 9.2302 05 8865 8.86 x85 6:388 05 no @8555 :5. 95 a
9 3858 88% 3:88 22» a 3 8282 83 20a Bum .88 a ea 82 a m 205 a 23» 88 Exam .2 2:5

:5 mcfimm
83MB commas common. 8 rank. oovmmn 88MB comma... 00 rank

 

 

8
O
3.; J.

 

E

s

E
00.06

 

(w) BummN

3&9:

     

 

 

ooow v.0; mmme 22>

162

.0E0.0m0_00>_800 0000000000 02009000 05 E 03.0.. 00000000 000 000 000mm: 05 00 0000000 000 5.00 40008 0000000900 05 00 m Er

A 03000 0000 000000000 0000 06 00003000 0&00 00030200000000 05 00 000G000 00 0300 00000000000 05. mm

.0_w00 000000000

000 3 000 Am 00.0000 000 ~ 0.0.80.0 0000 Ecowonto 030 0: .00 .300— 05 03 03000000 000020 00 00 00020003 00 000 3000003 .3
0000—0000000 0003 C 3000b 00000005 0.000 00000000000 05 00 038 05 :05 000000 3 00000000 303000 0003 05 0000030000 .2.
0000300000000 000% 05 00m00000 05 0000000 05 0030— 05 60300000 30000 .00 00%00 05 .00 00000005 00 00 A00+U§VU u

:0 0000 2: 0 0 0

.00000 $590000 00 523000? 2000000000 00 0:0 EH00 0:0 00 0000000300000 5 30000000000 0 00 000.0000 003:: 05 00 00 0

0900000 003 00000 00

E00 0 00:3 000000000 00 00000 05 .00 00000 30800500 005. 2002000 05 00 0000000000 003 0003000 :000w00003800 000 02000030
0000000 0002 0000000300 030000 0 0:0 warn—000 000000000 0003 0000 05 50000000000: 003000 0.0000w20030000 0050055 H

on 003 w0_ 0000 00.0 0003.000 0.200 .00 000800 000000008 05.

N2 ,00 00000200 w0_ 0 £3 A6 8 N0 .00 0w0_ 00.0 000030—00 0E00w0_00>_:000 000009000 00 00000 0000000 0000— 03 0000.0 0003 £0002 +

 

 

 

 

 

0.0 00 00 0 3 000 0000:2000
$0 000:2 00:: 080 0 :0E

0.0 R 00 00.0 000 00008000
0.2 000:2 00:: 32 m 220

no 00 03 00.0 0.02 000830
0.0: 000:2 :2 800 < 220

0.0 E 0: 00.0 000 0058000
000 000:2 oz 82 < 220

a 02 m:
E N03000- :0 + U
3.000 320.2 0.0000 0 c0 =0 30 000050 00:: 0.0.00
303000000000

 

0.080 a 20 82 :0 m 0.0 < 00: a 000 50:00 0: 20:8: 00.03360 50285 200: 0008000 0.0 2000

163

Field A 1999 Field A 2000

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4000004 400000‘
3500004 3500001 . '60
‘.:;:...ga: 1. 5 115
3000004. *ffi-arL AI‘I: 300000 0 70
agifientziflh““‘“ i A 25
zsooooa .A ‘ 2500001I
200000. 200000
.5555
1500001 “meow—r {m 0“”? m. {:70 ho- a To"
0 147 m‘xxxxxziuuuunhgu
1oooooj ‘ 102 1oooooi 5 o-
A
8 50000l O 57 50000
t: i A 12
'm 0 _,_ -» o+——-——--— —— —-—- -~—-
g 023400000700120140100130200 020400000100120140100100200
a aoooool _ 4000001 .
g Field B 1999 FieldB 2000
350000« 350000l
5‘93
_.___ “VA- ‘.6‘ °o°_ - ‘
WW 2°0°.:W*‘05545555 3000001
A “ .......OO 000° 000
W” :52.““,.--" m'fl‘ ”finfi .somiu .m
2W“ 0‘“ ‘0'. m1°a§*.o..
150000« .. 150000« .
i o 167 l . 0 176
ioooool . ‘ 122 1000001 A 131
o 77 j.’ o 86
5°°°°l A 32 5°°°°.; A 41
OL‘__._7_,__. . ~~T -- T 1.. . oin w w . r .
0 20 40 00 00 100 120 0 20 40 00 00 100 120
Lag distance (m)

Figure 7.3. Anisotropic semivariograms of soybean yield (Kg/ha) in Fields A and B in
1999 and in 2000 in different directions in degrees from East. The dashed line is the
sample variance. Parameters of the semivariogram models are shown in Table 7.2.
There was less yield variability along than across rows as indicated by the
anisotropy parameters. In Field B, yield spatial variability was highly structured, and
more so in 2000 than in 1999. Yield variability was greater along the north-south

direction as indicated by the shorter range of spatial autocorrelation perpendicular to the

row orientation both years.

164

Soybean Yield and Soil Attributes

Soybean yield and soil texture were strongly correlated in Field B both years,
weakly in Field A in 1999, and not linearly correlated in Field A in 2000 (Table 7.3).
Soils with high percentage of sand and low levels of clay and silt yielded significantly
less seed than finer soils. Soybean yield and soil texture were cross-correlated over a
range of 120 to 130 m, consistently across fields and over time (Figures 7.4 and 7.5). In
Field A, sand was more strongly cross-correlated with yield than silt or clay (Figure 7.4),
but in Field B the cross-correlograms for the three soil separates with yield were very
similar in structure, with sand negatively correlated with yield while the relation with
clay and silt was positive Gigure 7.5). Soil pH and yield were related in both fields in
2000 and only in Field B in 1999. High yield was associated with low pH in all cases
(Table 7.3), but the range of spatial correlation between the two attributes varied between
fields.
Table 7 .3. Pearson’s correlation coefficients for soybean seed yield with sand, clay, silt,

and soil pH and concentrations of P, K, Ca and Mg (Kg ha") in Fields A and B in 1999
and in 2000.

 

 

 

 

Soybean seed yield
Field A Field B

Soil attributesT l 999 2000 l 999 2000
Sand -O.36** NS -O.89** -0.77**
Clay 0.28" NS 0.86" 0.73"
Silt 0.27** NS 0.84" 0.73M
pH NS -O.39** -0.60** -O.56**
P NS NS 020* NS
K NS 0.32” 0.34" 0.29"
Ca NS -O.30** 0.60" 0.47"
Mg 0.31 ** -O.19* NS NS

 

*, ** Significant at the 0.05 and 0.01 probability levels, respectively.
1' Soil nutrients and soil texture were analyzed on samples collected at planting in 1999.

165

In Field A, pH and yield from samples separated by more than 50 m were not
linearly correlated (Figure 7.4), whereas in Field B the relation was maintained for up to
140 m (Figure 7.5). Soybean yield was poorly correlated with soil fertility, with great
variability between fields and years. In Field B, soils with higher levels of Ca and K
produced higher seed yield in 1999 and 2000, whereas in Field A higher seed yields were
obtained from soils with high levels of K and lower levels of Ca and Mg, although this
was true only in 2000. In 1999, yield was only correlated with Mg concentration in the
soil in Field A (Table 7.3). Yield was correlated with P only in Field B in 1999. The
spatial correlation between yield and soil nutrients was poor in Field A, except for Mg in
1999. The cross-correlation yield-Mg was almost identical to the cross-correlograms for
clay and silt (Figure 7.4). In Field B, Ca was cross-correlated with yield for up to 120 m

approximately, whereas P and K were not spatially correlated with yield (Figure 7.5).

 

 

 

 

 

 

 

 

 

 

 

1999 2000
1.0 ——o-— Sand 4. PH
....... 0...... Clay nun-0...... w
---v--- sm --+-- Ca
- ...... — ...... K
c 05 ‘9 M9 4 -v
3 x? .03...
£9
g .
8
(D
3
O
1.0 r f

 

 

 

 

O 50 100 150 200 0 50 100 150 200

Lag distance (m)

Figure 7 .4. Cross-correlograms of yield and soil texture (% sand, % clay, and % silt), and
soil fertility (pH, and Mg, Ca, K) in Field A in 1999 and in 2000. Cross-correlation at
zero lag distance equals the linear correlation coefficient.

166

 

 

 

 

 

9:01er I908

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

C

.9

E

o

t

o

‘r"

U)

3

O
5
Q

o 50 100 150 200 0 so 100 150 200

Lag distance (m)
Figure 7 .5. Cross-correlograms of yield and soil texture (% sand, % clay, and % silt), and
soil fertility (pH, and P, K, Ca in Kg ha!) in Field B in 1999 and 2000. Cross-correlation
at zero lag distance equals the linear correlation coefficient.
Soybean Yield and SCN Population

Soybean yield and SCN population density were more strongly correlated in Field
B than in Field A, and cyst population in the soil was more strongly and more ofien
related to yield than SCN infection levels in the roots. Correlations were negative in all
cases (Table 7 .4). The correlation between SCN and yield in Field A was very poor, and
in 2000, it was only significant for the number of cysts in the soil at planting. In Field B
however, SCN population densities at planting and over the season were relatively

strongly correlated with 1999 and 2000 seed yields. SCN population in the roots and

yield were only related in 1999. Even though the spatial correlation for yield and SCN

167

population density was poor in Field A, the structure of the cross-correlograms was very
similar for all SCN samples, with a range slightly above 50 m (Figure 7.6). Moreover,
cyst population density at planting in 2000 explained yield variability up to a separation

distance of about 50 m between observations as well (Figure 7.6).

Table 7.4. Pearson’s correlation coefficients for soybean seed yield and SCN population
density sarmfled at monthly intervals in Fields A and B in 1999 and in 2000.

 

 

 

 

 

 

Soybean seed yield
Cysts in the soil!" Larvae and adults in rootsL

SCN sample 1999 2000 1999 2000

Field A
Planting -O.l6* -O.20**
June§ NS NS NS NS
July -0.25** NS ~O.32** NS
August -O.18* NS NS NS
Harvest -0.23** NS NS

.Ei_el_d_B
Planting -O.48** -O.45**
June# -O.49** -O.36** -0.28** NS
July -O.45** -O.37** NS NS
August -O.65** -O.19* -0.22* NS
Harvest -O.45** -O.59** -O.l9*

 

*, ** Significant at the 0.05 and 0.01 probability levels, respectively.

1' SCN cysts extracted by elutriation and sugar flotation from 100 cm'3 of soil sub-
samples.

: SCN larval stages and mature females counted in 2 g of stained soybean roots. Root
samples were not collected at harvest in 2000.

§ In 1999, Field A samples were collected at 31 (June), 59 (July), 86 (August), and 126
(harvest) days after planting (DAP). In 2000, samples were collected at 17 (June), 45
(July), 73 (August), and 147 DAP (harvest).

# In 1999, Field B samples were collected at 44 (June), 73 (July), 100 (August), and 125
DAP (harvest). In 2000, samples were collected at 19 (June), 46 (July), 77 (August), and
155 DAP (harvest).

168

 

1-0 ....... O ...... 59 DAP (m)

 

 

 

---v--- 59 DAP
—...v.-.. m DAP
05 q , . — -. '—' ' 126 DAP

 

Cross-correlation

 

 

 

 

 

 

 

'1,0 T v v I 1 f

0 50 100 150 200 O 50 100 150 200
Lag distance (m)

Figure 7 .6. Cross-correlograms of yield and SCN population density in Field A in 1999
and in 2000. SCN population density was quantified in soil (cysts) and in soybean root
samples (all developmental stages) collected at the days after planting (DAP) indicated.
Cross-correlation at zero lag distance equals the linear correlation coefficient.

1999

 
  
 
    

 

—O— Planting
........ O----~ 44 DAP
--+-- 73 DAP
0.5 ‘ _.._v._... 100 DAP
125 DAP

 

 

 

 

 

 

Cross-correlation
O
C

 

 

 

Lag distance (m)

Figure 7.7 . Cross-correlograms of yield and SCN cyst population density in soil samples
in Field B in 1999 and in 2000. Soil samples were collected at the days after planting
(DAP) indicated. Cross-correlation at zero lag distance equals the linear correlation
coefficient.

169

1999

 

1.0

—O— 44 DAP
----o-~ 100 DAP
—+—- 125 DAP

 

 

 

Cross-correlation

 

 

 

O 50 160 150 200
Lag distance (m)

Figure 7 .8. Cross-correlograms of yield and SCN population density in soybean roots in
Field B in 1999. Root samples were collected at the days after planting (DAP) indicated.
Cross-correlation at zero lag distance equals the linear correlation coefficient.

The range of spatial correlation between yield and cyst densities in Field B was
about 130 m for all samples collected from planting to harvest in 1999. In 2000, there
was great variability in the cross-correlograms, but still the range for yield and cyst

densities was maintained (Figure 7.7). The level of infection, or number of SCN in the

roots was practically not spatially correlated with yield in Field B in 1999 (Figure 7.8).

DISCUSSION
In addition to genetics, crop yield is a function of biotic and abiotic factors
limiting plant productivity (Marschner, 1995). Ofien, the yield-limiting factors have
spatio-temporal structure that could be helpful in management decision-making (Pierce
and Sadler, 1997; Cassel et al., 2000). For example, linking spatial information of soil

texture, nutrients, and pest problems can allow for diagnostic determination of the

170

predominant factor(s) controlling crop production. This then becomes the basis for
developing precision input strategies (Sudduth, 1999).

Better yield was obtained from the SCN-susceptible soybean variety sown in
1999 than from the resistant variety sown in 2000. Soybean yield potential in the
presence of SCN varies among varieties and with level of infection, field topography, soil
properties, climate, and management practices (Koenning et al., 1988; Koenning et al.,
1995; Tylka et al., 1998; Niblack 1999; Koennning, 2000; Kravchenko and Bullock,
2000; Chen et al., 2001; Long and Todd, 2001). The differences in yield observed
between years in this study could have been a consequence of differences in weather
conditions and date of planting between years, in addition to the genetic differences
existent between varieties (Koenning et al., 1993; Noel and Edwards, 1996). The
difference in yield between fields, however, were probably due to differences inherent to
each field such as soil properties, or to other factors unique to each field such as SCN
population load.

Spatial variability analysis of the relationships among yield, SCN population
density, soil texture and soil fertility exhibited variable responses between the two fields
and among the parameters. Of all the variables examined, soil texture, soil pH and Ca
concentration had strong spatial correlation with yield. Analysis of cross-correlation
between a variable and another variable that is sampled at neighboring locations with
increasing distance provides insight into the spatial covariance structure (Davis, 1986). A
large degree of cross-correlation was observed between yield and soil texture, pH, and Ca
concentration in Field B. In general, soil texture and soil fertility were more variable and

were better spatially structured in Field B than in Field A (Chapter Four, Chapter Six).

171

The greater spatial structure and variability in yield in Field B was therefore expected, as
well as the poor spatial structure and rather uniform distribution of yield in Field A.

The strong spatial correlations in Field B generated a “good zone” on the south
side of the field, area of good yield, and lower pH, and a “problem zone” on the north
side, a sandier area of higher pH and Ca concentration where yield was extremely low. In
addition to differences in soil factors, these two areas differed significantly in SCN
population density. The cross-correlations between yield and SCN population densities
extended over approximately the same range as yield and soil properties, providing
evidence that SCN population in the soil contributed to the delineation of the good and
poor yield areas in Field B as well.

The relationships demonstrated in this work provide the basis for future
work on delineating management zones for SCN. In fields where soil properties and SCN
densities appear spatially structured, and where there is a history of yield spatial
variability due to the combined effect of SCN and unfavorable soil conditions,

management zone delineation could be an appropriate strategy to overcome yield losses.

172

CHAPTER EIGHT

GENERAL DISCUSSION AND CONCLUSION

The goal of this Dissertation was to understand the spatial distribution of the SCN
in soybean fields to assess the potential for developing SSM strategies for SCN. In order
to apply SSM strategies, it is necessary to know SCN spatial distribution, and how it
changes through time. Being able to detect the presence of the nematode before it reaches
damaging levels and to predict its location in a field will reduce effort and cost of control
practices while increasing their efficiency. In addition, knowing the role of soil nutrition
in the relationship plant-nematode interactions may provide tools for reducing or even
preventing low yield patches, therefore increasing productivity.

In Chapter Three, the magnitude, structure, and persistence in time of the spatial
distribution patterns of SCN cysts, eggs, and eggs per cyst were studied under field
conditions to test the hypothesis that the spatial distribution of SCN is sufficiently
structured and time-invariant to support the use of SSM in SCN infested fields. The
structure of SCN spatial distribution was further analyzed in Chapter Four in relation to
soil texture. For that purpose, the structure in soil texture spatial distribution and its
relationship to published soil survey maps were assessed within the study sites. The
extent to which the spatial variability in SCN cyst population density was related to soil
texture was also determined and quantified; and the extent to which this relationship held
between fields with similar soil types but different SCN population levels was assessed.
In Chapter Five, SCN population dynamics in soybean roots and in the surrounding soil

were described and analyzed spatially in relation to soil texture. Chapter Six focused on

173

the relationship between soil fertility and SCN population, and how did this relationship
reflected in tissue analysis to explain the differential degree of damage due to SCN
observed in infected soybean fields. Soil fertility variability was characterized in relation
to soil texture and tissue analysis, and SCN in the soil and in soybean roots was analyzed
in relation to soil fertility and tissue analysis. The purpose of Chapter Seven was to
identify the extent to which yield was related to soil texture, soil fertility, and SCN
population densities by correlating spatial information on these variables. Seed yield is
the ultimate interest of a soybean grower, and knowing the relationships among variables
within a field with soybean yield would be advantageous to the grower’s decision-making
ability.

While it is not known when SCN was introduced into the fields and what
undetermined factors may have contributed to the difference in SCN population density,
Field A and Field B presented me with the opportunity to study the relationships where
SCN population was involved under two different conditions fi'equently encountered by
soybean growers. Cyst population density was high in Field B and it increased during the
study, whereas in Field A, cyst density was much lower and remained constant over two
years. The within field variability in cysts, eggs per cyst, and eggs was large in both
fields, and even though the spatial structure in SCN population in general was poor, the
temporal variability was small. Several infections took place over the growing seasons of
1999 and 2000, but the greatest density increase occurred during the first month after
planting, followed by a decline in July- August (46 to 100 DAP). The low nematode
counts in roots recorded during the summer could be a reflection of the infective behavior

of SCN. Since the roots sampled in this study were collected close to the base of the

174

plant, an important portion of the SCN population in the roots may have been missed if
new infections were occurring at the root tips, where tissue is softer, rather than on older
portions of the roots. However, a decrease in SCN population density in roots
accompanied by a decline in cyst densities in the soil is an indication of reduced hatching
and consequently reduced new infestations, supporting the notion of early-induced
dormancy proposed by Yen et a1. (1995). Decrease hatching indicative of SCN dormancy
in midsummer has been observed in Iowa and north Missouri (Yen et al., 1995), whereas
in North Carolina, dormancy is induced at the end of the growing season (Hill and
Schmitt, 1989). The resistant soybean variety planted in 2000 did not prevent SCN
population density increase, cyst density at harvest was equal to or higher than at
planting, and the number of eggs per cyst was comparable to those observed at harvest in
1999. The spatial structure in SCN population varied with sampling times, but the
periodicity pattern in semivariograms appeared consistently from planting to harvest in
both fields. The difficulty in fitting an appropriate wave model to the empirical
semivariograms underestimated in some cases the spatial structure in SCN population.
Nonetheless, two levels were identified in the spatial structure of SCN population. A
short-range process described by the wave model, where cyst densities in the soil were
arranged in small clusters, and a larger range process described by spherical or
exponential models that represented the spatial arrangement of the small clusters. For
management decision-making purposes, the larger range process is of interest to detect
high infestation areas in a field. Once these areas are located, a more intensive sampling

may be necessary to identify the extent of the infestation.

175

Although at present the two fields are managed similarly, differences in
management histories could possibly have been a factor in the variability in SCN
populations currently observed between fields. Moreover, the differences in soil texture
and soil fertility between the two fields were also involved in defining SCN population
density and spatial distribution, given that soil texture, pH, and calcium concentration
were strongly correlated with SCN population density in the soil, but to a lesser extent
with SCN in the roots. The proportion of sand, clay, and silt in the soil were strongly
cross-correlated with SCN population density at different sampling times, indicating that
soil texture data at any given location in a field may provide some information about
SCN population density over a given separation distance. The separation distance
determined by the range of the cross-correlograms in Field A and Field B was 110 to 130
m. Soybean roots fiom plants grown in fine textured soil had higher SCN population
density, but the number of cysts in the soil was lower than in sandier soil. This indicates
that even though a larger number of infective juveniles were able to penetrate soybean
roots in fine soils, reproduction or survival was lower than in coarser soils. Soil texture is
a soil attribute relatively invariant in time that is less expensive to analyze than SCN.
Based on the consistency in time and across fields observed in the correlation between
soil texture and SCN population density, the information provided by texture analysis
could be used to identify SCN high-risk areas in a field. Therefore, the cost and effort of
sampling for SCN can be reduced if data on soil texture are available for a particular
field.

The nutritional status of the crop was correlated with soil fertility, soil texture,

and with SCN population density. Tissue nutrient analysis reflected the interactions of all

176

of these variables combined. In a field study such as this one, the variables under study
are beyond the researcher’s control, and conclusions are based on observations of natural
phenomena only. The results presented here on the relations between SCN and tissue
nutrient analysis constitute a valuable source of information for designing controlled
experiments and advance in the understanding of SCN role in plant nutrition. Despite the
differences between fields, however, some consistencies were observed. For example,
calcium concentration in leaf tissue was not related to SCN population density, and,
whereas magnesium was directly related to SCN population density at all sampling times,
zinc and boron were inversely correlated with SCN levels in both fields. Seed yield was
strongly correlated with soil texture, soil pH and calcium concentration, and with SCN
population density in the soil. The differences in yield observed between years could be
explained in terms of soybean variety, weather conditions, planting date, or previous
crop. Differences between fields however, were the result of the interactions of SCN and
soil attributes among other factors not accounted for in this project, such as field
topography and soil moisture.

The analyses of yield maps showed relatively homogeneous seed yield throughout
Field A, and two distinct areas with good and poor yield in Field B. The two zones
identified in Field B were very different with regards to the soil attributes measured. The
soil in the good yield area was sandier, with higher pH and calcium concentration,
whereas in the poorer yield section soil was finer, with lower pH and calcium
concentration. Moreover, SCN population density was consistently higher on the poor

yield area and remained low on the good yield portion of the field over the duration of the

177

study. In Field A, however, the correlation between different soil attributes was weaker
and SCN density was low through out the field.

The working hypotheses of this Dissertation were demonstrated to be true. SCN’s
spatial distribution was sufficiently structured and time invariant for SSM to be an
effective management strategy. In addition, SCN spatial distribution and population
density were related to soil properties such as soil texture, soil pH, and calcium
concentration in the soil that are easier to measure and manage. The relations among
SCN population density, soil properties and soybean yield were sufficiently strong to aid

in the management decision-making process. Therefore, SSM for SCN is plausible.

Soil texture
; Soil Fertility A

 

 

 

 

 

 

— SCN 4 » Soybean
\ Other factors /
(water. pathogens.
weeds. more)
- Dispersion - Tissue nutrients
- Host finding - Growth rate
- Root Penetration - Height
- Reproduction - Resistance to diseases
- Population density and pathogens
—’ W - Photosynthesis rate F
\W—J
SPA‘I1AL
orsrnreunou “El-D

 

 

 

 

 

 

Figure 8.1. SCN-soybean system. The diagram represents some of the factors and
elements involved in this system in a commercial soybean field.
The correlations and interactions observed during this study are represented in a

diagram (Figure 8.1). The success of SCN population in the soil and in soybean roots

178

depends on the presence of a host (soybean), and conversely, soybean development,
growth and yield are affected by the presence of SCN. These interactions rise from the
parasitic nature of the relationship. Environmental factors can modify the host-parasite
relationships positively or negatively. Among them, soil fertility and soil texture affect
the nutritional status of the plant, and SCN population density and spatial distribution.
Other factors not accounted for in this study, such as soil water availability, pathogens
and pests, weeds, and weather have an important role in this system as well affecting the
plant or SCN. The outcomes of the interaction of all the elements represented in Figure
8.1 are the spatial distribution of SCN population and soybean yield, and these will vary
by field because conditions vary by field as well.

The results of this work made a significant contribution towards the
understanding of SCN spatial distribution in soybean fields, providing evidence of the
underlying factors involved in determining seed yield, and laying the basis for further
research on cause and effect relations to advance in the understanding of SCN biology,
ecology, and management opportunities. The differential degree of SCN infection and
yield loss between fields, accompanied by differences in the spatial arrangement of soil
attributes observed provided the foundation for developing SSM of SCN through the

delineation of management zones.

179

APPENDIX A

180

ELUTRIATION PROCEDURE EFFICIENCY TO EXTRACT SCN
CYSTS FROM THREE SOIL TYPES

A variety of techniques, all of which are based on the flotation and sieving
principles developed by Cobb (1918), have been used to extract nematodes from soil. The
semi-automatic elutriator and the wet-sieving and decanting technique are currently the
most frequently used methods to extract soybean cyst nematode (SCN), Heterodera
glycines Ichinohe, cysts from the soil (Tylka, 1998; Smolik, 2001; Donald, 1999). In both
techniques, cysts are collected in a 250 um pore sieve (mesh #60). However, SCN
morphometrics indicate that cysts can be as small as 200 pm (Hirschmann, 1956).
Moreover, when cysts are extracted from sandy soils, a large amount of sand particles can
get retained in a #60 sieve, increasing the difficulty of separating the cysts from soil.
Another difficulty in cyst extraction fiom soil is to obtain a good sub-sample of soil. The
soil is usually mixed with water by hand to dislodge large clumps. The elutriator
enhances this step by mixing the soil and water with pressurized air. In addition, it can
significantly reduce the amount of work required to analyze a large number of samples
and the variability due to human error in the extraction process (Byrd et al., 1976).

Because of the variability of soil texture and structure, pre-treating the soil with a
soap solution to release cysts trapped in small soil aggregates has been proposed (Tylka,
personal communication). However, there is no published information. Pre-treating the
soil with a soap solution, removing the sieve #60 fi'om the elutriator setup, and collecting
cysts in a 75-um pore sieve (mesh #200) could potentially increase extraction efficiency.
Before I started my research, therefore, I tested efficiency of four methods of cyst

extraction using a semiautomatic elutriator to extract cysts from sandy loam, muck and

181

clay loam soils. The contamination level for each method, and the efficiency for

extracting cysts fi'om low to high inoculum levels were also tested for each soil type.

MATERIALS AND METHODS
Soil types, SCN inoculum and Experimental Design

The experimental soils used were clay loam collected from a plowed field at
MSU-Collins Road facility, muck obtained from MSU-Muck Farm Research Facility,
and sterilized sandy loam obtained fiom MSU-Greenhouse facilities. The experimental
soil was artificially inoculated with SCN cysts, mostly yellow and brown, obtained from
commercial soybean fields in Shiawassee and Saginaw Counties, MI. Cysts extracted
from soil were counted under a stereoscopic microscope and placed in test tubes in 10 ml
tap water at the right inoculum level. Inoculum levels were 0 (control), 50, 100 and 150
cysts. Inocula were kept at 4°C until used to inoculate the experimental soils for the
efficiency test within 48 hours of extraction.

Elutriator and Extraction Methods

The equipment used for this test was a semiautomatic elutriator manufactured at
the University of Georgia in 1999 (Research Services Instrument Shop, The University of
Georgia, Athens, GA), described by Byrd, et al. (1976) as North Carolina Elutriator (NC-
EL). A semi-automatic elutriator consists of four funnels where four soil samples are
simultaneously mixed with water using pressurized air (40 psi). As the funnels fill up, the
overflowing soil-water suspension is poured through an SSO-um pore sieve (mesh #20) to
retain large soil particles and plant debris. Sieves are positioned at an angle and

showerheads spray water on the sieves to help the soil pass through and into another

182

funnel that directs the flow to a sample splitter. The sample splitter is a metal bowl that
collects nematodes and small soil particles suspended in water and drains through 15
Tygon tubes into a sieve where cysts are retained.

Four extraction methods were tested: Method 1 consisted on placing a sieve #60
underneath the sieve #20 to collect cysts directly fiom the spout after mixing the soil with
water. Everything that passed through the sieve #60 was discarded. Method 2 consisted
on placing the sieve #20 at the spout and a sieve #200 to collect cysts fi'om the sampler
splitter. Seven aliquants were collected because the volume of water and the amount of
soil particles flowing from 15 tubes overflowed the sieve. Methods 3 and 4 consisted of
Methods 1 and 2, respectively, with pretreatment of the soil with a soap solution. The
soap solution was prepared by dissolving one 50 oz. box of Electr'asolTM dishwasher soap
in 18 L of hot water (Tylka, personal communication). Soil samples were prepared for
elutriation for all treatments as follows except when noted. Four one-liter plastic beakers
were filled with 350 ml of tap water. For methods 3 and 4 one hundred milliliters of soap
solution were placed in the beakers and then tap water was added to 350 ml. The
appropriate nematode inoculum was added and the volume was adjusted to 400 ml. To
measure soil volume independent of moisture or air space, soil was added into the beaker
until a final volume of 500 ml was reached. The suspension was thoroughly mixed, let sit
for 5 minutes, stirred again, and sit for another 5 rrrinutes before elutriation began. After
the elutriation cycle was over (approximately 4 minutes) the material collected in sieves
#60 or #200 was transferred to 50 m1 centrifuge tubes and spun for 5 minutes at 4000
rpm. The supernatant was discarded. The tubes were then filled with a sucrose solution

(454 g sugar in 1 L DI water) and the pellet was stirred with a spatula until it was

183

completely dispersed. Tubes were then spun for 30 seconds at 4000 rpm. The supernatant
was poured through a sieve #200 and cysts retained were transferred to test tubes after
rinsing off the sucrose solution with tap water. The procedure was based on the sugar
flotation centrifugation method described by Dunn (1969). Cysts were then counted
under a stereoscopic microscope and counts from Methods 2 and 4 were adjusted to
estimate the total number of cysts per sample if all 15 tubes had been used. The elutriator
was rinsed carefully with a fine hi gh-pressure handheld nozzle after each run to prevent
contamination with cysts that may have been trapped in the apparatus.

Efficiency tests were run separately for each type of soil. The experimental design
was a randomized complete block design blocked by method, with cyst inoculum levels
as the treatment factor with ten replications (runs). Treatments were randomly assigned to
each funnel in each run. The number of cysts recovered fiom the inoculum level zero was
subtracted from the number of cysts recovered from the other inoculum levels in each run
to account for contamination. The level of contamination was reported as the average
number of cysts among all runs recovered fi'om the inoculum level zero within each
method. Efficiency of each method for each inoculum level was calculated as the
percentage of cysts recovered from the inoculum.

Statistical Analysis

The hypothesis whether the four methods tested were equally efficient for the
three soil types was tested with ANOVA. The effect of each inoculum level on the
efficiency of each method was tested for each soil by calculating least significant
differences between means at the 5 % level. The statistical analysis was performed with

SAS Release 8 (SAS Institute, Cary, NC).

184

RESULTS

Method Efficiency by Soil Type
Clay Loam: Methods 2, 3 and 4 were equally efficient to extract cysts from clay loam,
and significantly more efficient than Method 1 (Table A.l). Logistically, either method
with soap worked well without complications. One person was enough to handle the
whole process effectively. The two methods without soap also ran well, except for the
centrifugation step after Method 1. The soil did not stick to the centrifuge tubes therefore
pouring off excess water without losing sample was difficult.
Muck: Method 4 was the most efficient for muck, significantly better than either method
using a sieve # 60 (Table A.1). Methods 1 and 3 required shower nozzles pointed away
from the sieves and two operators tapping the sieves continuously because the soil
clogged the sieve openings and the sample tended to overflow. Two centrifuge tubes
were needed per sample because of the large amount of soil trapped in the sieves with the
cysts. Flow was improved with the pretreatment of the soil with soap but still two
operators were required tapping the sieves continuously to prevent clogging. Methods 2
and 4 required very little tapping by one person and one centrifuge tube per sample.
Sandy Loam: Either method with soap was significantly more efficient to extract cysts
from sandy loam than either method without soap (Table A. 1). In method 1, the sand
particles clogged the sieves very easily causing them to overflow, and requiring two
operators tapping the sieves to facilitate the flow. Soap improved the operation, but still

some tapping was required. Both methods using the # 200 sieve presented no problems.

185

Table A. l. Efficiency (%) of four elutriation methods to extract SCN cysts from
artificially inoculated soil.

 

 

 

Extraction Method
Soil 1- #60'1' 2- #200L 3- #60 + Soapi 4-7 #200 + Soap
Clay Loam 43 i 40.2 b 61 i 65.4 a 68 i 57.5 a 58 :1: 50.4 a
Muck 15i22.3b 27:1:35.2b 19i24.0b 39i31.4a
Sandy Loam 22:17.1b 20i17.7b 51 i42.5a 60i38.7a

 

Efficiency means in a row followed by the same letter were not significantly different
(protected LSD, 5% level).
'1' Sieve #60: 250p. opening
I Sieve #200: 7511 opening
§ A soap solution was added to the soil suspension in Method 3 and 4.
Contamination

The level of contamination, that is the number of cysts recovered from the control, was
almost zero for sandy soil, low for muck, and high for clay loam (Table A2). For muck
and clay loam, contamination levels were lower with Methods 2 and 4.

Method Efficiency by Inoculum levels, by Soil Type
Clay Loam: For low inoculum levels (50 cysts/ 100 cm3 of soil), Methods 2 and 3 were
significantly more efficient than method 1. At a higher cyst inoculum (100 cysts/ 100
cm3 of soil) there were no significant differences in efficiency among methods. For 150
cysts/ 100 cm3 of soil, Methods 2, 3 and 4 were all more efficient than method 1 (Figure
A. 1).

Table A2. Average number of cysts recovered fiom the control sandy loam, muck and
clay loam samples, using different elutriation methodsxl'

 

 

 

Soil
Extraction Method Clay Loam Muck Sandy Loam
l- #601 37.3 4.4 0.6
2- #200§ 13.1 1.4 1.0
3- #60 + Soap# 35.3 5.4 0.1
4- #200 + Soap 10.3 1.2 0.5

 

'1‘ The number of cysts presented was the average of ten replications (runs).
1 Sieve #60: 2501: opening

§ Sieve #200: 7511 opening

# A soap solution was added to the soil suspension in Method 3 and 4.

186

Muck: At 50 cysts/ 100 cm3 of soil there were no differences among methods. For 100
cysts/ 100 cm3 of soil Method 4 was better than Methods 1 and 3. Method 2 was better
than 1 and not different from the other two methods. For 150 cysts/ 100 cm3 of soil
Method 4 was better than Methods 1 and 3; Methods 1, 2 and 3 were equally poor in

efficiency (Figure A.1).

 

140

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ClayLoam
120<
100-
so i I
I.
604 P
401 I
zoi
l;
0..
140
Muck —mreo
1204 —2-9200
.0 —3-#60+Soap
g 100- :214-t200+$oap
>
0 80+
8
g 601
5 40+ E
32 . .

201 .i 51

 

 

 

 

 

 

 

 

 

 

 

120 . Sandy Loam
100 -

00 J

4.. .2

20 l

o . ~-
50 100 150
Cysts added to 100 cm3 of soil

Fig. A. 1. Extraction efficiency (means 3: sd) of four elutriation methods to recover SCN
cysts from sandy loam, muck and clay loam artificially inoculated with 50, 100 and 150
cysts 100 cm'3 of soil.

187

Sandy Loam: For 50 and 100 cysts/ 100 cm3 of soil, Methods 3 and 4 were better than
Methods 1 and 2. Method 2 was not tested with 150 cysts/ 100 cm3 of soil, but Methods

3 and 4 were better than Method 1 (Figure A.l).

DISCUSSION

The North Carolina Elutriator (N C-EL) was adequate to extract SCN cysts from
sandy loam and clay loam soils, with efficiency of 60%, whereas the procedure presented
several difficulties when working with muck soil, with a great reduction in efficiency.
The method using the sieve #60 without soap treatment was the least efficient for all soil
types. The most efficient extraction method for the three soil types tested was to mix the
soil with soap solution and collect cysts in the sieve #200. The method with soap and a
sieve #60 was equally efficient regarding the percentage of cysts recovered fiom clay
loam and sandy loam, but logistically, this method required more labor than with the
sieve #200, especially with sandy loam soil.

The source of the contamination was probably cysts retained in the elutriator from
one run to the next. The highest level of contamination was observed with the clay loam.
However, this soil was obtained from a field that could have been infected with SCN
before it was taken to the lab. Therefore, even though contamination appeared lower with
soap and the #200 sieve than with the other methods, this portion of the test should be
conducted again with SCN-free soil.

Extraction efficiencies from clay loam and sandy loam did not vary significantly

with increasing inoculum level. There was great variability in extraction efficiencies fiom

188

muck, and the poor efficiency overall, made it difficult to draw conclusions from the
comparison of inoculum levels.

SCN is usually found in soils with much less organic matter content than that
found in muck. The results of this test indicate that for the soil types usually inhabited by
SCN, 60% cyst extraction efficiency can be obtained by using a semi-automatic elutriator

equipped with #200 sieves, and pre-treating soil samples with soap solution.

189

APPENDIX B

190

SOFTWARE FOR GEOSTATISTICS AND SPATIAL ANALYSIS

A great variety of software packages are available for spatial analysis,
geostatistics, and GIS. Some of them can be expensive, while others can be downloaded
from the Internet as freeware. The AI-GEOSTATS web site, a hub for geostatistics users
around the world (http://www.ai-geostats.org) offers a complete list of the software
available with a brief description of main functions and links to their homepages, from
where many of them can be downloaded.

There is not one program that has all of the tools required for a complete
geostatistical analysis. The spatial analyst rather, needs a combination of packages to
have a complete ‘geostatistical tool box’. The selection of the right programs to choose
should be made based on the tools required to achieve the research objectives, the
platform available (DOS, Unix, Windows, Linux), the user’s computer skills and
preferences, and affordability.

Software for geostatistics, spatial analysis, and GIS are upgraded frequently with
add-on codes, new versions, and new programs, constantly extending and improving the
tools offered. It is recommended that the spatial analyst test several packages to find the
one or the combination of packages that better fits the needs of the research project.

The software packages used for the geostatistical analysis presented in this
dissertation were Surfer 7, Variowin 2.1, and SAS 8. GSLIB was initially used for
semivariogram analysis, but it was soon replaced by the other packages, which are easier
to work with. Surfer 7 was used to calculate empirical semivariograms, model

semivariograms, kriging, and to draw contour maps. The overlaying tool only available in

191

Surfer, allowed showing more than one layer of information in one map. Variowin 2.1
was used mostly to calculate cross-correlograms. SAS 8 and GSLIB were used initially to
calculate empirical semivariograms. ArcView (ESRI) was also tried for its mapping and
GIS capabilities. Because of the lack of geostatistical tools and because it required high
expertise level to take firll advantage of its capabilities ArcView was not used further. For
GIS projects however, it is probably the most complete packages available
(http://www.esri.com). Following is a brief description of the main pros and cons I have
come across in the software tested during this research (Tables B.1- 4). The lists shown in
Tables B. 1 - 4 are meant to point out the obstacles encountered and highlight the positive
aspects of each package used and not as an exhaustive software analysis. For detailed

information and comparisons with more packages refer to the AI-GEOSTATS web site.

192

Table B. 1. Summary of some of the benefits and drawbacks of the software package

Surfer 7.1’

 

PROS

CONS

 

Windows platform, user fiiendly

Cross-validation not available in version 7,
new in version 8

 

Semivariogram modeling tools, with or
without drift

Limited basic statistics

 

Probably the most complete set of
semivariogram models offered

Univariate analysis only, no cross-
correlograms or cross-variography.

 

Interactive isotropic and anisotropic
semivariogram modeling

Does not provide empirical semivariogram
values

 

Multiple interpolation and contouring
tools available

Cost

 

Many fimctions for grid analysis

 

Excellent plotting quality and
capabilities

 

Excellent mapping quality and
capabilities

 

Data sets can be used fi'om Excel
spreadsheets

 

Data transformation tools

 

Imported maps, images, figures can be
digitized easily

 

User’s Manual is helpful and easy to
follow

 

 

 

'1‘ Version 8 available in 2002. Golden Software. Golden, Colorado.

http://www.goldensoftware.com

193

 

Table B.2. Summary of some of the benefits and drawbacks of the software package
Variowin 2.21.1’

 

 

 

 

 

 

PROS CONS
Windows platform Data sets should be carefully prepared as
simple text format (*.txt or *.dat)
Provides empirical semivariogram Restrictions on file name length
values
Calculates variograms, madograrns, Poor graphing quality
correlograms, standardized variograms,
covariance
Interactive modeling No interpolation tools
Bivariate analysis capabilities (cross- The manual is out of print, may become
correlo gram and cross-variography) available online soon

 

Semivariogram map

 

Free

 

 

 

 

T Yvan Pannatier. Shell International Exploration and Production. The Hague,
Netherlands. http://www-sst.unil.ch/research/variowin/index.html

Table B.3. Summary of some of the benefits and drawbacks of the software package
SAS/STAT.’[

 

PROS CONS

 

Excellent statistical analysis capabilities Limited semivariogram modeling
beyond spatial analysis

 

Variogram and ordinary kriging Knowledge on SAS-code is a must

 

Improvements should be seen in the near Requires SAS. License should be renewed
firture annually.

 

 

 

 

T The SAS Institute. Cary, North Carolina. http://www.SAS.com

194

Table B.4. Summary of some of the benefits and drawbacks of the software package
GSLIBT

 

PROS CONS

 

Powerful capabilities Executables and codes are needed to fully
benefit from GSLIB capabilities

 

One of the few packages capable of 3D Minor mistakes in data sets or variable inputs
geostatistical analysis will cause the program to stop running

 

Jack-knifing and other cross-validation Slow with large data sets
tools

 

Platform DOS, UNIX, new version with Data sets should be carefully prepared as

 

 

Windows interface. simple text format (*.txt or *.dat)

DOS and UNIX versions are free Programming skills required

Indicator semivariograms, and cross- Poor graphing quality (Postscript output)
variography

 

The book is indispensable, even to
accomplish the simplest tasks.

 

 

 

 

T Clayton V. Deutsch and Andre G. J oumel (1992). Stanford University.
http://www.gslib.com

195

APPENDIX C

196

Record of Deposition of Voucher Specimens“
The specimens listed on the following sheet(s) have been deposited in the named
museum(s) as samples of those species or other taxa, which were used in this research.
Voucher recognition labels bearing the Voucher No. have been attached or included in
fluid-preserved specimens.
Voucher No.: 2002-13

Title of thesis or dissertation (or other research projects):

Characterization of the spatial distribution of Heterodera glycines Ichinohe 1955

(N ematoda), soybean cyst nematode in two Michigan fields

Museum(s) where deposited and abbreviations for table on following sheets:

Entomology Museum, Michigan State University (MSU)

Other Museums:

Investigator’s Name(s) (typed)
Maria Felicitas Avendafio

 

 

Date 12/13/02

*Reference: Yoshimoto, C. M. 1978. Voucher Specimens for Entomology in North .
America.
Bull. Entomol. Soc. Amer. 24: 141-42.

Deposit as follows:
Original: Include as Appendix 1 in ribbon copy of thesis or dissertation.

Copies: Include as Appendix 1 in copies of thesis or dissertation.
Museum(s) files.

Research project files.

This form is available from and the Voucher No. is assigned by the Curator, Michigan
State University Entomology Museum.

197

Voucher Specimen Data

Pages

 

 

 

 

 

 

 

 

 

 

Page 1 ofl

 

 

 

 

 

 

 

 

 

\ . 2.5 t 98.6
3x \\\. \rxt‘ 832a. 38
32 Sam
$835 w swag: as e 588
8m 8233mm 3...: 26% 05 @0283— o§=o>< mama—om $.82
m 788 62 .8525 3093 3252 muoawumgfl
A508: a 38:... 335%? one
88 a .20: 53:22
3m: 5550 oafisaflm 23 ES... vogue—Bo
awe—088.5 - w .. .. - .. 398 82.. 3.523 89a @88be $30 859$» Senses:
m m m r M M .m. a voices—u :35 65° .8 86on
m m m. e .w .m m. m m w 98 com: 8 venue—.8 80589.“ Bu 8% .33
M w .m. m A A m N E

 

 

0

53:52

 

 

198

LITERATURE CITED

199

Acedo, J .R., V.H. Dropkin, and V.D. Luedders. 1984. Nematode population attrition and
histopathology of Heterodera glycines-soybean associations. J. Nematol. 16:48-57.

Agrios, G.N. 1997. Plant Pathology. Academic Press, New York.

Alby, T., J .M. Ferris, and V.R. Ferris. 1983. Dispertion and distribution of Pratylenchus
scribneri and Haplolaimus galearus in soybean fields. J. Nematol. 15:418-426.

Alston, D.G., and DP. Schmitt. 1987. Population density and spatial pattern of
Heterodera glycines in relation to soybean phenology. J. Nematol. 19:336-345.

Anand, S.C., S.B. Sharma, A.P. Rao-Arelli, and J .A. Wrather. 1994. Variation in parasitic
potential of Heterodera glycines populations. Crop Sci. 34: 1452-1454.

Anderson, T.R., T.W. Welacky, T.H. Olechowski, G. Ablett, and BA. Ebsary. 1988.
First report of Heterodera glycines in Ontario, Canada. Plant Dis. 72:453.

Anscombe, F. J. 1950. Soil sampling for potato root eel worm cysts. Ann. Appl. Biol.
37:286-295.

Atkinson, H.J., K.S. Blundy, M.C. Clarke, E. Hansen, G. Harper, V. Koritsas, M.J.
McPherson, D. O’Reilly, C. Scollan, S.R. Tumbull, and PE. Urwin. 1994. Novel
plant defenses against nematodes. p. 197-210. In F. Larnberti et al. (eds) Advances in
molecular plant nematology. Plenus Press, New York and London.

Avendano, M.F., F.J. Pierce, 0. Schabenberger, and H. Melakeberhan. 2001. Application
of geostatistical tools to assess the spatial distribution of Heterodera glycines.
Phytopathology 91 :S l 30 (abstr.).

Avendafio, M. F., Schabenberger, 0., Pierce, F. J ., Melakeberhan, H. 2003. Geostatistical
analysis of field spatial distribution patterns of soybean cyst nematode. Agronomy
Journal (accepted) Manuscript A02-106T

Barker, K.R. 197 8. Determining nematode population responses to control agents. p. 114-
125. In E. Zehr (ed.) Methods for evaluating plant fungicides, nematicides and
bactericides. American Phytopathological Society, St. Paul, MN.

Barker, KR. 1989. Yield relationships and population dynamics of Meloidogynes spp. on
flues-cured tobacco. Suppl. J. Nematol. 21: 597-603.

Barker, K.R., and SR. Koenning. 1989. Influence of soil moisture and texture on
Heterodera glycines/ Glycines max interactions. J. Nematol. 21 :550 (abstr.).

Barker, K.R., and OJ. Nusbaum. 1971. Diagnostic and advisory programs. p. 281-301. In
B.M. Zuckerrnan, W.F. Mai, and RA. Rohde (eds.) Plant parasitic nematodes, Vol 1.
Academic Press, New York.

200

Barker, K.R., and CL. Campbell. 1981. Sampling nematode populations. p. 451-474. In
B.M. Zuckennan, and RA. Rohde (eds.) Plant parasitic nematodes, Vol III.
Academic Press, New York.

Basso, B., J .T. Ritchie, F.J. Pierce, R.P. Braga, and J .W. Jones. 2001. Spatial validation
of crop models for precision agriculture. Agric. Syst. 68:97-112.

Bird, AR, and B.R. Loveys. 1975. The incorporation of photosynthates by Meloidogyne
javanica. J. Nematol. 7:111-113.

Bird, G.W., J .F. Davenport, and C.Chen. 1988. Potential role of Heterodera glycines in
dry bean production in Michigan. J. Nematol. 20: 628-629 (abstr.).

Blaxter, M. 1998. Caenorhabditis elegans is a nematode. Science 282: 2041-2046.

Blaxter, M.L., P. De Ley, J.R. Garey, L.X. Liu, P. Scheldeman, A. Vierstraete, J.R.
Vanfleteren et al. 1998. A molecular evolutionary framework for the phylum
nematoda. Nature 392: 71-75.

Blevins, D.G., V.H. Dropkin, and V.D. Luedders. 1995. Macronutrient uptake,
translocation, and tissue concentration of soybeans infested with the soybean cyst

nematode and elemental composition of cysts isolated from roots. J. Plant Nutr.
18:579-591.

Boag, B. 1998. Survey, surveillance and crop loss assessment. p. 123-140. In S.B.
Sharma (ed.) The cyst nematodes. Kluwer Academic Publishers, London.

Boag, B., and RB. Topham. 1984. Aggregation of plant parasitic nematodes and Taylor’s
Power Law. Nematologica 30:348-357.

Boerma, H.R., and RS. Hussey. 1992. Breeding plants for resistance to nematodes. . J.
Nematol. 24:242-252.

Bongers, T. 1990. The maturity index: an ecological measure of environmental
disturbance based on nematode species composition. Oecologia 83: 14-19.

Bongers, T., H. van der Meulen, and G. Korthals. 1997. Inverse relationship between the

nematode maturity index and plant parasite index under enriched nutrient conditions.
Appl. Soil Ecol. 6:195-199.

Bonner, M.J., and DP. Schmitt. 1985. Population dynamics of Heterodera glycines life
stages on soybean. J. Nematol. 17:153-158.

Bostian, A.L., D.P. Schmitt, and KR. Barker. 1984. Early growth of soybean as altered
by Heterodera glycines, Phenarrriphos and/or Alachlor. J. Nematol. 16: 41-47.

201

Bradley, C.A., W.L. Pedersen, S.E. Hart, and L.M. Wax. 1996. Effect of densities, seed
protectants, and herbicides on no-tilled drilled soybeans. Phytopathology 86:S69
(abstr.).

Bridge, J. 1996. Nematode management in sustainable and subsistence agriculture. Annu.
Rev. Phytopathol. 34:201-225.

Byrd, D.W.Jr., K.R. Barker, H. Ferris, C.J. Nusbaum, W.E. Griffin, R.H. Small, and GA.
Stone. 1976. Two semi-automatic elutriators for extracting nematodes and certain
fungi from soil. J. Nematol. 8: 206-212.

Byrd, D.W.Jr., T. Kirkpatrick, and KR. Barker. 1983. An improved technique for
clearing and staining plant tissues for detection of nematodes. J. Nematol. 15:142-
143.

Caballero, R., M. Arauzo, and P.J. Hemaiz. 1996. Accumulation and redistribution of
mineral elements in common vetch during pod filling. Agron. J. 88:801-805.

Cai, D., M. Kleine, S. Kifle, H.-J. Harloff, N.N. Snadal, K.A. Marcker, R.M. Klein-
Lankhorst, E.M.J. Salentijn, W. Lange, W.J. Stiekema, U. Wyss, F.M.W. Grundler,
and C. Jung. 1997. Positional cloning of a gene for nematode resistances in sugar
beet. Science 275: 32-34.

Cassel, D.K., O. Wendroth, and DR. Nielsen. 2000. Assessing spatial variability in an
agricultural experiment station field. Agron. J. 92:706-714.

Chen, S.Y., P.M. Porter, J.H. Orf, C.D. Reese, W.C. Stienstra, N.D. Young, D.D.
Walgenbach; P.J. Schaus, T.J. Arlt, and FR. Bretenbach. 2001. Soybean cyst
nematode population development and associated soybean yields of resistant and
susceptible cultivars in Minnesota. Plant Dis. 85: 760-766.

Chitwood, B.G., and MB. Chitwood. 1950. An introduction to nematology. Monumental
Printing, Co., Baltimore, MD.

Cobb, N. A. 1918. Estimating the nema population of the soil. Agric. Tech. Circ.l, US
Department of Agriculture, Washington, 1-48.

Cressie, N.A.C. 1993. Statistics for spatial data. Revised ed. J. Wiley and Sons, New
York.

CTIC- Conservation Technology Information Center. 2000. National Crop residue
management Survey News Release- Midwest Version [Online]. CTIC, West
Lafayette, IN. Available at http://www.ctic.purdue.edu/Core4/CT/CTSurvey/
NationalData.html (Posted 11 Oct. 2002, verified 9 Nov. 2002).

Davis, J .C. 1986. Statistics and data analysis in geology. 2nd ed. John Wiley and Sons,
New York.

202

Diers, B.W., P. Arelli, and SR. Cianzio. 1999. Management of SCN through
conventional breeding for resistance- Midwest perspective. p. 5. In First National _
Soybean Cyst Nematode Conference Proceedings. Orlando, FL.

Doerge, T. 1998. Defining management zones for precision farming. Pioneer Hi-Bred
International Co. Crop Insights, vol. 8, No 21.

Donald, P. 1999. University of Missouri Extension Nematology Lab. Available online at
http://agebb.missouri.edu/nemaext/neextra.htm (verified 1 Dec. 2002).

Donald P.A., K.A. Sudduth, and NR. Kitchen. 2001. Mapping soybean cyst nematode
field distribution. Phytopathology 91:8133 (abstr.).

Donald, P.A., W.W. Donald, A]. Keaster, R.J. Kremer, J.A. Kendig, B.S. Sims, and J.
Mihail. 1999. Changes in Heterodera glycines egg population density in continuous
Glycine max over four years. J. Nematol. 31: 45-53.

Doucet, ME, and P. Lax. 1999. Presence of the nematode Heterodera glycines
(N ernatoda: Tylenchida) associated with soybean in Argentina. Nematology 1:213-
216.

Dropkin, V.H., C.H. Baldwin, T. Gaither, and W. Nace. 1976. Growth of Heterodera
glycines in soybeans in the field. Plant Dis. Rep. 60:977-980.

Dropkin, V.H. 1980. Introduction to plant nematology. John Wiley and Sons, New York.

Duncan, L.W., and R. McSorley. 1987. Modeling nematode populations. p. 377-389. In

J .A. Veech, and D.W. Dickson (eds.) Vistas on nematology. Society of Nematology,
Hyattsville, MD.

Duncan, L.W., J .J . Ferguson, R.A. Dunn, and J.W. Noling. 1989. Application of Taylor’s
Power Law to sample statistics of Tylenchulus semipenetrans in Florida citrus. J.
Nematol. 21 :707-71 1.

Duncan, L.W., and J .W. Noling. 1998. Agricultural sustainability and nematode
integrated pest management. p. 251-287. In K.R. Barker, G.A. Pederson, and G.L.
Windham (eds.) Plant and nematode interactions. Agronomy Monographies 36. ASA,
CSSA, and SSSA, Madison, WI.

Dunn, R.A. 1969. Extraction of cysts of Heterodera species fi'om soils by centrifugation
in high density solutions. J. Nematol. 1:7 (abstr.).

Endo, B.Y. 1978. Feeding plug formation in soybean roots infected with the soybean cyst
nematode. Phytopathology 68: 1022-103 1 .

Endo, B.Y. 1992. Cellular responses to infection. p. 37-49. In R.D. Riggs, and J.A.
Wrather (eds) Biology and management of the Soybean Cyst Nematode. The
American Phytopathological Society, St. Paul, MN.

203

Epps, J .M., L.D. Young, and BE. Hartwig. 1981. Evaluation of nematicides and resistant
cultivar for control of soybean cyst nematode race 4. Plant Dis. 65:665-666.

Etterna, C. H. 1998. Soil nematode diversity: species coexistence and ecosystem function.
J. Nematol. 30:159-169.

Evans, K., J. Stafford, R. Webster, P. Halford, M. Russell, A. Barker, and S. Griffin.
1998. Mapping potato cyst nematode p0pulations for modulated applications of
nematicides. Aspects Appl.. Biol. 52:1-8.

Evans, K., R.M. Webster, P.D. Halford, A.D. Baker, and MD. Russell. 2002. Site
specific management of nematodes: Pitfalls and practicalities. J. Nematol. 34:194-
199.

FAQ (Food and Agriculture Organization of the United Nations) databases. 2002.
[Online] Available at http://www.fao.org/page/collection?subset=agriculture (verified
3 Dec 2002).

Ferris, H. 197 8. Nematode economic thresholds: derivation, requirements, and theoretical
considerations. J. Nematol. 8:225-263.

Ferris, H. 1984. Probability range in damage predictions as related to sampling decisions.
J. Nematol. 16:246-251.

Ferris, H. 1993. New frontiers in nematode ecology. J. Nematol. 25:374-382.

Ferris, H., and J.W. Noling. 1987. Analysis and prediction as a basis for management
decisions. p.49-85. In R.H. Brown, and B.R. Kerry (eds.) Principles and practice of
nematode control in crops. Academic Press, Orlando, FL.

Ferris, H., T.A. Mullens, and KB. Foord. 1990. Stability and characteristics of spatial
description parameters for nematode populations. J. Nematol. 22:427-439.

Fidler, J .H., and W.J. Bevan. 1963. Some soil factors influencing the density of cereal
root eelworrn (Heterodera avenae Woll.) populations and their damage to the oat
crop. Nematologica 9:412-420.

Fox, TC, and ML. Guerinot. 1998. Molecular biology of cation transport in plants.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 49:669-696.

Francl, L.J. 1986a. Spatial analysis of Heterodera glycines populations in field plots. J.
Nematol. 18:183-189.

Francl, L.J. 1986b. Heterodera glycines population dynamics and relation of initial
population to soybean yield. Plant Dis. 70:791-795.

Francl, L]. 1993. Multivariate analysis of selected edaphic factors and their relationship
to Heterodera glycines population density. J. Nematol. 25:270-276.

204

Gaston, LA, MA. Locke, R.M. Zablotowicz, and K.N. Reddy. 2001. Spatial variability
of soil properties and weed populations in the Mississippi Delta. Agron. J. 65:449-
459.

Gavassoni, W.L., G.L. Tylka, and GP. Munkvold. 2001. Relationships between tillage
and spatial patterns of Heterodera glycines. Phytopathology 91 :534-545.

Gee, G.W., and J .W. Bauder. 1986. Particle-size analysis. p. 383-411. In A. Klute (ed.)
Methods of soil analysis, Part 1 Physical and Mineralogical methods. Agronomy

Series No 9. American Society of Agronomy, Soil Science Society of America,
Madison, WI.

Golden Software Inc. 1999. User’s Guide. Golden, CO.

Goodell, P., and H. Ferris. 1981. Plant parasitic nematode distributions in alfalfa field. J.
Nematol. 12:136-141.

Goovaerts, P. 1997. Geostatistics for natural resources evaluation. Oxford University
Press, New York.

Grau, C.R., A.E. MacGuidwin, E.S. Oplinger, and NC. Kurtzweil. 1999. Relationship of
soil pH and population density of the soybean cyst nematode. p.24. In Conference
Proceedings, National Soybean Cyst Nematode Conference. Orlando, Fl.

Gupta, V.V.S.R. 1994. The impact of soil and crop management practices on the
dynamics of soil microfauna and mesofauna. p. 107-124. In C.E. Pankhurst, B.M.
Doube, V.V.S.R. Gupta, and PR. Grace (eds.) Soil biota: management in sustainable
farming systems. CSIRO Press, Melbourne, Australia.

Hanson, R.G., J.H. Muir, P.M. Sims, and J.K. Boon. 1988. Response of three soybean
cultivars to cyst nematode and KCL fertilization. J. Prod. Agric. 1:327-331.

Hartwig, BE. 1981. Breeding productive soybean cultivars resistant to the soybean cyst
nematode for the southern United States. Plant Dis. 65:303-307.

Haystead, A., and J.I. Sprent. 1981. Symbiotic nitrogen fixation. p. 345-364. In C.B.
Johnson (ed.) Physiological processes limiting plant productivity. Butterworths,
London, Boston.

Heal, O.W., and J. Dighton. 1985. Resource quality and trophic structure in the soil
system. p. 339-354. In A.H. Fitter, D. Atkinson, D.J. Read, and M.J. Usher (eds.)
Ecological interactions in the soil. Blackwell Scientific Publications, Oxford, UK.

Heald, CM. 1987. Classical nematode management practices. p. 100-104. In J .A. Veech,
and D.W. Dickson (eds.) Vistas on Nematology. Society of Nematologists, Inc.,
Hyattsville, MD.

205

Hill, NS, and DP. Schmitt. 1989. Influence of temperature and soybean phenology on
dormancy induction of Heterodera glycines. J. Nematol. 21:361-369.

Hirschmann, H. 1956. Comparative morphological studies on the soybean cyst nematode,
Heterodera glycines and the clover cyst nematode, H. trifolii (N ematoda:
Heteroderidae). Proc. Helminthol. Soc. Wash. 23: 140-151 .

Hirschmann, H. 1971. Comparative morphology and anatomy. p. 11-63. In B.M.
Zuckerrnan, W.F. Mai, and R.A. Rohde (eds.) Plant parasitic nematodes.Vol 1.
Academic Press, New York.

Hirunsalee, A., K.R. Barker, and MK. Beute. 1995. Effects of peanut-tobacco rotations
on population dynamics of Meloidogyne arenaria in mixed race populations. J.
nernatol. 27:178-188.

Howard, D.D., A.Y. Chambers, and GM. Lessman. 1998. Rotation and fertilization
effects on corn and soybean yields and soybean cyst nematode populations in a no-
tillage system. Agron. J. 90:518-522.

Huang, J. and KR. Barker. 1983. Influence of Heterodera glycines on Leghemoglobins
of soybean nodules. Phytopathol. 73:1002-1004.

Huck, M.G., C.M. Peterson, G. Hoogenboom, and CD. Busch. 1986. Distribution of dry
matter between shoots and roots of irrigated and non irrigated determinate soybeans.
Agron. J. 78:807-813.

Hung, Y. 1958. A preliminary report on the plant parasitic nematodes of soybean crop of
the Pingtung District, Taiwan, China. Agriculture Pest News 5:1-5.

Hussey, RS, and HR. Boerma. 1983. Influence of planting date on damage to soybean
caused by Heterodera glycines. J. Nematol. 15:253-258.

Ichinohe, M. 1955. Studies on the morphology and ecology of the soybean nematode
Heterodera glycines, in Japan. Report of the Hokkaido Agriculture Experimental
Station 48: 1-64.

Johnson, A.B., H.D. Scott, and RD. Riggs. 1993. Penetration of soybean roots by
soybean cyst nematode at high soil water potentials. Agron. J. 85:416-419.

Johnson, CS. 1989. Managing root-knot nematode on tobacco in the southern United
States. Suppl. J. Nematol. 21:604-608.

Johnson, CS. 1990. Control of Globodera tabacum solanacearum by rotating susceptible
and resistant flue-cured tobacco cultivars. Suppl. J. Nematol. 22:700-706.

Jones, B]. 1998. Plant Nutrition Manual. CRC Press, Boca Raton, FL.

206

Jones, F.G.W., D.W. Larbey, and D.M. Parrott. 1969. The influence of soil structure and
moisture on nematodes, especially Xiphinema, Longidorus. Trichodorus and
Heterodera spp. Soil Biol. Biochem. 1:153-165.

Journel, AG, and CH. Huijbregts. 1978. Mining geostatistics. Academic Press, New
York.

Koenning, SR. 2000. Density-dependent yield of Heterodera glycines-resistant and —
susceptible cultivars. J. Nematol. 32:502-507.

Koenning, S.R., and KR. Barker. 1995. Soybean photosynthesis and yield as influenced
by Heterodera glycines, soil type and irrigation. J. Nematol. 27:51-62.

Koenning, S.R., D.P. Schmitt, and KR. Barker. 1993. Effects of cropping systems and
population density of Heterodera glycines and soybean yield. Plant Dis. 77:780-786.

Koenning, S.R., D.P. Schmitt, and KR. Barker. 1995. Impact of crop rotation and tillage
system on Heterodera glycines population density and soybean yield. Plant Dis.
79:282-286.

Koenning, S.R., S.C. Anand, and J .A. Wrather. 1988. Effect of within-field variation in
soil texture on Heterodera glycines and soybean yield. J. Nematol. 20:373-3 80.

Koenning, S.R., and BS. Sipes. 1998. Biology. p.156-190. In S.B. Shanna (ed). The cyst
nematodes. Kluwer Academic Publishers, The Netherlands.

Koenning, S.R., and SC. Anand. 1991. Effects of wheat and soybean planting date on

Heterodera glycines population dynamics and soybean yield with conventional
tillage. Plant Dis. 75:301-304.

Kraus, R.G., G.R. Noel, and DJ. Edwards. 1982. Effect of pre-emergence herbicides and
Aldicarb on Heterodera glycines population dynamics and yield of soybean. J.
Nematol. 14:452 (abstr.).

Kravchenko, AN, and D.G. Bullock. 2000. Correlation of corn and soybean grain yield
with topography and soil properties. Agron. J. 92:75-83.

Kravchenko, AN, and D.G. Bullock. 2002. Spatial variability of soybean quality data as
a function of field topography: I Spatial data analysis. Crop Sci. 42:804-815.

Lauritis, J .A., R.V. Rebois, and LS. Graney. 1983. Development of Heterodera glycines
Ichinohe on soybean, Glycine max (L.) Merr., under gnotobiotic conditions. J.
Nematol. 15:272-280.

Lawn, DA, and Noel, GR. 1986. Field inter-relationships among Heterodera glycines,

Pratylenchus scribneri, and three other nematode species associated with soybean. J.
Nematol. 21:347-355.

207

Lehman, RS. 1994. Dissemination of phytoparasitic nematodes. p.4. In Florida
Department of Agriculture and Consumer Services, Division of Plant Industry.
Nematology Circular 208.

Loneragan, J .F . 1981. Distribution and movement of copper in plants. p. 165-188. In J .F.
Loneragan, A.D. Robson, and RD. Graham (eds.) Copper in soils and plants.
Academic Press, London.

Long, J .H., and TC. Todd. 2001. Effect of crop rotation and cultivar resistance on seed
yield and the soybean cyst nematode in full-season and double-cropped soybean.
Crop Sci. 41:1137-1143.

Luedders, V.D. and V.H. Dropkin. 1983. The effect of secondary selection on cyst
nematode (Heterodera glycines) reproduction on soybeans. Crop. Sci. 23:263-264.

Luedders, V.D., J .G. Shannon, and CH. Baldwin. 1979. Influence of rate and source of
potassium on soybean cyst nematode reproduction on soybean seedlings. Plant Dis.
Rep. 63:558-560.

Maggenti, AR. 1991. Nemata: Higher classification. p.147-187. In W.R. Nickle (ed.)
Manual of agricultural nematology. Marcel Dekker, New York.

Mai, W.F., and GS. Abawi. 1987. Interactions among root-knot nematodes and Fusarium
wilt fungi on host plants. Annu. Rev. Phytopathol. 25:317-338.

Maschner, H. 1995. Mineral nutrition of higher plants. Academic Press, New York.
Matheron, G. 1963. Principles of geostatistics. Econ. Geol. 58:1246-1266.

McKeague, J.A., and C. Wand. 1982. Soil structure: concepts, distribution and
interpretation. LRRI Contribution No 82-15, Research Branch, Agriculture Canada,
Ottawa, Ontario, Canada.

McSorley, R. (ed.). 1987. Bibliography of estimated crop losses in the United States due
to plant-parasitic nematodes. Suppl. J. Nematol. 1:6-12.

McSorley, R. 1998. Population dynamics. p. 109-133. In K.R. Barker, G.A. Pederson,
and G.L. Windham (eds.) Plant and nematode interactions. Agronomy Monographies
36. ASA, CSSA, and SSSA. Madison, WI.

McSorley, R., and D.W. Dickson. 1991. Determining consistency of spatial dispersion of
nematodes in small plots. J. Nematol. 23:65-72.

McSorley, R., and J .L. Parrado. 1982. Estimating relative error in nematode numbers
from single soil samples composed of multiple cores. J. Nematol. 14:522-529.

McSorley, R., W.H. Dankers, J .L. Parrado, and J .S. Reynolds. 1985. Spatial distribution
of the nematode community on perrine marl soils. Nematropica 15:77-92.

208

McSorley, R., and MS. Philips. 1993. Modeling population dynamics and yield loss and
their use in nematode management. p. 6-85. In K. Evans, D.L. Trudgill, and J .M.
Webster (eds.) Plant parasitic nematodes in temperate agriculture. CAB Int.,
Wallingford, England.

Melakeberhan, H. 1997. Plant, nematode and nutrient relations: An overview. Jpn. J.
Nematol. 27:41-52.

Melakeberhan, H. 1997b. Role of plant nutrition on alleviating nematode parasitism. p.
759-760. In T. Ando, K. Fujita, T. Mae, H. Matsumoto, S. Mori, and J. Sekiya (eds.)
Plant nutrition for sustainable food production and environment. Kluwer Academic
Publishers, Tokyo, Japan.

Melakeberhan, H. 1999. Effects of nutrient source on the physiological mechanisms of
Heterodera glycines and soybean genotypes interactions. Nematology 1:113-120.

Melakeberhan, H. 2002. Embracing the emerging precision agriculture technologies for
site-specific management of yield -limiting factors. J. Nematol. 34:185-188.

Melakeberhan, H., G.W. Bird, and R. Gore. 1997. Impact of plant nutrition on
Pratylenchus penetrans infection of Prunus avium rootstocks. J. Nematol. 29:381-
388.

Melakeberhan, H., R.C. Brooke, J .M. Webster, and J .M. D’Auria 1985. The influence of
Meloz’dogyne incognita on growth, nutrient content and physiology of Phaseon
vulgaris. Physiol. Plant Pathol. 26:259-268.

Melakeberhan, H., H. Ferris, and J. Dias. 1990. Physiological response of resistant and
susceptible Vitis vinifera cultivars to Meloidogyne incognita. J. Nematol. 22:224-230.

Melakeberhan, H., J .M. Webster, R.C. Brooke, and J .M. D’Auria. 1988. Effect of KNO3
on C02 exchange rate, nutrient concentration and yield of Meloidogyne incognita-
infected beans. Revue Nematologie 11:391-397.

Melakeberhan, H., J.M. Webster, R.C. Brooke, J.M. D’Auria, and M. Cackette. 1987.
Effect of Meloidogyne incognita on plant nutrient concentration and its influence on
the physiology of beans. J. Nematol. 19:324-330.

Melton, T.A., B.J. Jacobsen, and GR. Noel. 1986. Effects of temperature on

development of Heterodera glycines on Glycine max and Phaseolus vulgaris. J.
Nematol. 18:468-474.

Mengel, K., and G. Geurtzen. 1988. Relationship between iron chlorosis and alkalinity in
Zea mays. Physiol. Plant. 72:460-465.

Morales, F., A. Abadia, and J. Abadia. 1990. Characterization of the xanthophylls cycle

and other photosynthetic pigment changes induced by iron deficiency in sugar beet
(Beta vulgaris L.). Plant Physiol. 94:607-613.

209

Morgan, G.D., W.R. Stevenson, A.E. MacGuidwin, K.A. Kelling, L.K. Binning, and J.
Zhu. 2002. Plant pathogen population dynamics in potato fields. J. Nematol. 34:189-
193.

Mueller, T.G., F.J. Pierce, 0. Schabenberger, and DD. Wamcke. 2001. Map quality for
site-specific fertility management. Soil Sci. Soc. Am. J. 65:1547-1558.

Niblack, TL. 1999. Long term solutions to management of soybean cyst nematode. p.
295-299. In Proceedings of the World Soybean Research Conference VI. August 4-7
Chicago, Illinois.

Niblack, TL. 2002. Soybean cyst nematode HG-Type test. Pest Management and Crop

Development Bulletin. University of Illinois Extension. Available online at:
http://www.ag.uiuc.edu/cespubs/pest/articles/200209f.htm1 (Verified Nov. 29, 2002)

Niles, R.K., and D.W. Freckrnan. 1998. From the ground up: nematode ecology in
bioassessment and ecosystem health. p. 65-85. In K.R. Barker, G.A. Pederson, and
G.L. Windham (eds.) Plant and nematode interactions. Agronomy Series 36.
American Society of Agronomy, Madison, WI.

Nishizawa, T. 1984. Nematode survey on uplands fields in Java Island, with special
reference to soybean fields. p. 165-185. In Report on Japan-Indonesian Joint
Agriculture Research Project J R84-74, Japanese International Cooperative Agency.

Noe, J .P. 1998. Crop- and nematode-managernent systems. p. 159-171. In K.R. Barker,
G.A. Pederson, and G.L. Windham (eds.) Plant and nematode interactions. Agronomy
Series 36. American Society of Agronomy, Madison, WI.

Noe, JP, and KR. Barker. 1985. Relation of within-field spatial variation of plant-

parasitic nematode population densities and edaphic factors. Phytopathology 75:247-
252.

Noe, J .P., and CL. Campbell. 1985. Spatial pattern analysis of plant-parasitic nematodes.
J. Nematol. 17:86-93.

Noel, GR. 1992. History distribution and economics. p. 1-13. In Biology and
management of the soybean cyst nematode. American Phytopathological Society
Press, Saint Paul, MN.

Noel, GR, and DJ. Edwards. 1996. Population development of Heterodera glycines and
soybean yield in soybean-maize rotations following introduction into a noninfested
field. J. Nematol. 28:335-342.

Norton, DC. 1978. Ecology of plant-parasitic nematodes. John Wiley & Sons, New
York.

Norton, D.C., A.M. Golden, and F. Varon de Agudelo. 1983. Heterodera glycines on
soybean in Colombia. Plant Dis. 67:1389.

210

Nusbaum C.J., and KR. Barker. 1971. Population dynamics. p. 303-323. In B.M.
Zuckerrnan, W.F. Mai, and R.A. Rohde (eds.) Plant parasitic nematodes. Vol 1.
Academic Press, New York.

Opperman, C.H., G.N. Acedo, D.M. Saravitz, A.M. Skantar, W. Song, C.G. Taylor, and
MA. Conkling. 1994. Bioengineering resistance to sedentary endoparasitic
nematodes. p. 221-230. In F. Lamberti et al. (eds.) Advances in molecular plant
nematology. Plenus Press, New York and London.

Oteifa, BA. 1952. Potassium nutrition on the host in relation to the infection by root-knot
nematode, Meloidogyne incognita. Proc. Helminthol. Soc. Washington 19:99-104.

Oteifa, B.A., and A.Y. Elgindi. 1976. Potassium nutrition of cotton, Gossipium
barbadense, in relation to nematode infection by Meloidogyne incognita and
Rotylenchulus reniformis. p. 301-306. In XIIth Colloquium of International Potash
Institute Proceedings. Izmir, Turkey.

Panatier, Y. 1998. Variowin 2.21. User’s Guide. [Online] Available at
http://www.springer-ny.com/supplements/variowin.html (verified 3 Dec 2002)

Pierce, F.J., and P. Nowak. 1999. Aspects of Precision Agriculture. p.1-85. In D.L.
Sparks (ed.). Advances in Agronomy 67. Academic Press, New York.

Pierce, F.J., and B.J. Sadler. 1997. The state of site-specific management for agriculture.
American Society of Agronomy Miscellaneous Publication. ASA, CSSA, and SSSA
Publishers, Madison, WI.

Pierce, F.J., P.C. Roberts, and G. Mangold. 1994. Site-specific management: The pros,
the cons, and the realities. p. 17-21. In F.J. Pierce, and B.J. Sadler (eds.) Proceedings
of the International Crop Management Conference. Iowa State University, Ames,
Iowa.

Poinar, G.O.Jr. 1983. The natural history of nematodes. Prentice Hall, Inc., Englewood,
NJ.

Poskuta, J .W., V.H. Dropkin, and C.J. Nelson. 1986. Photosynthesis, photorespiration,

and respiration of soybean after infection with root nematodes. Photosynthetica,
20:405-10.

Prot, J.C. and C. Netscher. 1979. Influence of movement of juveniles on detection of
fields infested with Meloidogyne. p. 193-203. In F. Lamberti and CE. Taylor (eds.)
Root-knot nematodes (Meloidogyne species) Systematics biology and control.
Academic Press, New York.

Reinbott, T.M., and D.G. Blevin. 1991. Phosphate interaction with uptake and leaf
concentration of magnesium, calcium and potassium in winter wheat seedlings.
Agron. J. 83:1043-1046.

211

Riggs, RD. 1987. Nonhost root penetration by the soybean cyst nematode. J. Nematol.
19:251-254.

Riggs, R.D., and DP. Schmitt. 1988. Complete characterization of the race scheme for
Heterodera glycines. J. Nematol. 20:392-395.

Riggs, R.D., and J.A. Wrather. 1992. Biology and management of the soybean cyst
nematode. American Phytopathological Society Press, Saint Paul, MN.

Riggs, R.D., and TL. Niblack. 1993. Nematode pests of oilseed crops and grain legumes.
p. 209-258. In K. Evans, D.L. Trudgill, and J.M. Webster (eds.) Plant parasitic
nematodes in temperate agriculture. CAB Int., Wallingford, England.

Ritchie, S.W., and GD. Benson. 1994. How a soybean plant develops. Iowa State
University of Science and Technology. Cooperative Extension service, Ames, Iowa.
Special report # 53. Reprinted March 1994. (Available online at
http://www.agron.iastate.edu/soybean/beangrows.htrnl) (verified 10/26/02).

Roberts, PA. 1993. The future of nematology: integration of new and improved
management strategies. J. Nematol. 25:3 83- 394.

Robertson, GP, and D.W. Freckman. 1995. The spatial distribution of nematode trophic
groups across a cultivated ecosystem. Ecology 76: 1425-1432.

Robertson, G.P., J .R. Crum, and B.G. Ellis. 1993. The spatial variability of soil resources
following long-term disturbance. Oecologia 96:451-456.

Russell, RD. 1977. Plant root systems, their function and interaction with the soil.
McGraw-Hill, New York.

Sasser, J .N., and D.W. Freckman. 1987. A worldwide perspective on nematology: the
role of the society. p. 7-14. In J.A.Veech, and D.W. Dickson (eds.) Vistas on
Nematology. E.O.Painter, deLeon Springs, FL.

Schabenberger, O., and F .J . Pierce. 2002. Contemporary statistical models for the plant
and soil sciences. CRC Press LLC, Boca Raton, FL.

Schmitt, D.P., and GR. Noel. 1984. Nematode parasites of soybeans. p.13-60. In W.R.
Nickle (ed.) Plant and Insect nematodes. Marcel Decker, Inc., New York.

Schmitt, D.P., and RD. Riggs. 1989. Population dynamics and management of
Heterodera glycines. p. 253-267. In G.E. Russell (ed.) Agricultural Zoology Reviews.
Vol. 3. Wimbome, England.

Schmitt, D.P., and RD. Riggs. 1991. Influence of selected plant species on hatching of
eggs and development of juveniles of Heterodera glycines. J. Nematol. 23:1-6.

212

Seinhorst, J .W. 1966. The relationships between population increase and population
density in plant parasitic nematodes. I. Introduction and migratory nematodes.
Nematologica, 12:157-169.

Seinhorst, J .W. 1967. The relationship between population increase and population
density in plant parasitic nematodes. Nematologica 13:429-442.

Seinhorst, J .W. 1982. The distribution of cysts of Globodera rostochiensis in small plots
and the resulting sampling errors. Nematologica 28:285-297.

Seinhorst, J.W. 1970. Dynamics of populations of plant parasitic nematodes. Annu. Rev.
Phytopathol. 8:131-156.

Sikora, B.J., and GR. Noel. 1996. Hatch and emergence of Heterodera glycines in root
leachate from resistant and susceptible soybean cultivars. J. Nematol. 28:501-509.

Sinclair, J .B. 1994. Soybeans. p.99-103. In W. F. Bennet (ed.) Nutrient deficiencies and
toxicities in crop plants. American Phytopathological Society, St. Paul, MN.

Slack, D.A., R.D. Riggs, and ML. Harnblen. 1972. The effect of temperature and
moisture on the survival of Heterodera glycines in the absence of a host. J. Nematol.
4:263-266.

Smith, G.J., W.J. Wiebold, T.L. Niblack, P.C. Scharf, and D.G. Blevins. 2001.
Macronutrient concentrations of soybean infected with soybean cyst nematode. Plant
Soil 235121-26.

Smolik, J. 2001. South Dakota State University. Available online at
http://plantsci.sdstate.edu/planthealth/SCN/SCNExtr.html (Posted 8 Dec. 2001,
verified 1 Dec. 2002).

Soil Survey Division Staff. 1993. Soil survey manual. Soil Conservation Service. US.
Department of Agriculture Handbook 18.

Soil Survey Division, Natural Resources Conservation Service, United States Department
of Agriculture. 2001. Official Soil Series Descriptions [Online]. Available at
http://www.statlab.iastate.edu/soils/osd/ (Accessed 23 Mar 2001, verified 24 Sep.
2002)

Stirling, GR. 1991. Biological control of plant parasitic nematodes. CAB Int.
Wallingford, England.

Stover, RH. 1979. Flooding of soil for disease control. p.19-28. In DJ. Mulder (ed.) Soil
disinfestation. Elsevier, Amsterdam, The Netherlands.

Sudduth, K.A. 1999. Engineering technologies for precision farming. Presented at the
International Seminar on Agricultural Mechanization Technology for Precision

Farming, Suwon, Korea, 27 May. Available online at http://www.fse.missouri.edu/
mpac/pubs/eng_tech.pdf (verified Nov. 24, 2002).

Taylor, LR. 1961. Aggregation, variance and the mean. Nature 189:732-735.

Taylor, LR. 1984. Assessing and interpreting the spatial distribution 'of insect
populations. Annu. Rev. Entomol. 29:321-357.

Taylor, L.R., I.P. Woiwod, and J .N. Perry. 1979. The negative binomial as a dynamic
ecological model for aggregation, and the density dependence of k. J. Animal Ecol.
48:289-304.

Tefft, P.M., and L.W. Bone. 1985. Plant-induced hatching of eggs of the soybean cyst
nematode Heterodera glycines. J. Nematol. 17:275-279.

Thomas, W.K., J.T. Vidal, L. M. Frisse, M. M, and J. Baldwin. 1997. DNA sequences
from formalin-fixed nematodes: integrating molecular and morphological approaches
to taxonomy. J. Nematol. 29: 250-254.

Threlkeld, G.W., and J .E. Feenstra. 1974. Soil survey of Shiawassee County, Michigan.
US Dept. of Agriculture, Soil Conservation Service, Washington DC.

Todd, TC, and C.A.S. Pearson. 1988. Establishment of Heterodera glycines in three soil
types. Ann. Appl. Nematol. 2:57-60.

Topp, G.C., W.D. Reynolds, F.J. Cook, J.M. Kirby, and MR. Carter. 1997. Physical
attributes of soil quality. p. 21-58. In B.G. Gregorich, and MR. Carter (eds.) Soil
quality for crop production and ecosystem health. Developments in soil science 25.
Elsevier, New York.

Triantaphyllou, AC. 1975. Genetic structure of races of Heterodera glycines and
inheritance of ability to reproduce on resistant soybeans. J. Nematol. 7:3 56-364.

Trudgill, D.L. 1980. Effects of Globodera rostochiensis and fertilizers on the mineral
nutrient content and yield of potato plants. Nematologica 26:243-254.

Trudgill, D.L. 1987. Effects of rates of a nematicide and of fertilizer on the growth and
yield of cultivars of potato which differ in their tolerance of damage by potato cyst
nematodes (Globodera rostochiensis and G. Pallida). Plant Soil 104:235-243.

Tylka, G. 1998. SCN extraction protocols. Iowa State University. Available online at
http://www.extension.iastate.edu/Pages/plantpath/tylka/Frames.html (verified 1 Dec.
2002)

Tylka, G.L., C. Sanogo, and SK. Souhrada. 1998. Relationships among Heterodera

glycines population densities, soybean yields, and soil pH. J. Nematol. 30: 519
(abstr.).

214

 

 

Vi glierchio, DR. 1991. The world of nematodes. Agaccess, Davis, California.

Vitosh, M.L., J.W. Johnson, and DB. Mengel. 1997. Tri-State fertilizer
recommendations for corn, soybean, wheat and alfalfa. Purdue University
Cooperative Extension Service in cooperation with Michigan State University and
The Ohio State University. Last modified March 9, 2000. Available online at:
http://www.agry.purdue.edu/ext/forages/publications/ay9-32.htm#Tri-State (verified
Oct. 20, 2002).

Voronov, D.A., Y.V, Oanchin, and SE. Spiridonov. 1998. Nematode phylogeny and
embryology. Nature 395:28.

Vrain, TC. 1986. Role of soil water in population dynamics of nematodes. p. 101-128. In
K.J. Leonard, and WE. Fry (eds.) Plant disease epidemiology population dynamics
and management, vol. 1. Mac Millan, New York.

Wallace, HR. 1959. Movement of eelworrns. V. Observations on Aphelenchoides
ritzema-bosi (Scwartz 1912) Steiner 1932 on florists’ Chrysanthemums. Ann. Appl.
Biol. 47:350.

Wallace, M.K., and D.M. Hawkins. 1994 Applications of geostatistics in plant
nematology. Suppl. J. Nematol. 26:626-634.

Wallace, M.K., D.M. Hawkins, and DH. MacDonald. 1993. Correlation of edaphic
factors with plant-parasitic nematode population densities in a forage field. J.
Nematol. 25:642-653.

Wallace, M.K., J.H. Orf, and WC. Stienstra. 1995. Field population dynamics of soybean
cyst nematode on resistant and susceptible soybeans and their blends. Crop Sci.
35:703-707.

Wang, J., P.A. Donald, T.L. Niblack, G. Bird, J. Faglrihi, J.M. Ferris, C. Grau; et al.
2000. Soybean cyst nematode reproduction in the North central United States. Plant
Dis. 84:77-82.

Wardle, D.A., G.W. Yeates, R.N. Watson, and KS. Nicholson. 1995. The detritus food
web and diversity of soil fauna as indicators of disturbance regimes in agro-
ecosystems. Plant Soil 170:35-43.

Warner, F., R. Mather, G. Bird, and J. Davenport. 1994. Nematodes in Michigan 1.
Distribution of Heterodera glycines and other plant-parasitic nematodes in soybean.
Suppl. J. Nematol. 26:720-726.

Wasilewska, L. 1995. Directions of changes in communities of soil nematodes in man-
disturbed ecosystems. Acta Zoologica Fennica 196:271-274.

Webster, R., and B. Boag. 1992. Geostatistical analysis of cyst nematodes in soil. J. Soil
Sci. 43:583-595.

215

Webster, R., and TM. Burguess. 1980. Optimal interpolation and isarithmic mapping of
soil properties-III Changing drift and universal kriging. J. Soil Sci. 31:505-524.

Werner, D., J. Wilcockson, R. Tripf, E. M6rschel, and H. Papen. 1981. Limitations of
symbiotic and associated nitrogen fixation by developmental stages in the system
Rhyzobium japonicum with Glycine max and Azospirillum brasilensis with grasses, e.
g. Triticum aestivum. p 299-308. In H. Bothe, and A. Trebst (eds.) Biology of
Inorganic Nitrogen and Sulfirr. Springer Verlag, Berlin.

Wilhelm, N.S., J .M. Fisher, and RD. Graham. 1985. The effect of manganese deficiency
and cereal cyst nematode infection on the growth of barley. Plant Soil 85:23-32.

Winstead, N.N., C.B. Skotland, and J .N. Sasser. 1955. Soybean cyst nematode in North
Carolina. Plant Dis. Rep. 39:9-11.

Workneh, F., E. Villanueva, and CM. Rush. 2001. Relative within-field distribution
patterns of beet necrotic yellow vein virus and beet soilbome mosaic virus.
Phytopathology 91 :S96 (abstr.).

Workneh, F., X.B. Yang, and G.L. Tylka. 1999. Soybean brown stem rot, Phytophthora
sojae, and Heterodera glycines by soil texture and tillage relations. Phytopathology
89:844-850.

Wrather, J .A., T.R. Anderson, D.M. Arsyad, Y. Tan, L.D. Ploper, A. Porta-Puglia, H.H.
Ram, and J.T. Yorinori. 2001a. Soybean disease loss estimates for the top ten
soybean-producing countries in 1998. Can. J. Plant Pathol. 23:115-121.

Wrather, J .A., W. Stienstra, and SR. Koenning. 2001b. Soybean disease loss estimates
for the United States from 1996 tol998. Can. J. Plant Pathol. 23:122-131.

Wrather, J .A., S.C. Anand, and SR. Koenning. 1992. Management by cultural practices.
p. 125-131. In R.D. Riggs, and J .A. Wrather (eds.) Biology and management of the
soybean cyst nematode. The American Phytopathological Society, St. Paul, MN.

Wyse-Pester, D.W., L.J. Wiles, and P. Westra. 2002. The potential for mapping nematode
distributions for site-specific management. J. Nematol. 34:80-87.

Yeates, G.W. 1996. Nematode Ecology. Russian Journal of Nematology 4:71-75.

Yeates, G.W. 1999. Effects of plants on nematode community structure. Annu. Rev.
Phytopathol. 37:127-149.

Yeates, G.W., and T. Bongers. 1999. Nematode diversity in agroecosystems. Agric.
Ecosyst. Environ. 74:1 13-135.

Yeates, G.W., T. Bongers, R.G.M. de Goede, D.W. Freckman, and SS. Georgiera. 1993.
Feeding habits in soil nematode families and genera- an outline for soil ecologists. J.
Nematol. 25:315- 331.

216

Yen, J .H., T.L. Niblack, and W.J. Wiebold. 1995. Dormancy of Heterodera glycines in
Missouri. J. Nematol. 27:153-163.

Yokoo, T. 1936. Host plants of Heterodera shachtii Schmidt and some instructions.
Korea Agriculture Experimental Station Bulletin 8:47-174.

Young, L. 1998. Breeding for nematode resistance and tolerance. p. 187-207. In K.R.
Barker, G.A. Pederson, and G.L. Windham (eds.) Plant and nematode interactions.
Agronomy Series 36. American Society of Agronomy, Madison, WI.

Young, L.D. 1982. Reproduction of Tennessee soybean cyst nematode population on
cultivars resistant to Race 4. Plant Dis. 66:251-252.

Young, L.D. 1987. Effects of soil disturbance on reproduction of Heterodera glycines. J.
Nematol. 19:141-142.

Young, L. D. 1990. Survey of soybean cyst nematode races in Tennessee. Suppl. Journal
of Nematology 22: 672-675.

Young, L.D. 1992. Epiphytology and life cycle. p.27-36. In R.D. Riggs, and J .A. Wrather
(eds.) Biology and management of the Soybean Cyst Nematode. The American
Phytopathological Society, St. Paul, MN.

Young, L.D. 1998. Influence of soybean cropping sequences on seed yield and female
index of the soybean cyst nematode. Plant Dis. 82:615-619.

Young, L.D., and EB. Hartwig. 1992. Cropping sequence effects on soybean and
Heterodera glycines. Plant Dis. 76:78-81.

Young, L.D., and LG. Heatherly. 1990. Heterodera glycines invasion and reproduction
on soybean grown in clay and silt loam soils. J. Nematol. 22:618-619.

217