T .
A . nwnvxu Hubs,
. .

.2 ..
$3....

. .s . z:
(unﬁt. .0. .59...
, .5 ﬂammawﬁd
Hanan... . a ..
aw w% . .
. kvruiﬁuw , , .. ..
3r .ﬁﬁuoﬁ F. ..

ﬁg? . .w¢.m.u.,nuﬂ..rmmwwﬂ.m .
ﬁrﬁauﬁ . . . . 9 “mm.
&. g .
.. .
. ....3.....113u.
a 35.3%

I...

sens... i
i = v

a
u

g...
3 39p

.. 3.9.9.)...1.)
3. (.51!)

4 .
.2

d... L... .

nrb 3‘ *(1731‘

and... c

Lv 1....-
E .1... .13).!

 

 

Giana. .n «awmwmﬁﬂmmmmﬁa. “was- .3...

MICHIGAI‘IJIBMES
STATE UNIVERSITY
EAST LANSING, MICH 48824-1048

This is to certify that the
dissertation entitled

ORCHARD FLOOR MANAGEMENT SYSTEMS AND
ROOTSTOCK PERFORMANCE OF ORGANICALLY
MANAGED APPLES (Malus x domestica Borkh.)

presented by

Roberto Jose Zoppolo

has been accepted towards futﬁllment
of the requirements for the

PhD. degree in HORTICULTURE

 

 

 

MSU Is an Ammonia Action/Equal Opportunity Institution

 

 

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
.th Q 5] _. 00? -
J. 13 2V

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRCIDateDue.p65-p.15

ORCHARD FLOOR MANAGEMENT SYSTEMS AND ROOTSTOCK
PERFORMANCE OF ORGANICALLY MANAGED
APPLES (Malus x domestica Borkh.)

By

Roberto José Zoppolo

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Horticulture

2004

ABSTRACT
ORCHARD FLOOR MANAGEMENT SYSTEMS AND ROOTSTOCK
PERFORMANCE OF ORGANICALLY MANAGED
APPLES (Malus x domestica Borkh.)
By

Roberto José Zoppolo

Orchard ﬂoor management is a critical aspect of apple growing with enhanced
impact in organic fruit production. Weed management as well as the N cycling and its
synchrony (temporal connection between availability and demand) are major challenges,
and key aspects for sustainability. We studied the growth response of different rootstocks to
orchard floor management systems to determine if rootstock vigor could compensate for
stress imposed by vegetative cover competition. Shifts in soil, vegetation cover,
composition of the soil food web, and partitioning of C and N within the tree ecosystem
were measured to estimate biological sustainability. A total of 468 trees were planted in
May 2000 at Clarksville Horticultural Research Station, Michigan State University. They
are part of a 2.6 ha orchard certiﬁed organic by the Organic Crop Improvement
Association. The apple trees of the cv. ‘Paciﬁc Gala” were managed in the vertical axe
system with drip irrigation. Three rootstocks were evaluated: M.9 NAKB-337, a weak
dwarf clone; M.9 RN29, a dwarf clone, and Supporter 4, a semi-dwarf rootstock. The
orchard floor management systems used were: mulch of alfalfa hay, weed ﬂaming, and
strip tillage on each side of the tree row (“Swiss Sandwich System”). The rootstocks as
well as the treatments inﬂuenced the vigor of the trees. The trunk cross-sectional area of

Supporter 4 increased 35% and 21% more than that of M.9 NAKB 337 and M.9 RN 29,

respectively. Fruit yield in 2003 for Mulch and Flame treatments was highest for M.9
NAKB while for Sandwich maximum was achieved with M.9 RN29. There were no
differences between Flame and Sandwich even though vegetation-free-area was greater in
Flame. The vegetative cover underwent important shifts in species composition and number
of plants/m2 under the different ground floor treatments. Both C and N pools were
enhanced by Mulch and to a lower extent by Sandwich; conversely no increase took place
under Flaming. Soil food web was slightly enhanced under Sandwich which had a
signiﬁcantly higher number of nematodes and active fungal biomass.

To study the N dynamics 88 containers (wooden boxes of 1m x 1.2m x 0.5m) installed in
the ﬁeld with 0.6 m3 of soil were used for planting apple trees in May 2001. Five orchard
floor management systems were applied: Mulch, Flame, Sandwich, Partial, and Total
Cover of clover and rye. Boxes were lined in plastic and a system was designed to allow
100% of recovery of the water leaching through the soil in each container. This leachate
was repeatedly analyzed for inorganic N content during the 3 years of experiment.
Nitrogen-NO3' contents in the leachate from Mulch were more than twenty times higher
than in the other treatments. Carbon partitioning was affected and N content in wood
increased over 3 fold between trees from Total Cover and Mulch. Trees with greatest
growth from Mulch had the lowest rootzshoot ratio while those with least grth from
Total Cover had the highest. While Mulch gave superior results in tree performance, the
Sandwich was found to be a viable orchard ﬂoor management system with advantages over

the other treatments regarding sustainability and management efﬁciency.

Cepyright by
ROBERTO ZOPPOLO
2004

DEDICATION

This thesis is dedicated to my wife Virginia and my parents, César and Anna Ruth,

for their loving support and unconditional encouragement.

ACKNOWLEDGEMENTS

I would like to acknowledge the members of my Guidance Committee for their
advice, consideration, the time dedicated, and the opportunity of developing excellent
relationships. Especially I thank my major professor, Dr. Ron Perry who made it
possible for me to come to Michigan State University and gave me continuous advice and
support in the professional as well as personal level opening his family to mine. Dr.
Douglas Buhler who introduced me into the weed ecology world and taught me to look
differently to this portion of the fruit growing ecosystem. Dr. Jim Flore who promoted
new and challenging ideas that added meaning to the research project. Dr. Richard
Harwood whose keenness and wide experience were critical for a sustained development
of the program and the result analysis process. Dr. David Rothstein who through a
dynamic overview of the system gave me important concepts to envision natures cycles
in their whole dimension.

It would have been very difﬁcult to complete all the work done without the help
received from numerous students and colleagues. Special thanks go to Dario Stefanelli
with whom we built a strong team to go through numerous ﬁeld and laboratory
challenges as well as technical discussions; to Michael Jost, Steven Berkheimer,
Costanza Zavalloni, and Adriana Nikoloudi who were always ready to help, discuss,
suggest, and share. Thank you all for the friendship received.

My recognition and appreciation to the staff and personnel at Clarksville
Horticultural Research Station, specially Jerry Skeltis, Gail Byler, and Denise Ruwersma,

for their willingness and efﬁciency in performing the different activities that gave support

vi

to my ﬁeld work. I want to thank Jon Dahl and the staff at the MSU Soil and Plant
Nutrient Laboratory for their work. I want to thank Lynn Holcomb, Elaine Parker, and
Jeff Smeenk for their help in teaching me laboratory techniques.

Numerous friends from the Latinamerican Community and Horticultural
Organization of Graduate Students (Marlene, Mauricio, and Randy among others) helped
my family and I feel more at home, through their company and sharing, and for that I am
grateful.

The constant encouragement and conﬁdence received from my parents César and
Anna Ruth, and from the whole family was very important for me, and I deeply
appreciate it.

A very special recognition goes to my wife Virginia, and my children Diego and
Veronica, without whose continuous company and loving understanding it would have
been impossible to achieve this degree.

Finally I want to acknowledge the Instituto Nacional de Investigacién
Agropecuaria (INIA) from Uruguay who gave me this opportunity and the economical

support to pursue it.

vii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... xi

LIST OF FIGURES ................................................................................. xvi

CHAPTER 1. ROOTSTOCK PERFORMANCE UNDER DIFFERENT ORCHARD
FLOOR MANAGEMENT SYSTEMS FOR ORGANIC APPLE PRODUCTION. . . ...1

Abstract ........................................................................................ 2
Introduction ................................................................................... 3
Materials and methods ......................................................................... 7
Results ....................................................................................... 14
Discussion .................................................................................... 21
Conclusions ................................................................................. 27
Literature cited ............................................................................... 30

CHAPTER 2. EFFECT OF DIFFERENT GROUND FLOOR MANAGEMENT
SYSTEMS FOR ORGANIC APPLE PRODUCTION ON NITROGEN CYCLING

AND TREE PERFORMANCE ................................................................... 51
Abstract ...................................................................................... 52
Introduction ................................................................................. 53
Materials and methods ....................................................................... 55
Results ....................................................................................... 61
Discussion ................................................................................... 65
Conclusions ................................................................................. 73
Literature cited ............................................................................... 75

viii

CHAPTER 3. APPLE TREE AND SOIL COVER DRY MATTER PARTITIONING
UNDER FIVE ORCHARD FLOOR MANAGEMENT SYSTEMS IN ORGANIC

PRODUCTION ........................................................................................ 89
Abstract ...................................................................................... 90
Introduction ................................................................................. 91
Materials and methods ....................................................................... 92
Results ....................................................................................... 97
Discussion ................................................................................... 99
Conclusions ................................................................................ 1 02
Literature cited ............................................................................. 104

CHAPTER 4. ORCHARD FLOOR MANAGEMENT SYSTEMS AND THEIR
IMPACT ON RESIDENT VEGETATION IN ORGANIC APPLE PRODUCTION. . . 1 14

Abstract .................................................................................... 115
Introduction ................................................................................ 1 16
Materials and methods ..................................................................... 117
Results ...................................................................................... 123
Discussion ................................................................................. 125
Conclusions ................................................................................ 1 30
Literature cited ............................................................................. 132
SUMMARY AND CONCLUSIONS ........................................................... 137
APPENDIX A ...................................................................................... 143
APPENDIX B ...................................................................................... 153

ix

APPENDIX C ...................................................................................... 159

APPENDIX D ...................................................................................... 162

Table

Table

Table

Table

Table

Table

Table

LIST OF TABLES

CHAPTER 1

1.1 Carbon mineralization potential of soil. Cumulative amount of COz-C (ug of

COz-C * g of soil ' 1) evolved during incubation in laboratory at 25°C and no
light from soil of different positions of the sandwich treatment. Values are
means of six replicates. Samples were collected April 24th 2003. Analysis of
variance was carried out for each date aﬁer incubation. Values followed by the
same letter are not signiﬁcantly different for p_<_0.05 (LSMEANS
test) ....................................................................................... 34

1.2 Nitrogen mineralization potential. Nitrogen available in soil (mg Kg’l) after

different days of incubation in laboratory at 25°C and no light from soil of the
sandwich treatment in its three positions (0 — 10 cm depth) Values are means
of six replicates. Sample collected April 24th 2003. Analysis of variance was
carried out for each date. Values followed by the same letter are not
signiﬁcantly different for p30.05 (LSMEAN S test). ....................................... 34

1.3. Soil food web. Microbial Biomass of bacteria and fungi (in ug/g soil) and

protozoa (in number/g soil) at 0-30 cm depth in the treatment row. Each value
is mean of 6 replicates. Samples were collected November 10th 2003. Values
followed by different letters in the same row are signiﬁcantly different for
pS0.05 (LSMEANS test) ................................................................................... 35

1.4. Microbial Biomass of bacteria and fungi (in jig/g soil) and protozoa (in

number/g soil) at 0-30 cm depth in row and strip positions of the Sandwich
treatment. Each value is mean of 6 replicates. Samples were collected
November 10th 2003. Values followed by different letters in the same row are
signiﬁcantly different for p50.05 (LSMEAN S test). ....................................... 36

1.5. Nematodes population characterization at 0-30 cm depth in the row by

treatment (in Number of nematodes/ g soil). Each value is mean of 6 replicates.
Samples were collected November 10th 2003. Values followed by different
letters in the same row are signiﬁcantly different for p50.05 (LSMEANS
test) .................................................................................................................... 36

1.6. Yearly Trunk Cross Sectional Area Increase (TCAI) of ‘Paciﬁc Gala’ in cm2

by rootstock for the three treatments from 2001 to 2003 measured 30 cm above
the graft union. Each value is an average of 12 trees. Analysis of variance was
carried out for each year. Values followed by different letters in the same
column or row are signiﬁcantly different for pS0.05 (LSMEAN S test). ......... 37

1.7. Branch ﬁnal length of ‘Paciﬁc Gala’ in cm, by rootstock for the three

treatments from 2001 to 2003. Each value is an average of 12 trees. Analysis
of variance was carried out for each year. Values followed by different letters

xi

in the same column or row are signiﬁcantly different for p50.05 (LSMEANS
test) ................................................................................................................... 38

Table 1.8. The effects on tree vigor, ﬂowering, and fruit set of ‘Paciﬁc Gala’ for 2003
by: A. orchard ﬂoor management systems, and B. rootstocks. Each value is an
average of 12 trees. Analysis of variance was carried out for each variable.

Values followed by different letters in the same row are signiﬁcantly different
for [930.05 (LSMEANS test) ............................................................................. 39

Table 1.9. Leaf N content (% dry weight) of ‘Paciﬁc Gala’, by treatment and rootstock
from 2001 to 2003 (n=36 in 2001, and n=12 in 2002 and 2003) Analysis of
variance was carried out for each year. Values followed by different letters in
the same column are signiﬁcantly different for p50.05 (LSMEANS test) ....... 39

Table 1.10. Three-year average nutrient content for leaves of ‘Paciﬁc Gala’ sampled from
the middle portion of one-year-old branches in mid-August 2001 to 2003, in
trees under different orchard ﬂoor management systems. Each value averages
12 trees. Values followed by different letters in the same column are
signiﬁcantly different for p30.05 (LSMEANS test) ......................................... 40

Table 1.11. The effect on quantitative and qualitative characteristics of ‘Paciﬁc Gala’ in
2003 of: A. orchard ﬂoor management systems, and B. rootstocks. Each value
is an average of 12 trees. Analysis of variance was carried out for each
variable. Values followed by different letters in the same row are signiﬁcantly
different for p50.05 (LSMEANS test) .............................................................. 40

CHAPTER 2

Table 2.1. Content of organic matter in soil (%) for 0-30 cm depth for two positions in SS
treatment (n=16). Analysis of variance was carried out for each date and no
signiﬁcant differences were found for pS0.05 (Tukey’s adjusted) ................... 78

Table 2.2. Nitrogen content in soil (ppm of NO3'—N ) in 0-30 cm depth for the two
positions of SS of boxes with trees (n=16). Analysis of variance was carried
out for each date. Values followed by different letters in the same row are
signiﬁcantly different for p50.05 (Tukey’s adjusted) ....................................... 78

Table 2.3. Average N concentration (mg NO3'—N L'l) for the whole year in the leachate
collected from boxes under each soil management system and with trees
grafted on rootstock: A. Supporter 4, and B. M.9 NAKB 337. (MU = Mulch,
FL = Flame, SS = Sandwich System, PC = Partial Cover, TC = Total Cover).
Analysis of variance was carried out for each year. Values in the same row
followed by the same letter are not signiﬁcantly different for p50.05 (Tukey’s
adjusted) ............................................................................................................ 79

xii

Table 2.4. Number of ﬂower clusters at time of full bloom for two rootstocks and by
orchard ﬂoor management system (MU = Mulch, FL = Flame, SS = Sandwich
System, PC = Partial Cover, TC = Total Cover). A. May 2002 B. May 2003.
Results of analysis of variance are shown for rootstock and treatment averages.
Values in the same row or column followed by the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted) ....................................... 79

Table 2.5. Nitrogen content (% dry weight) in apple leaves sampled in August of each
year, for each orchard ﬂoor management system (MU = Mulch, FL = Flame,
SS = Sandwich System, PC = Partial Cover, TC = Total Cover). Analysis of
variance was carried out for each year. Values in the same row followed by the
same letter are not signiﬁcantly different for p_<_0.05 (Tukey’s adjusted) ......... 80

Table 2.6. Nutrient 3-year average content in apple leaves sampled in August of each
year, for each orchard ﬂoor management system (MU = Mulch, FL = Flame,
SS = Sandwich System, PC = Partial Cover, TC = Total Cover). Values in the
same row followed by the same letter are not signiﬁcantly different for p30.05
(Tukey’s adjusted) ............................................................................................. 80

CHAPTER 3

Table 3.1. Content (g/tree) of total C and N per plant of ‘Buckeye Gala’ grafted on M.9
NAKB 337 harvested in November 2003 for each soil management system.
Each value is average of 4 replicates. Values in the same row followed by the
same letter are not signiﬁcantly different for p $0.05 (Tukey’s adjusted). . ...107

Table 3.2. Total dry weight of biomass of vegetative cover under each orchard ﬂoor
management system, amount of total C and N contained in it, and ON ratio.
Each value is average of 4 replicates. Values in the same row followed by the
same letter are not signiﬁcantly different for p 50.05 (Tukey’s adjusted). . ...107

CHAPTER 4

Table 4.1. Number of plants/m2 that were hand weeded from the Sandwich treatment for
each year of experiment ............................................................. 134

Table 4.2. Nitrogen concentration (% in dry weight of shoot and root), R/S ratio (dry

weight basis) and estimation of N content in a pure stand of the species (g
N/m’) Each value is a mean of 5 samples .......................................... 134

xiii

APPENDIX A

Table A]. Day of the year (DOY) with corresponding date in which treatment
applications of ﬂoor management were performed ............................... 144

Table A2. Carbon mineralization potential. Cumulative amount of COz-C (ug of COz-C
* g of soil") evolved during incubation in laboratory at 25°C and no light from
soil of different treatments. Values are means of 6 replicates. Analysis of
variance was carried out for each date. Values in the same row followed by
different letter are signiﬁcantly different for p50.05 (Tukey’s adjusted).
MU=Mulch, FL=Flame, SS=Sandwich ............................................ 145

Table A3. Nitrogen mineralization potential. Sample collected April 13'h 2001 (*). Total
N (N03 + NH4) available in soil (mg Kg‘l) after different days of incubation in
laboratory at 25°C and no light (0 —- 10 cm depth) from different treatments.
Values are means of six replicates. Analysis of variance was carried out for
each date and no signiﬁcant differences were found for p50.05 (LSMEANS
test). MU=Mulch, F L=Flame, SS=Sandwich ..................................... 146

Table A. 4. Nutrient content for leaves of ‘Paciﬁc Gala’ sampled from the middle
portion of one-year-old branches in mid-August 2001 to 2003, and 3—year
average, in trees under different orchard ﬂoor management systems. Each
value is an average of 12 trees ...................................................... 147

APPENDIX B

Table B]. Day of the year (DOY) with corresponding date in which treatment
applications of ﬂoor management were performed .............................. 154

Table 3.2. Nitrogen content (% dry weight) in alfalfa hay at moment of application to
boxes with Mulch ..................................................................... 154

Table 3.3. Ratio of C :N in alfalfa hay mulch for two dates in 2003 and for two different
layers according to time of application ............................................. 155

Table B. 4. Nutrient content for leaves of ‘Buckeye Gala’ sampled from the middle

portion of one-year—old branches in mid-August 2001 to 2003, in trees under

different orchard ﬂoor management systems. Each value is an average of 12

trees ..................................................................................... 156
APPENDIX D

Table D1. Field experiment. Day of the year (DOY) with corresponding date in which
treatment applications were performed ............................................. 163

xiv

Table D2. Container experiment. Day of the year (DOY) with corresponding date in
which treatment applications were performed .................................... 164

Table D3. Field experiment. Composition and density of vegetative cover. List of species
and their occurrence (plants/m2) for each date of survey as identiﬁed in Mulch
treatment ................................................................................ 165

Table D4 Field experiment. Composition and density of vegetative cover. List of species
and their occurrence (plants/m2) for each date of survey as identiﬁed in Flame
treatment ................................................................................ 166

Table D5. Field experiment. Composition and density of vegetative cover. List of species
and their occurrence (plants/m2) for each date of survey as identiﬁed in
Sandwich treatment ................................................................... 168

Table D6. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in
Mulch treatment ....................................................................... 170

Table D7. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in
Flame treatment ........................................................................ 171

Table D8. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in
Sandwich treatment ................................................................... 172

Table D9. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in
Partial Cover treatment ............................................................... 173

Table D.10. Container experiment. Composition and density of vegetative cover. List of

species and their occurrence (plants/m2) for each date of survey as identiﬁed in
Total Cover treatment ................................................................ 175

XV

LIST OF FIGURES
CHAPTER 1

Figure 1.1. Content of soil organic matter (SOM) in the row position (expressed as % dry
weight) by sampling date. A. for 0 — 10 cm depth. B for 0 — 30 cm depth. Each
point represents a mean of six samples (iSE). Analysis of variance was
carried out for each date. Values with the same letter are not signiﬁcantly
different for pS0.05 (LSMEANS test) .............................................................. 41

Figure 1.2. Content of soil organic matter (SOM) in the three positions of Sandwich
(expressed as % dry weight) by sampling date. A. for 0 — 10 cm depth. B. for 0
- 30 cm depth. Each point represents a mean of six samples (iSE). Analysis
of variance was carried out for each date. Values with the same letter are not
signiﬁcantly different for p50.05 (LSMEANS test) ......................................... 42

Figure 1.3. Content of NO3'— N in soil for position row (expressed as ppm) by sampling
date. A. for 0 — 10 cm depth . B. for O — 30 cm depth Each point represents a
mean of six samples (:tSE). Analysis of variance was carried out for each date.
Values with the same letter are not signiﬁcantly different for p50.05
(LSMEANS test) ............................................................................................... 43

Figure 1.4. Carbon mineralization. Cumulative ug COz-C evolved per g of soil during
155 days of laboratory incubation at 25°C and no light for soil from row
position of three orchard ﬂoor management systems. Samples were collected
on April 24th 2003 ............................................................................................. 44

Figure 1.5. Mineralized C of soil ﬁom rows, after 77 and 155 days of laboratory
incubation as affected by orchard ﬂoor management system. Samples were
collected April 24th 2003. Bars with the same letter are not signiﬁcantly
different for pS0.05 (LSMEAN S test) .............................................................. 45

Figure 1.6. Nitrogen mineralization potential. Nitrogen available in soil (mg Kg") after
different days of incubation in laboratory at 25°C and no light from soil of the
row position (0 — 10 cm depth) from different treatments. Each value is a mean
of six samples. Samples were collected on April 24th 2003 .............................. 46

Figure 1.7 Volumetric soil moisture content (as %) measured with TDR for the average
of 0-45 cm depth, on the tree row during years 2002 and 2003 ........................ 47

Figure 1.8. Cumulative trunk cross sectional area increase (TCAI) of ‘Paciﬁc Gala’ in
cm2 between 2000 and 2003 as affected by: A. orchard ﬂoor management
systems, and B. rootstocks. Analysis of variance was performed between
rootstocks and treatments. Letters indicate signiﬁcant differences for 1250.05
(LSMEAN S test) ............................................................................................... 48

xvi

Figure 1.9. Fruit set of ‘Paciﬁc Gala’ grafted on three different rootstocks and under
different orchard ﬂoor management systems for year 2003. It shows signiﬁcant
interaction between orchard ﬂoor management systems and rootstocks .......... 49

Figure 1.10 First harvest. Fruit production in kg/plant of ‘Paciﬁc Gala’ by rootstock and
for the three treatments (n=12). Year2003. It shows signiﬁcant interaction
between orchard ﬂoor management systems and rootstocks ............................ 50

CHAPTER 2

Figure 2.1. Diagram of the system designed for leachate collection. The wooden boxes
for the apple trees were lined with plastic ﬁlm before ﬁlling and a funnel (F)
installed from which a plastic pipe (PP) conducted the leachate to a jug (J)
inside a 20 l. bucket (B) with the lid placed just above soil
level ....................................................................................... 81

Figure 2.2. Content of organic matter in soil (SOM) for 0-30 cm depth of boxes with
trees by date for different treatments (n=16). (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Analysis of
variance was carried out for each date. Values followed by different letters
within date are signiﬁcantly different for p50.05 (Tukey’s adjusted) ............ 81

Figure 2.3. Nitrogen content in soil (ppm of N03'—N ) in 0-30 cm depth by treatment
(MU = Mulch, FL = Flame, SS = Sandwich System, PC = Partial Cover, TC =
Total Cover) for boxes with trees. Each value is average of 16 replicates :k SE.
Analysis of variance was carried out for each date. Values followed by
different letters in the same row are signiﬁcantly different for p50.05 (Tukey’s
adjusted) ............................................................................................................ 82

Figure 2.4. Seasonal pattern of N content in soil (ppm NHI—N) in 0 - 30 cm depth as
affected by orchard ﬂoor management system. Each value is average of 16
replicates :t SE. Analysis of variance carried out for each date gave no

signiﬁcant differences for pS0.0S (Tukey’s adjusted) ..................................... 83
Figure 2.5. Soil moisture measured with TDR for 0 - 45 cm depth for 2002 .................... 84
Figure 2.6. Soil moisture measured with TDR for O - 45 cm depth for 2003 .................... 84

Figure 2.7. Volume of leachate (liters/m2) recovered ﬁ'om boxes with apple trees during
three consecutive seasons (2001-2003) for different soil management systems
(MU = Mulch, FL = Flame, SS = Sandwich System, PC = Partial Cover, TC =
Total Cover). Each bar represents the average of 16 replicates. Analysis of
variance was carried out for each date, bars with same letter have no
signiﬁcant differences for p50.05 (Tukey’s adjusted) ..................................... 85

xvii

Figure 2.8. Total N03'—N (g./m2 per year) collected in the leachate from boxes during the
whole year for each soil management system (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover) Each bar represents
the average of 16 replicates :t SE. Analysis of variance was carried out for each
date, bars with same letter have no signiﬁcant differences for pS0.05 (Tukey’s
adjusted) ............................................................................................................ 85

Figure 2.9. Trunk cross sectional area increase (TCAI) between April 2001 and
November 2003 measured 25 cm above the graft union for two rootstocks and
by orchard ﬂoor management treatment (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Each bar represents
an average of 8 replicates. Results of analysis of variance are shown for
rootstock effect under each treatment. Bars with the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted) ....................................... 86

Figure 2.10. Total growth in length (cm) of all growing branches during 2002 and
number of branches, for two rootstocks under each orchard ﬂoor management
system treatment (MU = Mulch, FL = Flame, SS = Sandwich System, PC =
Partial Cover, TC = Total Cover). Bars are average of 8 replicates :t SE.
Results of analysis of variance are shown for rootstock under each treatment.
Bars with the same letter are not signiﬁcantly different for p50.05 (Tukey’s
adjusted) ............................................................................................................ 87

Figure 2.11. Figure 2.11. Tree volume estimation (m3), after growth in 2003, for trees
grafted on rootstocks Supporter 4 and M.9 NAKB 337 under different soil
management systems (MU = Mulch, FL = Flame, SS = Sandwich System, PC
= Partial Cover, TC = Total Cover). Bars are average of 8 replicates t SE.
Results of analysis of variance are shown for rootstock under each treatment.
Bars with the same letter are not signiﬁcantly different for p50.05 (Tukey’s
adjusted) .......................................................................................................... 88

CHAPTER 3

Figure 3.1. Trunk cross sectional area increase (TCAI) between April 2001 and
November 2003 measured 25 cm above the graft union for two rootstocks and
by orchard ﬂoor management treatment (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Each bar represents
an average of 8 replicates. Results of analysis of variance are shown for
rootstock effect under each treatment. Bars with the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted) ............................. 108

Figure 3.2. Shoot and root dry matter accumulated per tree between April 2001 and
November 2003 for ‘Buckeye Gala’ grafted on M.9 NAKB 337 as affected by
orchard ﬂoor management treatment (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Each bar represents

xviii

an average of 4 replicates 2% SE. Complete bars (shoot + root) with same capital
letter indicate total dry matter of trees that are not signiﬁcantly different for
p50.05 (Tukey’s adjusted). Bar portions with the same lower case letter are
not signiﬁcantly different for p50.05 (Tukey’s adjusted) ........................ 109

Figure 3.3 Root/shoot ratio of 3-year-old ‘Buckeye Gala’ grafted on M.9 NAKB 337 as
affected by orchard ﬂoor management treatment (MU = Mulch, FL = Flame,
SS = Sandwich System, PC = Partial Cover, TC = Total Cover). Each bar
represents an average of 4 replicates :1: SE ......................................... 109

Figure 3. 4. Regression between the rootzshoot ratio (R:S) and the C:N ratio of the total
tree biomass for ‘Buckeye Gala’ grafted on M.9 NAKB 337. Each value is an
average of 4 replicates ................................................................ 110

Figure 3.5. Nitrogen concentration in tissue (% in dry weight) in different plant parts
from ‘Buckeye Gala’ trees on rootstock M.9 NAKB 337 harvested in
November 2003 under different orchard ﬂoor management systems (MU =
mulch, FL = ﬂaming, SS = Sandwich system, PC = Partial cover, TC = Total
cover). Each bar represents an average of 4 replicates i SE. Analysis of
variance was carried within each tree part. Bars with the same letter are not
signiﬁcantly different for pS0.05 (Tukey’s adjusted) ............................ 111

Figure 3.6. Total N content (in g of N/tree) in different plant parts ﬁ'om ‘Buckeye Gala’
trees on rootstock M.9 NAKB 337 harvested in November 2003 from different
soil management systems (MU = mulch, FL = ﬂaming, SS = Sandwich system,
PC = Partial cover, TC = Total cover). Each bar represents an average of 4
replicates :1: SE. Analysis of variance was carried within each tree part. Bars
with the same letter are not signiﬁcantly different for p50.05 (Tukey’s
adjusted) ................................................................................ 112

Figure 3.7. Dry weights (in g/mz) of vegetative soil covers separated for root and shoot
portions in each orchard ﬂoor management system, and rootzshoot ratio (R/S)
of that biomass. Each value is mean of 4 replicates. Analysis of variance
results shown are for total dry biomass of vegetative cover (root + shoot); bars
with the same letter are not signiﬁcantly different for p50.05 (Tukey’s
adjusted) ................................................................................ 113

CHAPTER 4

Figure 4.1 Field Experiment. Number of species identiﬁed n the vegetative soil cover for
each orchard ﬂoor management during 5 surveys. Each bar is mean of 6
replicates ............................................................................... l 35

xix

Figure 4.2. Field Experiment Plants/m2 counted for all species in the vegetative soil cover
for each orchard ﬂoor management during 5 surveys. Each column is mean of
6 replicates :t Std error ............................................................... 135

Figure 4.3. Container Experiment Number of species identiﬁed in the vegetative soil
cover for each orchard ﬂoor management during 5 surveys. Each bar is mean
of 4 replicates .......................................................................... 136

Figure 4.4. Container Experiment Plants/m2 counted for all species in the vegetative soil
cover for each orchard ﬂoor management during 5 surveys. Each bar is mean
of4 replicates i Std error136

APPENDIX A

Figure A.1. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural
Experiment Station during 2000. Data were collected by the Michigan
Automated Weather Network ....................................................... 148

Figure A.2. Ten—day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural
Experiment Station during 2001. Data were collected by the Michigan
Automated Weather Network ....................................................... 148

Figure A.3. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural
Experiment Station during 2002. Data were collected by the Michigan
Automated Weather Network ....................................................... 149

Figure A.4. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural
Experiment Station during 2003. Data were collected by the Michigan
Automated Weather Network ....................................................... 149

Figure A.5. Diagram of the application of the orchard ﬂoor management systems to the
tree rows ................................................................................ 150

Figure A.6. Monthly soil moisture average measured by TDR (% v/v) and precipitation
(mm) recorded at Clarksville Horticultural Experiment Station during 2002-
2003. MU=Mulch, F L=Flame; SS=Sandwich .................................... 151

Figure A.7. Monthly irrigation applied and precipitation (mm) occurred at Clarksville
Horticultural Experiment Station during 2001. Precipitation data was collected
by an automated weather station of the Michigan Automated Network ....... 151

XX

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

A.8. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2002. Precipitation data was collected
by an automated weather station of the Michigan Automated Network ....... 152

A9. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2003. Precipitation data was collected
by an automated weather station of the Michigan Automated Network ....... 152

APPENDIX B

8.1. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2001. Precipitation data was collected
by an automated weather station of the Michigan Automated Network ....... 157

8.2. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2002. Precipitation data was collected
by an automated weather station of the Michigan Automated Network ....... 157

8.3. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2003. Precipitation data was collected
by an automated weather station of the Michigan Automated Network ....... 158

APPENDIX C

C.l. Irrigation and total precipitation (mm) recorded monthly at Clarksville
Horticultural Experiment Station during 2001. Data was collected by an
automated weather station of the Michigan Automated Network .............. 160

C2. Irrigation and total precipitation (mm) recorded monthly at Clarksville
Horticultural Experiment Station during 2002. Data was collected by an
automated weather station of the Michigan Automated Network .............. 160

C3. Irrigation and total precipitation (mm) recorded monthly at Clarksville
Horticultural Experiment Station during 2003. Data was collected by an
automated weather station of the Michigan Automated Network .............. 161

Figure C.4. Regression between tree total dry matter weight (g) and root:shoot ratio... 1 61

xxi

CHAPTER 1

CHAPTER 1

ROOTSTOCK PERFORMANCE UNDER DIFFERENT ORCHARD FLOOR
MANAGEMENT SYSTEMS FOR ORGANIC APPLE PRODUCTION
Abstract
Orchard floor management has a high impact on soil condition and thus on tree
performance. This research was initiated to investigate the effect of alternative orchard
ﬂoor management systems, compatible with organic production, on soil parameters and
apple rootstocks of diverse vigor. Alfalfa hay mulch, propane ﬂame burner and Swiss
Sandwich system (a combination of resident vegetation and tilled strips) were compared
from 2001 to 2003. These treatments applied to the tree rows, which provided different
vegetation-free area, were followed in regard to their applicability and effect on soil
organic matter, nitrogen availability, moisture, and soil food web composition.
Laboratory incubations to determine C and N mineralization potentials were performed.
Growth and productivity parameters of ‘Paciﬁc Gala’ grafted on M.9 NAKB 337, M.9
RN29, and Supporter 4 were measured. Both C and N pools were enhanced by Mulch
and to a lower extent by Sandwich; conversely no increase took place under Flame.
Nitrate-N content in soil under Mulch increased between 5 and 10 times compared to
Sandwich and Flame. Mulch required important input of material, Flame was highly
sensible to timing in regard to efﬁciency and side effects; Sandwich was easy to apply
and maintain. Signiﬁcant interaction was found between orchard ﬂoor management and
rootstock. Growth and yield were in general higher for Mulch but did not express as big a
difference as would be expected from the inputs. Even though vegetation-free area was

greater in Flame it did not differ from Sandwich, except for M.9 RN29 which had a better

performance in Sandwich. Supporter 4 was more vigorous and less precocious than M.9

NAKB 337.

INTRODUCTION

Worldwide efforts towards developing more sustainable agricultural systems and
the importance of organic production of fruits and vegetables have increased steadily in
the last few decades (Yusseﬁ, 2004; Dimitri, 2002). Compared with conventional
agriculture, contemporary organic production has fewer available tools (e.g. herbicides,
insecticides), thus there is a need to develop and adjust methods and practices acceptable
for organic production and that concomitantly render the systems more sustainable. In the
organic production approach, soil is a basic part of the farming system and is considered
the main support for life. As stated by some of the main thinkers identiﬁed with the
origins of the Organic Movement, the key to the health of soil and its fertility, is to
observe the Rule of Return which states the importance of encouraging the presence of
humus (Conford, 2001). Understanding the soil as a living organism and valuing the
processes carried on by all organisms that live within it, is the basis for its sustainable
management. Soil organisms inﬂuence every aspect of decomposition and nutrient
availability (Magdoff, 2000).

In fruit production, orchard ﬂoor management is a key component of the system
with an increased importance under organic production conditions (Bloksma, 2000;

Tinsley, 2000)). Orchard ﬂoor management has as its main objectives to maintain and/or

increase the soil fertility, physical and biological properties, and to supply nutrients and
water to trees in a synchronized way. Practices of ﬂoor management include soil covers
and their management, as well as weed management.

Soil covers can be either live (e.g. cover crops) or dead organic matter (hay,
straw, wood chips) or inert materials (plastic, paper, cardboard) that are seeded or put on
the soil surface to achieve some beneﬁcial effect. In numerous cases their use can
improve sustainability of fruit production. The beneﬁcial effects of soil covers have been
studied under various conditions. These effects, depending on the type of soil cover used,
include improvement of soil organic matter content, erosion control, soil moisture
retention, competition with weed species, soil fertility enhancement, increased numbers
of beneﬁcial insects, improved fruit quality, and ease of equipment movement and
circulation (Childers, 1972; Hogue and Neilsen, 1987; Hipps and Samuelson, 1991). One
of the main challenges to fruit growers is to maximize the beneﬁts of soil covers while
minimizing the disadvantages. For example, one main downside effect of cover crops as
soil covers is competition for water and nutrients with the main crop. Other factors to
take into account are costs and timing of hand labor required to install and maintain soil
covers.

Fruit trees need a vegetation free area to reduce or prevent competition for
nutrients and achieve maximum growth and production. Some authors have investigated
the effect of the time during which the tree is provided with that area (Merwin and Ray,
1997), and others have focused on its spatial arrangement or size (Parker, 1990; Welker
and Glenn, 1991; Merwin and Ray, 1997). Most of the studies include the tree trunk

inside the vegetation free area, generally with the trunk centered in that space.

In organic production, the lack of authorized herbicides has given a renewed
importance to weed control. Mechanical control is the most common weed control
practice used in organic production but has a high risk of trunk damage and requires a
considerable cost of hand labor or the need of sophisticated and expensive machinery.
The importance and concern of this problem is shown through numerous studies and
surveys in which growers give high priority to solving weed aspects in organic fruit
production (Sooby, 1999; Webster, 2000).

In the search for more sustainable production systems, one key aspect to consider
is the change of approach from weed control to weed management (Fryer, 1983). Weed
management emphasizes integration of preventive techniques, scientiﬁc knowledge, and
management skills. A better understanding of the ecology and biology of weeds can help
in understanding weediness and thus permit the design of appropriate management
(Buhler, 1999).

Soil cover application and weed control must be coordinated. To maximize the
beneﬁts of soil cover it is necessary to increase the time and coverage of the vegetation,
and implement vegetation-free areas only when weed competition with the fruit tree is
critical. Keeping the tree strips weed free just for “cosmetic” reasons or “tradition” is not
compatible with the idea of soil management under organic production (Weibel, 2002).
To solve practical aspects of providing a vegetation-free area and at the same time also
reducing damage to trunks, this area can be relocated away from the tree trunks. Once
located away from the trunk, and with a smaller dimension and during a shorter period of
the season, the trees are expected to face more stressful conditions. New practice

protocols need to be developed that will be acceptable under organic standards increasing

the ecological sustainability of the system, easing the work load and expenses, and
achieving for the grower a comparable or higher success than current conventional
practices.

Another element of vital importance in the fruit production system is the type of
rootstock used. Rootstocks are the part of the plant in direct contact with the soil. They
are selected for adaptation to soil conditions and according to the growth characteristics
that they impose to the scion. Rootstock evaluation is generally performed using standard
practices that tend to give ideal conditions for the growth of the tree. Although there is a
wide range of rootstock vigor, information on their response under different levels of
stress generated by different types of orchard ﬂoor management is lacking.

Considering the importance of managing soil covers and the need for alternatives
from conventional weed control practices it would be of great value to develop
information about the applicability and impact on soil and trees of orchard ﬂoor
management systems compatible with organic production. For this purpose three
contrasting systems with different spatial distribution and surface per tree of vegetative-
free-area, achieved through diverse practices (shading, ﬂaming, and filling), were used to
evaluate performance of three rootstocks of different vigor. The objectives of this study
were: 1) to determine the practical applicability of each orchard ﬂoor management
system; 2) estimate their impact on soil conditions and tree performance; 3) evaluate if
rootstocks with different vigor would have a diverse response to the anticipated changes

away from ideal “standard” growing conditions.

MATERIALS AND METHODS

Site description

The orchard is located at the Horticultural Experiment Station of Michigan State
University (latitude of 42° 84’ N and longitude of 84° 24’ W) in the vicinity of
Clarksville, MI. The experiment was conducted from 2000 to 2003. Previous farming
consisted of a conventional soybean-corn-corn-alfalfa rotation for two cycles until 1999,
at which point soil preparation began with sowing of buckwheat followed by a mixture of
red clover and endophytic rye. The ﬁeld experiment covers an area of 0.38 ha of the 2.5
ha of total surface that is certiﬁed by the Organic Crop Improvement Association
(OCIA). The predominant soil type is a Kalamazoo sandy clay loam (Typic Hapludalfs)
with mild slopes of less than 3%, and a particle size composition of 53.1% sand, 23.1%
silt, and 23.8% clay. Daily temperatures and precipitation during the whole experiment
were recorded with an automated weather station of the Michigan Automated Weather
Network, located approximately 560 in west of the orchard (Figures A.1 thru A.4,

Appendix A).

Plant material

One-year-old ‘Paciﬁc Gala’ apple trees (Malus x domestica Borkh.) grafted on
three different rootstocks (M.9 NAKB 337, M.9 RN29, and Supporter 4) were planted in
April 2000. The clonal rootstock M.9 NAKB 337 has the vigor of a weak M.9 growing
under 40% of the size of a seedling (Marini, 2000; Robinson, 2003). M.9 RN29 is a clone

slightly more vigorous than the NAKB 337. Supporter 4, which is a cross of M.9 x M.4,

is a more vigorous rootstock, with a size comparable to M.26. Trees were uniform in size
and had no differences in trunk cross sectional area at planting time.

Planting distance was adjusted according to the vigor of each rootstock and the
acquired knowledge of their performance under Michigan conditions (Perry, 2002).
Distance between rows was 4.6 m while the distance between trees in the row ranged
from 1.4 m for M.9 NAKB 337, to 1.7 m in the case of M.9 RN29, and 2.0 m for
Supporter 4.

Plants were trained to a vertical axe with the use of rubber bands and clothes pins
to bend branches (Perry, 2000). Minimal pruning was applied, mainly to single out the
leaders or main branches. Trees arranged in single rows of same rootstock and distances
were established as guard rows so that each data row alternated with buffer rows. A trellis

system with two wires and a galvanized metal pole to support each tree were installed.

Orchard ﬂoor management

Chicken compost was applied in 1999 at a rate of 1250 kg/ha (1100 lb/acre) and
lime at a rate of 2250 kg/ha (2000 lb/acre). A mixture of red mammoth clover and
endophytic rye was sown at planting time in April 2000. To discourage weed
development and enhance clover growth, mowing was performed regularly in all alleys
of the orchard according to best management farming practices. Three different orchard
ﬂoor management systems were applied to the trees at the beginning of the season in
2001 (Figure A.5, Appendix A):

1. Mulch (MU): A l-m-strip on each side of the tree row was covered with alfalfa hay

mulch. During the growing season the thickness of the cover was over 15 cm to provide a

shading effect that would suppress weeds, and to maintain moisture. Mulch was hand-
applied in early spring and fall.

2. Flame (FL): A-l-m strip on each side of the tree row was kept free of vegetation
through the use of a propane ﬂame burner that provided 800,000 BTU, and was equipped
with a metal cover to concentrate heat and prevent warm upward spells that cause tree
damage. Minimal amounts of water were applied just behind the burner onto the orchard
ﬂoor to reduce the risk of the heat affecting tree branches, or ﬁre spreading to the
neighboring mulch. The tractor speed was kept between 1.6 and 3.2 km/h, adjusted
according to the density of weeds to be controlled. Weed ﬂaming treatments were applied
when weed height approached 10 cm (for dates see Table A. l. A., Appendix A).

3. Sandwich system (SS): an adaptation of a Swiss system (Weibel, 2002), consisted of a
central strip that spread 30 cm to each side of the tree row in which vegetation was
allowed to grow. A 70-cm-wide strip was tilled on each side of that vegetated central
strip. The initial tilling was applied with a rotovator to cut down and incorporate the
vegetation from the previous season. The following tilling applications were made with a
spring tooth harrow to keep the strip weed-free. In the vegetated central strip, hand
weeding was performed when weeds, known to host apple pests, dominated the strip such
as curly dock (Rumax crispus L.), common lambsquarters (Chenopodium album L.),
redroot pigweed (Amaranthus retroﬂexus L.) and thistles (Cirsium sp.). Tilling started in
spring, and was repeated when weeds were 10 cm tall (for dates see Table A.l.B.,

Appendix A).

Mulch decay and ON ratio

The mulch was sampled at the time of application to determine the C/N ratio. In Spring
2003 a plastic net with 1 cm x 1 cm grid was laid down to separate mulch material
recently applied from that applied the previous year. Decomposition rate of the mulch
was monitored monthly for each age material until November. The samples were taken
by cutting a vertical cylinder of 7.6 cm in diameter from the mulched area and the depth
of each layer was recorded. The dry weight of the cores was measured in the laboratory
after oven-drying for 48 hours at 60 ° C, and sub-samples were taken for C and N

determination.

Irrigation system

Drip irrigation with drippers of 2.3 L/h every 0.6 m was installed in May 2001.
The lines were suspended from the lowest wire of the trellis system at a height of 0.7 m
to reduce the potential of damage while controlling weeds. All orchard ﬂoor management
systems received equal irrigation throughout the season to keep the available soil water

over 50% (Figure A.6 thru A.9, Appendix A).

Tree growth and productivity measjurements

Trunk cross sectional area was calculated from the diameter of the trunk measured
at 30 cm above the graft union. The measurements were taken at the end of the growing
season. Branch growth was measured on four branches chosen at the beginning of spring.
The ﬁrst three were selected to represent the bottom, middle, and lower-top parts of the

tree, and the fourth branch was the leader. Branches selected for each position were

10

similar in size at the beginning of the measuring period and had a similar angle of
insertion in the trunk. Branch length was measured weekly between June and August.
Number of ﬂower clusters in the whole tree was counted before full bloom, and total
number of fruits as well as weight was measured per tree at harvest in 2003 (ﬁrst
cropping season).

In March — April 2001 there was little pruning needed. The branches cut during
the following seasons (2002 and 2003) were weighed as a complementary measure to
estimate growth and vigor of trees.

At the end of each season, the total height of the tree as well as two orthogonal
diameters of the canopy at 0.7 m from the soil, were measured and used to estimate the

canopy volume.

Nutritional status

A composite sample of 10 leaves per tree was taken in early August of each year
from the middle portion of one-year-old branches. Relative chlorophyll content was
measured using a SPAD-502 meter (Spectrum Technologies Inc, Plainﬁeld IL). Leaves
were then rinsed with distilled water, dried for 48 h at 60°C and analyzed for mineral

nutrient composition.

Soil sampling and measurements
Soil was sampled 4 times per year, each year for each plot. Each sample was a
composite obtained by mixing 20 cores taken from the 0-10 cm depth. Same procedure

was followed for 0-30 cm depth. Samples were stored in plastic bags at 4 ° C for not more

11

than one month. In-row samples were taken not farther than 20 cm from the tree line.
Alley sub-samples were mixed to make a composite sample from both sides of the tree
row. In the case of SS treatment, a third sample was taken from the tilled area by mixing
sub-samples from the strips on each side of the tree row. Soil moisture was measured by
time domain reﬂectometry (TDR) using a Mini Trase 6050X3 (Soilrnoisture Equipment
Corp., Goleta, CA) with 30, 45, and 60 cm long stainless steel rods permanently installed
in the ﬁeld in the tree rows, halfway between two trees. One measurement at each depth
was taken for each of the 6 plots per treatment weekly in 2002 and every other week in
2003.

Inorganic N (NO3' and NH?) was determined by the MSU Soil and Plant Nutrient
Laboratory following the procedure described by (Kenney, 1982) using a Lachat
automated colorimetric analyzer (Lachat Instruments Inc. Milwaukee, WI).

Organic matter content was measured by loss on ignition of 3 g of dry soil at 400°C for 8
hours using a mufﬂe furnace.

Carbon mineralization potentials were obtained through laboratory incubation of
fresh soil samples collected on April 13th 2001 and April 24th 2003. The soil samples
were sieved through a 5 mm screen and sub samples of 50 g dry soil equivalent were
weighed into 100 ml specimen containers. Moisture was adjusted to 50% of water
holding capacity. The specimen containers were placed in mason jars with 10 ml distilled
water to maintain humidity, and these were sealed with lids that had rubber septa. The
jars were kept in a dark room at 25°C and C02 content of the headspace in the jar was
determined on days 0, 21, 30, 50, 77, 100 and 150 with an infrared gas analyzer (LI-

COR, Lincoln, NE) following the procedure described by (Paul et al., 2001)

12

Nitrogen mineralization potentials were determined from fresh samples of same
date as C mineralization. After sieving through a 5 mm screen, sub samples of 20 g dry
soil weight equivalent were put into 100 ml specimen containers. Soil moisture of
samples was adjusted to 50% of water holding capacity estimated with the funnel method
(Paul et al., 2001). The specimen containers were kept in a dark room at 25°C within a
plastic box with water in the bottom to maintain humidity. One set of samples was

analyzed for each time interval: 0, 30, 70 and 165 days. At the end of the respective
incubation period, inorganic N (NH;r + NO3_) was extracted with 1N KCl and aliquots

run on a Lachat automated colorimetric analyzer (Lachat Instruments Inc. Milwaukee,
WI).

Soil food web studies were performed with the soil sample collected on
November 10th 2003 from 0-30cm. Samples were cooled down immediately after putting
them in plastic bags and sent overnight to Soil Foodweb Inc. (Corvallis, OR) for
immediate analysis of population composition of soil microorganisms through

measurement of biomass of bacteria, fungi, protozoa, and nematodes.

Experimental design and statistical aﬂlvsis

The experiment had a randomized complete block design with six replicates of the
three applied treatments. The plots consist of 12 trees to which the main factor, ground
ﬂoor management, was randomly assigned. Subplots are formed by four trees of each
rootstock (split-plot). The two central trees of the subplot are the data trees. Analysis of
variance was performed using MIXED procedure with SAS (Version 8, SAS Institute,

Cary, NC, USA) to detect treatment effects. When treatment effects were signiﬁcant,

13

means were separated using Least-squares means test (LSMEANS test) with p50.05

unless stated.

RESULTS

Soil parameters

Soil organic matter content

Prior to the establishment of orchard ﬂoor systems, soil analysis showed no
differences between the plots. Organic matter content increased over three years with
amount of change affected by orchard ﬂoor management system (Figure 1.1). Beginning
in December 2002 the Mulch (MU) had signiﬁcantly higher OM content (3.36 %) than
Flame (FL) and Sandwich (SS) for the depth of 0-10 cm (Figure 1.1. A). For the deeper
sampling (0—30 cm), the differences were less consistent with MU, which was highest in
only two dates (April and November 2003) (Figure 1.1. B).

Soil sampling of the SS revealed that OM content was signiﬁcantly higher in alley
at 0-10 cm depth than the row and tilled strip portions of this treatment but only for the
last sampling date (Figure 1.2. A). Conversely, OM at 0-30 cm depth (Figure 1.2. B) was

higher in the row than the tilled strip.

14

Soil Nitrogen content

From the second sampling set, regardless of depth, the N03' - N content in the soil
of the MU treatment was signiﬁcantly higher in the row than the other two treatments for
the duration of the experiment (Figure 1.3).

Nitrate-N content varied among sampling time and location within the SS

treatment (data not shown).

C mineralization potential

The orchard ﬂoor management systems had an effect on C mineralization
potential, as it can be seen in the 2003 sampling (Table A2, Appendix A). Differences
are clearer for the tree row where as the time of incubation increased so too did the
differences (Figure 1.4). At the end of the period of 155 days, the soil from the row had a
higher accumulation of CO2 for MU, followed by the SS which did not differ
signiﬁcantly (Figure 1.5). The lowest accumulation of CO2 occurred in the FL row, with a
value signiﬁcantly different from MU. When comparing the positions within the SS

(Table 1.1) both, alley and strip had signiﬁcantly higher accumulation than row.

N mineralization potential

The amount of N that potentially could be mineralized did not differ among
treatments or location in the plot at the beginning of the experiment (Table A3,
Appendix A). In 2003 for the row position (Figure 1.6) the MU treatment gave higher

values of N potential than SS and FL for the initial stages of the incubation. There were

15

no signiﬁcant differences between treatments for the net increment at the end of the
incubation (day 71 to 165). The amount of N mineralized in the alley of the SS treatment
was greater than tilled strip and row (Table 1.2).

The values of N mineralization did not have a substantial change ﬁom 2001 to
2003 in the case of the alley, but went to less than half of the original average for the row

ofFL and SS.

Soil water content

Measurements performed with TDR (Figure 1.7) to the depth of 45 cm showed
that MU and FL had signiﬁcant higher soil moisture than SS during part of the season of
2002. In 2003 there were no signiﬁcant differences between the treatments even though

the tendency of lower values for SS was the same as in the previous year.

Soil food web

In the analysis of population composition of soil microorganisms amounts of over
500 ug of total bacteria per g of soil were found for all treatments with no signiﬁcant
differences between them for active or total bacterial biomass (Table 1.3). The amount of
fungal, total or active biomass, did not differ between treatments. Sandwich system had
signiﬁcantly higher values than the FL for active fungal biomass at p<0.l, while MU had
an intermediate value that did not differ from any of the other treatments. Bacterial and

fungal biomass were highest in the row than tilled strip in the SS (Table 1.4).

16

The protozoa population at 0-30 cm depth did not differ among treatments except
for the ciliates, which were higher in MU (Table 1.3). No signiﬁcant differences appeared
between positions of SS (Table 1.4).

The nematode population was characterized by a wide diversity and low numbers
(Table 1.5). Mulch and SS had signiﬁcantly higher number of bacterial feeders than FL,
and SS had higher number of root feeders than MU and FL. There was lower total

number of nematodes in the FL treatment.

Tree parameters
Trunk Size
Values of TCAI of trees in the MU system were found to be slightly more
vigorous than SS and FL. Treatment differences were only signiﬁcant in 2002 (Table 1.6
& Figure 1.8.A). Rootstocks differed in vigor with Supporter 4 the most vigorous, M.9

RN29 next, followed by the weakest M.9 NAKB 337 (Figure 1.8.8).

Branch growth

Total annual branch growth was not signiﬁcantly different between the three
orchard ﬂoor management systems in 2001 and in 2003 (Table 1.7). Mulched trees in
2002 encouraged more branch growth than FL (intermediate) and SS systems. At the end
of 2003 season, branch length was greater for Supporter 4 than other rootstocks.

The interaction for the branch growth between the orchard ﬂoor management

systems and the rootstocks was signiﬁcant in 2001when the maximum growth under SS

17

was achieved by M.9 RN29 while under MU and FL there were no differences between

M.9 RN29 and Supporter 4. This interaction was not signiﬁcant for 2002 and 2003.

Tree size
Tree size reported as canopy volume, at the end of the experiment was greater for
MU than for SS. Flame, with an intermediate value, did not differ signiﬁcantly from

either (Table 1.8). Supporter 4 had greater size tree than M.9 RN29 and M.9 NAKB 337.

Pruning weigl_rt

Orchard ﬂoor management systems and rootstocks inﬂuenced the amount of
wood (branches) removed during pruning in winter in 2003 (Table 1.8). More wood was
removed from the MU treatment, followed by FL and SS. As in trunk measurements,
more wood was removed from the more vigorous Supporter 4 than M.9 RN29, followed

by M.9 NAKB 337 (Table 1.8.3).

Flower Density

Flower density expressed here in number of ﬂower clusters counted at full bloom
per tree was greater on MU treatment than FL and SS in 2003 (Table 1.8.A). Trees on
Supporter 4 were less precocious than M.9 RN29 and M.9 NAKB 337 in both years.
Comparable results were found for the ratio of ﬂower density/1“ CA (Table 1.8.8). No
signiﬁcant interaction between orchard ﬂoor management systems and rootstocks was

found for ﬂower density.

18

EntiLset

The orchard ﬂoor management system had an impact on fruit set according to the
rootstock. The interaction treatment by rootstock was signiﬁcant (Figure 1.9). The
increase of set that occurred for M.9 RN29 in the SS compared to the other treatments
was greater than for the other rootstocks. Supporter 4 had the lowest values of mu set,
varying between 1.8 and 8.4 % for the different treatments. Rootstock M.9 RN29 had
values that went from 8.3 to 24.1 % and for M.9 NAKB 337 the values went from 14.1 to
15.4% showing a lower variation than both previous, under the different soil management

systems.

Nutritional status

Rootstock treatments had no effect on foliar N levels for all three growing
seasons. Management systems did not affect foliar N levels in 2001 (Table 1.9). Values
in 2001 averaged 2.1 % N and rose to 2.4% in 2002. N levels in leaves in 2002 and 2003
were higher in the MU and FL treatments compared to the lowest values in the SS.
Values for N separated more distinctly in 2003 with levels for MU greater than FL
followed by SS. In none of the years was there a signiﬁcant interaction between the
orchard ﬂoor management and the rootstocks for the N content in leaves.

Increase of the content of P, K, Ca, and Mg occurred from year to year between
2001 and 2003, same as in N (Table A4, Appendix A). Such trend was followed to a
lower extent by the contents of Fe and Mn which increased signiﬁcantly their content
from 2001 to 2002, but did not differ between 2002 and 2003. Phosphorus levels in

leaves from trees in the MU did not increase over the years.

19

The orchard ﬂoor management systems also had an effect on nutrient levels.
Averaging the three years, the values of K and Mn were higher under MU while those of
P and Zn had lower levels than the other treatments (Table 1.10). Differences between

FL and SS were less, being the contents of K the only ones with higher values for SS.

Y_iel_d

The ﬁrst signiﬁcant production of ‘Paciﬁc Gala’ was registered in 2003 (Table
1.11). The interaction of treatment by rootstock was signiﬁcant in the case of yield
(Figure 1.10). Trees managed in the MU treatment produced more hit than SS (except
for M.9 RN29), followed by the FL treatment (Table 1.11.A). Cropping was greatest on
the two M.9 rootstocks over Supporter 4 (Table 1.11.B).

The same trend as for kg/plant was seen, between orchard ﬂoor management
systems, for yield efﬁciency as expressed in grams per cm2 of TCA (Table 1.11.A) but
differences were not statistically signiﬁcant. Comparing rootstocks, ‘Paciﬁc Gala’ on
Supporter 4 had signiﬁcantly less production (61 g/cmz) than on M.9 NAKB 337 or M.9

RN29 (278 and 241 g/cm2 respectively) (Table 1.11.B).

20

DISCUSSION

The orchard ﬂoor management systems made signiﬁcant impacts on trees and
soil. The change in amount of organic matter in soil is a slow process that takes years
unless large quantities are added. Organic matter content did not change through the
years in the alleyways where the treatments remained unchanged and where vegetation
was mowed. The three orchard management systems focused on an area down the tree
rows, which was approximately 1.5 m in width, where they made the greatest impact. The
amount of material added through the mulching process to the row generated an expected
(Merwin et al., 1994; Sanchez et al., 2003) increase of SOM (Figure 1.1) especially in
the ﬁrst layer of soil (0 — 10 cm) for MU. It is reasonable that differences for 0 - 30 cm
were less noticeable since that soil was not disturbed and the mulch was just placed on
the soil surface.

Some of the results were surprising, considering that management effects cannot
be determined usually until after at least 10 years have elapsed (Diaz Rossello, 1992a;
Paul et al., 2001). In the SS, the values of SOM for the three positions: row, strip and
alley, for the 0-30 cm depth, showed signiﬁcant differences in the second year after
treatment installation. The differences varied but persisted until the end of the evaluation
and the tilled strip position was always the one with lower SOM content. Strangely
enough the differences were almost non existent for 0-10 cm depth. It has been well
established that tilling increases soil aeration and loss of organic matter (Brady and Wei],
2002) even though the importance of those changes will depend on the original state of

the soil prior to disruption (Calderon et al., 2000). Since the ﬁlling performed in the strip

21

was very shallow (equal or less than 10 cm), we were not expecting changes in content of
SOM in the deeper proﬁle if no differences occurred in the surface (Needelrnan et al.,
1999). Soil organic matter content for alley and row position, in 0-30 cm depth, varied
achieving differences in some years. When differences occurred, SOM was higher in the
row than alley.

The initial amounts of total N in soil for April 2001 showed no difference
between the treatment plots where the contrasting soil management systems would be
installed. After installation, the treatments started to have an impact on the soil condition
and differences appeared in June 2001. From then on, the MU kept higher levels of total
N in soil due to the continuous decomposition and mineralization of the alfalfa hay
(Cookson et al., 2002). As it was expected (Donahue et al., 1971; Sarrantonio, 2003), the
high content in protein and N of alfalfa, that clearly reﬂects in the low C:N ratio (between
20 and 15) assured a good supply of this critical nutrient. Signiﬁcant differences for the
row between the treatments, both at 0 ~10 and at 0-30 cm depth were constantly present
throughout the experiment. The mulch provided more organic matter and it was efﬁcient
in conserving soil moisture through the growing season consistent with other numerous
reports (Langord, 1937; Hogue and Neilsen, 1987).

The mineralization potential of the soil determined through incubation conﬁrmed
the impact of the orchard ﬂoor management systems. The active C pool was enhanced
and followed a similar pattern to SOM. The orchard soil management systems, even in
the case of FL increased the active C pool compared to the values present at the moment
of planting the trees after many years of agriculture production under the soybean-com

rotation. The increase of SOM produced by the addition of alfalfa hay yielded an active C

22

pool 33 % higher for the MU treatment compared with FL. Coinciding with results from
tart cherry orchards where living ground cover (including weeds) was found to affect
positively the active C pool (Sanchez et al., 2003). The herbaceous vegetation in the row
of the SS made a signiﬁcant contribution increasing that reservoir by 13 % compared
with FL.

The size of the active N pool was also affected by the orchard ﬂoor management
systems. The importance of N held in SOM has been stated in regard to its value as a
source for plant uptake (Powlson and Jenkinson, 1990; Diaz Rossello, 1992b; Bloksma,
2000). We can estimate that N increased in a signiﬁcant amount in the MU row due to the
enhanced active N pool. The possibility of adding enough N through this practice for the
trees is conﬁrmed through their growth and N content in leaves (Hanson, 1996).

Quality of the organic material added on surface to the soil has an important
impact on the mineralization process (Handayanto et al., 1997; Wardle and Lavelle,
1997; Brady and Wei], 2002) The difference of the active N pool within the SS treatment
for the different positions (Table 1.6) could be caused by the effect of the ﬁlling and of
the diverse cover composition of row and alley. The different plants generate residue of
different type and quality that goes into the soil, and at the same time have a different
uptake. The mix of legumes and grasses in the alleyway grew with little restriction, other
than occasional mowing, but likely was ﬁxing N in higher quantities than the understorey
in the row of the SS, where a higher number of grasses and broadleaf species were

uptaking more N and ﬁxing less.

23

The SS treatment surely provided a different environment than MU or FL, with
lower levels of N and slightly lower soil moisture consistent with previous research
(Hogue and Neilsen, 1987).

Results found for the soil food web show also an impact of the orchard ﬂoor
management systems. Total bacterial biomass was in levels over 500 ug of total bacteria
per g of soil for the whole orchard. This meant an increase over 3 fold, compared to the
original levels of the orchard. 0n the contrary, levels of fungi, protozoa, and nematodes
have not improved so rapidly and higher levels could have an enhancement on nutrient
cycling (Ingharn et al., 1985; Garcia and Moron, 1992; Yeates, 1998). A signiﬁcant
increase in number of ciliates was found for the MU treatment (Table 1.3). This is
sometimes associated with anaerobic conditions generated because of compaction, which
clearly was not the case under our experiment. The increased water content in the soil due
to the thick layer of hay preventing evaporation might have generated anaerobic
conditions that could explain the ciliates’ increase.

The active fungal biomass tended to be higher in the SS with signiﬁcant
differences for p _<_ 0.10. The vegetated row of the SS had high levels of active bacterial
and fungal biomass (Table 1.4). The total ﬁrngal biomass was signiﬁcantly higher in the
row than in the tilled strip of the SS. This clearly suggests that the diverse cover of the
vegetation of the in-row is important in generating an adequate environment for
multiplication of microorganisms. Recent studies (Bird, 2003) at the same experimental
site have demonstrated how higher populations of microorganisms are present in the ﬁrst
few cm of the soil proﬁle and even more in the litter when compared with deeper layers

of the soil. Clearly, the differences among treatments and positions within those

24

treatments suffer a dilution effect when using the 0-30 cm depth to characterize the soil
food web.

The diversity of plant species in the row of the SS treatment also promoted a
higher number of root feeder nematodes (Table 1.5). This could be a negative aspect that
will need to be taken into account and a management strategy developed to prevent
damage to the crop (Merwin and Stiles, 1989). The number of fungal/root feeder
nematodes also tended to be higher, although not signiﬁcantly.

When we consider the tree’s response, we ﬁnd differences among the treatments
as well as among the rootstocks. Trees performed best in the MU system as measured by
tree vigor, foliar N, ﬂowering and number of fruits per plant. Vigor as measured by TCAI
did not show signiﬁcant differences in 2003 but did in the previous year when trees from
MU treatment were found to be more vigorous than FL and SS. Nevertheless, there was a
trend towards higher vigor of the MU conﬁrmed by other parameters. For fruit
production in kg/plant (Figure 1.10) there were higher values for the MU, but those
differences were not statistically signiﬁcant for all rootstocks. When comparing FL and
SS, the differences in tree response are less. In spite of the lack of competition from
weeds, and higher water availability, the trees under the FL did not grow and produce
differently from the SS. The vegetation-free area for trees in FL was 30% more extended
than in SS. The spatial arrangement of this area also differed. For FL (as for MU) the
vegetation-free area was centered on the trunk, while for SS it was in two lateral strips
not including the tree trunk. Previous studies have reported the impact of the vegetation-
free area on growth of peach and apple trees (Welker and Glenn, 1989; Parker, 1990).

(Merwin and Ray, 1997) working with newly planted apple trees showed that weed-free

25

areas above 2 m2 per tree, in general, would not give a strong enough response to justify
the cost necessary to increase that weed-free area. In our experiment, all trees had
slightly over 2 m2 per tree, except for M.9 NAKB 337 in the SS with 1.9 m2. We could
probably be too near to the threshold in this last case and a wider tilled strip might have a
more desirable impact. In regard to weed-free period, clear differences were between MU
and the other treatments. While MU had an effective 100 % of the season weed-free, in
the case of FL and SS the weed-free period was from May till August when ﬂaming and
ﬁlling applications were stopped. This covers the period recognized by most authors as
the critical one (Neilsen and Hogue, 1985; Merwin and Ray, 1997). The timing of
treatment applications could be something to adjust but we wouldn’t expect important
changes from this in regard to tree performance improvement.

Leaf nutrient content in apple trees has been reported to be affected by the orchard
ﬂoor management (Neilsen and Hogue, 1985; Merwin and Stiles, 1994). The leaf N
content of trees reﬂected the higher N availability of MU, although they were all in the
optimum range (Hanson, 1996). Considering all treatments, the levels of K tended to be
above optimum range, while Mg reached minimum requirements and Fe, Cu, and Mn
appeared slightly deﬁcient (Hanson, 1996). In regard to K concentrations the variations
found in this experiment coincide with others (Merwin and Stiles, 1994) being highest
under MU. In the case of P various reports state an increase of its availability in soils
under mulch (Delver, 1980; Merwin and Stiles, 1994) and others a decrease when
ground cover was suppressed (Atkinson and White, 1980; Neilsen and Hogue, 1985). In
our study the unchanged level of P under MU could be due to a dilution effect caused by

the increased N availability and growth, as well as by a smaller root size (see Chapter 3).

26

As demonstrated in rootstock trials elsewhere (Marini, 2000), trees on Supporter 4
were more vigorous than intermediate M.9 RN29 and weakest growing M.9 NAKB 337.
Consistent with those reports, ‘Paciﬁc Gala' on Supporter 4 was less precocious than the
M.9 clonal rootstocks. For fruit production, we found a signiﬁcant interaction treatment
by rootstock. While M.9 NAKB 337 was the highest yielding rootstock for MU and FL
treatments for one season, M.9 RN29 was the one to perform best under the SS. More
seasons of cropping are needed to study this further, but results suggest that this rootstock
demonstrates a potential capability to overcome higher stress conditions than the standard
M.9 NAKB 337 being more adequate to ﬁt into potentially more stressful sandwich

system (less soil moisture and less N available to the trees).

CONCLUSIONS

The organic matter content of soil changes relatively slowly. This research
demonstrated that the orchard ﬂoor management system treatments had a positive impact
on organic matter content. Among the treatments, Mulch at the 0-10 cm depth had the
greatest impact. This effect was inconsistent. at 0-30 cm depth.

Both C and N pools were enhanced by the Mulch and to a lower extent by
Sandwich, conversely no increase took place under Flame. Nitrate-N content in soil
increased signiﬁcantly for both depths in the row under Mulch.

The results obtained from this study conﬁrm that tree grth and yield are best in

the Mulch. There were no differences between Flame and Sandwich even though

27

vegetation-free area was greater in the Flame. Spatial distribution of that area, not

including the tree trunk, had no important consequences.
The high levels of N03_-N in soil under Mulch indicate a higher risk of leaching

for this orchard ﬂoor management system. The use of alfalfa hay should be adjusted
carefully to prevent it and the potential environmental damage. Either by alternating or by
mixing sources with different N content and C:N ratio, an adequate supply could be
reached. Availability and cost of mulching material, application logistics and secondary
effects like increase of mice and ﬁre risk, remain as draw backs of this orchard ﬂoor
management system that should be taken into account.

Flame was the most sensitive orchard ﬂoor management system to timing for a
successful result. Risk of heat damage to trees obliges to have a careful monitoring of
environmental conditions and treatment application.

We think that the SS is a viable orchard ﬂoor management strategy. The easiness
of application of this system and its versatility show advantages when compared with the
other orchard ﬂoor management systems. Even though some indicators might not be as
ideal as it would be desired, like the increased levels of root feeder nematodes, there are
numerous possibilities for adjusting the system for Michigan conditions. The restrained

width and depth of the tilling strip seems to have a low impact on soil and until now no
consistent difference were measured for soil organic matter or N03’-N content when

compared with row.
The rootstock performance was within the expected values with a more vigorous

Supporter 4, followed by M.9 RN29 and M.9 NAKB 337. There was a signiﬁcant

interaction between treatment and rootstock due to a better performance of the M.9 RN29

28

under SS while M.9 NAKB337 was the best for MU and FL treatments. This conﬁrms in
a ﬁrst step, the idea of being able to compensate certain levels of competition through the
use of a more vigorous rootstock. These observations should continue so as to conﬁrm

these results on the long term.

29

LITERATURE CITED

Atkinson, D. and G. C. White (1980). Some effects of orchard soil management on the
mineral nutrition of apple trees. In: 'Mineral nutrition of fruit trees'. D. Atkinson,
J. E. Jackson, R. 0. Sharples and W. M. Waller. Butterworths. London - Boston.
pp: 241-254.

Bird, G. W. (2003). Soil quality dynamics associated with transitioning from a Michigan
conventional com-soybean production system to an organic apple orchard.
Michigan State University, East Lansing. 18 p.

Bloksma, J. (2000). Soil management in organic fruit growing. Conference 'Organic fruit
opportunities and challenges', Ashford, Great Britain.

Brady, N. C. and R. R. Weil (2002). The nature and properties of soils. Upper Saddle
River, NJ, Prentice Hall. 960 p.

Calderon, F. J ., L. E. Jackson, K. M. Scow and D. E. Roston (2000). Microbial responses
to simulated tillage in cultivated and uncultivated soils. Soil Biology and
Biochemistry 32(2000): 1547-1559.

Childers, N. F. (1972). Modern fruit science. Somerville, NJ, Somerset Press, Inc. 912 p.

Cookson, W. R., I. S. Comforth and J. S. Rowarth (2002). Winter soil temperature (2-15
°C) effects on nitrogen transformations in clover green manure amended or

unamended soils; a laboratory and ﬁeld study. Soil Biology and Biochemistry
34(10): 1401-1415.

Delver, P. (1980). Uptake of nutrients by trees grown in herbicide strips. In: 'Mineral
nutrition of fruit trees'. D. Atkinson, J. E. Jackson, R. 0. Sharples and W. M.
Waller. Butterworths. London - Boston. pp: 229-240.

Diaz Rossello, R. (19923). Evolucién de la materia organica en rotaciones de cultivos con
pasturas. Revista INIA Investigaciones Agronémicas, Uruguay 1(1): 103-110.

Diaz Rossello, R. (1992b). Evolucién del nitrogeno total en rotaciones con pasturas.
Revista IN IA Investigaciones Agronémicas, Uruguay 1(1): 27-35.

Donahue, R. L., J. C. Shickluna and L. S. Robertson (1971). Soils, an introduction to
soils and plant growth. Engelwood Cliffs, NJ, Prentice-Hall, Inc. 587 p.

Fryer, J. D. (1983). Recent research on weed management - New light on an old practice.

In: 'Recent advances in weed research'. W. W. Fletcher. The Gresham Press. 01d
Woking, Surrey. pp: 181-198.

30

Garcia, A. and A. Morén (1992). Estudios de C, N y P en la biomasa microbiana del
suelo en tres sistemas de rotacion agricola. Revista INIA Investigaciones
Agronémicas, Uruguay 1(1): 111-126.

Handayanto, E., G. Cadisch and K. E. Giller (1997). Regulating N mineralization from
plant residues by manipulation of quality. In: 'Driven by nature'. G. Cadisch and
K. E. Giller. CAB International Publishing. New York. pp: 175-185.

Hanson, E. J. (1996). Fertilizing fruit crops, E-852. Michigan State University, East
Lansing, MI. 20 p.

Hipps, N. A. and T. J. Samuelson (1991). Effects of long-term herbicide use, irrigation
and nitrogen fertiliser on soil fertility in an apple orchard. J. Sci. Food Agric.
55(1991): 377-387.

Hogue, E. J. and G. H. Neilsen (1987). Orchard ﬂoor vegetation management.
Horticultural reviews 9: 377-430.

Ingharn, R. E., J. A. Trofymow, E. R. Ingham and D. C. Coleman (1985). Interactions of
bacteria, fungi, and their nematode grazers: effects on nutrient cycling and plant
growth. Ecological Monographs 55(1): 119-140.

Langord, L. R. (1937). Effect of sod, cultivation and straw mulch upon orchard soil
moisture content. Proceedings of the American Society for Horticultural Science
3 5: 3 13-3 14.

Marini, R. P., J .L. Anderson, B.H. Barritt, G.R. Brown, J. Cline, W.P. Cowgill Jr., P.A.
Domoto, D.C. Ferree, J. Garner, G.M. Greene, C. Harnpson, P. Hirst, M.M.
Kushad, E. Mielke, C.A. Mullins, M. Parker, R.L.Perry, J .P. Prive, T. Robinson,
C.R. Rom, T. Roper, J .R. Schupp, E. Stover, R. Unrath. (2000). Performance of
'Gala' apple on 18 dwarf rootstocks; A ﬁve year summary of the 1994 NC-140
Semi-dwarf rootstock trial. J. of the American Pomological Society 54(2): 92-
107.

Merwin, I. A. and J. A. Ray (1997). Spatial and temporal factors in weed interference
with newly planted apple trees. HortScience 32(4): 633-637.

Merwin, I. A. and W. C. Stiles (1989). Root-lesion nematodes, potassium deﬁciency, and
prior cover crops as factors in apple replant disease. J. Amer. Soc. Hort. Sci.
1 14(5): 724-728.

Merwin, I. A. and W. C. Stiles (1994). Orchard groundcover management impacts on
apple tree growth and yield, and nutrient availability and uptake. J. Amer. Soc.
Hort. Sci. 1 19(2): 209-215.

31

Merwin, I. A., W. C. Stiles and H. M. van Es (1994). Orchard groundcover management
impacts on soil physical properties. J. Amer. Soc. Hort. Sci. 119(2): 216-222.

Needelrnan, B. A., M. M. Wander, G. A. Bollero, C. W. Boast, G. K. Sims and D. G.
Bullock (1999). Interaction of tillage and soil texture: Biologically active soil
organic matter in Illinois. Soil Sci. Soc. Am. J. 63: 1326-1334.

Neilsen, G. H. and E. J. Hogue (1985). Effect of orchard soil management on the growth
and leaf nutrient concentration of young dwarf red delicious apple trees. Canadian
J. Plant Science 65: 309-315.

Parker, M. L. (1990). The response of ﬁ'uit trees to orchard ﬂoor management.
Horticulture. East Lansing, Michigan State University: 136.

Paul, E. A., S. J. Morris and S. Bohm (2001). The determination of soil C pool sizes and
turnover rates: biophysical ﬁ'actionation and tracers. In: 'Assessment methods for
soil carbon'. R. Lal, J. M. Kimble, R. F. Follett and B. A. Stewart. CRC Press
LLC. Boca Raton. pp: 193-206.

Perry, R. (2000). The keys to maintaining productive Vertical Axe trees. The Fruit
Growers News 39(5): 21.

Perry, R. (2002). Spacing tree fruit.
http://www.hrt.msu.edu/department/Perry/Spacing Fruit/Spacing Frpit Index.ht
_rp. January 10, 2004.

Powlson, D. S. and D. S. J enkinson (1990). Quantifying inputs of non-fertilizer nitrogen
into an agro-ecosystem. In: 'Nutrient cycling in terrestrial ecosystems'. A. F.
Harrison, Ineson, P. Elsevier Applied Science. UK. pp: 56-68.

Robinson, T. L. (2003). Rootstock as a key component to high density orchards.
Conference 'Orchard systems workshop’, Geneva, NY. Int. Dwarf Fruit Tree
Assoc.

Sanchez, J. E., C. E. Edson, G. W. Bird, M. E. Whalon, R. R. Harwood, K. Kizilkaya, J.
E. Nugent, W. Klein, A. Middleton, T. L. Loudon, D. R. Mutch and J. Scrimger
(2003). Orchard ﬂoor and nitrogen management inﬂuences soil and water quality
and tart cherry yields. J. Amer. Soc. Hort. Sci. 128(2): 277-284.

Sarrantonio, M. (2003). Soil response to surface-applied residues of varying carbon-
nitrogen ratios. Biol. Fertil. Soils 37: 175-183.

Sooby, J. (1999). Meeting the Research Needs of Organic Farmers.
http://www.ofrf.m/research/needs.htrnl. May 18, 2004.

32

Tinsley, A. (2000). Nutrition of trees in organic systems. Conference 'Organic fruit
Opportunities and challenges', Ashford, Great Britain.

Wardle, D. A. and P. Lavelle (1997). Linkages between soil biota, plant litter quality and
decomposition. In: 'Driven by nature'. G. Cadisch and K. E. Giller. CAB
International Publishing. New York. pp: 107-124.

Webster, T. (2000). Apple and pear scion varieties and rootstocks for organic tree fruit
production. Conference 'Organic fruit opportunities and challenges', Ashford,
Great Britain.

Weibel, F. (2002). Soil management and in-row weed control in organic apple
production. The Compact Fruit Tree 35(4): 118-121.

Welker, W. V. and D. M. Glenn (1989). Sod proximity inﬂuences the growth and yield of
young peach trees. J. Amer. Soc. Hort. Sci. 114(6): 856-859.

Welker, W. V. and D. M. Glenn (1991 ). Growth response of young peach trees to
distribution pattern of vegetation-free area. HortScience 26(9): 1141-1142.

Yeates, G. W. (1998). Feeding in free-living soil nematodes: a functional approach. In:
'The physiology and biochemistry of free-living and plant-parasitic nematodes'. R.
N. P. a. D. J. Wright. Cab International. Palmerston North, New Zealand. pp: 245-
269.

33

Table 1.1 Carbon mineralization potential of soil. Cumulative amount of C02-C (ug of
CO2-C * g of soil ' 1) evolved during incubation in laboratory at 25°C and no light from
soil of different positions of the sandwich treatment. Values are means of six replicates.
Samples were collected April 24th 2003. Analysis of variance was carried out for each
date after incubation. Values followed by the same letter within the row are not
signiﬁcantly different for p50.05 (LSMEANS test).

 

 

 

 

 

 

 

Day Row Strip Alley
21 248b 284 ab 320 a
30 331 b 390 a 435 a
50 466 b 555 a 614 a
77 645 b 787 a 859 a
100 788 b 953 a 1029 a
155 1070 b 1297 a 1371 a

 

Table 1.2 Nitrogen mineralization potential. Nitrogen available in soil (mg Kg") after
different days of incubation in laboratory at 25°C and no light from soil of the sandwich
treatment in its three positions (0 — 10 cm depth) Values are means of six replicates.
Sample collected April 24th 2003. Analysis of variance was carried out for each date.
Values followed by the same letter within the row are not signiﬁcantly different for

1950.05 (LSMEANS test).

 

 

Day Row Strip Alley
O 16.54 b 19.31 b 32.76 a
30 36.59 b 42.28 b 52.54 a
70 53.08 b 57.21 b 78.31 a
165 69.48 c 83.94 b 92.15 a

 

34

Table 1.3. Soil food web. Microbial Biomass of bacteria and fungi (in rig/g soil) and
protozoa (in number/ g soil) at 0-30 cm depth in the treatment row. Each value is mean of
6 replicates. Samples were collected November 10th 2003. Values followed by different
letters in the same row are signiﬁcantly different for 150.05 (LSMEANS test).

 

 

 

 

Microorganisms Mulch Flame Sandwich
Active Bacteria (AB) 29.16 a 33.08 a 39.18 a
Total Bacteria (TB) 531.67 a 503.93 a 723.90 a
AB / TB 0.08 a 0.08 0.09 a
Active Fungal (AF) 7.85 ab 5.86 b 10.51 a
Total Fungal (TF) 79.97 a 100.46 a 94.34 a
AF / TF 0.10 a 0.07 a 0.12 a
AP / AB 0.27 a 0.19 a 0.27 a
TF / TB 0.21 a 0.23 a 0.20 a
Protozoa

Flagellates 3947.86 a 2419.58 a 2129.40 a
Amoebae 5506.97 a 2639.87 a 3735.00 a
Ciliates 1958.67 a 181.48 b 195.95 b

 

35

Table 1.4 . Microbial Biomass of bacteria and ﬁrngi (in rig/g soil) and protozoa (in
number/g soil) at 0-30 cm depth in row and strip positions of the Sandwich treatment.
Each value is mean of 6 replicates. Samples were collected November 10th 2003. Values
followed by different letters in the same row are signiﬁcantly different for pS0.05
(LSMEAN S test).

 

 

 

 

Row Strip

Active Bacteria (AB) 39.18 a 26.79 b
Total Bacteria (TB) 723.90 a 496.96 b
AB / TB 0.09 a 0.08 a
Active Fungal (AF) 10.51 a 6.57 b
Total Fungal (TF) 94.34 a 66.94 b
AF/TF 0.12 a 0.10 a
AP / AB 0.27 a 0.25 a
TF / TB 0.20 a 0.21 a
Protozoa

Flagellates 2129.40 a 4824.23 a
Amoebae 3735.00 3 4702.99 a
Ciliates 195.95 a 201.67 a

 

Table 1.5 Nematodes population characterization at 0-30 cm depth in the row by
treatment (in Number of nematodes/g soil). Each value is mean of 6 replicates. Samples
were collected November 10th 2003. Values followed by different letters in the same row
are signiﬁcantly different for p30.05 (LSMEANS test).

 

 

Nematodes Mulch Flame Sandwich
Bacterial feeders 0.79 A 0.21 b 0.71 a
Fungal feeders 0.03 A 0.02 a 0.04 a
Fungal/Root feeders 0.02 A 0.02 a 0.08 a
Predatory 0.02 A 0.01 a 0.02 3
Root feeders 0.01 B 0.04 b 0.1 1 a
Total 1.03 A 0.34 b 1.11 a

 

36

Table 1.6. Yearly Trunk Cross Sectional Area Increase (TCAI) of ‘Paciﬁc Gala’ in cm2
by rootstock for the three treatments from 2001 to 2003 measured 30 cm above the graft
union. Each value is an average of 12 trees. Analysis of variance was carried out for each
year. Values followed by different letters in the same column or row are signiﬁcantly
different for p30.05 (LSMEAN S test).

 

 

 

 

Rootstock Mulch Flame Sandwich ROOtStOCk
average
TCAI 2001
Supporter4 1.7 1.5 1.5 1.6 a
M.9-RN29 1.5 1.1 1.7 1.4ab
M.9 — NAKB 337 1.5 1.2 1.0 1.2 b
Treatment 1 .6 1 .2 1 .4
TCAI 2002
Supporter4 4.4 3.5 3.6 3.8 a
M.9 - RN 29 4.7 2.8 3.0 3.5 ab
M.9 — NAKB 337 3.6 3.0 2.7 3.1 a
Treatment 4.2 a 3.1 b 3.1 b
TCAI 2003
Supporter4 7.5 7.2 5.9 6.9 a
M.9 - RN 29 5.4 5.3 4.9 5.2 b
M.9 - NAKB 337 5.4 4.7 4.1 4.7 b
Treatment 6. 1 5.8 5.0

 

37

Table 1.7. Branch ﬁnal length of ‘Paciﬁc Gala’ in cm, by rootstock for the three
treatments from 2001 to 2003. Each value is an average of 12 trees. Analysis of variance
was carried out for each year. Values followed by different letters in the same column or
row are signiﬁcantly different for p30.05 (LSMEANS test).

 

 

 

 

Rootstock Mulch Flame Sandwich Rootstock Avg
a) 2001
Supporter 4 72.4 64.3 69.1 68.4 a
M.9 - RN 29 71.2 64.6 74.4 70.0 a
M.9 - NAKB 337 61.0 61.6 56.1 59.5 b
Treatment Avg 67.9 63.5 66.3
b) 2002
Supporter 4 61.3 56.8 55.7 57.9 a
M.9 - RN 29 58.3 55.5 51.2 55.0 a
M.9 - NAKB 337 51.9 51.0 45.8 49.6 b
Treatment Avg 57.1 a 54.5 ab 50.9 b
c) 2003
Supporter4 46.4 51.7 52.0 50.1 a
M.9 - RN 29 44.6 48.1 43.3 45.3 b
M.9 - NAKB 337 44.9 45.4 45.8 45.3 b
Treatment Avg 45.3 48.4 47.1

 

38

Table 1.8. The effects on tree vigor, ﬂowering, and fruit set of ‘Paciﬁc Gala’ for 2003
by: A. orchard ﬂoor management systems, and B. rootstocks. Each value is an average of
12 trees. Analysis of variance was carried out for each variable. Values followed by
different letters in the same row are signiﬁcantly different for p50.05 (LSMEAN S test).

 

 

 

 

 

A TREATMENTS

units Mulch Flame Sandwich
Canopy volume m3 3.65 a 2.85 ab 2.48 b
Pruning wood weight gr 100 a 64 b 55 b
No. of ﬂower clusters number 194 a 69 b 73 b
No. clusters/TCA No./om2 15.2 a 6.4 b 6.3 b
Fruit set * % 13.2 8.2 15.5
B ROOTSTOCKS

units Supporter 4 M.9 RN29 M.9 NAKB337
Canopy volume or3 3.10 ab 3.17 a 2.71 b
Pruning wood weight gr 83 a 78 a 57 b
No. of ﬂower clusters number 94 b 121 a 122 a
No. clusters/T CA No./cm2 6.3 b 10.5 a 11.1 a
Fruit set * % 6.1 16.2 14.6

 

* signiﬁcant interaction rootstock by treatment

Table 1.9. Leaf N content (% dry weight) of ‘Paciﬁc Gala’, by treatment and rootstock
from 2001 to 2003 (n=36 in 2001, and n=12 in 2002 and 2003) Analysis of variance was
carried out for each year. Values followed by different letters in the same column are
signiﬁcantly different for p50.05 (LSMEAN S test).

 

 

 

 

 

2001 2002 2003

TREATMENTS
Mulch 2.14 2.42 a 2.68 a
Flame 2.10 2.39 a 2.43 b
Sandwich 2.10 2.25 b 2.24 c

ROOTSTOCKS
Supporter 4 2.34 2.48
M.9 RN29 2.36 2.38
M.9 NAKB 337 2.36 2.49

 

Table 1.10. Three-year average nutrient content for leaves of ‘Paciﬁc Gala’ sampled
from the middle portion of one-year-old branches in mid—August 2001 to 2003, in trees
under different orchard ﬂoor management systems. Each value averages 12 trees. Values
followed by different letters in the same column are signiﬁcantly different for p30.05
(LSMEAN S test).

 

 

 

P (%) K (%) Mn (ppm) Zn (ppm)
Mulch 0.21 b 2.04 a 42.42 a 14.34 b
Flame 0.25 a 1.90 b 31.53 b 19.81 a
Sandwich 0.27 a 1.98 ab 30.34 b 18.88 3
Std error 0.013 0.041 1.849 1.606

 

Table 1.11. The effect on quantitative and qualitative characteristics of ‘Paciﬁc Gala’ in
2003 of: A. orchard ﬂoor management systems, and B. rootstocks. Each value is an
average of 12 trees. Analysis of variance was carried out for each variable. Values
followed by different letters in the same row are signiﬁcantly different for pS0.05
(LSMEANS test).

 

 

 

 

 

 

 

A TREATMENTS
unit Mulch Flame Sandwich
Total production * kg/plant 3.37 1.27 2.08
No. of fruit No./plant 28 a 9 b 14 b
Average weight g/fruit 132 102 123
Weight/T CA g/cm2 272 128 180
B
ROOTSTOCKS
unit Supporter 4 M.9 - RN 29 M.9 - NAKB 337
Total production * kg/plant 0.94 2.78 3.00
No. of fruit No./plant 8 b 21 a 21 3
Average weight g/fruit 80 b 129 a 148 a
Weight/TCA g/cm2 61 b 241 a 278 a

 

* signiﬁcant interaction between rootstock and orchard ﬂoor management system

40

SOM (%)

SOM (%)

4.0 g

3.5 .

3.0

2.5

2.0 ‘

1.5 i

 

I - e- Mulch +Flame --D-- Sandwich

0.5 ‘

0.0

4.0 ,

3.5

3.0 '

2.5 1

2.0

1.5 A

1.0

0.5

0.0

F“ _

j— . - l

APR/2001 MAY/2002 AUG/2002 DEC/2002 APR/2003 JUL/2003 AUG/2003 NOV/"2003

 

- 0- Mulch +Flame "Cl-- Sandwich

f 7

APR/2001 MAY/2002 AUG/2002 DEC/2002 APR/2003 J UL/2003 AUG/2003 NOV/2003

Figure 1.1. Content of soil organic matter (SOM) in the row position (expressed as % dry
weight) by sampling date. A. for 0 — 10 cm depth. B for 0 — 30 cm depth. Each point
represents a mean of six samples (:tSE). Analysis of variance was carried out for each
date. Values with the same letter are not signiﬁcantly different for pS0.05 (LSMEANS

test).

41

4.0

 

3.5 -

3.0 _

2.5 .

2.0 .

 

MID-l

1.5

SOM (%)

1.0

-0' Row --D--Strip +Alley

0.5 —

 

 

0.0 w
APR/2001 MAY/2002 AUG/2002 DEC/2002 APR/2003 JUL/2003 AUG/2003 NOV/2003

 

4.0

3.5 . B

3.0 -

 

2.5 . a a

2.0 ~

SOM (%)

1.5 --

 

1.0:

0.5 - 0- Row --1:1-- Strip +Alley

 

 

0.0

 

APR/2001 MAY/2002 AUG/2002 DEC/2002 APR/2003 JUL/2003 AUG/2003 NOV/2003

Figure 1.2. Content of soil organic matter (SOM) in the three positions of Sandwich
(expressed as % dry weight) by sampling date. A. for 0 — 10 cm depth. B. for 0 — 30 cm
depth. Each point represents a mean of six samples (iSE). Analysis of variance was
carried out for each date. Values with the same letter are not signiﬁcantly different for
pS0.0S (LSMEANS test).

42

 

 

 

 

 

 

 

 

 

140
- o- Mulch +Flame --C1-- Sandwich 3 A
120
’¥
0 0
100 ' ‘
' s
A a ' g
E ,' -
‘3‘ 80 « . ‘
a a I “
Z i a ' o ' |
' n . . o I
-' 60 - . i .
m o s
O o . a
z a l b | a
40 ~
20 J
0 - . . . - . . . ' .., . . . .
APR/2001 JUN/2001 AUG/2001 NOV/2001 MAY/2002 AUG/2002 DEC/2002 APR2003 JUN/2003 AUG/2003 NOV/2003
Date
60
- 0- Mulch +Flame "D" Sandwich a
. g B
50 ”l o. s
a ’ ‘
, 0
a . . a
% \
A 40 " “ O ‘0 0;
E ' | 0 s o s
o u
g 0' “ a " ‘s x s
o ‘ a
Z 30 — . “ {0 ‘o a o. ‘
l t o ‘ 0 ‘
o o s
o M a ab “ a ' ‘f 0
O 20 _ r ' ‘
Z ~_. ‘{ '0 b b “ a
s o
a. -. Q ‘ . a .0 b b E
10 - ~. r; . °°°°° . b
b .e... 0". .0. .......... .9 b
b .Q. b ."
0 w , y - . - - . -
APR/2001 JUN/2001 AUG/2001 NOV/2001 MAY/2002 AUG/2002 DEC/2002 APR2003 JUN/2003 AUG/2003 NOV/2003
Date

Figure 1.3. Content of N03'— N in soil for position row (expressed as ppm) by
sampling date. A. for 0 — 10 cm depth . B. for 0 - 30 cm depth Each point represents a
mean of six samples (:tSE). Analysis of variance was carried out for each date. Values
with the same letter are not signiﬁcantly different for p50.05 (LSMEANS test).

43

 

l 400

1200

l 000

800

400

ugC02 — C per g of soil

200

 

600 '1

 

  
  

o mulch
A ﬂame
D sandwich

 

 

1'

50 100 150 200

Days of incubation

Figure 1.4. Carbon mineralization. Cumulative 11g C02-C evolved per g of soil during
155 days of laboratory incubation at 25°C and no light for soil from row position of three
orchard ﬂoor management systems. Samples were collected on April 24th 2003.

44

 

1400 - I 0-77 days 13 78-155 days

1200 -
1000 .
800 .
600

400 ‘

ugC02 -— C per g of soil

200 4

 

 

 

Mulch Flame Sandwich
TREATMENTS

Figure 1.5. Mineralized C of soil from rows, after 77 and 155 days of laboratory
incubation as affected by orchard ﬂoor management system. Samples were collected
April 24th 2003. Bars with the same letter are not signiﬁcantly different for p50.05
(LSMEAN S test).

45

 

100

 

0103+NH0—N <mgkg“)

 

 

 

0 i I T r r l
0 20 40 60 80 100 120 140 160 180

Days of incubation

Figure 1.6. Nitrogen mineralization potential. Nitrogen available in soil (mg Kg") after
different days of incubation in laboratory at 25°C and no light from soil of the row
position (0 - 10 cm depth) from different treatments. Each value is a mean of six
samples. Samples were collected on April 24th 2003.

46

30

 

 

 

 

 

25
e 20-
>
e".
o
B 15 ~
0§
o
E
'5 10
m
- 0 - Mulch
5 +Flame
~-D--Sandwich
0
«0’ (\\<5 «(9’ «9? $9?) chit” 0,6» \Q\\ . ‘)\\°’ (00:) «(l9 g3 0,?) \Q\<\
2002 2003
Date

Figure 1.7 Volumetric soil moisture content (as %) measured with TDR for the average
of 0-45 cm depth, on the tree row during years 2002 and 2003.

47

 

 

.—
N
l
r——1
pr

 

p—H

TCAI (cmz)

 

 

 

 

 

 

 

Mulch Flame Sandwich
TREATMENT

 

 

.—
N
H—cm

or

0"

TCAI (cmz)

 

 

 

 

 

 

Supporter 4 M.9 RN 29 M.9 NAKB 337
ROOTSTOCK

Figure 1.8. Cumulative trunk cross sectional area increase (TCAI) of ‘Paciﬁc Gala’ in
cm2 between 2000 and 2003 as affected by: A. orchard ﬂoor management systems, and
B. rootstocks. Analysis of variance was performed between rootstocks and treatments.
Letters indicate signiﬁcant differences for {230.05 (LSMEANS test).

48

 

30
C1 Supporter 4
I M.9 RN 29

25 ' IM.9 NAKB 337

Fruit set (%)

 

 

 

Mulch Flame Sandwich
TREATMENTS

Figure 1.9. Fruit set of ‘Paciﬁc Gala’ grafted on three different rootstocks and under
different orchard ﬂoor management systems for year 2003. It shows signiﬁcant
interaction between orchard ﬂoor management systems and rootstocks.

49

 

Cl Supporter 4
l M.9 RN 29
I M.9 NAKB 337

 

 

 

Mulch Flame Sandwich
TREATMENTS

Figure 1.10 First harvest. Fruit production in kg/plant of ‘Paciﬁc Gala’ by rootstock and
for the three treatments. Each bar is average of 12 trees. Year 2003. It shows signiﬁcant
interaction between orchard ﬂoor management systems and rootstocks.

50

CHAPTER 2

51

CHAPTER 2

EFFECT OF DIFFERENT GROUND FLOOR MANAGEMENT SYSTEMS FOR
ORGANIC APPLE PRODUCTION ON NITROGEN CYCLING AND TREE
PERFORMANCE

Abstract
Genetic and environmental conditions determine performance of trees. This study was
conducted to evaluate the potential of dwarf and semi-dwarf rootstocks to compete for
resources under different orchard ﬂoor management systems that comply with organic
production regulations. Apple trees of ‘Buckeye Gala’ grafted on M.9 NAKB 337 and
Supporter 4 were planted in 0.6m3 wooden boxes. A system was designed to collect total
leachate for its analysis. Orchard ﬂoor treatments evaluated were: alfalfa hay mulch,
propane ﬂame burner, Swiss Sandwich system (a combination of resident vegetation and
tilled strips), and partial or total soil cover with a mix of red clover and rye grass. These
systems were compared from 2001 to 2003 in regard to their impact on soil, leachate, and
tree performance. Mulch had highest leachate volume with higher total N content more
than ten times that of the other treatments. Flame had the least increase in volume of
leachate indicating lower water inﬁltration rates than the treatments with some type of
soil cover. ‘Buckeye Gala’ on Supporter 4 was more vigorous with a trunk cross sectional
area increase over 40% higher than on M.9 NAKB 337. Conversely trees on M.9 NAKB
337 were more precocious. The signiﬁcant interaction between orchard ﬂoor
management system and rootstock indicates the potential of compensating stressing

conditions, within a certain range, with a more vigorous rootstock.

52

INTRODUCTION

Orchard ﬂoor management practices aim at improving natural processes which
will enhance crop health, productivity and environmental quality (Landis et al., 2002).
This is achieved through conserving and boosting beneﬁcial chemical, physical and
biological properties of the soil, thus providing an appropriate environment for tree
growth. Organic matter is part of the foundation for a healthy and productive soil
(Magdoff and van Es, 2000). It feeds soil life and contains large quantities of plant
nutrients, acting as a slow-release nutrient storehouse, specially for N (Brady and Weil,
2002). In fruit growing one of the main challenges is to manage the N supply for trees.
and under organic production the accent for this should be on building organic matter
rather than on fast mineralization (Bloksma, 2000).

The N transformation processes that take place in the soil and their timing will
vary greatly according to the practices followed to maintain or increase the organic
matter and soil’s quality. Concerns about N as a pollutant have been growing in the last
decades. Contamination of surface water as well as deep water resources through the
leaching of N from fertilizers used on agricultural land is a potential hazard that needs to
be addressed not only by efﬁcient practices but also by an increased understanding of N
cycling. In organic production, the use of chemical fertilizers, which are in general,
highly soluble and readily available, is prohibited. The use of organic forms might seem
less risky, but problems of potential N loss to the environment are not completely
eliminated in organic systems. The N release from different organic sources varies with

the quantity and quality of the material used and the soil environment (Myers et al., 1997;

53

Wardle and Lavelle, 1997; Magdoff and van Es, 2000; Brady and Weil, 2002).
Synchronizing the N availability with the plant’s demand is highly important for higher
efﬁciency and lower probabilities of leaching. Proper selection and management of the
sources will have an important effect on N availability.

Another factor that affects N availability, in the case of fruit trees, is competition
from understorey vegetation growing in the same area. In organic ﬁ'uit production, as for
all organic systems, the need for more sustainable approaches of weed management
compatible with the organic protocols has been identiﬁed as a major constraint (Sooby,
1999). Thus, efforts have been put into the development of alternative orchard ﬂoor
management systems (Weibel, 2002). The soil cover can attain a critical role in the N
dynamics and is increasingly important in the organic approach.

Research has been done to determine the adaptability of rootstocks to various soil
types and conditions (F erree and Carlson, 1987). While differences among some
rootstocks in regard to water utilization have been measured (Carlson, 1967) we are still
missing information on the response of many clones. Nutrient uptake can also be affected
by rootstocks however, in the case of nitrogen and phosphorous Kennedy et al. (1980)
reported they were more inﬂuenced by environmental factors than by rootstock genotype.
The strong relationship between the genetic and environmental conditions in determining
the root system will have an effect on mineral nutrition of the tree and its performance.

We hypothesized that rootstocks of different vigor will have diverse ability to
compete and compensate at a higher level of nutrient and/or water stress. Through
various soil covers and understorey management the timing and amount of N and water

available can be changed and their consumption will be different according to the

54

 

 

rootstock. The overall objectives of this work were to: 1) assess the effects of orchard
ﬂoor management systems on N availability and potential leaching; 2) evaluate
performance of dwarf and semi-dwarf rootstocks in their ability to compete for resources;
3) determine the impact of rootstocks and orchard ﬂoor management systems on nutrient

and water relations.

MATERIALS AND METHODS

Site description

The experiment was conducted at the Clarksville Horticultural Experiment Station
of Michigan State University in the vicinity of Clarksville, MI (latitude of 42° 84’ N and
longitude of 84° 24’ W). The orchard has a total surface of 2.5 ha and is being certiﬁed
by the Organic Crop Improvement Association (OCIA). Daily temperatures and
precipitation during the whole experiment were recorded with an automated weather
station of the Michigan Automated Weather Network, located approximately 560 m west
of the orchard (Figure A.2 thru A.4 , Appendix A)

Wooden boxes of 1 m x 1.2 m and 0.5 m high were used for planting. The boxes
rested on the soil surface and they were arranged in two rows at the orchard with plants at

1.4 m x 8 m spacing.

55

 

 

Plant material

One-year-old “Buckeye Gala” apple trees (Malus x domestica Borkh.) were
planted in April 2001 in the boxes that had been previously lined with plastic and ﬁlled
with soil. The soil used was homogeneous, taken from the orchard site being A horizon
of Kalamazoo sandy clay loam (53.1% sand, 23.1% silt, and 23.8% clay). Plant residues
present were eliminated. One-half of the trees were grafted on M.9 NAKB 337 which has
the vigor of a weak M.9 growing to 40% of the size of a seedling, and the remaining trees
were grown on Supporter 4, which is a cross of M.9 x M.4, a more vigorous rootstock
than M.9, with a size comparable to M.26 (Childers, 1972; Marini, 2000; Robinson,
2003). Trees had similar caliper within each rootstock at planting time.

A trellis system with two wires was installed and galvanized metal poles were
used for support of each tree. Plants were trained to a Vertical Axe with the use of rubber
bands and clothes pins to bend branches (Perry, 2000a). Minimal pruning was necessary,

mainly to single out the leaders or main branches.

Orchard ﬂoor management

Five different orchard ﬂoor management systems were established among boxes
in spring of 2001 at planting time:
1. Mulch (MU): The whole area of the box (1.2 m2) was covered with a mulch of alfalfa
hay. During the growing season the thickness of the cover was over 15 cm to provide a
shading effect that would suppress weeds, and maintain moisture. Mulch was hand-

applied in spring and fall.

56

2. Flaming (FL): The entire area of the box (1.2 m2) was kept free of vegetation through
the use of a propane ﬂame burner hand torch that provided 150.000 BTU. Minimal
amounts of water were applied just behind the burner to reduce the risk of heat affecting
tree branches or ﬁre spreading to the neighboring mulch. Flame treatments were applied
when weed height approached 10 cm, 4 to 7 times according to season (Table B.1.A,
Appendix B). This was considered the control treatment.

3. Sandwich System (SS): an adaptation of the Swiss Sandwich system (Weibel, 2002),
consisted of a central strip that spread 20 cm to each side of the tree row in which
volunteer vegetation was allowed to grow. A 40 cm wide strip was tilled on each sideof
that vegetated central strip. The tilling was performed with a hand hoe to a depth of 7 cm.
In the vegetated central strip, hand weeding was performed selectively to prevent
development of curly dock (Rumex crispus L.), common lambsquarters (Chenopodium
album L.), redroot pi gweed (Amaranthus retroﬂexus L.) and thistles (Cirsr'um sp.). In the
lateral tilled strips, bell beans (Viciafava) were sown (45 kg.ha'l in 2001 and 130 kg.ha’1
in 2002) in the fall as a green manure. Tilling started in spring, and was repeated when
weeds got 10 cm tall (4 to 6 times according to season, Table 3.1.3, Appendix B).

4. Partial Cover (PC). Just after the planting of trees a mixture of red mammoth clover
and endophytic rye was sewn. The cover was allowed to grow in a central strip that
spread 20 cm to each side of the tree trunk. The remaining area on each side (40 cm wide
strip), was tilled at the same time as SS.

5. Total Cover (TC). As in the previous treatment, a mixture of red mammoth clover and
endophytic rye was sewn just after planting the trees. In this case, the mixture was

allowed to grow in the whole area of the box.

57

Irrigation system

Drip irrigation was installed in May 2001. The lines were suspended from the
lowest wire of the trellis system at a height of 0.3 m to prevent potential damage to them
when controlling weeds. Drippers of 3.8 L/h supplied water to the ﬁve soil management
treatments with the goal of maintaining available soil water over 50% (Figure 31 thru

33., Appendix B).

Tree growth gnd productivity measurements

Trunk cross sectional area was calculated from the diameter of the trunk measured
at 25 cm above the graft union. These measurements were taken at the end of each
growing season. Branch growth was measured on ﬁve branches chosen at the beginning
of the spring. The ﬁrst four were selected to represent the bottom, middle, and lower-top
parts of the tree, and the ﬁfth branch was the leader. Branches sampled in each position
were similar in size at the beginning of the measuring period and had a similar angle of
insertion to the trunk. Branch length was measured weekly between June and August. In
2002 at the end of the season all growing branches were measured in all trees. Number of
ﬂower clusters in the whole tree was counted before full bloom each year, and total
number of fruits as well as weight was measured per tree for the ﬁrst harvest in 2003.

In March — April 2001 and 2002 there was little pruning needed. The branches cut
during the following season (2003) were weighed as a complementary measure to

estimate growth and vigor of trees.

58

At the end of each season the total height of the tree as well as two orthogonal
diameters of the canopy at 0.5 m were measured and used to estimate the canopy volume.
In 2002 and 2003 the total growth in length of the tree was measured by adding

the individual length of each branch of that year or the amount it had grown.

Nutritional status

A sample of 20 leaves per tree was taken in early August of each year from the
middle portion of one-year-old branches. Leaves were rinsed with distilled water, dried
for 48 h at 60 ° C and analyzed for mineral composition.

Relative chlorophyll content was measured using a SPAD-502 meter (Spectrum
Technologies Inc, Plainﬁeld IL). This measurement was taken on the same leaves
sampled for mineral composition. In 2003, between June and September, variations in

leaf chlorophyll abundance were followed through six measurements.

Leachate collection

To analyze the potential leaching of the different treatments, the wooden boxes
were lined with a plastic ﬁlm before ﬁlling them with soil. In the bottom, a hole was
made and a funnel installed, sealing all the unions, so that leachate could be collected
(Figure 2.1). This water drained through a short pipe into a jug that was in a 20 liter
bucket put below the bottom level of the wooden box. Regularly, and according to the
water regime, the jugs were emptied, the volume of leachate measured, and a sample

taken to analyze N content. The bucket lid and pipe were painted to diminish light

59

penetration that would allow algae growth. As a preventive strategy, once a year the jugs

were rinsed with alcohol.

Soil sampling_and measurements

Soil was sampled individually for each box, three times in 2001 and 2002 and
four in 2003, by mixing 3 cores taken for the 0-30 cm depth. Samples were stored in
plastic bags at 4°C for less than a month until analysis. Soil moisture was measured by
time domain reﬂectometry (TDR) using a Mini Trase 6050X3 (Soilrnoisture Equipment
Corp., Goleta, CA) with 30 and 45 cm long stainless steel rods permanently installed in
the boxes in the tree rows. Inorganic N (N03‘ and NH4+) was determined by the MSU
Soil and Plant Nutrient Laboratory following the procedure described by (Kenney and
Nelson, 1982) using a Lachat automated colorimetric analyzer (Lachat Instruments Inc.
Milwaukee, WI).
Organic matter concentration was measured by loss on ignition of 3g of dry soil at 400°C
for 8 hours using a mufﬂe fumace. Carbon and N mineralization potentials were obtained

through laboratory incubation (see Chapter 1 for detailed methodology).

Experimental desigp and statistical analysis

The experiment was set as a completely randomized design with eight replicates
of the ﬁve orchard ﬂoor management systems. The plots consisted of 1 tree planted on a
wooden box.. Orchard ﬂoor management system treatments and rootstocks were
randomly assigned. Additional boxes of two of the treatments, SS and TC, were installed

without trees.

60

Analysis of variance was performed using MIXED procedure with SAS (Version
8, SAS Institute, Cary, NC, USA) to detect treatment effects. After using ANOVA to
identify treatment effects, least squares means were compared using the Tukey-Kramer

test with p_<_0.05 unless stated otherwise.

RESULTS
Soil parameters

Soil Organic Mm

No signiﬁcant differences were found for the SOM content among the treatments
except for the last sampling date (Figure 2.2). In November 2003 SOM in Mulch (MU)
was not signiﬁcantly different from Total Cover (TC) and Partial Cover (PC) but was
signiﬁcantly higher than Flaming (FL) and Sandwich (SS). No interaction was found
between treatment and rootstock for this variable. In the case of SS no signiﬁcant

differences were found between the row and strip content of organic matter (Table 2.1).

Nitrogen in soil

The effect of alfalfa hay application (Table 8.2 and 33, Appendix B) on the
content of nitrate in soil was measurable throughout the experiment (Figure 2.3). The MU
had a higher content at all times, approaching on different dates as much as a ten-fold
increase compared with the rest. The SS started with higher nitrate content in soil than
TC and PC, but from November 2001 on, differences were only slight. For SS, the row
tended to lower values of nitrate in soil than the strip (Table 2.2) although signiﬁcant

differences were measured only in August 2002 and April 2003. The rootstocks averaged

61

for all treatments did not generate signiﬁcant changes in the values of N in soil (data not
shown).

There was a clear seasonal pattern of the N forms in soil. The content of N03'—N
in soil decreased towards the end of the season (November sampling) when NHf—N

became a higher portion of the available N in soil (Figure 2.4).

Soil water content

The soil moisture measured to a depth of 45 cm had signiﬁcant differences
between treatments at certain dates (Figure 2.5 and Figure 2.6). Soil moisture content was
highest for MU than other treatments. It was closely followed by FL in 2002. During

some short periods SS, PC, and TC had slightly lower content.

Leachate

The orchard ﬂoor management systems had an effect on the volume of leachate
collected from boxes with apple tree in two of three years (Figure 2.7) and impacted the
N concentration similarly. The highest volume came from MU for 2001 and 2003. In
2001 MU was followed by FL and SS with approximately 35% less volume and ﬁnally
PC and TC which had an average of 54% less volume leached. In 2003 except for MU
which was the highest, there were no differences among the rest of the treatments. From
2001 to 2003, FL had a volume increase of 32 %, while SS almost doubled that increase
(59 %), and PC as well as TC increased the leachate in a much higher percentage (118
and 154 % respectively) going from values signiﬁcantly lower than the ﬁrst two

treatments, to slightly higher ones, even though not statistically different.

The N03'—N concentration in leachate of 2001 had values that overall, were
higher than in 2002 and 2003. During the 3 years, MU had the highest values for
concentration of NO3'—N in the leachate for both rootstocks (Table 2.3).

Rootstocks had a limited inﬂuence on N concentration with no signiﬁcant
differences among them.

The impact of the orchard ﬂoor management systems reached the total amount of
N, calculated from volume and concentration of each sample, with MU far above the
other treatments (Figure 2.8). The maximum difference was in 2003 when MU had a
content in leachate of N03'—N for the total year, 80 times that of SS. In general, and for

all treatments, the amounts were higher for 2001 and diminished in the following years.

Tree parameters

Trunk size

Soil management treatments as well as rootstocks bad inﬂuence on the tree
growth measured as trunk cross sectional area increase (TCAI) (Figure 2.9). The growth
under TC was the lowest and under MU it was the highest although not signiﬁcantly
different from FL in the case of Supporter 4. This explains the signiﬁcant interaction
treatment by rootstock. At the highest TCAI values, the difference for MU between the
two rootstocks was not a signiﬁcant value, showing a different behavior than for FL, SS
and PC, in all of which Supporter 4 had a higher TCAI than M.9 NAKB 337. In the case
of TC, at the lower extreme of values, there was no signiﬁcant difference between

Supporter 4 and M.9 NAKB 337, the same as MU just mentioned above.

63

Branch growth

In 2002 and 2003 the orchard ﬂoor management systems gave signiﬁcant
differences for total branch growth as well as number of branches (Figure 2.10). The
highest growth and number was for MU. Then followed FL, SS, and PC, which
generally did not differ signiﬁcantly between them, and lastly was TC with signiﬁcantly
lower branch grth and number of branches. Supporter 4 was found to be more

vigorous in branch growth than M.9 NAKB 337.

Tree size

Orchard ﬂoor management systems and rootstock had inﬂuence on tree volume
generating signiﬁcant differences in 2002 as well as 2003. The ﬁnal tree size in 2003 of
MU was greatest but not signiﬁcantly different from FL and SS, while TC had the

smallest trees (Figure 2.11).

Number of ﬂower 01%

Orchard ﬂoor management had no effect on count of ﬂower clusters in 2001 but
did impact signiﬁcantly in 2002 and 2003 (Table 2.4). The treatment effect was stronger
in the last season when MU had maximum number of ﬂowers, followed by FL and SS,
and ﬁnally with lower values PC and TC.

Rootstock had a signiﬁcant impact on number of ﬂower clusters as well.

Supporter 4 was below the ﬂowering potential showed by M.9 NAKB 337.

64

Nutritional status

The different systems of orchard ﬂoor management impacted the level of
nutrients. The foliar N content in 2001 was slightly under the recommended range
(Hanson, 1996) and in good stand for 2002 (Table 2.5). No signiﬁcant differences were
found between treatments during those two years. In 2003 while the value for MU
increased slightly (2.32 %) for the other treatments it diminished generating signiﬁcant
differences.

The treatment MU had lower values of P than the other soil management systems
(Table 2.6). This behavior was similar for Ca and Mg with a trend to increase content as
years passed (Table B.4, Appendix B). In the case of K the impact was opposite to what
was found in the previous nutrients, having MU the highest values. For K as well, all
treatments showed a trend to increase content with years.

Soil management had no effect on some micronutrients like in the case of Fe, Al,
and B (Table B.4, Appendix B). For Zn higher levels were found in PC, TC and SS,
while in the case of Mn the opposite occurred with MU and FL having higher content

(Table 2.6).

DISCUSSION

Soil pmeters
Generally, changes in soil organic matter (SOM) are slow, over time, as

inﬂuenced by different soil management practices (Diaz Rossello, 1992; Paul et al.,

65

2001). The effects will even take longer when the whole A horizon up to a depth of 30
cm is studied, which generates almost a dilution effect (see Chapter 1). Also we must
consider that during the ﬁlling process of the boxes soil mixing occurred and natural
macroporosity was affected. Nevertheless, we started ﬁnding differences after the period
of 3 years in the last soil sampling. The effect of the addition of large amounts of alfalfa
hay straw increased, as it would be expected the SOM content (Merwin et al., 1994;
Sanchez et al., 2003). On the contrary the continuous burning of the understorey without
giving it much opportunity of developing green biomass in the FL lead to a lower content
of SOM. The addition of a vegetated strip and tillage in the SS still has SOM values
lower than MU. In the case of PC and TC with an intermediate value it seems as if the
active growth of the cover with a dense root system of the species involved (red
mammoth clover and endophytic rye) could have been the cause of slightly higher values.

Filling of the boxes generated a disruption of the soil with an enhancement of
mineralization (Magdoff and van Es, 2000; Brady and Weil, 2002) that could be seen
through the values of N03'—N in soil (Figure 2.3) which were high for all treatments.
Levels were still high in August 2001. But those values were quickly changed by effect
of the different soil management systems imposed. The trees increased the N uptake as
they developed their root system. The rye grass and clover cover in TC and PC started
actively absorbing the readily available N that was present in the soil solution. The
development of the understorey in the SS composed of the resident vegetation was slower
and less dense, while for FL it was almost null, and absent in MU (See Chapter 4). This
gradient of soil surface coverage by an actively absorbing live cover is reﬂected in a

comparable gradient of N content in soil inversely proportional to coverage. In 2003

66

differences persisted only between MU and the rest of the treatments. The ability of the
contained system to supply the required N diminished, except in the case of the MU
where the alfalfa hay provided a continuous source for N. The presence of legumes ﬁxing
N as well as an increased biomass produced by the understorey in PC and TC over that
of SS, likely explained the higher NO3'—N values.

The comparison of strip and row positions for the SS showed a slight tendency to
higher values for the strip (Table 2.2). It is reasonable to expect a lower N content in the
row where there is a denser root system of trees and vegetative cover competing to
uptake nutrients. At the same time there is abundant information about the increase of
mineralization due to tilling (Stevenson and Cole, 1999; Brady and Weil, 2002), so
enhanced levels in the tilled strip could be expected. Research conducted under similar
conditions of tilling showed that when the soil was loosened, its average mineral N
content in the 0-30 cm layer was enhanced only slightly in two successive years
(Steinmann, 2002) coinciding with what was found in this research.

The treatments showed differences in water availability. Even though we
attempted to provide a differential amount of water to achieve comparable water moisture
levels across treatments, this was hard to do under the experiment’s conditions. The fast
consumption of water in the proﬁle during days of high demand and the relatively limited
soil volume made it necessary to use short and frequent irrigation applications. The
characteristics of the mulch treatment helped to provide a more stable and efﬁcient water
supply to the tree as seen in other reports and reviews (Hogue and Neilsen, 1987; Merwin
et al., 1994). In the case of FL which was the closest to MU in water content, soil

moisture surface evaporation accounted for more loss in FL than MU. Soil moisture

67

content was lowest for TC and PC, in which, levels that could have restricted plant
development were reached during short periods (Kenworthy, 1949; Cripps, 1971).
Neither SS nor F L showed important differences with PC or TC and could also have gone
through short periods of drought stress. While understorey consumption and general
evaporation could explain the lower moisture content in SS, for FL it could be due to
what we understand was the reduction in inﬁltration of the soil surface, higher
temperatures and an increase in evaporation due to the exposed soil surface (Hogue and
Neilsen, 1987; Glenn and Welker, 1989; Merwin et al., 1996).

One of the primary effects of the mulch besides conserving soil moisture was on
N availability. The alfalfa hay used as source for the mulch had a high N content and thus
a low C/N ratio (Table 8.2 and B3, Appendix B). The estimations considering the
addition of an amount of 30 Tn/ha of dry weight add approximately to 650 kg N/ha if we
use an estimate of 2.18% N in the hay. These are very high quantities to be added every
year, and generate a potential environmental problem as seen ahead when discussing the
leachate results.

An initial strategy was to limit the application of hay to spring or fall for short
term N availability (Hu et al., 1997). However the initial addition of mulch in June 2001
did not generate a signiﬁcant difference of N03'—N in soil, probably because of the
already high level of N due to the active mineralization caused by soil disruption. The N
content in the hay itself reduced after the ﬁrst month of application and then stabilized
following episodes of rainfall and soil microbial utilization. This is reﬂected in a C/N

ratio that gets smaller with time (Table B.3, Appendix B) due to a loss of C as CO2 which

68

is liberated through microbial respiration while the N is recycled in the microbial biomass

(Brady and Wei], 2002).

Leachate

The higher volume collected from MU coincides with studies that show that a
mulch cover of straw increases inﬁltration (Hogue and Neilsen, 1987). At the same time
the insulating capacity of the mulch cover reducing evaporation and preventing water
consumption other than by the trees, kept the soil moisture at higher levels. This reduced
the amount of water needed to saturate the soil proﬁle before leaching would start.

For FL, as it was mentioned earlier, the development of a crust that slowed down
inﬁltration was clear. Behavior of the bare soil under ﬂaming applications decreased soil
water inﬁltration rates which was similarly found with use of continuous herbicides
(Glenn and Welker, 1989).

Yearly increase of the leachate volumes could be explained by the improvement
of soil macroporosity caused by the extension of roots of the vegetative cover (Brady and
Weil, 2002). The change in precipitation can not be considered the reason of this increase
since, on the contrary its total volume was less in 2003 than 2001 and distribution during
the year did not vary greatly (Figure B.1 thru B.3, Appendix B). The relative increase
between years was different for each treatment. Mulch had the smallest increase in the
ﬁrst year which would be due to the already mentioned capacity of the mulch to increase
inﬁltration and reduce evaporation, and thus, masking part of the increase of inﬁltration
due to macroporosity development. Flame on the contrary was the only treatment that on

the second year had a reduction of the volumes leached as a consequence of the

69

 

 

superﬁcial crust formed that obviously reduced water inﬁltration. The treatments with a
vegetation cover were those that increased inﬁltration the most, as found in other studies
(Zoppolo and Pieroni, 1985).

The N03'—N concentration in the leachate reﬂected the same variation of content
in soil which was previously discussed. In 2003, the values diminished for all treatments,
but the concentration for MU was 20 times greater than the average for the other orchard
ﬂoor management systems. During every year there was a common maximum peak of
NO3'—N between the end of October and the middle of November. The N03.—-N content
in soil at the end of the year was very low and NHX—N turned into an important
proportion of the available N. These values conﬁrmed the high leaching potential of N03"
and the higher activity of ammoniﬁers at lower temperatures giving a seasonal pattern
(Donahue et al., 1971). Similar results for variation in the N forms during the season were
obtained for the ﬁeld experiment (Chapter 1).

Trees in the MU treatment had the maximum beneﬁt of high leachate volume,
which possessed high concentrations of N. The continuous decomposition of alfalfa hay
was clearly the source of all the N. The N content of this hay was very high with a C/N
ratio around 15. This gave ideal conditions to the microorganisms to decompose this
material and through a high mineralization generate a net liberation of N into the soil.
Clearly, for each case, the material used for mulching could be adjusted according to the
needs of the orchard, which gives an additional tool in managing the nutrient cycling.

The contained condition with a shallower proﬁle for root development clearly

differs from what would be ﬁeld conditions. The values obtained of N in leachate

70

shouldn’t be extrapolated to ﬁeld conditions but used as an indicator of the relative

potential of leaching that each orchard ﬂoor management system has.

Tree performapcp

As expected, Supporter 4 gave more vigorous trees than M.9 NAKB 337 (Marini,
2000; Perry, 2000b). No signiﬁcant differences were found for the TCA of Supporter 4
between MU, FL, and SS treatments. Conversely, for M.9 NAKB 337 there were
signiﬁcant differences between MU and SS as well as FL (in this case for pSOJO). The
fact that performance of these rootstocks did not differ when comparing FL and SS, has
important management implications. Considering FL as our control (comparable to weed
suppression by herbicide in a conventional system) we can state that trees under SS had
an acceptable performance in relation to growth.

The signiﬁcant interaction rootstock by treatment tells us of a different variation
in the response of each rootstock when comparing treatments. The difference between
TCA of the rootstocks reached the maximum value under SS where during the 3 years the
percentage of Supporter 4 over M.9 NAKB 337 got to an average of 33 %. This was
followed by FL (25 %) and PC (17 %) and ﬁnally by MU and TC in which the difference
of TCA between the two rootstocks was 13 % and 12 % respectively. It seems as if in the
case of MU the rootstock M.9 NAKB 337 had no constraint and achieved maximum
growth. When limiting conditions were not extreme, Supporter 4 seems to have been able
to overcome better the restrictions and achieve a higher portion of its potential growth.
But when conditions turned extremely competitive in TC, not even Supporter 4 could

overcome them and returned to the level of 12 % thicker trunk than M.9 NAKB 337.

71

 

Research conducted by (Parker, 1990) showed as well that growth response of even a
vigorous rootstock like MM.111 was restricted by an increased competition due to
smaller vegetation-free-area or a more aggressive grass. Considering strictly water
availability (Fernandez, 1992) demonstrated how rootstocks with diverse vigor have
different response and capability to overcome water stress soil conditions.

From the analysis of the cumulative TCAI, branch growth, and canopy volume we
reach similar conclusions as with TCA where the difference between rootstocks under
MU and TC was not signiﬁcant or much smaller than under the other treatments in which
Supporter 4 was able to grow better than M.9 NAKB 337. Flower density was greatest
for MU in 2003, a reﬂection perhaps of increased canopy volume. There was no
difference in ﬂowering between FL and SS.

The N content in leaves indicated that levels were in general at adequate values or
slightly under for Michigan (Hanson, 1996). The difference due to the continuous supply
of N given by the alfalfa hay mulch was clearly measured in 2003. Probably at this time
the amount of N in the system for the other treatments which had no or very low N
amendment had started to be limiting. The N content in leaves for 2003 was slightly
lower than in 2002 for all treatments except MU and reached values under the
recommended range (Hanson, 1996). The challenge will increase as trees mature and
demands accelerate in the future.

In regard to the nutritional status, the increase of K levels under MU treatment
followed the trend stated in other reports (Wander and Gourley, 1943; Merwin and Stiles,
1994), while the lower levels for P under MU seem to contradict what has been found by

other researchers (Wander and Gourley, 1943; Hogue and Neilsen, 1987). This could

72

. /—_...

result from a dilution effect due to the higher growth under MU and a lower rootzshoot
ratio (see Chapter 3), and not necessarily from a reduction in P availability in the soil

under the alfalfa hay.

CONCLUSIONS

The effect of orchard ﬂoor management systems on SOM and NO3'—N was
limited except for a clear increase of available N due to the alfalfa hay applied in Mulch.
This was conﬁrmed by the N03-—N concentration in the leachate, which increased more
than 10 fold for Mulch compared to the other treatments, as well as by the total content of
NO3'—N. The alfalfa hay proved to be a sufﬁcient source of N and its supply could even
exceed the systems requirements or be lost through leaching.

Total volume of annual leachate was highest for Mulch during the whole
experiment. Water inﬁltration rates varied according to orchard ﬂoor management
systems. During the three year experiment a higher increase in inﬁltration occurred for
Total Cover and Partial Cover than for Sandwich and Mulch, and least for Flame. These
results conﬁrm the importance of a soil cover in improving inﬁltration rates compared
with bare soil, and the impact that a vegetative cover can have on the quantity and quality
of leachate according to its coverage and composition.

‘Buckeye Gala’ grafted on Supporter 4 was more vigorous than on M.9
NAKB337. Conversely trees on M.9 NAKB 337 were more precocious than those on

Supporter 4.

73

Orchard ﬂoor management systems also impacted tree growth. Vigor was greatest
on Mulch and weakest on Total Cover, with some exceptions. There was a signiﬁcant
interaction between orchard ﬂoor management treatment and rootstock for trunk cross
sectional area as well as for canopy volume values. Supporter 4 had a proportionally
enhanced response under Flame, Sandwich, and Partial Cover than the response of M.9
NAKB337. Conversely under Mulch and Total Cover there was a lower differential
between the two rootstocks. This trend was consistent for most of the parameters
measured. This indicates the potential of compensating stressing conditions, within a

certain range, with a more vigorous rootstock.

74

LITERATURE CITED

Bloksma, J. (2000). Soil management in organic fruit growing. Conference 'Organic fruit
opportunities and challenges', Ashford, Great Britain.

Brady, N. C. and R. R. Weil (2002). The nature and properties of soils. Upper Saddle
River, NJ, Prentice Hall. 960 p.

Carlson, R. F. (1967). Growth response of several rootstocks to soil water. HortScience
2(3): 108-110.

Childers, N. F. (1972). Modern fruit science. Somerville, NJ, Somerset Press, Inc. 912 p.

Cripps, J. E. L. (1971). The inﬂuence of soil moisture in apple root growth and root :
shoot ratios. HortScience 46: 121-130.

Diaz Rossello, R. (1992). Evolucién de la materia organica en rotaciones de cultivos con
pasturas. Revista IN IA Investigaciones Agronomicas, Uruguay 1(1): 103-110.

Donahue, R. L., J. C. Shickluna and L. S. Robertson (1971). Soils, an introduction to
soils and plant growth. Englewood Cliffs, NJ, Prentice-Hall, Inc. 587 p.

Fernandez, R. T. (1992). Mechanisms of drought tolerance in apple as inﬂuenced by
rootstock. Horticulture. East Lansing, Michigan State University: 141.

Ferree, D. C. and R. F. Carlson (1987). Apple rootstocks. In: 'Rootstocks for fruit
production'. R. C. Rom and R. F. Carlson. John Wiley & Sons, Inc. New York.
pp: 107-143.

Glenn, D. M. and W. V. Welker (1989). Orchard soil management systems inﬂuence
rainfall inﬁltration. J. Amer. Soc. Hort. Sci. 114(1): 10-14.

Hanson, E. J. (1996). Fertilizing ﬁ'uit crops, E-852. Michigan State University, East
Lansing, MI. 20 p.

Hogue, E. J. and G. H. Neilsen (1987). Orchard ﬂoor vegetation management.
Horticultural reviews 9: 377-430.

Hu, 8., N. J. Grunwald, A. H. C. van Bruggen, G. R. Gamble, L. E. Drinkwater, C.
Sherman and M. W. Demment (1997). Short-term effects of cover crop
incorporation on soil carbon pools and nitrogen availability. Soil Sci. Soc. Am. J.
61: 901-911.

75

Kenney, D. R. and D. W. Nelson (1982). Nitrogen -- Inorganic Forms. In: 'Methods of
Soil Analysis. Part 2. Chemical and Microbiological Properties'. D. R. Kenney.
American Society of Agronomy and Soil Science Society of America. Madison,
WI. pp: 643-698.

Kenworthy, A. L. (1949). Soil moisture and growth of apple trees. American Society for
Horticultural Science 54: 29-39.

Landis, J. N., J. E. Sanchez, G. W. Bird, C. E. Edson, R. Isaacs, R. H. Lehnert, A. M. C.
Schilder and S. M. Swinton (2002). Fruit crop ecology and management. East
Lansing, Michigan State University. 102 p.

Magdoff, F. and H. M. van Es (2000). Building soils for better crops. Washington DC,
J arboe Printing. 230 p.

Marini, R. P., J .L. Anderson, B.H. Barritt, G.R. Brown, J. Cline, W.P. Cowgill Jr., P.A.
Domoto, D.C. Ferree, J. Garner, G.M. Greene, C. Hampson, P. Hirst, M.M.
Kushad, E. Mielke, C.A. Mullins, M. Parker, R.L.Perry, J .P. Prive, T. Robinson,
C.R. Rom, T. Roper, J .R. Schupp, E. Stover, R. Unrath. (2000). Performance of
'Gala' apple on 18 dwarf rootstocks; A ﬁve year summary of the 1994 NC-140
Semi-dwarf rootstock trial. J. of the American Pomological Society 54(2): 92-
107.

Merwin, I. A., J. A. Ray, T. S. Steenhuis and J. Boll (1996). Groundcover management
systems inﬂuence fungicide and nitrate-N concentration in leachate and runoff
from a New York apple orchard. J. Amer. Soc. Hort. Sci. 121(2): 249-257.

Merwin, I. A. and W. C. Stiles (1994). Orchard groundcover management impacts on
apple tree growth and yield, and nutrient availability and uptake. J. Amer. Soc.
Hort. Sci. 119(2): 209-215.

Merwin, I. A., W. C. Stiles and II. M. van Es (1994). Orchard groundcover management
impacts on soil physical properties. J. Amer. Soc. Hort. Sci. 119(2): 216-222.

Myers, R. J. K., M. van Noordwijk and P. Vityakon (1997). Synchrony of nutrient release
and plant demand: plant litter quality, soil environment and farmer management
options. In: 'Driven by nature'. G. Cadisch and K. E. Giller. CAB International
Publishing. New York. pp: 215-229.

Parker, M. L. (1990). The response of ﬁ'uit trees to orchard ﬂoor management.
Horticulture. East Lansing, Michigan State University: 136.

Paul, E. A., S. J. Morris and S. Bohm (2001). The determination of soil C pool sizes and
turnover rates: biophysical fractionation and tracers. In: 'Assessment methods for
soil carbon'. R. Lal, J. M. Kimble, R. F. Follett and B. A. Stewart. CRC Press
LLC. Boca Raton. pp: 193-206.

76

Perry, R. (2000a). The keys to maintaining productive Vertical Axe trees. The Fruit
Growers News 39(5): 21.

Perry, R. (2000b). Planting fruit trees in 2000. Crop Advisory Team Alert 15(1): 2-4.

Robinson, T. L. (2003). Rootstock as a key component to high density orchards.
Conference 'Orchard systems workshop', Geneva, NY. Int. Dwarf Fruit Tree
Assoc.

Sanchez, J. E., C. E. Edson, G. W. Bird, M. E. Whalon, R. R. Harwood, K. Kizilkaya, J.
E. Nugent, W. Klein, A. Middleton, T. L. Loudon, D. R. Mutch and J. Scrimger

(2003). Orchard ﬂoor and nitrogen management inﬂuences soil and water quality
and tart cherry yields. J. Amer. Soc. Hort. Sci. 128(2): 277-284.

Sooby, J. (1999). Meeting the Research Needs of Organic Farmers.
http://www.ofrf.org/research/needs.htrnl. May 18, 2004.

Steinmann, H. H. (2002). Impact of harrowing on the nitrogen dynamics of plants and
soil. Soil and Tillage Research 65(1): 53-59.

Stevenson, F. J. and M. A. Cole (1999). Cycles of soil - Carbon, nitrogen, phosphorus,
sulftrr, micronutrients. New York, John Wiley & Sons, Inc. p.

Wander, I. W. and J. H. Gourley (1943). Effect of heavy mulch in an apple orchard upon
several soil constituents and the mineral content of foliage and fruit. Proceedings
of the American Society for Horticultural Science 42: 1-6.

Wardle, D. A. and P. Lavelle (1997). Linkages between soil biota, plant litter quality and
decomposition. In: 'Driven by nature'. G. Cadisch and K. E. Giller. CAB
International Publishing. New York. pp: 107-124.

Weibel, F. P. (2002). Soil management and in-row weed control in organic apple
production. The Compact Fruit Tree: 118-121.

Zoppolo, R. J. and J. L. Pieroni (1985). Efectos de diferentes sistemas de manejo de suelo
sobre el regimen hfdrico y la produccion de un monte de manzanos bajo riego y
en secano. Orientacién Granjera. Montevideo, Uruguay, Universidad de la
Repr’rblica, Facultad de Agronomia: 133 p.

77

Table 2.1. Content of organic matter in soil (%) for 0-30 cm depth for two positions in SS
treatment (n=16). Analysis of variance was carried out for each date and no signiﬁcant
differences were found for p50.05 (Tukey’s adjusted).

 

 

Date Row Strip
MAY/2002 1 .87 1.96
AUG/2002 1 .84 1.85

APR/2003 1.30 1.33
JUN /2003 1.98 1.90
AUG/2003 1 .99 1 .93
NOV/2003 1.78 1.76

 

Table 2.2. Nitrogen content in soil (ppm of N03'—N ) in 0-30 cm depth for the two
positions of SS of boxes with trees (n=16). Analysis of variance was carried out for each

date. Values followed by different letters in the same row are signiﬁcantly different for
pS0.0S (Tukey’s adjusted).

 

 

Date Row Strip
JUN/2001 56.01 56.01
AUG/2001 32.33 -
NOV/2001 3.29 1.30
MAY/2002 8.89 8.95
AUG/2002 4.45 b 8.00 a
APR/2003 30.42 b 39.45 a
JUN/2003 2.62 5.35
AUG/2003 1 .55 2.70
NOV/2003 2.31 2.37

 

78

Table 2.3. Average N concentration (mg N03'—N L”) for the whole year in the leachate
collected from boxes under each soil management system and with trees grafted on
rootstock: A. Supporter 4, and B. M.9 NAKB 337. (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Analysis of variance was
carried out for each year. Values in the same row followed by the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted).

A

 

 

 

 

 

Supporter 4 MU FL SS PC TC
2001 105 a 73 be 95 ab 35 cd 15 d
2002 99 a 6 b 7 b 3 b 7 b
2003 30a 1 b 1 b 2 b 2 b

B
M.9 NAKB 337 MU FL SS PC TC
2001 80 a 47 bc 58 ab 38 cd 22 d
2002 79 a 10 b 3 b 3 b 6 b
2003 44 a 1 b 0 b 3 b 5 b

 

Table 2.4. Number of ﬂower clusters at time of full bloom for two rootstocks and by
orchard ﬂoor management system (MU = Mulch, FL = Flame, SS = Sandwich System,
PC = Partial Cover, TC = Total Cover). A. May 2002 B. May 2003. Results of analysis
of variance are shown for rootstock and treatment averages. Values in the same row or
column followed by the same letter are not signiﬁcantly different for pS0.05 (Tukey’s

 

 

 

 

 

 

 

 

adjusted).
A
2002 MU FL PC TC Rootstock avg
Supporter 4 24 19 3 4 13 b
M.9 NAKB 337 41 50 21 28 35 a
Treatment avg 33 a 35 a 26 ab 12 b 16 b
B
2003 MU FL PC TC Rootstock avg
Supporter 4 389 289 130 70 227 b
M.9 NAKB 337 421 329 155 112 264 a
Treatment avg 405 a 309 b 278 b 143 c 91 c

 

 

Table 2.5. Nitrogen content (% dry weight) in apple leaves sampled in August of each
year, for each orchard ﬂoor management system (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Analysis of variance was
carried out for each year. Values in the same row followed by the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted).

 

MU FL SS PC TC
2001 1.87 1.91 1.91 1.87 1.91
2002 2.29 2.39 2.22 2.21 2.19
2003 2.32 a 2.08 b 2.02 b 1.96 b 1.95 b

 

 

Table 2.6. Nutrient 3-year average content in apple leaves sampled in August of each
year, for each orchard ﬂoor management system (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Values in the same row
followed by the same letter are not signiﬁcantly different for p30.05 (Tukey’s adjusted).

 

 

MU FL 88 PC TC
P(%) 0.18 c 0.30 ab 0.31 a 0.30 ab 0.27 b
K(%) 1.98 a 1.74 b 1.74 b 1.64 b 1.56 c
Mn(ppm) 43.12 a 40.72 ab 39.03 b 38.73 b 38.78 b
Zn(ppm) 16.82 b 16.55 b 18.19 ab 20.65 a 19.82 a

 

80

ll‘l;J
i‘fi
I“;

‘ .

.‘ 1’14 if": 121335101: :51 if?

<
4

Figure 2.1. Diagram of the system designed for leachate collection. The wooden boxes
for the apple trees were lined with plastic ﬁlm before ﬁlling and a funnel (F) installed
from which a plastic pipe (PP) conducted the leachate to a jug (J) inside a 20 l. bucket (B)
with the lid placed just above soil level.

 

 

 

 

 

2.5
2.0
g 1.5
2
O
(I)
1.0
0.5
-o-MU +FL --o--ss —-x--Pc —<>-TC
0.0
MAY AUG APR JUN AUG NOV
2002 2003

Figure 2.2. Content of organic matter in soil (SOM) for 0-30 cm depth of boxes with
trees by date for different treatments (n=16). (MU = Mulch, FL = Flame, SS = Sandwich
System, PC = Partial Cover, TC = Total Cover). Analysis of variance was carried out for
each date. Values followed by different letters within date are signiﬁcantly different for
p50.05 (Tukey’s adjusted).

81

 

180

- of?" MU a

 

160

r—n —-
N «b
O O

N03'—N in soil (ppm)
525

  

 

 

 

 

 

80
a
60 r
40 b .-
20 «
0 . , .
JUN AUG NOV MAY AUG APR JUN AUG NOV
2001 2002 2003

Figure 2.3. Nitrogen content in soil (ppm of N03'—N ) in 0-30 cm depth by treatment
(MU = Mulch, FL = Flame, SS = Sandwich System, PC = Partial Cover, TC = Total
Cover) for boxes with trees. Each value is average of 16 replicates :h SE. Analysis of
variance was carried out for each date. Values followed by different letters within date
are signiﬁcantly different for p50.05 (Tukey’s adjusted).

82

 

 

 

 

 

7 I a
6 _
”g 5
0..
3 4
z.
+
E 3 ,
Z
2 _
! --D--SS
l —-u- PC
-°—TC
O . .
APR JUN AUG NOV
Month of2003

Figure 2.4. Seasonal pattern of N content in soil (ppm NHf—N) in 0 — 30 cm depth as
affected by orchard ﬂoor management system. Each value is average of 16 replicates i

SE. Analysis of variance carried out for each date gave no signiﬁcant differences for
p50.05 (Tukey’s adjusted).

83

30*

--o--MU —+—FL --o~ss --o<- PC +TC

% soil moisture v/v

 

0 4 , r r r
b e v\ b-
’\\‘D «\‘W «\‘q A“ “’0’ Po, oé‘b <19” 036’ “\b 010 a"? Oi" \°\
Date

Figure 2.5. Soil moisture measured with TDR for 0 - 45 cm depth for 2002

--o--MU +11 --o--ss -x- PC +TC

% soil moisture v/v
G

 

O i ’ T’ A 1 ”v7 7 '7" T ' 7' ' V ’T’ '7 7' 4 ' T ' 7' ‘ 7" '

x x x ‘x e e c
all") ‘9?) of“) bi") ‘9'» 00 0"? “9 «\O «19 ‘9 ‘8‘ 08” <b\”’ a“ a“ P @‘Q

0,
Date

Figure 2.6. Soil moisture measured with T DR for 0 - 45 cm depth for 2003

84

 

UMU DFl. ISS BPC ITC 3

Volume of leachate (l/mz)

 

 

 

2001 2002 2003

Figure 2.7. Volume of leachate (liters/m2) recovered from boxes with apple trees during
three consecutive seasons (2001-2003) for different soil management systems (MU =
Mulch, FL = Flame, SS = Sandwich System, PC = Partial Cover, TC = Total Cover).
Each bar represents the average of 16 replicates. Analysis of variance was carried out for
each date, bars with same letter have no signiﬁcant differences for pS0.05 (Tukey’s
adjusted).

 

a a DMU ClFL ISS BPC ITC

N03‘—N (ng2 per year)

 

    

 

2001 2002 2003
Figure 2.8. Total NO3'—N (g./m2 per year) collected in the leachate from boxes during the
whole year for each soil management system (MU = Mulch, FL = Flame, SS = Sandwich
System, PC = Partial Cover, TC = Total Cover) Each bar represents the average of 16
replicates :l: SE. Analysis of variance was carried out for each date, bars with same letter
have no signiﬁcant differences for pS0.05 (Tukey’s adjusted).

85

 

14

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a DSupporter4
‘2 a I M.9 NAKB 337
l
10 . I
a a
e, b
5 6 b a
l— I a
4 . 1
2
0
MU FL 83 PC TC

Figure 2.9. Trunk cross sectional area increase (TCAI) between April 2001 and
November 2003 measured 25 cm above the graft union for two rootstocks and by orchard
ﬂoor management treatment (MU = Mulch, F L = Flame, SS = Sandwich System, PC =
Partial Cover, TC = Total Cover). Each bar represents an average of 8 replicates. Results
of analysis of variance are shown for rootstock effect under each treatment. Bars with the
same letter are not signiﬁcantly different for p50.05 (Tukey’s adjusted).

86

2500 120

 

 
 
 
  

E Total growth Supporter 4
-Total growth M.9 NAKB 337 ’3‘
A 100 o
g 2000 -» ”A“ No. branches Supporter 4 E
f) —o— No. branches M.9 NAKB 337 8.
3‘ .8
g 1500 g
v c:
E a
e .r:
on 1000 E
“g a
£3 E“
-- 500 » g
S o
O r...
[-2 O

0

 

 

 

 

 

 

 

 

 

MU FL SS PC TC

Figure 2.10. Total growth in length (cm) of all growing branches during 2002 and
number of branches, for two rootstocks under each orchard ﬂoor management system
treatment (MU = Mulch, FL = Flame, SS = Sandwich System, PC = Partial Cover, TC =
Total Cover). Bars are average of 8 replicates :1: SE. Results of analysis of variance are

shown for rootstock under each treatment. Bars with the same letter are not signiﬁcantly
different for p50.05 (Tukey’s adjusted).

87

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3.5 ~ -
ElMU DFL ISS IPC ITC
a
3 a
l
2.5 a
p
E
g 1.5 -
a.)
2
[-* l
0.5
O
Supporter 4 M.9 NAKB 337
ROOTSTOCK

Figure 2.11. Tree volume estimation (m3), after growth in 2003, for trees grafted on
rootstocks Supporter 4 and M.9 NAKB 337 under different soil management systems
(MU = Mulch, FL = Flame, SS = Sandwich System, PC = Partial Cover, TC = Total
Cover). Bars are average of 8 replicates 2!: SE. Results of analysis of variance are shown
for rootstock under each treatment. Bars with the same letter are not signiﬁcantly
different for pS0.05 (Tukey’s adjusted).

88

CHAPTER 3

89

CHAPTER 3

APPLE TREE AND SOIL COVER DRY MATTER PARTITIONING UNDER FIVE
ORCHARD FLOOR MANAGEMENT SYSTEMS IN ORGANIC PRODUCTION

Abstract
Apple trees respond to environmental conditions and adapt the allocation of resources for
growth and production to achieve best performance. This research was conducted to
determine the effect of organically accepted orchard ﬂoor management practices on C
and N partitioning. Forty ‘Buckeye Gala’ trees grafted on M.9 NAKB 337 were planted
in 0.6m3 wooden boxes. The treatments evaluated were: alfalfa hay mulch, propane ﬂame
burner, Swiss Sandwich system (a combination of resident vegetation and tilled strips),
and partial or total soil cover with a mix of red clover and rye grass. After 3 years the
boxes were disassembled and trees as well as all vegetative soil cover was harvested.
Shoots and roots were set apart, dried, and C and N measured. Growth of trees under
Mulch was higher with 128% more dry matter than trees under Total Cover which were
the ones with least growth. Mulch trees had the lowest rootzshoot ratio (0.77) while those
from Total Cover had the highest (1.14). In the Sandwich System the highest values of
C:N ratio were obtained (85), while the lowest were in Mulch (51). The treatments with
an active vegetative cover Sandwich, Partial Cover, and Total Cover had substantial
increase of the total biomass weight considering trees and soil cover. The orchard ﬂoor

management systems impacted the C and N allocation.

90

INTRODUCTION

The concept of sustainable agriculture has been growing from the early 19805 and
getting broader over time (Harwood, 1990). But many of the principles around it evolved
during the early 19008 and include mainly: the interrelation of all parts of a farming
system, the existence of important biological balances, and the need to maximize desired
biological relationships.

Organic production has a holistic approach sharing many of these principles that
guide sustainable agriculture. The main changes in management when transitioning from
conventional to organic fruit growing have to do with the systems approach. Basic in the
system is the soil, to which important efforts are aimed so as to improve its fertility
through the enhancement of microbial activity (Magdoff and van Es, 2000; Brady and
Wei], 2002).

For the fruit tree, the inﬂuence of orchard ﬂoor management has been studied for
decades (The Duke of Bedford and Pickering, 1911; Hogue and Neilsen, 1987). With the
global approach compatible with organic principles, it is of key importance to understand
the way in which the interaction between the tree and the soil cover takes place and how
this system fits in a higher scale within the agro ecosystem. Recent publications have
shown the importance of intensive agriculture practices in regard to their global warming
potential and their relation to C sequestration (Robertson et al., 2000).

Newly designed orchard ﬂoor management systems like the Swiss Sandwich

system (Weibel, 2002) are being evaluated and until now appear as a viable alternative

91

for organic production (see Chapter 1). Information on its impact on whole plant and on
how the distribution of resources takes place in the tree is lacking.

The objectives of this work were to: l) quantify the impact of five orchard floor
management systems on growth of newly planted apple trees; 2) determine the effect of
treatments on C & N allocation to vegetation in the system and; 3) evaluate the

applicability of the sandwich system from a sustainable perspective.

MATERIALS AND METHODS

Site description

The experiment was conducted at the Clarksville Horticultural Experiment Station
of Michigan State University in the vicinity of Clarksville, MI (latitude of 42° 84’ N and
longitude of 84° 24’ W). Daily temperatures and precipitation during the whole
experiment were recorded with an automated weather station of the Michigan Automated
Weather Network, located approximately 560 m west of the orchard (Figure A.2 thru

A.4, Appendix A).

Plant material

One-year-old “Buckeye Gala” apple trees (Malus x domestica Borkh.) were
planted in April 2001 in the boxes that had been previously ﬁlled with soil taken from top
layers at the orchard site (Kalamazoo sandy clay loam with a particle size composition of

53.1% sand, 23.1% silt, and 23.8% clay). Plant residues present were eliminated, and

92

some soil mixing occurred. Trees were grafted on M.9 NAKB 337 and all had similar
caliper at planting time.

A trellis system with two wires was installed and a galvanized metal pole was
used for support of each tree. Plants were trained to a vertical axe with the use of rubber
bands and clothes pins to bend branches (Perry, 2000). Minimal pruning was necessary,

mainly to single out the leaders or main branches.

OrclLam ﬂoor management

Five different orchard ﬂoor management systems were established among boxes
in spring of 2001 at planting time:
1. Mulch (MU): The whole area of the box (1.2 m2) was covered with a mulch of alfalfa
hay. During the growing season the thickness of the cover was over 15 cm to provide a
shading effect that would suppress weeds, and maintain moisture. Mulch was hand-
applied in spring and fall. The mulch was sampled at the time of application to determine
the C/N ratio. In Spring 2003 a plastic net with 1 cm x 1 cm grid was laid down to
separate mulch material recently applied from that applied the previous year.
Decomposition rate of the mulch was monitored monthly for each age material until
November. The samples were taken by cutting a vertical cylinder of 7.6 cm in diameter
from the mulched area and the depth of each layer was recorded. The dry weight of the
cores was measured in the laboratory after oven drying for 48 hours at 60 ° C, and sub-
samples were taken for C and N determination.
2. Flaming (FL): The entire area of the box (1.2 m2) was kept free of vegetation through

the use of a propane ﬂame burner hand torch that provided 150.000 BTU. Minimal

93

amounts of water were applied just behind the burner to reduce the risk of heat affecting
tree branches or ﬁre spreading to the neighboring mulch.

Flame treatments were applied when weed height approached 10 cm (4 to 7 times
according to season).
3. Sandwich system (SS): an adaptation of the Swiss Sandwich system (Weibel, 2002),
consisted of a central strip that spread 20 cm to each side of the tree row in which
vegetation was allowed to grow. A 40 cm wide strip was tilled on each side of that
vegetated central strip. The tilling was performed with a hand hoe to a depth under 7 cm.
In the vegetated central strip hand weeding was performed selectively to prevent
development of curly dock (Rumex crispus L.), common lambsquarters (Chenopodium
album L.), redroot pigweed (Amaranthus retroﬂexus L.) and thistles (Cirsium sp.). In the
lateral tilled strips, bell beans (Viciafava) were sown (45 kg.ha'l in 2001 and 130 kg.ha’l
in 2002) in the fall as a green manure. Tilling started in spring, and was repeated when
weeds got 10 cm tall (4 to 6 times according to season).
4. Partial cover (PC). Just after the planting of trees a mixture of red mammoth clover
and endophytic rye was sewn. The cover was allowed to grow in a central strip that
spread 20 cm to each side of the tree trunk. The remaining area on each side (40 cm wide
strip), was tilled at the same time as SS.
5. Total cover (TC). As in the previous treatment, a mixture of red mammoth clover and
endophytic rye was sewn just after planting the trees. In this case, the mixture was

allowed to grow in the whole area of the box.

94

Irrigation system

Drip irrigation was installed in May 2001. The lines were suspended from the
lowest wire of the trellis system at a height of 0.3 m to prevent potential damage to them
when controlling weeds. Drippers of 3.8 L/h were inserted to the lines to have one per
box, and irrigation could be independently turned on and off for each of the ﬁve ﬂoor
management treatments. Irrigation was applied accordingly to keep the available soil

water over 50% (Figure C.1 thru C.3, Appendix C).

Treggrowth mea_surements

Trees were measured to assess growth and development during three years (2001-
2003). Those measures included trunk size, branch growth, number of ﬂower clusters,
weight of wood pruned, height of the tree, as well as canopy volume. For details of these

measurements and their results see Chapter 2.

Dry matter partitioning

At the end of the third season, four replicates of the boxes with trees were
disassembled and destructively harvested to evaluate dry matter partitioning.

Above ground biomass of the soil cover from each box was cut at soil surface and
bagged separately to be dried and weighed. The trees were cut at the graft union and then
separated into three portions: one-year-old shoots, central stern (trunk), and branches

(two-year-old and older).

95

The wooden boxes were cut open and the soil washed off. The root system of
each tree was carefully divided from the rest of roots of the soil cover and bagged
separately. In the case of MU, all the alfalfa hay present was also collected individually
for each box.

All the material was oven dried at 60°C until constant dry weight was achieved,
which in general was around the 7th day. Dry weight was registered.

The dried plant material was homogenized (wood from trees previously chipped),
and alter homogenizing, a sample was taken for grinding and analyzed for C and N
content. Carbon and N content was determined for all samples using a Carlo Erba

NA] 500 Series 2 N/C/S analyzer (CE Instruments Milan, Italy).

Experimental design and statistical analysis

The experiment was set as a completely randomized design with eight replicates
of the ﬁve applied treatments. The plots consisted of 1 tree planted on a wooden box of l
m x 1.2 m and 0.5 m high. The boxes were arranged in two rows with plants at 1.4 m x 8
m spacing and the treatments and rootstocks randomly assigned.

Analysis of variance was performed using MIXED procedure with SAS (Version
8, SAS Institute, Cary, NC, USA) to detect treatment effects. After using ANOVA to
identify treatment effects, least squares means were compared using the Tukey-Kramer

test with p50.05 unless stated otherwise.

96

RESULTS

Tree vigor as expressed by trunk cross sectional area increase (TCAI) was lowest.
under Total cover (TC) and greatest under Mulch (MU) (Figure 3.1).

The orchard ﬂoor management systems had a signiﬁcant effect on the biomass
accumulation of trees and its partitioning. These results are consistent with the data
previously presented for TCAI. Tree root dry weight values were smaller for TC (Figure
3.2) than for MU, Flaming (FL) and Sandwich (SS). There was no signiﬁcant difference
in root and total tree (root + shoot) dry weight between MU, FL, and SS. For the shoot,
including trunk and branches, MU and FL had higher values, followed by SS, PC and TC
A similar response was found for total tree dry weight (root + shoot), with the difference
that SS was not statistically different from either MU or FL.

Orchard ﬂoor management systems inﬂuenced the rootzshoot (R/S) ratio. This
parameter increased from MU (0.77) to TC (1.14) but signiﬁcant differences were found
only between MU and the three treatments with relatively greater roots: SS, PC, and TC.
In this case FL had an intermediate value with no differences from the rest (Figure 3.3).

The regression analysis between total dry weight of the trees and the R:S ratio
gave an R2 = 0.58 (Figure C.4, Appendix C). In the case of C:N ratio of the whole tree
versus R:S there was a signiﬁcant correlation with an R2 = 0.77 (Figure 3.4).

The soil management treatment created no difference in C concentration but
affected N concentration in the tissue of the different tree portions. For root, branch (2-
year-old and older), and trunk tissue the N concentration was lower in SS although not

statistically different from FL, PC and TC (Figure 3.5). In one-year-old branches

97

differences in N concentration were more extended with MU the highest and SS the
lowest. Mulch had the highest N concentration in all tissues. In general, N concentration
was highest in one-year-old wood followed by roots and lowest in the trunk.

The orchard ﬂoor management impacted on total C content of apple trees. Carbon
content was highest for MU, FL, and SS, and lowest in TC (Table 3.1). Total N content in
trees was affected as well and showed differences for root, branches, and trunk values
(Figure 3.6). Total N content of the plant was signiﬁcantly higher for MU even though
not statistically different with FL (Table 3.1), while TC had the lowest content.

Dry weights of the soil cover varied with the ﬂoor management treatment.
Maximum total biomass (above and under ground) was produced in TC, followed by SS
and PC, but the R/S ratio was more than double for SS (Figure 3.7).

The diverse soil cover composition from the different orchard ﬂoor management
systems was reﬂected in differences in the values of C and N concentration, as well as in
the total content of each in the biomass of the treatments (Table 3.2). Vegetation of SS
had slightly lower N concentration both for roots and shoots. The C/N ratio was highest

for the soil cover of SS followed by PC and TC.

98

DISCUSSION

The difference in grth achieved by the trees in the MU was consistent with
results found in other research (Langord, 1937; Hogue and Neilsen, 1987; Merwin et al.,
1995; Sanchez et al., 2003). The competition imposed by the total vegetative cover of soil
restricted the apple trees’ grth as it could be expected (The Duke of Bedford and
Pickering, 1911; Harrison, 1945; Merwin and Stiles, 1994). The alfalfa hay mulch gave
higher N availability and conserved soil moisture, enhancing trees growth.

Orchard ﬂoor management systems did not have as great an impact on root
biomass as compared to shoots. The impact of the orchard ﬂoor management in the
above-ground portion of trees ended with ahnost a three fold difference between MU and
TC. In contrast, root biomass in MU was less than double compared to TC. The change
imposed by the competition of the soil cover and the variations in other soil parameters
(see Chapter 1 & 2) generated a shiﬁ in the partitioning of tree resources. The analysis of
the R:S ratio supports the results from other research (Cripps, 1971). The treatment under
best growing conditions yielded the lowest R:S ratio. The response of the tree to the
higher soil moisture and nutrient availability near the soil surface by adjusting location
(shallower rooting), number (lower), and diameter of roots has been reported (Levin et
al., 1979; Layne et al., 1986). These results support the ﬁndings in this study that show
how under MU, a signiﬁcantly lower proportion of the resources were devoted to root
growth compared with TC. Conversely under TC the effort of the tree was into increasing
soil exploration trying to compensate lower N and water availability generating a

proportional increase of root size so as to fully explore the soil volume and eventually

99

obtain more N and/or water (Cripps, 1971; Atkinson, 1980). Regression showing the
negative relationship between dry matter production of trees and the R:S ratio reinforces
this idea. The same happens with the regression between C:N ratio of tree wood and R:S
ratio, which shows in this case a positive relationship with an increase of C:N ratio when
the tree partitions more resources to the root.

The other three treatments had an intermediate response compared to MU and TC.
The vegetation-free-area of FL and its characteristics were the closest to conventional
practices and thus considered as the control treatment. Even though FL had higher total
tree dry weight it was not statistically different from SS and PC. This suggests that a
smaller weed-free-area can be tolerated more readily under ﬁeld conditions than what
traditionally is done, as shown in other research (Merwin and Ray, 1997). We conclude
as well, that a spatial distribution of the weed-free-area that does not include the tree
trunk can also be applicable with good results.

The effect of N availability from the different systems is reﬂected partially in the
N concentration in tree wood. Results show a higher N concentration for trees under MU.
With the dry weight and the values of C and N concentration, the C:N ratio for the trees
was calculated. Many authors have used the plant C:N ratio as an estimate of the nitrogen
use efﬁciency (Maranville and Madhavan, 2002; Zavalloni, 2004) which is deﬁned as
the amount of dry matter produced per unit of N absorbed (Robertson, 1997). Higher C:N
ratio will indicate higher nitrogen use efﬁciency. Trees under SS were those that achieved
highest efﬁciency while for MU the ratio was the lowest. For this treatment the high
abundance of N in the system could have induced luxury consumption thus the lower

efﬁciency in its use.

100

Considering also the understorey there is almost a duplication of the values of
root dry weight for SS, PC, and TC but almost no change for FL and MU. Taking into
account the tree, together with the understorey vegetation, the SS treatment produced the
largest root biomass in the tree-soil cover-system, as well as above ground biomass. This
suggests greater C ﬁxation of the system, as well as an increased rhizosphere that will
promote microbial activity. The concern about global warming and increase of C02
content in the atmosphere has promoted the idea of a currency dependant on C ﬁxation
with special compensations that is still going through major adjustments (Tomich et al.,
2002). The impact of different soil management systems and ecosystem changes have
been studied mainly related to agriculture crops and forest areas (Komer, 1995; Thomley
and Cannell, 1996; Robertson et al., 2000). New considerations might have to be taken in
this respect at the moment of selecting an orchard ﬂoor management system in the pursuit
of sustainability and facing a development of the C market. It is clear that the relative
importance of the ground cover biomass diminishes as trees get to their maximum sizes.

Differences seen in C:N ratio of the soil cover could also be considered as an
indicator of the average N use efﬁciency of the whole vegetative mass. But we must take
this value also as a quality parameter of the material that eventually is being incorporated
on the soil surface or into the soil proﬁle as plants grow, lose leaves and roots, or die. The
cover composition of TC and PC which initially was only of red mammoth clover and rye
had a shift in species composition with the growth of resident vegetation (see Chapter 4
for details). Conversely, in the case of SS, there was eventually legume specimen that
appeared as part of the resident vegetation. Nevertheless, after 3 years of the initial

installation, differences could still be seen in the C:N ratio that reﬂect the disparity in

101

composition and thus in mineralization quality of the organic matter supplied by each
cover (Smethurst and Narnbiar, 1995; Handayanto et al., 1997). This is of great
importance when it comes to the adjustment of fertilization and nutrient supply to the

apple trees.

CONCLUSIONS

The different orchard ﬂoor management systems had an effect on the grth of
trees. While Mulch promoted maximum tree growth, Total Cover was the one that gave
the least growth. There were almost no signiﬁcant differences between Flame and
Sandwich.

Trees responded to the conditions imposed by the orchard ﬂoor management
system with a different partitioning of resources which is reﬂected in the variation of the
R:S ratio. The trees with greatest grth under Mulch had the lowest ratio while the trees
with least growth from Total Cover had the highest R:S ratio.

Considering C:N ratio as an estimator of nitrogen use efﬁciency the Sandwich
was the treatment with better performance. The low C:N ratio for Mulch might be
indicating luxury N consumption consistent with high values of N leaf content and N05—
N in leachate (See Chapter 2).

The treatments with an active vegetative cover Sandwich, Partial Cover, and Total
Cover had substantial increase of the total biomass weight considering trees and soil

cover. This is a characteristic of the system that will need more research under a

102

sustainable approach. The Sandwich appears with advantage over the other treatments
with a more balanced tree growth. We conclude that the Sandwich system is a good
option for organic fruit growers in Michigan and research should continue for

improvements in its design.

103

LITERATURE CITED

Atkinson, D. (1980). The distribution and effectiveness of the roots of tree crops. In:
'Horticultural Reviews'. J. J anick. AVI Publishing Company, Inc. Kent, UK. pp:
424-488.

Brady, N. C. and R. R. Weil (2002). The nature and properties of soils. Upper Saddle
River, NJ, Prentice Hall. 960 p.

Cripps, J. E. L. (1971). The inﬂuence of soil moisture in apple root grth and root :
shoot ratios. HortScience 46: 121-130.

Handayanto, E., G. Cadisch and K. E. Giller (1997). Regulating N mineralization from
plant residues by manipulation of quality. In: 'Driven by nature'. G. Cadisch and
K. E. Giller. CAB International Publishing. New York. pp: 175-185.

Harrison, C. M. (1945). Establishment and management of sods in orchards. Fruit
Growers Conference, East Lansing. 6 p.

Harwood, R. R. (1990). A history of sustainable agriculture. In: 'Sustainable agriculture
systems'. C. A. Edwards, R. Lal, P. Madden, R. Miller and G. J. House. Soil and
Water Conservation Society. Ankeny, IA. pp: 3-19.

Hogue, E. J. and G. H. Neilsen (1987). Orchard ﬂoor vegetation management.
Horticultural reviews 9: 377-430.

Komer, C. (1995). Towards a better experimental basis for up scaling plant responses to
elevated C02 and climate warming. Plant Cell Environment 18(10): 1101-1110.

Langord, L. R. (1937). Effect of sod, cultivation and straw mulch upon orchard soil
moisture content. Proceedings of the American Society for Horticultural Science
3 5: 3 1 3—3 14.

Layne, R. E. C., C. S. Tan and R. L. Perry (1986). Characterization of peach roots in fox
sand as inﬂuenced by sprinkler irrigation and tree density. J. Amer. Soc. Hort.
Sci. 111(5): 670-677.

Levin, 1., R. Assaf and B. Bravdo (1979). Soil moisture and root distribution in an apple
orchard irrigated by tricklers. Plant and Soil 52: 31-40.

Magdoff, F. and H. M. van Es (2000). Building soils for better crops. Washington DC,
J arboe Printing. 230 p.

104

Maranville, J. W. and S. Madhavan (2002). Physiological adaptations for nitrogen use
efﬁciency in sorghum. Plant and Soil 245: 25-34.

Merwin, I. A. and J. A. Ray (1997). Spatial and temporal factors in weed interference
with newly planted apple trees. HortScience 32(4): 633-637.

Merwin, I. A., D. A. Rosenberger, C. A. Engle, D. L. RistandM. Fargione (1995).
Comparing mulches, herbicides, and cultivation as orchard groundcover
management systems. HortTechnology 5(2): 151-158.

Merwin, I. A. and W. C. Stiles (1994). Orchard groundcover management impacts on
apple tree growth and yield, and nutrient availability and uptake. J. Amer. Soc.
Hort. Sci. 119(2): 209-215.

Perry, R. (2000). The keys to maintaining productive Vertical Axe trees. The Fruit
Growers News 39(5): 21.

Robertson, G. P. (1997). Nitrogen use efﬁciency in row-crop agriculture: crop nitrogen
use and soil nitrogen loss. In: 'Ecology in Agriculture'. L. E. Jackson. Academic
Press, New York. pp: 347-365.

Robertson, G. P., E. A. Paul and R. R. Harwood (2000). Greenhouse gases in intensive
agriculture: contributions of individual gases to the radiative forcing of the
atmosphere. Science 289: 1922-1925.

Sanchez, J. E., C. E. Edson, G. W. Bird, M. E. Whalon, R. R. Harwood, K. Kizilkaya, J.
E. Nugent, W. Klein, A. Middleton, T. L. Loudon, D. R. Mutch and J. Scrimger
(2003). Orchard ﬂoor and nitrogen management inﬂuences soil and water quality
and tart cherry yields. J. Amer. Soc. Hort. Sci. 128(2): 277-284.

Smethurst, P. J. and E. K. S. Narnbiar (1995). Changes in soil carbon and nitrogen during
the establishment of a second crop of Pinus radiata. Forest Ecology and
Management 73(1-3): 145-155.

The Duke of Bedford, K. G., F.R.S. and S. U. Pickering, M.A., F RS. (1911). Thirteenth
Report of the Woburn experimental ﬁ'uit farm. London, The Amalgamated Press,
Ltd. 1-151 p.

Thomley, J. H. M. and M. G. R. Cannell (1996). Temperate forest responses to carbon

dioxide, temperature and nitrogen, a model analysis. Plant Cell Environment
19(12): 1331-1348.

Tomich, T. P., H. d. Foresta, R. Dennis, Q. Ketterings, D. Murdiyarso, C. Palm, F. Stolle,

Suyanto and M. v. Noordwijk (2002). Carbon offsets for conservation and
development in Indonesia. Am. J. Alternative Agriculture 17(3): 125-137.

105

Weibel, F. (2002). Soil management and in-row weed control in organic apple
production. The Compact Fruit Tree 35(4): 118-121.

Zavalloni, C. (2004). Evaluation of nitrogen-fertilizer uptake, nitrogen-use and water-use
efﬁciency in sweet cherry (Prunus avium L.) on dwarﬁng and standard
rootstocks. Department of Horticulture. East Lansing, Michigan State University:
163 p.

106

Table 3.1. Content (g/tree) of total C and N per plant of ‘Buckeye Gala’ grafted on M.9
NAKB 337 harvested in November 2003 for each soil management system. Each value
is average of 4 replicates. Values in the same row followed by the same letter are not
signiﬁcantly different for p $0.05 (Tukey’s adjusted).

 

 

 

 

 

 

 

 

 

 

 

 

Mulch Flame Sandwich Partial Cover Total Cover
Total C (g/tree) 1244 a 1 169.7 ab 943.0 ab 848.4 bc 551.7 c
Total N (g/tree) 23.5 a 16.9 ab 10.8 be 11.8 be 7.6 c
C/N ratio 53 b 69 ab 87 a 72 a 73 a

 

 

 

 

 

Table 3.2. Total dry weight of biomass of vegetative cover under each orchard ﬂoor
management system, amount of total C and N contained in it, and C/N ratio. Each value
is average of 4 replicates. Values in the same row followed by the same letter are not
signiﬁcantly different for p $0.05 (Tukey’s adjusted).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mulch Flame Sandwich Partial Cover Total Cover
Biomass (g/mz) 8 c 27 c 2104 ab 1840 b 2685 a
TotalC(g/m2) 3 c 8 c 658 b 653 b 895 a
TotalN(g/m2) 0.1 c 0.3 c 17.7 b 20.4 b 30.0 a
UN ratio 19 c 26 be 38 a 32 ab 30 b

 

 

107

 

 

a Cl Supporter 4
‘2 ' a IM.9NAKB337

 

 

o-——4

 

TCAI (cmz)

 

 

 

 

 

 

 

 

 

 

 

   

FL SS PC

TREATMENTS

Figure 3.1. Trunk cross sectional area increase (TCAI) between April 2001 and
November 2003 measured 25 cm above the graft union for two rootstocks and by orchard
ﬂoor management treatment (MU = Mulch, FL = Flame, SS = Sandwich System, PC =
Partial Cover, TC = Total Cover). Each bar represents an average of 8 replicates. Results
of analysis of variance are shown for rootstock effect under each treatment. Bars with the
same letter are not signiﬁcantly different for p_<_0.05 (Tukey’s adjusted).

108

 

 

 

3500 W“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A DShoot
3000 , I AB IRoot
a I AB
A 2500 - ab
0 T BC
V 2000 A
t— b‘;r
0
3::
M
E 1500 -
b
D
1000 -
500 4
0 ,
MU FL 88 PC TC
TREATMENTS

Figure 3.2. Shoot and root dry matter accumulated per tree between April 2001 and
November 2003 for ‘Buckeye Gala’ grafted on M.9 NAKB 337 as affected by orchard
ﬂoor management treatment (MU = Mulch, FL = Flame, SS = Sandwich System, PC =
Partial Cover, TC = Total Cover). Each bar represents an average of 4 replicates :1: SE.
Complete bars (shoot + root) with same capital letter indicate total dry matter of trees that
are not signiﬁcantly different for p50.05 (Tukey’s adjusted). Bar portions with the same
lower case letter are not signiﬁcantly different for p50.05 (Tukey’s adjusted).

 

 

l .3 .. -5 .......

 

 

a
r .2 a
C a
.g ab
‘- I.I ‘
H
o
O
5, r .0
b
o
o
04 0.9
b
+ R/S
0.8 * l
0.7 .
MU FL SS PC TC

Figure 3.3 Root/shoot ratio of 3-year-old ‘Buckeye Gala’ grafted on M.9 NAKB 337 as
affected by orchard ﬂoor management treatment (MU = Mulch, FL = Flame, SS =
Sandwich System, PC = Partial Cover, TC = Total Cover). Each bar represents an
average of 4 replicates t SE.

109

 

100
90 -
80
70
60 -
50
40 -

y = 67.428x"°7"‘
30 -

R2 = 0.7735
20 ~«

10

 

 

 

0.7 0.8 0.9 1 1.1 1.2 1.3
R : S

Figure 3. 4. Regression between the rootzshoot ratio (R:S) and the C:N ratio of the total
tree biomass for ‘Buckeye Gala’ grafted on M.9 NAKB 337. Each value is an average of
4 replicates.

110

1.6 - -~—- ---

 

IMU ElFL ISS IPC ITC

Nitrogen concentration (% dry weight)

 

 

Root Trunk Branches l-year-old branches

Figure 3.5. Nitrogen concentration in tissue (% in dry weight) in different plant parts
from ‘Buckeye Gala’ trees on rootstock M.9 NAKB 337 harvested in November 2003
under different orchard ﬂoor management systems (MU = mulch, FL = ﬂaming, SS =
Sandwich system, PC = Partial cover, TC = Total cover). Each bar represents an average
of 4 replicates :1: SE. Analysis of variance was carried within each tree part. Bars with the
same letter are not signiﬁcantly different for p50.05 (Tukey’s adjusted).

lll

l6

 

IMU ClFL ISS IPC ITC
14 J

ab

12.

10:

Total N content (g/tree)
op

 

 

   

6:
4-
2w
0— "Z/Z

W
S.

Trunk Branches l-year-old branches

Figure 3.6. Total N content (in g of N/tree) in different plant parts from ‘Buckeye Gala’
trees on rootstock M.9 NAKB 337 harvested in November 2003 from different soil
management systems (MU = mulch, FL = ﬂaming, SS = Sandwich system, PC = Partial
cover, TC = Total cover). Each bar represents an average of 4 replicates :1: SE. Analysis
of variance was carried within each tree part. Bars with the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted).

112

2500 7 6
- Root :1 Shoot -D- R/S ratio 1

 

 

 

    

a I 1
1.5
2000 {
NE | 4
R1 1500 1‘
in if 3 .3
a; t‘.‘
1000 ' (I)
b
D l: 2 2
500 ‘
4 l
b b
0 , .L : 2.- 2.. 0

 

 

Mulch Flame Sandwich Partial Cover Total Cover

Figure 3.7. Dry weights (in g/mz) of vegetative soil covers separated for root and shoot
portions in each orchard ﬂoor management system, and root:shoot ratio (R/S) of that
biomass. Each value is mean of 4 replicates. Analysis of variance results shown are for
total dry biomass of vegetative cover (root + shoot); bars with the same letter are not
signiﬁcantly different for p50.05 (Tukey’s adjusted).

113

CHAPTER 4

114

CHAPTER 4

ORCHARD FLOOR MANAGEMENT SYSTEMS AND THEIR IMPACT ON
RESIDENT VEGETATION IN ORGANIC APPLE PRODUCTION
Abstract
The design of more sustainable fruit growing systems requires a better understanding of
the dynamics of vegetation growing under the trees. We investigated the effect of ﬁve
orchard ﬂoor management systems on population composition and density of the
vegetative cover in an organic apple orchard. Alfalfa hay mulch, propane ﬂame burner,
Swiss Sandwich system (a combination of resident vegetation and tilled strips), were
followed for three years in their impact on resident vegetation. Complementary to the
ﬁeld experiment, a container experiment was performed to evaluate the impact of the
previous management systems plus a partial and a total soil cover of a mix of red clover
and rye grass, on the N cycling. Nitrogen content of the most frequently found species
was determined. The treatments with live vegetative soil cover, and especially Swiss
Sandwich had the highest species diversity, due to the increase of herbaceous perennial
and biennial species. In Flame the annual species were dominant. A range in
concentration between 0.9 and 3.7% of N in the tissue was measured for the different
species present. Biomass production and total N content varied among species having no
correlation with the previous parameter. The presence of active vegetative cover in
Sandwich, Partial Cover, and Total Cover reduced the leachate volume and nitrate
content compared to Mulch and increased inﬁltration compared to Flame. The orchard
ﬂoor management systems changed the composition and density of the vegetative cover

and impacted the total N uptake and nitrate content in leachate.

115

INTRODUCTION

Orchard ﬂoor management is one of the key pieces of the fruit growing system.
The importance increases when we take the organic production approach in which soil is
the basis that sets the production outcome (Magdoff and van E8, 2000). The efforts in
achieving a healthy soil take a good part of the grower’s concerns for successful organic
production.

Ground ﬂoor management has as one of its main objectives to control unwanted
vegetation growing near the trees. This area of focus in organic fruit production calls for
new adjustment of the strategies. Today there is need for more than controlling weeds. A
wider approach that aims to manage the vegetative cover growth (generally with the
incorporation of live soil covers) with a system vision is more compatible with the
organic farming principles. The use of mulches for example incorporates materials that
can achieve the weed control and add some other advantages (Hogue and Neilsen, 1987).
Viewing all the plants other than the fruit trees, also as a desirable part of the system,
requires that we evaluate their participation in it, and that considering the pros and cons
of having them we ﬁnd the ways to maximize the former and minimize the latter through
a correct management. In the search for more sustainable production systems, one key
aspect to consider is this change of approach from weed control to weed management.
Weed management emphasizes integration of preventive techniques, scientiﬁc
knowledge, and management skills. A better understanding of weed biology and ecology
can help in understanding weediness and thus in the design of appropriate management

(Buhler, 1999).

116

One of the main drawbacks of vegetative cover (herbaceous vegetation growing
under the fruit trees) is known to be its competition with the ﬁ'uit crop for nutrients, and
specially water and nitrogen (The Duke of Bedford and Pickering, 1911; Hogue and
Neilsen, 1987). Plant species use different strategies to compete with others and this
explains part of the population composition changes that takes place in weed succession
(Gleeson and T ilman, 1990; Tihnan and Wedin, 1991a; Craine et al., 2002). When
interference takes place through disruptive practices as those used for weed control the
response can vary greatly through the spectrum of species present and while some might
be hindered, others might be enhanced in their multiplication or growth.

The aim of this study was to obtain information from contrasting ground ﬂoor
management systems applied in organic apple production to improve weed management
strategies. For this purpose, population compositions of the vegetative cover and its shifts
as well as surface coverage were monitored in two experiments: ﬁeld and containers. We
looked at indicators of N consumption for different species and the implications of the
soil and cover management in the trees performance. Changes in N cycling were

followed.

MATERIALS AND METHODS

Site description

The experiments were conducted at the Clarksville Horticultural Experiment

Station of Michigan State University in the vicinity of Clarksville, MI (latitude of 42° 84’

117

 

N and longitude of 84° 24’ W). The ﬁeld experiment covers an area of 0.38 ha of the 2.5
ha of total surface that is certiﬁed by the Organic Crop Improvement Association
(OCIA). The predominant soil type is a Kalamazoo sandy clay loam (Typic Hapludalfs)
with a particle size composition of 53.1% sand, 23.1% silt, and 23.8% clay, and mild
slopes of less than 3%. Daily temperatures and precipitation during the whole experiment
were recorded with an automated weather station of the Michigan Automated Weather
Network, located approximately 560 m west of the orchard (Figure A.l thru A.4,

Appendix A).

Field experiment
0mm ﬂoor management

The site where the orchard was planted went through a rotation soybean-com-
com-alfalfa previous to buckwheat sown in 1999. Chicken compost was applied this last
year at a rate of 1250 kg/ha and lime at a rate of 2250 kg/ha.

A mixture of red mammoth clover and endophytic rye was sowed at the time of
apple tree plantation in April 2000. In all alleys of the orchard mowing was performed
regularly according to best management farming practices to discourage weed
development and enhance clover growth. Three different management systems were
applied to the tree rows at the beginning of the season in 2001:

1. Mulch (MU): A l-m-strip on each side of the tree row was covered with alfalfa hay
mulch. During the growing season the thickness of the cover was over 15 cm to provide a
shading effect that would suppress weeds, and maintain moisture. Mulch was hand-

applied in early spring and fall.

118

2. Weed ﬂaming (FL): A l-m-strip on each side of the tree row was kept free of
vegetation through the use of a propane ﬂame burner that provided 800.000 BTU through
four torches, and was equipped with a metal cover to prevent tree damage. Minimal
amounts of water were sprayed just behind the burner to reduce the risk of heat affecting
tree branches or ﬁre spreading to the neighboring mulch. The tractor speed was kept
between 1.6 and 3.2 km/h, adjusted according to the density of weeds to be burned.
Flame treatments were made when weed height approached 10 cm (Table D.1.A,
Appendix D).

3. Sandwich System (SS): an adaptation of the Swiss Sandwich system (Weibel, 2002),
consisted of a central strip that spread 30 cm to each side of the tree row in which
vegetation was allowed to grow. A 70-cm-wide strip was tilled on each side of that
vegetated central strip. The initial tilling was applied in early spring with a rotovator to
destroy and incorporate the vegetation from the previous season. From then on, a spring
tooth harrow was used to keep the strip vegetation-free. Tilling was repeated when weeds
grew to 10 cm height (Table D.1.B, Appendix D). In the vegetated central strip hand
weeding was performed when weeds such as curly dock (Rumex crispus L.), common
lambsquarters (Chenopodium album L.), redroot pigweed (Amaranthus retroﬂexus L.)
and thistles (Cirsium sp.), known to be hosts for some of the important apple pests,
appeared. Track was kept of these plants registering the number that was pulled out by
plot. In the lateral tilled strips, bell beans (Vicia fava) were sown in the fall as a green

manure.

119

Vegetative cover

Three areas of l m x 0.7 m adjacent to the tree line in each plot were surveyed in
November 2001, April and October 2002, and May and November 2003. During each
survey all plants in the sampled area were counted and identiﬁed. The MSU Herbarium
performed identiﬁcation of several samples.

Adult individuals of the main occurring species were unearthed from guard rows.
They were measured to determine size and to estimate area of coverage. Five plants per
species were washed and rinsed with distilled water, the root cut apart from shoot, and
each part separately dried at 60°C until constant weight. Dry weights were registered, and

after grinding a sample was taken to determine total N content by the Kjeldhal method.

Soil and tree measurements

For a complete description of measurements on soil and trees see Chapter 1. 1

Experimental design and statistical analysis

The ﬁeld experiment was set as a randomized complete block design with six
replicates of the three applied treatments. The plots consist of 12 trees to which the main
factor, ground ﬂoor management, was randomly assigned. Subplots given by each of the
three rootstocks evaluated consist of four trees each (split-plot). Data was taken from the
two central trees of each subplot.

Analysis of variance was performed using MIXED procedure with SAS (Version
8, SAS Institute, Cary, NC, USA) to detect treatment effects. After using ANOVA, least

squares means were compared using the Tukey-Kramer test with p50.05 unless stated

120

otherwise. For the count of species, data was analyzed using GLIMMIX procedure with

Poisson distribution with SAS (same as above).

Container experiment

Oriard ﬂoor management

Five ﬂoor management systems (three identical to the ﬁeld experiment) were
applied to wooden boxes of 1m x 1.2m x 0.5m boxes where apple trees were planted in
2001:
1. Mulch (MU) managed the same as in the ﬁeld.
2. Flaming (FL) In this case the entire area of the box (1.2 m2) was kept free of
vegetation through the use of a propane ﬂame burner hand torch that provided 150.000
BTU (Table D.2.A, Appendix D).
3. Sandwich system (SS): same as in the ﬁeld but adapted to the box area: the central
strip spread 20 cm to each side of the tree row in which vegetation was allowed to grow;
and a 40 cm wide strip was tilled on each side of that vegetated central strip. The tilling
was performed with a hand hoe to 7 cm depth or less. Tilling started in spring, and was
repeated when weeds grew to 10 cm tall (Table D.2.B, Appendix D).
4. Partial cover (PC). Just after the planting of trees a mixture of red mammoth clover
and endophytic rye was sewn. The cover was allowed to grow in a central strip that
spread 20 cm to each side of the tree trunk. The remaining area on each side (40 cm wide

strip), was tilled at the same time as SS.

121

 

5. Total cover (TC). As in the previous treatment, a mixture of red mammoth clover and
endophytic rye was sewn just after planting the trees. In this case, the mixture was

allowed to grow in the whole area of the box.

Soil and tree mea_suremergs
For a complete description of measurements on soil, leaching, and trees see

Chapter 2.

Experimental design and statistical analysis

The container experiment was installed at the side of the ﬁeld experiment, so
under the same weather conditions. It was set as a completely randomized design with
ﬁve treatments (same three from the main experiment plus two more), having eight
replicates of each for each of two rootstocks. The plots for this experiment consisted of 1
tree planted on a wooden box of l m x 1.2 m and 0.5 m high. The boxes were lined in
two separate rows, and the treatments and rootstocks randomly assigned.

Analysis of variance was performed using MIXED procedure with SAS (Version
8, SAS Institute, Cary, NC, USA) to detect treatment effects. After using ANOVA, least
squares means were compared using the Tukey-Kramer test with p50.05 unless stated
otherwise. For the count of species, data was analyzed using GLIMMIX procedure and

Poisson distribution with SAS (Version 8).

122

RESULTS

Field experiment

Vegiative cover composition

The number of species identiﬁed was affected by the orchard ﬂoor management
system (Figure 4.1). In the MU (Table D3, Appendix D) there was a lower number than
in the other treatments, and this number remained the same or got even smaller
throughout the experiment. Plants/m2 also was the least in MU (between 2 and 20). In FL
the number of species identiﬁed was similar to SS at the beginning and didn’t change by
2003. The regular burning of weeds had a selective effect towards certain species (Table
D4, Appendix D). Most of them were controlled and just a few species like dandelion
(T araxacum oﬁcinale Weber in Wiggers) or some grasses remained to be of concern. The
number of plants/m2 went from 60 at the beginning to 190 for the ﬁnal survey.

The highest number of species was identiﬁed in the SS with a maximum of 38
species in May of 2003 among the six plots under this groundcover management system
(Figure 4.1 and Table D.5, Appendix D). The vegetated central strip gave place to an
increased number of species in October 2002 and May 2003 that slightly diminished in
the sampling of November 2003. Part of the tilled strip was included as well in the
sampling area, and in that surface different species that could proﬁt from the growth
conditions and lack of competence, alternated in the intent to colonize the weed free area.
Number of plants/m2 was the highest in SS with values that started at 100 and ended up at

370 in the survey of November 2003, doubling the density of FL (Figure 4.2).

123

The zero threshold applied to four species in the SS gave a good control
preventing their seeding. The number of plants that needed to be hand weeded diminished
every year (Table 4.1), and by the third one, this practice was carried out in a very short
time. Conversely, in the case of other species as wild carrot (Daucus sp.), horseweed
[Conyza canadiensis (L.) Cronq.], and some grasses like orchardgrass (Dactylis
glomerata L.) their population increased continuously.

Other species that were observed to develop and increase their occurrence in the
vegetated strip of SS (Table D.5, Appendix D) were: red sorrel (Rumex acetosella L.),
chickweed [Stellaria media (L.) Vill.], hairy bittercress (Cardamine hirsuta L.), henbit
(Lamium aplexicaule L.), white clover (T rifolium repens L.), and annual bluegrass (Poa
annua L.).

The species more commonly found in the plots were analyzed to determine N
content (Table 4.2). These results showed a variation from plants with high N
concentration (legumes, larnbsquarter [Chenopodium album L.], curly dock [Rumex
crispus L.], etc) to plants with very low content (grasses, old cinquefoil [Potentilla
.simplex Michx.], red root pigweed [Amaranthus retroﬂexus L.], etc). The R/S ratio of the

species analyzed had a wide variation with values between 0.1 and 4.3.

Container experiment

Vegetative cover composition

As for the ﬁeld, the number of species that were identiﬁed varied over time and

according to the orchard ﬂoor management system shifting in different way (Tables D6

124

thru D.10, Appendix D). All treatments increased in species diversity over time except
MU, which was also the least diverse (Figure 4.3). Species diversity was greatest in SS,
PC, and TC with FL intermediate for the ﬁnal survey.

Additionally, herbaceous cover density was lowest in the MU treatment (Figure
4.4). Plant density for TC was generally lower than SS and PC with a value between MU
and the other three treatments. In the ﬁnal survey of November 2003 plant density was

greatest in SS.

DISCUSSION

The impact of orchard ﬂoor management systems on some of the soil parameters
(see Chapter 1) was clearer in the ﬁeld than in the boxes where distinction between two
depths could not be made (see Chapter 2). The increase of SOM and N content due to the
MU was higher for the shallower depth of 0-10 cm than for the 0-30 cm as reported in
previous research (N eedehnan et al., 1999) . Conversely the buffer capacity of the system
was restrained in the case of the boxes and thus for some other variables the differences
appeared enhanced in that experiment, as it occurred with soil moisture.

In regard to the effect on resident vegetation, results were coincident for both
experiments. The vegetative cover population went through changes due to the impact of
the ground cover management systems. The thick layer of alfalfa hay (MU treatment)
with a year round average depth of approximately 15 cm was sufﬁcient to suppress most
of the weed species and just some isolated individuals of henbit, chickweed, or

orchardgrass appeared very sporadically. The initial higher number of species and

125

number of plants was associated with the area just by the trunk of the trees where the
mouse guards left a small surface uncovered by the mulch that eventually got enough
light for the growth of weeds. Two exceptions to this general rule were quackgrass
[Elymus repens (L.) Gould] and red sorrel. Both species multiply by rhizomes at very
high rates under good growing conditions. High colonization by these species was mainly
in guard rows since they got into the MU from adjacent treatments, and it occurred only
slightly in treatment rows. Once they were able to establish themselves, they found
excellent conditions to grow in the environment created by the mulch.

For those species in which meristems remain covered or under soil, the efﬁciency
of the FL was not as good as for others and they could not be eliminated (Rifai et al.,
1999). A clear example was dandelion with high regrowth and increasing number.
Concurrently, the clean area gave possibility to certain broadleaf species like chickweed,
henbit, mustard, or purslane speedwell (Veronica peregrina L.) to take advantage of the
lack of competition and brieﬂy colonized for the period before the last ﬂame treatment.
The size and species composition of the weed seed bank can have a rapid decline if new
seed introduction does not occur (Buhler et al., 1997). As the seeds germinate, and no
new seeds are exposed, since there is no soil movement, the variety of species that can
take advantage of this vegetation-free-area should be less in the subsequent years.
Grasses were another example of plants that survived the FL treatment when they
developed over a certain size, since the growing points got covered and obtained enough
protection under their own biomass. In this case, as in many other techniques, the proper
timing of the ﬂaming is very important for a successﬁrl application (Rifai et al., 1999).

An added problem of a late application is that even though the propane ﬂame burner

126

might dry the plants, an increased amount of biomass is left on the ﬁeld. With subsequent
treatment, accumulated dry material burns immediately, generating a higher risk of
damaging trees or causing ﬁre. Even though FL was one of the treatments with higher
density, these plants generally didn’t survive due to ﬂaming that occurred shortly after
the seedling stage. With the proper timing, in the successive years, the weed control
turned easier due to a cumulative effect of the FL.

The zero threshold strategy (Norris, 1999) which consisted in the elimination of
every specimen not allowing it to grow and before it could multiply, was applied for
curly dock, red root pi gweed, larnbsquarter, and thistle. This was effective in reducing the
infestation of those species both in the tilled strip as in the vegetated central strip.
Clearly, this practice can be carried out more easily in restricted areas. It could be argued
that over larger areas it might be impracticable or too costly.

The tilling performed in the strip of Sandwich was able to maintain the designed
vegetation-free-area. Plant density of the vegetative cover in S8 was among the highest
values. This should be careﬁrlly considered since the highest proportion of those plants
counted, germinated in the tilled strip and were killed with subsequent tillage. Species
established in the vegetated central strip, passed through the life cycle undisturbed. These
species, such as wild carrot, horseweed, and orchardgrass proliferated in the central strip,
as they propagated by sexual and asexual means. Some of these species could achieve an
important height causing an increased competition or phytosanitary problems which is
species dependent (Tworkoski and Glenn, 2001).

Another group of species that increased their presence in the Sandwich included

those of low or moderate stature (under 40 cm) that would not interfere with the

127

development of the tree canopy. Examples of these plants in our research were: red
sorrel, chickweed, mustards, henbit, white clover, and some grasses like annual
bluegrass. The encouragement of low stature species is more beneﬁcial and will reduce or
prevent the growth of undesirable species. It was very clear, mainly in the case of red
sorrel, that when it grew and expanded its coverage, clusters were formed, and not many
other species were able to establish. In other cases manual intervention might be needed,
but once the surface is covered by the low growing species, selective weeding will be less
necessary in the central vegetated strip of the SS.

The N content in the plants is another important variable to consider in relation to
the biomass production and potential competition for uptake with the trees (Tilman and
Wedin, 1991b; Engelaar et al., 1995; Koutroubas et al., 2000). Plants with an average N
content or slightly above average like curly dock, or even with low N content like yellow
foxtail [Setaria glauca (L.) Beauv.] or quackgrass but with high biomass production, can
turn into a more signiﬁcant problem of competition than plants with high N content with
low biomass.

It is clear that to be able to grow and multiply faster, the plants use different
mechanisms to out compete the other species. When it comes to compete under low N
levels in soil, low-activity leaves and roots, and high root:shoot ratios are among the
functional traits of non-legumes that produce and maintain large amounts of biomass
(Craine et al., 2002). Craine and collaborators found that the low-N species maintained
greater biomass than high-N species more by producing tissues with low N

concentrations and greater longevity than by acquiring more N.

128

But nutrient competition is not the only concern about a permanent cover near the
trees. Clear effects of other interference mechanisms have been shown for example in
corn and soybeans (Bhowmik and Doll, 1984). The incorporation of residues of common
lambsquarters, redroot pigweed, velvetleaf, or yellow foxtail into the soil, increased the
inhibition of growth of the crops independent of N and P nutrition, suggesting that the
inhibitory effects were allelopathic and not related to nutrient uptake. This is another
aspect of the system that can vary and generate completely different results according to
the species involved and existing in the orchard. Having a permanent vegetation in the
central strip will increase the possibility of enhanced allelopathic effects that can occur
on all phases of the nitrogen cycle which involve plants and microorganisms (Rice,
1992). Little information is available in this regard.

The values of N found in leachate (see Chapter 2) show the importance of
checking the N balance not only to be sure the trees are receiving enough supply, but to
prevent excessive amounts that could become a problem in the environment. The
presence of a live cover as well as plant diversity, has been reported to substantially
reduce leaching problems (Smethurst and Nambiar, 1995; Merwin et al., 1996; Sanchez
et al., 2004). It can also improve soil water inﬁltration rates as found in this study and by
other researchers (Atkinson and White, 1981; Glenn and Welker, 1989). This can be
more important than the additional water consumption. Thus, the differences in volume
of leachate followed more the inﬁltration capacity of the treatments than their
consumption capacity. Root characteristics like depth, density, and growth habit have a
strong impact on marcoporositiy as well as on soil food web development both of which

can strongly affect the N cycling (Merwin et al., 1994; Rothstein, 2000).

129

CONCLUSIONS

The orchard ﬂoor management systems impacted the vegetative cover population
composition. The treatments with live vegetative soil cover, and especially Sandwich had
the highest species diversity, which is a desirable attribute towards sustainability. In these
treatments mostly herbaceous perennial and biennial species increased their participation
in the soil coverage. Annual species dominated vegetative cover in Flame. Attention
should be given to this aspect due to the possibility of increase in tall growing species,
which would not be beneﬁcial.

Density of vegetative cover was highest for Sandwich, Partial Cover, and Flame.
While in the ﬁrst two cases a high percentage was made up of established plants, for
Flame it was mainly because of seedlings. Thus, while surface coverage was important in
Sandwich and Partial Cover, it was almost insigniﬁcant for Flame.

Density of vegetative cover was very low for Mulch reﬂecting the effectiveness of
the treatment but concerns arise in regard to quality and quantity of the material
necessary for achieving it and its practical applicability and cost. Risk of weed invasion
should also be considered.

The applications with propane ﬂame burner were the most sensitive to timing for
a successful result. Maximizing thermal units for greater effectiveness without damaging
trees was critical in operation. Environmental conditions such as air temperature or
humidity, and treatment application characteristics such as tractor speed or irrigation

usage had to be constantly monitored and adjusted.

130

The zero threshold strategy performed in Sandwich was successful in reducing the
problem of species like curly dock, red root pigweed, lambsquarter, and thistles.

The presence of active vegetation in Sandwich, Partial Cover, and Total Cover
reduced the leachate volume and N03'—N content compared to Mulch and increased
inﬁltration compared to Flame. These are part of the advantages of having a live cover on
the soil. Total N uptake by the vegetative cover as estimated for pure stands varied with
species showing the importance of considering its composition because of the impact this

can have on competition with trees.

131

LITERATURE CITED

Atkinson, D. and G. C. White (1981). The effects of weeds and weed control on
temperate fruit orchards and their environment. In: 'Pests, pathogens and
vegetation: The role of weeds and wild plants in the ecology of crop pests and
diseases.‘ J. M. Thresh. Pittman. London. pp: 415-428.

Bhowmik, P. C. and J. D. Doll (1984). Allelopathic Effects of Annual Weed Residues on
Growth and Nutrient Uptake of Corn Zea-Mays and Soybeans Glycine-Max.
Agronomy Journal 76(3): 383-388.

Buhler, D. D. (1999). Expanding the Context of Weed Management. Bingharnton, NY,
Food Products Press. 289 p.

Buhler, D. D., R. G. Hartzler and F. Forcella (1997). Implications of weed seedbank
dynamics to weed management. Weed Science 45: 329-336.

Craine, J. M., D. Tilman, D. Wedin, P. Reich, M. Tjoelker and J. Knops (2002).
Functional traits, productivity and effects on nitrogen cycling of 33 grassland
species. Functional Ecology 16: 563—574.

Engelaar, W. M. H. G., E. J. W. Visser, B. W. Veen and C. W. P. M. Blom (1995).
Contents, uptake rates and reduction of nitrate of Rumex palustris and Plantago
major spp. major grown on compacted soil. Functional Ecology 9(2): 334-339.

Gleeson, S. K. and D. Tilman (1990). Allocation and the Transient Dynamics of
Succession on Poor Soils. Ecology 71(3): 1144-1155.

Glenn, D. M. and W. V. Welker (1989). Orchard soil management systems inﬂuence
rainfall inﬁltration. J. Amer. Soc. Hort. Sci. 114(1): 10-14.

Hogue, E. J. and G. H. Neilsen (1987). Orchard ﬂoor vegetation management.
Horticultural reviews 9: 377-430.

Koutroubas, S. D., D. S. Veresoglou and A. Zounos (2000). Nutrient use efﬁciency as a
factor determining the structure of herbaceous plant communities in low-nutrient
environments. J. Agron. Crop Sci. 184(4): 261-266.

Magdoff, F. and H. M. van Es (2000). Building soils for better crops. Washington DC,
J arboe Printing. 230 p.

Merwin, I. A., J. A. Ray, T. S. Steenhuis and J. B011 (1996). Groundcover management

systems inﬂuence fungicide and nitrate-N concentration in leachate and runoff
from a New York apple orchard. J. Amer. Soc. Hort. Sci. 121(2): 249-257.

132

Merwin, I. A., W. C. Stiles and H. M. van Es (1994). Orchard groundcover management
impacts on soil physical properties. J. Amer. Soc. Hort. Sci. 119(2): 216-222.

Needelman, B. A., M. M. Wander, G. A. Bollero, C. W. Boast, G. K. Sims and D. G.
Bullock (1999). Interaction of tillage and soil texture: Biologically active soil
organic matter in Illinois. Soil Sci. Soc. Am. J. 63: 1326-1334.

Norris (1999). Ecological implications of using thresholds for weed management. In:
'Expanding the context of weed management'. Food Product Press. Binghamton,
NY. pp: 3 1-58.

Rice, E. L. (1992). Allelopathic effects on nitrogen cycling. In: 'Allelopathy - Basic and
applied aspects'. V. Rizvi. Chapman & Hall. London. pp: 31-58.

Rifai, N., M. Lacko-Bartosova and R. Somr (1999). Weed control by ﬂaming and hot
steam in apple orchards. Plant Protection Science 35(4): 147-152.

Rothstein, D. E. (2000). Spring ephemeral herbs and nitrogen cycling in a northern
hardwood forest: an experimental test of the vernal dam hypothesis. Oecologia
124(3): 446-453.

Sanchez, J. E., R. R. Harwood, T. C. Willson, K. Kizilkaya, J. Smeenk, B. Parker, E. A.
Paul, B. D. Knezek and P. Robertson (2004). Managing soil carbon and nitrogen
for productivity and environmental quality. Agronomy Journal 96: 769-7 75.

Smethurst, P. J. and E. K. S. Nambiar (1995). Changes in soil carbon and nitrogen during
the establishment of a second crop of Pinus radiata. Forest Ecology and
Management 73(1-3): 145-155.

The Duke of Bedford, K. G., F.R.S. and S. U. Pickering, M.A., F.R.S. (1911). Thirteenth
Report of the Woburn experimental ﬁ'uit farm. London, The Amalgamated Press,
Ltd. pp: 1-151.

Tilman, D. and D. Wedin (1991a). Dynamics of Nitrogen Competition Between
Successional Grasses. Ecology 72(3): 1038-1049.

Tilman, D. and D. Wedin (1991b). Plant Traits and Resource Reduction For Five Grasses
Growing on a Nitrogen Gradient. Ecology 72(2): 685-700.

Tworkoski, T. J. and D. M. Glenn (2001). Yield, shoot and root growth, and
physiological responses of mature peach trees to grass competition. HortScience
36(7): 1214-1218.

Weibel, F. P. (2002). Soil management and in-row weed control in organic apple
production. The Compact Fruit Tree: 118-121.

133

Table 4.1. Number of plants/m2 that were hand weeded from the Sandwich treatment for
each year of experiment.

2001 2002 2003

 

redroot pi gweed 175 123 3
curly dock 36 23 9
lambsquarter 1 5 8 l

 

Table 4.2. Nitrogen concentration (% in dry weight of shoot and root), R/S ratio (dry
weight basis) and estimation of N content in a pure stand of the species (g N/m’) Each
value is a mean of 5 samples.

Common name %N in Shoot %N in Root R/S ratio Total g N/m2

 

annual bluegrass 2.43 0.94 0.22 0.52
bell bean 3.14 2.20 0.76 9.84
blacknightshade 2.09 1 .45 0.65 7.98
buckhom 2.09 1.19 0.14 1.03
chickweed 2.44 1.05 0.10 9.83
clover 2.69 2.70 0.59 11.28
crabgrass 1.19 1.49 0.15 4.84
curly dock 2.73 2.58 4.32 109.21
dandelion 2.66 1.42 1.27 3.42
downey brome 2.42 1.07 0.18 3.28
green foxtail 2.19 1.18 0.12 20.78
hairy bittercress 2.05 0.10 0.05 2.78
henbit 3.17 1.36 0.24 2.62
horseweed 1.46 0.98 0.25 1 1.01
Italian ryegrass 1.89 0.99 0.49 2.56
lambsquarter 2.80 0.62 0.12 8.42
old cinquefoil 1.33 1.38 0.41 32.38
orchardgrass 1.71 1.13 0.13 36.38
prickly lettuce 2.27 1.09 0.14 15.92
quackgrass 1.26 0.97 0.26 86.29
red sorrel 3.72 1.79 0.21 1.86
redroot pigweed 1.36 0.78 0.13 5.49
sheperd's purse 2.27 1.40 0.14 14.88
wild carrot 2.74 1.38 0.65 13.32
yellow foxtail 0.91 1.88 0.84 131.59

 

134

35 '- EElMulch UFlame ISandwich
l

 

Number of species

 

 

 

 

 

 

 

NOV 2001 APR 2002 OCT 2002 MAY 2003 NOV 2003

Figure 4.1 Field Experiment. Number of species identiﬁed n the vegetative soil cover for
each orchard ﬂoor management during 5 surveys. Each bar is mean of 6 replicates.

 

 

 

 

 

“E :2: ' IMulch DFlame ISandwich %
i a
E 350 %
q;- 300 —- %
1. 250 4 /
'8 /
g 200 -‘ %
Z 150 A~ %

100 i %
50 - %

Figure 4.2. Field Experiment Plants/m2 counted for all species in the vegetative soil cover
for each orchard ﬂoor management during 5 surveys. Each column is mean of 6
replicates 3c Std error.

135

35.. IMU ElFL ISS IPC ITC

Number of species

x\\\\\\\\\\\\\\\\‘§\\\\\\\\\\\\\\V

     

?
g
/
¢
/
¢
¢
/
¢
¢
¢
/
¢
/
/
¢
%
¢
¢
¢
2
a

 

NOV 2001 APR 2002 OCT 2002 MAY 2003 NOV 2003 Accumulated
sp.

Figure 4.3. Container Experiment Number of species identiﬁed in the vegetative soil
cover for each orchard ﬂoor management during 5 surveys. Each bar is mean of 4
replicates.

800
1
l IMU. ClFL. ISS. IPC. ITC.

Number of plants per m2

 

 

   

APR 2002 OCT 2002 MAY 2003 NOV 2003

Figure 4.4. Container Experiment Plants/m2 counted for all species in the vegetative soil
cover for each orchard ﬂoor management during 5 surveys. Each bar is mean of 4
replicates d: Std error.

136

SUMMARY AND CONCLUSIONS

137

SUMMARY AND CONCLUSIONS

This study was conducted as part of the Organic Apple Project that includes
aspects of soil quality, pest and disease management, tree and ground ﬂoor management,
marketing, and education. The objective of our work was to investigate the effect of
alternative orchard ﬂoor management systems, compatible with organic production, on
soil parameters and apple rootstocks of diverse vigor. Impacts on N cycling and
vegetative cover population, as well as on soil food web were also studied.

The Mulch, done with alfalfa hay, had clear effects increasing soil organic matter
content, N availability in the soil proﬁle, and conservation of soil moisture. Differences
in organic matter were highly signiﬁcant for 0-10 cm depth, and tended to disappear
when considering 0-30 cm in the proﬁle. In the case of N03'—N the differences were
marked even at 0-30 cm depth, consistent with the high mobility of N03'. The 8011 food
web composition reﬂected the different conditions generated by the orchard ﬂoor
management systems. The increased conservation of soil moisture under mulch gave a
higher level of ciliates, which associate with low oxygen niches. This should be a call of
attention in regard to other steps of the N cycling. Numerous anaerobic micro sites with a
high N supply could be giving place to the emission of N20 at values worth considering.
This is an area in which research will be valuable.

Higher values of active fungal biomass and bacterial feeder nematodes were
found under Sandwich followed by Mulch. Probably these differences will increase with
time since these two systems have resources to support higher microbial activity than

Flame in which addition of organic matter or rhizosphere volume is certainly lower.

138

The inﬂuence of the soil cover on the inﬁltration rates was clearly measured
through the container experiment. Mulch had the highest rates throughout the experiment,
while Total Cover generated the maximum increase in the 3-year period through a dense
root system that enhanced macropore development. The inﬂuence of the cover on
inﬁltration increase was greater than on water consumption in regard to determining the
volumes of leachate. Nevertheless, counting with irrigation is of critical importance to be
able to consider ﬂoor management practices like the Sandwich System in areas where
precipitation is limited or very irregular.

Refraining from the use of fertilizer in all treatments was an extra pressure put on
the system. This strategy gave the Mulch treatment an advantage by the liberation of
nitrogen and its higher availability. These advantages in regard to N as well as those on
water supply for the Mulch reﬂected in higher growth and yield than in the other
treatments, for most rootstocks. Nevertheless differences weren’t as signiﬁcant as it could
have been anticipated considering the external inputs that were added for the mulching.
Both quantity and quality of the material used for the mulching has an important effect on
the efﬁciency of the system as well as on the cost. Even though the maintenance is
relatively low, the cost of the material and hand labor for its application in the ﬁeld can
be an important constraint, as well as the need of redressing to maintain the litter
thickness to suppress weeds. The type of material used for mulching will affect the
success on vegetation control and the amount and timing of N supplied. The possibility of
adding sufﬁcient N through the mulching process was demonstrated and even the
potential of leaching into the environment. Content of N in the leachate under mulch was

10 times that of the other orchard ﬂoor management systems. When values are up scaled

139

from plot to orchard level, the amounts of a potential leaching and environmental
recharge are around 30 to 40 kg.ha’l under the Mulch, which is lower to values seen
under agricultural land. Preliminary observations conducted in some of the leachate
samples showed that inorganic N was just part of the total N present. Organic N forms
were found to be more than half of the total N for many of them. This is an area of
research to be pursued.

The ﬂaming obliged to a continuous and detailed monitoring of environmental
conditions and treatment application to diminish risk of heat damage to trees. Thus,
timing of the applications was very important for achieving a good result. The use of
propane gas, even though in small amounts, is a concern for sustainable systems since it
is derived from off-farm sources. Under Flame, similar undesirable effects as for
continuous herbicide application were seen in regard to the development of a superﬁcial
crust of soil with reduced inﬁltration capacity.

The broader vegetation-free area that the Flame had over the Sandwich System
did not give measurable advantages to the trees in Flame. The effect of the soil cover of
Sandwich had a beneﬁcial impact on the soil, showing a tendency to increase the labile C
pool and total soil organic matter content compared to Flame. The spatial distribution of
the vegetation-free area of the Sandwich, which did not include the tree trunk, facilitated
cultivation. A simple spring-tooth harrow was used in a fast operation with high time and
energy efﬁciency and low trunk damage potential. The shallow and narrow strip tilling
had no impact until recently on soil properties. The width of the strips however, both

tilled and the central vegetated one, is one of the variables that will need to be studied

140

further to obtain the best relationship of vegetation-free area according to type of soil and
rootstock vigor.

The rootstocks had a strong effect on vigor, Supporter 4 being the most vigorous,
and M.9 NAKB 337 the least. ‘Paciﬁc Gala’ on M.9 RN29 was closer to Supporter 4 in
vigor but had a ﬂower number comparable to the other M.9 clone and much higher than
Supporter 4. The ﬁeld and container experiments supported one another with
complimentary information in regard to rootstock performance as well as in other areas.
There was a signiﬁcant interaction of orchard ﬂoor management system and rootstock.
While for the ﬁrst and only harvest ‘Paciﬁc Gala’ on M.9 NAKB 337 had the highest
yields for Mulch and Flame, it was on M.9 RN 29 that it performed best under Sandwich.
This interaction was also present among trees in the container experiment. The trunk
cross sectional area increase of ‘Buckeye Gala’ between 2001 and 2003 did not differ
among rootstocks for Mulch and Total Cover. Conversely, for Flame, Sandwich, and
Partial Cover, there were signiﬁcant differences between the two rootstocks. This
conﬁrms in a ﬁrst step, the idea of being able to compensate certain levels of competition
through the use of a more vigorous rootstock. These observations should continue so as
to conﬁrm these results on the long term.

With good supply of N and water, trees under Mulch allocated lower percentage
of their resources on root growth than on shoots (R:S ratio = 0.77). Treatments with a
vegetative soil cover developed greater root systems reaching values of 1.14 for R:S
ratio. Sandwich had the highest value for C:N ratio of full plant which could be indicating
a higher efﬁciency in the use of N and more balanced condition. While Mulch could have

been under luxury consumption of N and thus the low value (ratio of 30), it seems that

141

the trees under Total Cover had a high value more due to a lack of N, than to an efﬁcient
work of the tree machinery in ﬁxing C for each unit of N.

The diversity and number of species found in the vegetative cover of Sandwich
seems to be an additional advantage to promote a more stable system, with higher
beneﬁcial insect numbers, and a diverse microbial population, which will be enhancing
the nutrient cycling. This study shows that the Swiss Sandwich System is a viable option
in the initial years of an organic orchard for growers in Michigan, and research should

continue with trees in full production, for the adjustment in the design of the system.

142

APPENDIX A

143

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A]. Day of the year (DOY) with corresponding date in which treatment
applications of ﬂoor management were performed.
A. Flaming
I I st 2116 1 3rd 1 4111 l 5111 6th 7th
2001
DOY 159 173 194 215
Date June 7 June 21 July 12 August 2
2002
DOY 142 159 179 193 207 250
Date May 21 June 7 June 27 July 11 July25 Septtgmber
2003
DOY 151 169 178 185 203 218 239
Date May 30 June 17 June 26 July 3 July 21 August 5 August 26
B. Tilling
I 1 st I 2nd I 3rd I 4m l 51h 61h
2001
DOY 160 171 202 219 250
Date June 8 June 19 July 20 August 6 September 6
2002
DOY 163 184 212 247
Date June 1 1 July 2 July 30 September 3
2003
DOY 144 169 191 199 218 239
Date May 23 June 17 July 9 July 17 August 5 August 26

 

 

 

 

 

 

 

 

144

 

 

Table A2. Carbon mineralization potential. Cumulative amount of C02-C (11g of C02-C
* g of soil") evolved during incubation in laboratory at 25°C and no light from soil of
different treatments. Values are means of 6 replicates. Analysis of variance was carried
out for each date. Values in the same row followed by different letter are signiﬁcantly
different for p50.05 (Tukey’s adjusted). MU=Mulch, FL=Flame, SS=Sandwich.

Year 2001 * (sampled April 13‘“)

 

 

Alley Row
Day MU FL SS MU FL SS
21 265 252 264 247 234 243
30 358 346 356 332 317 329
50 515 496 507 482 458 479
70 644 616 633 617 581 608
99 785 748 764 755 717 737
149 929 866 894 921 868 887

 

Year 2003 (sampled April 24th)

 

 

Alley Row
Day MU FL SS MU FL SS
21 333 312 320 250 195 248
30 452 418 435 339 270 331
50 655 588 614 518 391 466
77 924 791 859 774 558 645
100 1111 a 938 a 1029 a 943 a 678 b 788 ab
155 1479 a 1201 a 1371 a 1262 a 946 b 1070 ab

 

* Sampling and analysis courtesy of the CS Mott Chair of Sustainable Agriculture

145

Table A3. Nitrogen mineralization potential. Sample collected April 13th 2001 (*). Total
N (N03 + NH4) available in soil (mg Kg") after different days of incubation in laboratory
at 25°C and no light (0 — 10 cm depth) from different treatments. Values are means of six
replicates. Analysis of variance was carried out for each date and no signiﬁcant
differences were found for p50.05 (LSMEANS test). MU=Mulch, FL=Flame,

SS=Sandwich.

 

 

Alley Row
Day MU FL SS MU FL SS
0 14.48 14.70 16.58 35.80 42.22 37.58
30 44.30 44.40 46.20 72.77 75.20 72.22
70 76.99 83.66 76.95 101.74 109.40 96.30
150 100.62 106.80 100.35 118.51 126.84 139.13

 

 

 

 

(*) Sampling and analysis courtesy of the CS Mott Chair of Sustainable Agriculture

146

36m 3.? >0: :d 3.56 3.6 wm.ww 3.? mass NNdm omém wmdm FIBRE—am

 

 

 

 

 

3.3 a L .NN co. "N awd ooém Nm.w mmém wmdw mods NNém wadm NVSN “ESE
mud“ 3.2 3.2 $8 mmdm $.w 093 coda NYE. oodv wmﬁv $.mm £232
moom meow SON moom Noom SON Sam 88 Ben moom NOON Sam
:N :0 om :2

 

33 Eng 5 macoBscEBE

 

 

 

one mg mg 3.2 m2 who 43 a; 2: 5o 85 w; ﬁscﬁm
35 mg mg a: 2: Ed 8d a; a: mac Rd w; 682.2
So 25 3o :2 3; Ned 8N mi a: 18 85 ed 532
88 88 88 88 88 82.. 88 88 Sam 88 88 88

w: 6 u a

 

><3 “GDP—0Q a: muQOESﬁOhowv/H

 

.805 2 me 095% 5 ﬂ 38> comm @889? EoEomewE Soc P858 Hubbub have: much 5 .owEQE Sofm 28 .88 8 Sam
mama—<63“ E 85655 20-53-25 mo coupon 0628 05 Bow BEES .225 050mm. CO 352 eom 38:8 E0322 .v .< 033.

147

1: Monthly prec1p + Max temp - 0 - Min temp

30 - 250
200 ’"
A
v E
9... v
g 150 8
H 1 '5
g , .8
g i 100 -§
0 8
E— o.
r 50
* 0

 

 

 

 

 

 

 

 

 

 

 

Figure A.l. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural Experiment Station
during 2000. Data were collected by the Michigan Automated Weather Network.

[:1 Monthly prec1p + Max temp - 0 - Mrn temp

 

35 . 160
- 140
6 120 A
O
V 100 E
d) v
+5 8
8 30 .5
3 .§
8 F 60 .9
133 8
40 a:
» 20
, 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure A.2. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural Experiment Station
during 2001. Data were collected by the Michigan Automated Weather Network.

148

35 . :1 Monthly precip + Max temp - 0 - Min temp . '20

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

30 1
f"
6 20 1 80 E
. V
a. 15 a,
:5 I0 1 60 g
g 5 W E
O. 0 1- 40 8
a , ., a:
CD -5 L 5’ , a
E“ I. v ‘ 20
-10 .
-15 wall +77. +4 - ~l-x sums-L1 4M4»; M 1 l . 1 , o
o“ of” 60' a“ o” o” of” a“ c)" a)" a“ o“
\\ '\>\ “’9 °-\\ 5\\ b\\ «\\ %\ q\\ \ \ \\\ \ \

Figure A.3. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural Experiment Station
during 2002. Data were collected by the Michigan Automated Weather Network.

[:3 Monthly precip + Max temp - 0 - Min temp

 

 

 

 

 

35 . 140
1 120
,r - 100 E
U 1
2. v
o ‘ 80 c:
g 1 .9.
- H
8 L«50 33
8 j .9
O
E 40 e
I— ‘ CL.
°‘ 0' ‘0 ‘.' i 20
-l5 5 I . L 1
-20 ,p_,_-..r:rr +- +01 + 1 1+; +é T 1 1—1“ l-L +4 1-+ 4 17.1-10

 

 

 

 

 

 

 

66" 66” 8'" 8” 66" 8”
\SW \QO’ \QOI \60 \Q0 \Q9
\\ 0 «1 a .,\ 6

6°” 66" 66" 65’
0 '1' 0 0
\° 9 x“ \Q5
«\ a) 01 \c\

Figure A.4. Ten-day average for maximum and minimum daily temperatures of air (°C)
and monthly precipitation (mm) recorded at Clarksville Horticultural Experiment Station
during 2003. Data were collected by the Michigan Automated Weather Network.

149

\ /
>7
/

\
r” \

Flamed area —I
p

 

 

Vegetated
central strip

 

 

 

 

 

 

 

 

 

j >/ (J, y: Tilled strips
, '4 Ml? All“;

'51

 

21nd 2m ‘ ’

‘___ , H 0.6 rad—4}
‘ P 0 7 m 0.7 m
MULCH FLAME SANDWICH

Figure A.5. Diagram of the application of the orchard ﬂoor management systems to the
tree rows.

150

30 140

 

 

 

 

 

 

1:]Precipitation - 0 - MU +FL "O“SS
25 - 120
20 . _ . 2 100
3% p .'-. d(i: E
E __ 1:1" ' 80 g
in F“ ‘5
.5 15 _ .g .g
5 ° o ' F 60 .9
'— T r
‘2 10 , T _ H ‘g
40
5 [b [I .I] [1 I720
0 [:1 .ﬁ ﬁn Cl ﬂ 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

’9 '\ ‘1 ‘8 ‘ ’9 '\ ‘1 ‘1 ‘
riff 89° r e“ 583’ dig-tiff a“ as 60‘ saggy
2002 2003

Figure A.6. Monthly soil moisture average measured by TDR (% v/v) and precipitation
(mm) recorded at Clarksville Horticultural Experiment Station during 2002-2003.
MU=Mulch, FL=F lame; SS=Sandwich.

 

 

 

 

 

 

 

 

 

 

 

 

 

160
Ilrrigation
140 4. DPrecipitation —
120 ‘
100 ,.
E 80 .—

1 — r—- F'—
60 -

1
40 ~
20 4 I I

0 U. _J.j_ D 1 -.L_c, .._._L_.. . _ l_1 1_J __

 

swearsaaeaxr
Q X
\‘9 Q Vb Y. Y9 9°qu 06 6°46 of

Figure A.7. Monthly irrigation applied and precipitation (mm) occurred at Clarksville

Horticultural Experiment Station during 2001. Precipitation data was collected by an
automated weather station of the Michigan Automated Network.

151

140 - Ilnigation
DPrecipitation
120 .

100 7‘

80‘

(mm)
J

60 ~

40 ,.

20 if H L [I
0 if, D07 I _ Y I l '7" " T T Y ' ' ‘l

9,5 <59 «56’ Y'Q& 64" 9“" 5°“ 88” V“ 3" ”'5‘ ‘9‘
6 ,8»:
\§ Q0 ‘H V. f 0" +0 f
Figure A. 8. Monthly irrigation applied and precipitation (mm) at Clarksville

Horticultural Experiment Station during 2002. Precipitation data was collected by an
automated weather station of the Michigan Automated Network.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

160 -
140 . Ilrrigation
DPrecipitation
120 . _'
100 f
E 80
60
40 .
0 . 121.. E1, . [:l- ._l_.,. ,T a, If 4 a
$651 56“? $10 v? ‘56) 365"” 3&7 ’09:? 8‘63? 0&5)“

Figure A. 9. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2003. Precipitation data was collected by an
automated weather station of the Michigan Automated Network.

152

 

APPENDIX B

153

Table B. 1. Day of the year (DOY) with corresponding date in which treatment
applications of ﬂoor management were performed.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A. Flaming
2001
DOY 159 173 194 221
Date June 7 June 21 July 12 August 8
2002
DOY 142 159 193 208 250
Date May 21 June 7 July 11 July26 September 6
2003
DOY 151 169 178 185 203 218 239
Date May 30 June 17 June 26 July 3 July 21 August 5 August 26
B. Tilling
I 1 51 I 2nd I 3rd I 4111 i 5H1 61h
2001
DOY 160 171 202 221 250
Date June 8 June 19 July 20 August 8 September 6
2002
DOY 163 183 212 221 247 262
Date June 11 July 1 July 30 August 8 September 3 September 18
2003
DOY 151 169 198 233
Date May 30 June 17 July 16 August 20

 

 

 

 

 

 

 

 

Table 8.2. Nitrogen content (% dry weight) in alfalfa hay at moment of application to
boxes with Mulch.

 

 

 

Sample 2001 2003
A 2.04 2.04
B 2.23 2.13
C 2.25 2.35
Average 2.18 2.17

 

154

 

 

Table 3.3. Ratio of C:N in alfalfa hay mulch for two dates in 2003 and for two different
layers according to time of application.

 

Time of application

 

 

 

Date of sampling Aug-2002 Jun-2003

June 2003 15.02 17.63

November 2003 13.60 15.67
Average by age 14.31 16.65

 

155

 

W: v.2 mom 5.x. Non Wm ﬂ: mow moo QNm ﬁwm were 8.50 8o.

 

 

 

 

 

 

 

 

 

 

v.3 Q: wém ms ado Qw ode Woo o.ow w.wm o.mv v.3” 8.60 ca
md— o.m _ mom M: 0.3 VS mac o. EL Non Qov Ndv o.mm £03628
3: m4: vow 9w m.mo_ ﬁn v.3 mno— won NS» 13‘ won oEaE
ed o 0.2 m. 5 En odm ms v.3 mom Emu mow mow mom £232
mooN NooN Sow moon Noom SoN moom moom Sow mooN NooN Sou
:N so um 52
33 Eng 5 3552822
m
mmo omo mmo m: ooo woo ~o.~ hm.“ mo; too So 2o 850 8%
Go Go amo mm; 5: $6 SN no; on; and So 26 .850 .53
wmo omo mmo em; mm; woo Ed on." ow; avo nmo m _ .o sogwgm
omo omo NNo mm; m 2 mod SN R4 we; :3 36 m mo 082m
end 36 5o co." oo.— Nod om.~ :.N 34 "No So m _.o 5232
moon Noon Sow. moom mooN Som moom moom Sow mooN Noom Sow
m2 8 x a

 

33 :5ch 5 3532882

<

.82“ No me owmeozw 5 mm 33> scam $8893 Eon—ommcwﬁ Soc @8280 EEEoE been much E .moom 8 Som
63920728 E 8:285 20.33-05 oo newton 22:8 05 Bob BEES .EmO PAS—03m. mo 82*: com 88:8 H5532 .v .m 2an

156

250 1

; Ilrrigation

' DPrecipitation
200 3

l

l
150 ~

(mm)
1

1

100
50 ‘* I:I
0 I D‘j‘” D r " "f —‘* ""' r ‘T r— , . Uﬁﬂ i

*9 '\
{590$ ‘o‘sﬁ ~99 Y'Qo 445%
\ (‘0

 

 

 

 

 

 

 

8°°° 9% s4? 9" 36‘ <- as
Y’ 3" 0° 04" a
‘9 Q 0
Figure 3.1. Monthly irrigation applied and precipitation (mm) at Clarksville

Horticultural Experiment Station during 2001. Precipitation data was collected by an
automated weather station of the Michigan Automated Network.

 

 

 

 

250
lln'igation
Cl Precipitation
200 l
l
150
100 1 _
l _
SOJ ‘ n I]
:1: DH _t__i, ‘ to ,D
~9 '\ A o \
9&6 ‘oéﬁ ‘95» ‘30 ‘9 \so x&‘\ a»? 6&0" ‘50” 06.96 66106
\‘D' Q6 Y” $- 00 $04 06;

Figure 8.2. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2002. Precipitation data was collected by an
automated weather station of the Michigan Automated Network.

157

250 ,1

. Ilrrigation
200 i DPrecipitation
l
150 5
100 3
1
50 ‘
0 li a D [:1 , EU . H at _,T

V
is $6 $6
9‘3 0

(mm)
]

 

 

 

 

‘9 '\ 0

Figure 3.3. Monthly irrigation applied and precipitation (mm) at Clarksville
Horticultural Experiment Station during 2003. Precipitation data was collected by an
automated weather station of the Michigan Automated Network.

158

 

APPENDIX C

159

250 -

 

 

 

 

 

 

I Irrigation
D Precipitation
200 ‘
150 __
a i r
5 l
100 -
50 ~
I M
0 '* Vﬂ'li 7' ’i ‘J ’T’ ’ I T liﬁ

 

6 6° *9 «1% 9° \‘3 ‘ ~06 ~05 ‘5‘ 6"
$5 S 0 \Q ‘0
\‘9 Qapf ‘5» Y9 ‘5 8 $6, «9 00x9 eodéé‘ of?

Figure 01. Irrigation and total precipitation (mm) recorded monthly at Clarksville
Horticultural Experiment Station during 2001. Data was collected by an automated
weather station of the Michigan Automated Network.

250 ;
Ilrrigation

DPrecipitation

200 i

150 -

E .

5 l

l

100 i

 

 

 

 

 

50 i
0. ﬂﬂ

'\ \
ff yrs Varese.» area a .6
‘5 ﬁg? é

 

Figure C.2 Irrigation and total precipitation (mm) recorded monthly at Clarksville

Horticultural Experiment Station during 2002. Data was collected by an automated
weather station of the Michigan Automated Network.

160

(mm)
]

250 ~
1 Ilrrigation
200 l DPrecipitation
\
150 l
l
100 7“
50 .3 I

.\ 0 N p p
Q0 ‘5‘5‘ 85} $9 é}, ‘0
Y” o 69 0
a“ Y' 43‘ 0

 

 

 

Figure C.3. Irrigation and total precipitation (mm) recorded monthly at Clarksville
Horticultural Experiment Station during 2003. Data was collected by an automated
weather station of the Michigan Automated Network.

 

3500
,\ 3000 + .
if 2500 - ° ,
.§> 2000 -
E; 1500 .
o 1000 ~
.E 500
0

 

 

 

0.7 0.8 0.9 1 1.1 1.2

Root:Shoot ratio

Figure C. 4. Regression between tree total dry matter weight (g) and root:shoot ratio.

y = - 3329.7 x = 5643.2 R2 = 0.58

161

APPENDIX D

162

Table D. 1. Field experiment. Day of the year (DOY) with corresponding date in which
treatment applications were performed.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A. Flaming
I 1 st 2nd I 3rd I 4th I 5th I 6m 7th
2001
DOY 159 173 194 215
Date June 7 June 21 July 12 August 2
2002
DOY 142 159 179 193 207 250
Date May 21 June 7 June 27 July 11 July25 September 6
2003
DOY 151 169 178 185 203 218 239
Date May 30 June 17 June 26 July 3 July 21 August 5 August 26
B. Tilling
I 15! I 2nd I 3rd I 4th I 5th 6th
2001
DOY 160 171 202 219 250
Date June 8 June 19 July 20 August 6 September 6
2002
DOY 163 184 212 247
Date June 11 July 2 July 30 September 3
2003
DOY 144 169 191 199 218 239
Date May 23 June 17 July 9 July 17 August 5 August 26

 

 

 

 

 

 

 

 

163

 

 

Table D2. Container experiment. Day of the year (DOY) with corresponding date
which treatment applications were performed.

in

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A. Flaming
I 151 1 2nd I 3rd I 4th I SEE I 6th I 7th
2001
DOY 159 173 194 221
Date June 7 June 21 July 12 August 8
2002
DOY 142 159 193 208 250
Date May 21 June 7 July 11 July26 September 6
2003
DOY 151 169 178 185 203 218 239
Date May 30 June 17 June 26 July 3 July 21 August 5 August 26
B. Tilling
If lsl I 2nd I 3rd I 4th 1 5111 I 6th
2001
DOY 160 171 202 221 250
Date June 8 June 19 July 20 August 8 September 6
2002
DOY 163 183 212 221 247 262
Date June 11 July 1 July 30 August 8 September 3 September 18
2003
DOY 151 169 198 233
Date May 30 June 17 July 16 August 20

 

 

 

 

 

 

 

 

 

164

Table D.3. Field experiment. Composition and density of vegetative cover. List of species
and their occurrence (plants/m2) for each date of survey as identiﬁed in Mulch treatment.

Date of survey
NOV APR OCT MAY NOV

 

 

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003
annual bluegrass Poa annua L. 0.2 0.0 0.0 0.0 0.0
black nightshade Solanum nigrum L. 0.2 0.0 0.0 0.0 0.0
buckhom Plantago lanceolata L. 0.1 0.0 0.0 0.0 0.0
Canada bluegrass Poa compressa L. 0.0 0.0 0.0 0.1 0.0
chickweed “35%?“ media (L) 5.3 1.4 0.8 0.5 0.4
common sowthistle Sonchus oleraceus L. 0.0 0.0 0.0 0.1 0.0
crabgrass gig?“ “mama” (L') 8.3 0.0 1.4 0.0 0.0
curly dock Rumex crispus L. 0.2 0.2 0.0 0.0 0.0
dandelion xggm’" 0177“"“19 0.2 0.2 0.0 0.1 0.0
downey brome Bromus tectorum L. 0.0 0.0 0.0 1.3 0.0
ﬁeld pennycress Nilaspi arvense L. 0.0 0.0 0.0 0.1 0.0
grass (not identiﬁed) 0.6 0.0 0.0 0.2 0.0
ground ivy Glechoma hederacea L. 0.0 0.0 0.0 0.7 0.0
hairy bittercress Cardamine hirsuta L. 0.7 0.0 0.0 0.0 0.0
henbit Lamium aplexicaule L. 0.9 0.2 0.4 0.0 0.8
horseweed €333“ cam’dm“ (L) 0.0 0.0 0.0 0.1 0.0
lambsquarters Chenopodium album L. 0.1 5.1 0.0 0.6 0.0
orchardgrass Dactylis glomerata 1.6 0.8 0.1 0.1 0.1
quackgrass Elymus repens (L.) Gould 0.0 0.0 4.0 15.4 0.9
red clover Tnfolium pratense L. 1.3 0.7 0.0 0.0 0.0
red sorrel Rumex acetosella L. 0.2 0.0 0.0 0.0 0.0
rye-grass Lolium perenne L. 0.2 0.0 0.0 0.0 0.0
wild carrot Daucus carota L. 0.1 0.1 0.0 0.3 0.0
TOTAL plants (number / m’) 20.1 8.7 6.7 19.4 2.1
N° Species 16 8 5 l3 4

 

165

Table D.4 Field experiment. Composition and density of vegetative cover. List of species
and their occurrence (plants/m2) for each date of survey as identiﬁed in Flame treatment.

Date of survey

 

NOV APR OCT MAY NOV

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003
annual bluegrass Poa annua 0.9 3.9 1.6 12.9 19.3
bald brome Bromus racemosus L. 0.0 0.0 0.0 0.1 0.0
black nightshade Solanum nigrum L. 0.1 0.0 0.0 0.0 0.0
broad leaf (not identiﬁed) 0.0 0.0 0.0 0.0 9.8
broadleaf dock Rumex obtusifolius L. 0.0 0.0 0.3 0.0 0.0
buckhom Plantago lanceolata L. 0.2 0.4 0.0 0.0 0.1
bull thistle $22M W’gare (33“) 0.0 0.0 0.2 0.0 0.0
Canadian thistle Cirsium arvense (L) Scop. 0.0 0.0 0.0 0.1 0.0
chicory Cichorium intybus L. 0.5 0.0 0.0 0.0 0.0
chickweed Stellaria media (L.) Cyrillo 20.2 46.2 80.4 52.8 47.2
common sowthistle Sonchus oleraceus L. 0.0 0.0 0.0 0.3 0.0
corn chamomile Anthemis arvensis L. 0.0 0.0 0.0 0.7 0.0
crabgrass géﬁgfma sanguinalis (L') 9.4 0.1 1.7 0.0 1.4
curly dock Rumex crispus L. 2.0 4.2 3.3 1.4 0.3
dandelion aggw’" ”mam"? 1.6 2.9 4.2 4.5 22.9
downey brome Bromus tectorum L. 0.0 0.0 0.0 2.3 0.0
forget-me-not Myosotis arvensis (L.) Hill 0.0 0.3 0.0 0.0 0.0
grass (not identiﬁed) 2.9 0.0 0.0 0.1 6.7
green foxtail gi’gi‘ggw“ (L) 0.0 0.0 0.0 0.0 0.5
ground ivy Glechoma hederacea L. 0.0 0.0 0.0 0.2 0.0
hairy bittercress Cardamine hirsuta L. 14.7 5.1 3.7 33.5 37.4
henbit Lamium aplexicaule L. 1.2 4.3 3.3 13.5 29.4
horseweed Egg" madm“ (L.) 0.2 0.1 1.0 134.5 0.4
Italian ryegrass Lolium multzﬂorum Lam. 0.0 0.6 0.0 0.0 0.0
Kentucky bluegrass Poa pratensis L. 0.0 0.0 0.0 0.2 0.0
lambsquarters Chenopodium album L. 0.2 20.5 0.0 2.1 0.0
marsh yellowcress Rorippa palustris L. 0.0 0.0 0.1 0.0 0.0

166

 

mayweed chamomille Anthemis cotula L. 0.2 0.0 0.0 0.0 0.0
oldﬁeld cinquefoil Potentilla simplex Michx. 0.2 0.2 0.0 0.1 0.0
orchardgrass Dactylis glomerata L. 5.1 2.0 1.0 1.0 1.0
prickly lettuce Lactuca scariola L. 0.0 0.0 0.0 0.0 0.2
purslane Portulaca oleracea L. 0.0 0.0 0.2 0.6 6.9
purslane speedwell Veronica peregrina L. 0.0 0.0 0.0 6.7 3.7
quackgrass Elymus repens (L.) Gould 0.0 0.0 0.6 0.7 0.0
red clover Trifolium pratense L. 1.0 2.5 0.1 0.2 0.0
red sorrel Rumex acetosella L. 0.6 0.2 1.2 3.4 1.8
redroot pigweed Amaranthus retroﬂexus L. 0.0 0.0 0.3 0.0 0.0
roughstalk bluegrass Poa trivialis L. 0.0 0.0 0.0 0.3 0.0
rye-grass Lolium perenne L. 0.0 0.0 0.9 0.2 0.0
sheperd's purse (Cl‘agﬁléaﬁléursa-pastoris 0.9 1.0 0.0 1.3 0.9
white campion Lychnis alba Mill. 0.0 0.2 0.0 0.1 0.0
white clover T nfolium repens L. 0.0 0.0 0.0 0.7 0.1
wild carrot Daucus carota L. 0.4 0.2 0.0 2.9 0.2
wild oat Avenafatua L. 0.0 0.0 0.1 0.1 0.0
wild strawberry gliiirsfemgmiam 0.1 0.5 0.1 0.0 0.0
yellow rocket Barbarea vulgaris L. 0.0 0.0 0.2 0.0 0.0
yellow wood sorrel Oxalis stricta L. 0.0 0.0 0.0 0.2 0.0
TOTAL Plants (“umber / m’) 62.1 95.2 104.2 277.5 190.1
N0 Species 21 20 21 31 20

 

167

Table D.5. Field experiment. Composition and density of vegetative cover. List of species
and their occurrence (plants/m2) for each date of survey as identiﬁed in Sandwich

treatment.

Date of survey

 

NOV APR OCT MAY NOV

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003
annual bluegrass Poa annua L. 1.7 4.8 2.3 15.1 25.4
bell bean Viciafava L. 7.5 0.0 0.0 0.0 0.0
black medic Medicago lupulina 1. 0.0 0.0 0.1 0.0 0.0
black nightshade Solanum nigrum L. 0.0 0.0 0.6 0.0 0.0
blackberry Rubus sp. L. 0.0 0.0 0.1 0.1 0.0
broad leaf (not identiﬁed) 0.0 0.0 0.2 0.0 0.2
buckhom Plantago lanceolata L. 0.1 0.2 0.5 0.6 2.4
bull thistle $2?” ”“15”"? (33“) 0.1 0.0 0.0 0.0 0.0
Canada bluegrass Poa compressa L. 0.0 0.0 0.0 0.6 0.0
chicory Cichorium imybus L. 0.2 0.0 0.0 0.0 0.0
chickweed “3553:“ media (L) 34.0 55.8 123.0 52.7 61.2
common sowthistle Sonchus oleraceus L. 0.0 0.0 0.1 1.3 0.0
comspeedwell Veronica arvensis L. 0.0 0.0 0.0 1.1 0.0
crabgrass giggaria sanguinalis (L) 3.6 0.0 2.3 0.2 0.1
curly dock Rumex crispus L. 0.6 2.1 1.0 1.1 0.0
dandelion xggw’" ”memale 0.5 0.3 2.3 1.5 3.1
downey brome Bromus tectorum L. 0.0 0.0 0.0 8.9 0.0
ﬁeld bindweed Convolvulus arvensis L. 0.0 0.2 0.0 0.0 0.0
ﬂeabane rough Erigeron strigosus Muhl. 0.0 0.0 0.1 0.0 0.0
grass (not identiﬁed) 1.0 0.0 0.0 0.4 16.5
green foxtail :33: grid“ “9 0.0 0.0 0.0 0.0 1.0
ground spurge Euphorbia prostrata Ait. 0.0 0.0 0.2 0.0 0.0
hairy bittercress Cardamine hirsuta L. 18.3 31.6 18.6 10.0 123.6
hawkweed Hieracium sp. L. 0.0 0.0 0.0 0.3 0.0
hedge mustard gfgzb'iu’" ”mama“? (1") 0.0 0.0 0.0 0.1 0.0
henbit Lamium aplexicaule L. 19.3 23.7 14.4 12.2 39.6
horseweed CW2" “madmjs (L') 0.0 0.0 0.9 166.0 34.2

Cronq.

168

 

Kentucky bluegrass Poa pratensis L. 0.0 0.0 0.0 1.2 0.0
lambsquarters Chenopodium album L. 0.2 14.2 0.1 0.4 0.0
oilseed radish Raphanus sativus L. 9.4 0.0 0.0 0.0 0.0
oldﬁeld cinquefoil Potentilla simplex Michx. 0.0 0.0 0.0 0.3 0.3
orchardgrass Dactylis glomerata L. 1.0 0.4 0.7 0.2 2.7
pineapple weed nggaggnﬁmmmdes 0.0 0.0 0.0 3.9 0.0
prickly lettuce Lactuca scariola L. 0.0 0.0 0.1 0.0 0.0
purslane Portulaca oleracea L. 0.0 0.0 0.2 0.7 0.2
purslane speedwell Veronica peregrina L. 0.0 0.0 0.0 2.5 0.0
quackgrass Elymus repens (L.) Gould 0.0 0.0 0.2 4.3 1.5
red clover T rifolium pratense L. 1.7 1.3 2.1 1.2 1.7
red sorrel Rumex acetosella L. 0.2 0.0 50.2 92.1 40.3
redroot pigweed Amaranthus retroﬂexus L. 0.0 0.0 0.8 0.0 0.0
rough cinquefoil Polentilla norwegica L. 0.0 0.0 0.2 0.0 0.0
rye-grass Lolium perenne L. 0.0 0.2 1.5 0.2 0.0
sheperd's purse grzgﬁlégigmsa-pmtoris 0.6 1.1 0.0 3.8 0.2
silvery cinquefoil Potentilla argentea L. 0.0 0.0 0.2 0.0 0.0
slender rush Juncus tenuis Willd. 0.0 0.0 0.1 0.0 0.0
smooth brome Bromus inermis Leyss. 0.0 0.0 0.0 0.3 0.0
velvet leaf ”13:33” ’heol’hm“ 0.0 0.0 0.0 0.1 0.0
white campion Lychnis alba Mill. 0.0 0.9 0.1 0.2 0.0
white clover T rifolium repens L. 0.0 0.0 3.0 14.4 1.1
wild carrot Daucus carota L. 0.1 1.0 1.1 45.3 11.0
wild oat Avenafatua L. 0.0 0.0 0.6 0.2 0.0
wild strawberry gaggewrgmam 0.0 0.2 0.0 0.0 0.1
yellow hawkweed 531:?“ c““”""‘“’" 0.0 0.0 0.3 0.2 0.0
yellow rocket Barbarea vulgaris L. 0.0 0.0 0.3 0.1 0.0
yellow wood sorrel Oxalis stricta L. 0.0 0.0 0.2 1.7 0.2
TOTAL plants (number / m’) 99.9 137.7 228.7 445.3 366.4
N° Species 19 16 36 38 22

 

169

Table D6. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in Mulch
treatment.

Date of survey

 

NOV APR OCT MAY NOV

 

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003
annual bluegrass Poa annua L. . 0.0 0.0 0.7 0.2
buckhom Plantago lanceolata L. x 0.0 0.0 0.0 0.0
chickweed Stellaria media (L.) Cyrillo x 0.0 8.9 13.5 1.3
dandelion gggw’" ”ﬂaw"? x 0.0 0.0 0.2 0.0
downey brome Bromus tectorum L. . 0.0 0.2 0.4 0.0
grass (not identiﬁed) . 0.0 0.9 0.0 0.0
hairy bittercress Cardamine hirsuta L. x 0.0 0.0 0.0 0.0
henbit Lamium aplexicaule L. x 0.0 11.7 2.6 1.5
horseweed €332“ “Mdems (L.) 0.0 0.0 6.9 0.0
lambsquarters Chenopodium album L. . 0.2 0.0 0.4 0.0
quackgrass Elymus repens (L. ) Gould . 0.0 0.0 0.9 0.0
red clover T rifolium pratense L. x 0.0 0.0 0.0 0.2
red sorrel Rumex acetosella L. x 0.0 12.8 3.9 8.0
redroot pigweed Amaranthus retroﬂexus L. . 0.0 0.0 4.6 0.0
rye-grass Lolium perenne L. x 0.0 0.0 0.0 0.0
sheperd's purse ((Ezgﬁlégizursa-pastoris 0.0 0.0 0.4 0.0
white clover T rifolium repens L. . 0.0 0.7 1.9 0.0
wild carrot Daucus carota L. x 0.0 0.0 1.3 0.0
TOTAL plants (number/m2) . 0.2 35.2 37.6 11.1
N° Species 9 1 6 13 5

 

x = species present but number of plants not counted

170

Table D7. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in Flame

treatment.

Date of survey

 

APR OCT MAY NOV
2002 2002 2003 2003

 

0.0 0.0 5.6 5.8
0.6 0.6 0.0 0.0
83.3 90.3 28.6 5.3
0.0 0.8 0.0 0.0

1.9 2.2 1.4 0.8
0.0 0.0 2.5 0.0
0.0 0.0 0.0 1.9
0.0 508.1 3.9 90.0
10.0 58.3 0.8 0.0
0.0 0.0 600.8 3.3

0.6 0.0 0.0 0.0
0.0 0.0 0.0 43.6

0.0 0.0 0.0 0.6
0.0 0.0 1.7 0.0

0.0 6.4 5.3 5.8
1.4 0.0 0.0 0.0
0.8 1.9 3.9 0.6
0.6 1.4 0.6 0.0
1.9 0.0 11.9 3.3

0.0 0.0 1.7 0.3
0.0 0.0 3.3 0.0
0.0 0.0 17.8 0.8

 

NOV
Common name Scientiﬁc name 2001
annual bluegrass Poa annua L.
buckhom Plantago lanceolata L.
chickweed Stellaria media (L.) Cyrillo x
crabgrass géggarta sangumalzs (L.)
dandelion Taraxacum oﬁicinale Weber x
downey brome Bromus tectorum L.
grass (not identiﬁed)
hairy bittercress Cardamine hirsuta L.
henbit Lamium aplexicaule L. x
horseweed 80:3? canadenszs (L.)
lambsquarters Chenopodium album L.
xxﬁuc Anthemis cotula L.
332:5;“1 Cerastium vulgalum L.
53:13:11 Veronica peregrina L.
quackgrass Elymus repens (L.) Gould
red clover T rifolium pratense L. x
red sorrel Rumex acetosella L.
rye-grass Lolium perenne L. x
sheperd's purse 54:13?“ bursa—pastorls (L.) x
white campion Lychnis alba Mill.
white clover T rifolium repens L.
wild carrot Daucus carota L.
TOTAL plants (number/m2)
N° Species 6

101.1 670.0 689.7 162.2
9 9 15 13

 

x = species present but number of plants not

counted

171

Table D8. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in Sandwich
treatment.

Date of survey

 

NOV APR OCT MAY NOV

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003
annual bluegrass Poa annua L. 1.2 0.0 7.8 12.8
bell bean Viciafava L. 0.0 84.2 0.0 0.0
black nightshade Solanum nigrum L. 0.0 2.0 9.5 0.0
broadleaf (not identiﬁed) 0.0 1.0 0.0 0.0
buckhom Plantago lanceolata L. 0.8 1.5 0.0 2.3
chickweed Stellaria media (L.) Cyrillo x 61.8 32.7 78.7 134.5
crabgrass 33:?” WNW,“ (L.) 0.0 1.3 0.0 0.3
curly dock Rumex crispus L. 0.0 0.0 0.2 0.2
dandelion T araxacum oﬂicinale Weber x 0.7 1.8 2.7 1.5
downey brome Bromus tectorum L. 0.0 0.0 6.2 0.0
grass (not identiﬁed) 0.0 0.0 0.2 9.0
green foxtail Setaria viridis (L.) Beauvois 0.0 1.0 0.0 0.7
hairy bittercress Cardamine hirsuta L. 2.2 0.0 1.0 11 1.3
henbit Lamium aplexicaule L. x 7.5 22.2 3.8 17.0
horseweed 53:31: T! canadensis (L') 0.0 6.7 315.0 45.2
Italian ryegrass Lolium multiﬂorum Lam. 0.0 0.0 0.0 0.8
lambsquarters Chenopodium album L. 0.5 0.2 0.0 0.0
xxﬁlle Anthemis cotula L. 0.0 0.0 0.0 3.8
mousear chickweed Cerastium vulgatum L. 0-0 0~0 0-0 0-5
oldﬁeld cinquefoil Potentilla simplex Michx. 0-0 0-7 0-0 0-5
prickly lettuce Lactuca scariola L. 0.0 0.0 0.5 0.0
purslane Portulaca oleracea L. 0.0 2.0 0.0 0.0
purslane speedwell Veronica peregrina L. 0.0 0.0 2.0 0.0
quackgrass Elymus repens (L.) Gould 0.0 0.0 1.2 4.7
red clover T rifolium pratense L. x 1.7 1.2 4.8 3.0
red sorrel Rumex acetosella L. 1.3 114.8 210.5 67.5
redroot pigweed Amaranthus retroﬂexus L. 0.0 0.2 0.0 0.0
roughstalk bluegrass Poa trivialis L. 0.0 0.0 0.0 0.7

172

rye-grass Lolium perenne L. x 0.8 1.8 5.3 4.8

sheperd's purse camel!“ b“"“‘p"“o"s(L') x 5.8 0.0 12.0 0.2

 

Medic.
slender rush Juncus tenuis Willd. . 0.0 0.3 0.0 0.0
white clover T rifolium repens L. . 0.0 3.7 14.0 3.3
wild carrot Daucus carota L. . 0.2 0.8 25.5 4.2
wild oat Avenafatua L. . 0.0 0.0 2.2 0.0
wild strawberry 11;; :ﬁgﬁ'f’gmma 0.2 0.0 0.0 0.0
yellow foxtail fledgbla Iutescens (Weigel) 0.0 0.5 0.0 0.3
yellow rocket Barbarea vulgaris L. . 0.0 0.0 4.5 0.0
yellow wood sorrel Oxalis stricta L. . 0.0 0.5 0.0 0.2
TOTAL plants (number/m2) . 84.7 281.0 707.5 429.3
N° Species 6 13 22 21 25

 

x = species present but number of plants not counted

Table D9. Container experiment. Composition and density of vegetative cover. List of

I I 2 D O O 0
spec1es and then occurrence (plants/m ) for each date of survey as identiﬁed 1n Partial
Cover treatment.

Date of survey

 

 

NOV APR OCT MAY NOV

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003
annual bluegrass Poa annua L. . 0.0 0.0 48.1 19.6
black nightshade Solanum nigrum L. . 0.0 0.4 11.5 0.0
buckhom Plantago lanceolata L. x 0.9 0.6 0.0 1.1
chickweed Stellaria media (L.) Cyrillo x 21.7 14.6 98.1 127.8
crabgrass gig?“ sangumm (L.) 0.0 3.3 0.0 0.0
dandelion Taraxacum oﬂicinale Weber . 0.4 0.4 1.1 0.4
downey brome Bromus tectorum L. . 0.0 0.2 0.7 0.0
grass not identiﬁed . 0.0 1.3 0.0 0.0
green foxtail Setaria viridis (L.) Beauvois . 0.0 1.1 0.0 1.7
hairy bittercress Cardamine hirsuta L. . 0.0 7.4 1.9 87.4

173

 

henbit Lamium aplexicaule L. x 6.5 1.9 41.7 12.6
horseweed 533:3. “Mdems (L) 0.0 3.9 38.9 0.0
Italian ryegrass Lolium multiﬂorum Lam. 0.0 0.0 0.0 0.6
Kentucky bluegrass Poa pratensis L. 0.0 0.0 0.6 0.0
lambsquarters Chenopodium album L. 6.3 0.7 2.2 0.0
mayweed Anthemis cotula L. 0.0 0.0 0.0 16.9
chamomllle
mousear chickweed Cerastium vulgatum L. 0.0 0.0 6.9 0.7
oldﬁeld cinquefoil Potentilla simplex Michx. 0.0 0.7 0.4 0.4
purslane Portulaca oleracea L. 0.4 0.0 0.0 0.0
purslane speedwell Veronica peregrina L. 0.0 0.0 8.3 0.2
quackgrass Elymus repens (L.) Gould 0.0 0.0 19.1 19.6
red clover T rifolium pratense x 23.0 9.1 3.7 5.2
red sorrel Rumex acetosella L. x 1.5 31.5 72.4 39.6
{)‘l’ughsmlk Poa trivialis L. 0.0 0.0 0.0 0.7
uegrass
rye-grass Lolium perenne L. x 18.9 10.7 17.6 1.5
sheperd's purse 311$?" bursa’pas’oris (L) x 4.6 0.0 17.2 2.2
silvery cinquefoil Potentilla argentea L. 0.0 0.0 0.2 0.0
white campion Lychnis alba Mill. 0.0 0.0 1.1 0.0
white clover T rifolium repens L. 0.0 0.0 31.1 0.2
wild carrot Daucus carota L. x 0.9 0.7 64.8 3.9
wild oat Avenafatua L. 0.0 1.9 4.6 0.0
wild strawberry gighirsﬁ’emgmana 0.2 0.0 0.4 0.0
yellow foxtail 15186:? ’“tesce’” (weigel) 0.0 3.1 0.0 1.1
yellow wood sorrel Oxalis stricta L. 0.0 0.6 0.0 0.0
TOTAL plants (number/m2) 85.2 94.1 492.6 343.3
N° Species 8 12 20 24 21

 

x = species present but number of plants not counted

174

Table D.10. Container experiment. Composition and density of vegetative cover. List of
species and their occurrence (plants/m2) for each date of survey as identiﬁed in Total
Cover treatment.

Date of survey
NOV APR OCT MAY NOV

 

 

Common name Scientiﬁc name 2001 2002 2002 2003 2003

annual bluegrass Poa annua L. . 0.0 0.0 6.7 0.3

broadleaf (not identiﬁed) . 0.0 0.3 0.5 0.0

buckhom Plantago lanceolata L. x 0.0 0.5 0.0 0.3
Stellaria media (L.)

chickweed Cyrillo x 0.2 30.0 17.5 1 1.5

crabgrass Digitaria sangu malts 0") 0.0 4.0 0.0 2.5
Scop.

dandelion “mam" 0177c‘”“16 0.5 1.2 3.0 1.2
Weber x

downey brome Bromus tectorum L. . 0.0 0.0 0.7 0.0

grass (not identiﬁed) . 0.0 0.2 0.0 0.7

Setaria viridis (L.)

green foxtail 0.0 2.8 0.0 0.2

Beauvois
hairy bittercress Cardamine hirsuta L. . 0.0 13.3 0.5 0.0
henbit Lamium aplexicaule L. . 0.5 3.0 5.8 40.0

Conyza canadensis (L.)

 

horseweed 0.0 0.2 62.3 28.5
Cronq.
Italian ryegrass Lolium multiﬂorum Lam. . 0.0 0.0 0.0 1.5
lambsquarters Chenopodium album L. . 2.8 1.0 0.8 0.0
oldﬁeld cinquefoil Potentilla simplex Michx. . 0.0 0.0 0.2 0.3
purslane speedwell Veronica peregrina L. . 0.0 0.0 2.8 0.0
quackgrass Elymus repens (L.) Gould . 0.0 1.8 12.2 9.2
red clover T rifolium pratense L. x 33.3 16.5 6.3 8.7
red sorrel Rumex acetosella L. x 1.2 44.2 73.5 25.5
roughstalk bluegrass Poa trivialis L. . 0.0 0.0 0.0 0.7
rye-grass Lolium perenne L. x 15.0 18.2 7.5 1.8
sheperd's purse fifﬁ’;£f”salmmm 0.0 0.0 3 .8 0.7
slender rush Juncus tenuis Willd. . 0.0 0.2 0.0 0.0
white campion Lychnis alba Mill. . 0.0 0.0 3.5 0.0
white clover T rifolium repens L. . 0.0 0.0 28.3 0.0
wild carrot Daucus carota L. x 0.5 0.3 3.8 1.8
wild oat Avenafatua L. . 0.0 0.0 5.3 0.0
wild strawberry Fragaria virginiana Duch . 0.5 0.0 0.0 0.0
yellow foxtail fveé‘gééuﬁgim . 0.0 2.2 0.0 2.3
TOTAL plants (number/m2) . 54.5 139.8 245.2 137.7
N° Species 7 9 18 20 19

 

x = species present but number of plants not counted

175

l1111111111111llll