'"T'a sun was»: IES

:_ m U) llllllllllllllllllllllll'lllllllllllllllllll

3 1293 014052

 

This is to certify that the

thesis entitled

The Effects of Habitat Manipulation on Vegetation
Characteristics and Avian Communities on Conservation
Reserve Program Fields in Gratiot County, Michigan

presented by

Alison Jane Pearks

has been accepted towards fulfillment
of the requirements for

MaSter Of Science degree in Fish. & Wildl.

5m CDWL‘JIC;
Major professor

Date i/‘II‘IQ

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University l

 

 

 

 

 

PLACE N RETURN BOX to romovothlo ohookout from your rooord.
TO AVOID FINES Mum on or bdoro duo duo.

nutmeg} DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ 3 . 5 l j-
:LJ-
fiF—Ifi

MSU In An Afflrmotivo Action/Equal Opportunity Institution
. Wanna-m

 

 

 

 

THE EFFECTS OF HABITAT MANIPULATION ON VEGETATION
CHARACTERISTICS AND AVIAN COMMUNITIES ON CONSERVATION
RESERVE PROGRAM FIELDS IN GRATIOT COUNTY, MICHIGAN

By

Alison Jane Pearks

A THESIS

Submitted to
Michi an State University
in partial ful llment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Fishgries and Wildlife
an
Ecology and Evolutionary Biology Program

1995

ABSTRACT

THE EFFECTS OF HABITAT MANIPULATION ON VEGETATION
CHARACTERISTICS AND AVIAN COMMUNITIES ON CONSERVATION
RESERVE PROGRAM FIELDS IN GRATIOT COUNTY, MICHIGAN

By

Alison Jane Pearks

Avian populations and vegetation characteristics were examined on 18
Conservation Reserve Program (CRP) fields subjected to varying mowing
treatments. Six fields had a 4-year history of complete mowing in late summer, 6
fields had not been previously manipulated and were mowed in late summer in
strips totaling 1/3rd of the field acreage; and 6 fields had no history of mowing
and served as controls. Manipulated fields were characterized by low live
vegetation height and live canopy; strip mowed fields tended towards grass
canopy while whole mowed fields had more forb canopy. Control fields were
characterized by tall live vegetation and dead canopy. Avian diversity and
relative abundance were significantly reduced on manipulated fields compared to
control fields. Strip mowed fields showed higher abundances than whole mowed
fields, but similar diversity. Overall number of nests was reduced on manipulated
fields, but nest success was equivalent to control fields. If mowing is deemed
necessary for weed control, strip mowing is recommended over whole field

mowing.

Copyright by
ALISON JANE PEARKS
1995

Dialogue...

"Morals are your agreement with yourself to abide by your own rules. To thine
own self be rue or you spoil the game."

"Crazy."

"Thertiy vary the rules and play a different game. You cannot exhaust her infinite
vane

Lazarus Long in '
by Robert A. Heinlein

ACKNOWLEDGEMENTS

This project was funded by the Michigan Agricultural Experiment Station,
Federal Aid in Vifildlife Restoration Project W-127-R administered by the Michigan
Department of Natural Resources, Wildlife Division, and the Michigan Chapters of
Pheasants Forever. Special thanks is extended toJ. Beall and J. DesRoucher
for soliciting support from Pheasants Forever Chapters, and to J. DesRoucher
and F. Whitaker for their valuable assistance in mowing fields. I would also like
to thank P. Squibb, C. Bennett, Jr., and R. Elden from the Michigan DNR for
supporting the project and A. Stewart for providing maps. A special thank you is
extended to R. Payne and S. Curtiss from the Farm Service Agency for
assistance in locating study sites and J. Swanson from the Soil Conservation
Service for valuable information provided pertaining to the Conservation Reserve
Program. A very special thank you to all landowners for their outstanding
cooperation and allowing the use of their property for research.

My committee members were a valuable source of insight, knowledge, and
support, and I thank them for their guidance and trust. To my major advisor and
mentor Dr. Scott Winterstein, I would like to express my deep appreciation for the
many battles fought on my behalf, and for the strong confidence in my abilities to
make it through this program. If it were not for your faith I simply would not be
here, and I cannot thank you enough for the opportunity to prove myself. The
advice and guidance you have given me in the last 3 years are words I will carry

with me for the rest of my life. Most importantly, I thank you for allowing me to

climb all my mountains - I am stronger from my trips up and down. To Dr. Rique
Campa, I thank you for the knowledge I did not know I needed in order to
succeed in this endeavor. Your patience and feedback were essential to this
process and I greatly appreciate the chance to work with you. To Dr. Don
Beaver, I also thank you for your wisdom and support, and for the valuable time
you invested in this work.

Special thanks to my interns Jenn Barrett, Jeff Pype, Dawn Alex, and Jill
DeShaaf for their assistance in data collection. It was fun working with all of you
and I hope you learned as much as I did. To volunteers Andrew Pearks, Gil
DeShaaf, and Don Gibbon, sincere thanks are extended for the time and effort
you put into field work. You made the job much easier.

Finally, to the people who were on the outskirts of this project but were no
less instrumental to the process - my family and friends. Mum, Dad, your love,
encouragement, and support formed the base of my effort and I thank you from
the bottom of my heart for all you have done for me. To my brother, Drew, I
thank you not only for the month of your life you gave to help with field work, but
for your support over the years from all those long distance phone calls. To my
many, many friends who were my family in absentia, thank you for your strength,

for your encouragement, and for listening.

vi

TABLE OF CONTENTS

List of Tables
List of Figures

Introduction
Historical Back round of the Conservatio Reserve Program
Habitat Manipu ation

Hypothesis and Objectives
Study Area
Experimental Design

Methods
Vegetation Characteristics
Avran Diversity and Relative Abundance
Avian Productivity

Analysis

Results
Vegetation Characteristics
Avian Diversity
Avian Relative Abundance
Avian Productivity

Discussion
Vegetation Characteristics
Vegetation Characteristics and Avian Diversity
Vegetation Characteristics and Avian Relative Abundance
Vegetation Characteristics and Avian Productivity

Recommendations

Appendices
Appendix I. Statisical evidence for the effect of field age on avian

diversity and relative abundance.
Appendix II. Top 5 vegetation species per treatment in Gratiot
County, Michigan, 1994-1995.

vii

Appendix III. Bird species observed on Conservation Reserve Program
fields in Gratiot County, Michigan, 1991-1995.

Appendix IV. Number of each avian species seen per treatment per
census period in Gratiot Coun , Michigan, 1994 and 1995.

Appendix V. Number of nests y specres found per treatment in
Gratiot County, Michigan, 1994 and 1995.

Literature Cited

viii

Page

50
52
53

LIST OF TABLES

Page
Table 1. Range and mean field size of whole mowed fields, stri mowed
fields, and control fields in Gratiot County, Michigan, 1994 and 995. 8
Table 2. Mean values of vegetation characteristics on whole mowed
fields in May, July, and post-manipulation in 1994 and 1995 in Gratiot
County, Michigan. 16
Table 3. Mean values of vegetation characteristics on whole mowed
fields in 1994 and 1995 in Gratiot County, Michigan. 17
Table 4. Mean values of vegetation characteristics on strip mowed
fields in May, July, and post-manipulation in 1994 and 1995 in Gratiot
County, Michigan. 18
Table 5. Mean values of vegetation characteristics on strip mowed
fields in 1994 and 1995 in Gratiot County, Michigan. 19

Table 6. Mean vegetation characteristic values for whole mowed
fields (Whl) and control fields (Con) in Gratiot County, Michigan, 1994-5. 21

Table 7. Mean vegetation characteristic values for strip mowed
fields (Strp) and control fields (Con) in Gratiot County, Michigan, 1994-5. 22

Table 8. Mean vegetation characteristic values for whole mowed
fields §Whl) and strip mowed fields (Strp) in Gratiot County, Michigan,
1994- . 23
Table 9. Mean avian diversities (Shannon-Weaver Index) on whole

mowed fields and control fields in 1994 and 1995 in Gratiot County,

Michigan. 24

Table 10. Mean avian diversities (Shannon-Weaver Index) on strip
mowed fields and control fields in 1994 and 1995 in Gratiot County,
Michigan. 25

Table 11. Mean avian diversities (Shannon-Weaver Index) on strip

mowed fields and whole mowed fields in 1994 and 1995 in Gratiot
County, Michigan. 26

ix

Page
Table 12. Mean avian diversities (Shannon-Weaver Index) excluding
red-winged blackbirds on whole mowed fields and control fields in 1 94
and 1995 in Gratiot County, Michigan. 27

Table 13. Mean avian diversities (Shannon-Weaver Index) excludin
red-winged blackbirds on strip mowed fields and control fields In 19
and 1995 in Gratiot County, Michigan. . 27

Table 14. Mean avian diversities (Shannon-Weaver Index) excluding
red-winged blackbirds on whole mowed fields and strip mowed fields
in 1994 and 1995 in Gratiot County, Michigan. 28

Table 15. Mean avian relative abundances (birds/ha on whole
mowed fields and control fields in 1994 and 1995 in ratiot County,
Michigan. 29

Table 16. Mean avian relative abundances (birds/ha on strcip
mowed fields and control fields in 1994 and 1995 in ratiot ounty,
Michigan. 30

Table 17. Mean avian relative abundances (birds/ha) on strip mowed
geldhs and whole mowed fields in 1994 and 1995 in Gratiot County, 31
IC igan.

Table 18. Mean avian relative abundances (birds/ha) excluding
red-winged blackbirds on whole mowed fields and control fields in
1994 and 1995 in Gratiot County, Michigan. 32

Table 19. Mean avian relative abundances (birds/ha) excluding
red-winged blackbirds on strip mowed fields and control fields in
1994 and 1995 in Gratiot County, Michigan. 33

Table 20. Mean avian relative abundances (birds/ha) excluding
red-winged blackbirds on whole mowed fields and strip mowed fields in
1994 and 1995 in Gratiot County, Michigan. 34

Table 21. Number of active nests, successful nests, and percent of
successful nests found on strip mowed fields, whole mowed fields, and
control fields in 1994 and1995 in Gratiot County, Michigan. 35

Table 22. Mean nest success r field type combined by year on whole
mowed fields, strip mowed fiel s, and control fields in Gratiot County,
Michigan. 36

Table 23. Number of active nests, successful nests, and percent of
successful nests excluding red-winged blackbirds found on strip mowed
fields, whole mowed fields, and control fields in 1994 and 1995 in Gratiot
County, Michigan. 37

Table 24. Mean nest success excluding red-winged blackbirds per field

type combined b year on whole mowed fields, strip mowed fields, and
control fields in ratiot County, Michigan. 37

X

Appendix II. Top 5 vegetation species per treatment in Gratiot
County, Michigan, 1994-1995.

Appendix III. Bird species observed on Conservation Reserve Program
fields in Gratiot County, Michigan, 1991-1995.

Appendix IV. Number of each avian s ies seen per treatment
per census period in Gratiot County, ichigan, 1994 and 1995.

Appendix V. Number of nests by species found per treatment in
Gratiot County, Michigan, 1994 and 1995.

xi

Page
47

48

49

51

LIST OF FIGURES

Figure 1. Map of the state of Michigan showing the location of
Gratiot County.

Figure 2. Diversity (Shannon-Weaver Index) regression analysis
com arisons by field age and the year the data were recorded for
the rst, second, and third birding penods, respectively.

Figure 3. Relative abundance (birds/ha) r ression analysis

com arisons by field age and the year the ata were recorded for
the rst, second, and third birding periods, respectively.

xii

Page

46

47

INTRODUCTION

Historical Background of the Conservation Reserve Program

As agricultural production has increased over time, the extensive native
grasslands which once covered most of North America have been replaced by
croplands and pastures. Monoculture fields now dominate a once heterogeneous
landscape. These changes have greatly impacted wildlife communities as their
habitats diminished in size and diversity. Loss of nesting sites, feeding areas and
cover led to drastic reductions in wildlife populations, both in numbers and in
diversity (Bemer 1988).

Increasing agricultural development also brought about detrimental
environmental effects. Soil erosion and chemical runoff rose dramatically and
water quality decreased substantially. At one point, an estimated 3.1 billion tons
of soil eroded from croplands on an annual basis. Off-farm damages of $5 - $18
billion were reported annually, as well as $1.84 billion in on-farm damages (US.
General Accounting Office 1989).

To counter these problems, the federal government implemented several
set-aside programs. The first was the Cropland Adjustment Act (CAA) in 1934,
which simply removed land from production (Edwards 1984). It was replaced in
1936 with the Agricultural Conservation Program (ACP), an annual set-aside
program requiring farmers to plant cover crops such as grasses or legumes to
conserve soil (Bemer 1988). The Soil Bank Act of 1956 created the

Conservation Reserve (CR) - a multiyear, cover crop program - and the Acreage

1

2
Reserve (AR) - an annual, non-cover crop program. The AR was discontinued in
1959 (Edwards 1984).

The Emergency Feed Grain Program (FGP), implemented in 1961, and
the Wheat Program (WHP), started in 1962, were both annual programs initiated
to boost land retirement participation by increasing acreage payments (Bemer
1988). Cover crops were not required, and even though participation increased,
wildlife population levels remained low (Erickson and Wrebe 1973). In 1966, the
Cropland Adjustment Program brought back multiyear retirement plans with cover
crops, but the FGP and WHP proved to be more economically attractive to the
landowners (Bemer 1988). The 1984 Payment-ln-Kind (PIK) program combined
the time length of an annual cycle with some cover crop requirements, but did not
benefit wildlife or landowners, decrease erosion, or improve water quality (Cutler
1934).

In 1985 the Food Security Act (Food Bill) created the current Conservation
Reserve Program (CRP). It too is a multiyear set-aside program requiring a
cover crop to be planted and targets highly erodible areas, or land that
contributes to a water quality problem (US. Dept. of Agriculture 1990). The Food
Bill was amended in 1990 by the Food, Agriculture, Conservation, and Trade Act
(FACTA) to increase opportunities for landowners to retire environmentally
sensitive land (US. Dept. of Agriculture 1990).

As of the 12th signup period (1993), about 15 million hectares are enrolled
in the CRP, with a total acreage goal of 18 million hectares. Presently there are
roughly 377,000 contracts, or 8% of all US cropland, in CRP, with 93% of that
planted to grass or trees. An estimated 700 million tons of soil are prevented
from eroding annually, decreasing the pre-CRP amounts by 22% (Osborn 1993).
The CRP has also decreased sedimentation, preserved the long-term productivity

of the land, and decreased chemical runoff (US. General Accounting Office

1989).

Benefits of the presence of CRP fields to wildlife are well documented
(Taylor 1980, Berthelsen et al. 1989, Burger et al. 1990, Stouffer 1990, Campa et
al. 1991 Campa et al. 1992, Campa et al. 1993, Campa etal. 1994, Campa et al.
1995). For example, 16 grassland bird species that were in long-term decline are
now more abundant on CRP lands than they are elsewhere (Outdoor Life 1994).
It is also apparent that the continuing enrollment of lands in the CRP is important
to avian diversity, as the "age" of the field - the length of time it has been planted
to cover - influences bird use of a field. Younger fields tend to have higher
diversity and density, whereas older fields have greater productivity (Millenbah
1993). Age influence can also be seen in small mammal use of the fields, with
younger fields having more diversity and older fields having greater abundance
(Furrow 1994). The canopy makeup of a field changes by age, with younger
fields (1 - 3 years of enrollment) showing relatively large amounts of bare ground,
live cover, and forb cover. Older fields (4 - 6 years of enrollment) are
characterized by more total canopy coverage, grasses, and litter cover (Campa
et. al 1993).

Habitat Manipulation

Key requirements for wildlife are availability and adequate interspersion of
cover, food and water. Juxtaposition and interspersion of vegetation types and
structures increases potential habitat, particularly in an environment limited for
size. The creation of edge tends to increase vegetation and the relative
abundance of wildlife (National Research Council, Committee on Impacts of
Emerging Agricultural Trends on Fish and Wildlife Habitat 1982).

In a managed system, periodic manipulations to revitalize vegetation and

maintain a balance between successional stages may be necessary for long-term

4
maintenance of viable wildlife habitat (Noss 1991, Schenck and Williamson
1991). Periodic disturbance of grasslands in 3 - 5 year intervals has been found
to enhance wildlife production by more than 100%, and winter cover/spring
nesting vegetation benefits from a 5 - 10 year manipulation cycle (Sousa 1987,
Schenck and Williamson 1991). Research has also indicated that interspersion
of areas of unmanipulated climax grassland within areas of periodic disturbance
tends to perpetuate and increase grassland wildlife populations (Schenck and
Williamson 1991).

CRP fields are subject to periodic manipulation as a form of weed control,
but the methods used and the amount disturbed vary from one participating area
to the next. There have been few data gathered on how these management
practices may affect the vegetation structure on the fields or wildlife populations
utilizing the fields. One brief, informal study showed that mowing on an annual
basis reduces weed content and increases legume mass (Fee 1995), but the
impact of this on wildlife was not addressed. Avian species in particular could be
the most affected, as vegetation structure variations have a potentially large
affect on nesting, feeding, and winter cover habitat.

Information on how habitat manipulation impacts avian communities and
vegetation composition could provide insight on which method is the most
beneficial to maintaining avian densities and relative abundance. These data
could also provide background on how useful habitat manipulations are to avian

conservation plans in general.

HYPOTHESIS AND OBJECTIVES

The primary null hypothesis for this study is that manipulated (mowed)
CRP fields do not differ significantly from unmanipulated CRP fields with respect
to avian productivity, relative abundance, and diversity; and with respect to

vegetation characteristics. ‘

1 Specific objectives are to:
1) quantify vegetation characteristics of manipulated and unmanipulated CRP
fields;
2) determine the effects of manipulations (annual whole field mowing, annual strip
mowing, and no manipulations) on avian diversity, relative abundance, and
productivity; and
3) provide management recommendations for increasing avian habitat and
habitat quality for species that utilize CRP fields.

STUDY AREA

Gratiot County is located in central Michigan (T 9, 10, 11, 12N; R 1, 2, 3,
4W) (Figure 1). Precipitation averages 76.1cm annually, 62% of that in the April -
September period. Winter snow fall averages 104.90m annually. Temperatures
range from an average low in the winter of - 4.2 C to an average summer high of
20.9 C. Agriculture involves 85% of the land (personal communication from the
Farm Service Agency, December 1995), with corn, soybeans, and wheat being
the major crops. Soils in the area that are involved with agriculture include
Capac, Parkhill, Lenawee, Corunna, Selfridge and Dixboro - all moderate to poor
drainers; loamy soil, clay loam, sandy loam, or loamy sand; and mostly level.
Soils that have native vegetation on them include Plainfield, Riverdale, and
Vestaburg. All are loamy sands, but Plainfield is well drained, whereas Riverdale
and Vestaburg are poorly drained (US. Dept. of Agriculture 1979).

Among field types, a significant difference was found between sizes of
control fields and sizes of whole mowed fields (KW, P < 0.02) and strip mowed
fields (KW, P < 0.1). This may impact avian diversity, relative abundance, and
productivity measurements through reduction of species composition (Herkert
1991). Mean sizes of whole mowed fields and strip mowed fields were not
significantly different (KW, P > 0.2) (Table 1).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1. Map of the state of Michigan showing the location of Gratiot County.

8

Table 1. Range and mean field size of whole mowed fields, strip mowed fields,
and control fields in Gratiot County, Michigan, 1994 and 1995.

 

 

 

Field size
Range (ha) Mean size“ (ha)
whole mowed 2.4 - 13.6 5.325 A
strip mowed 3.2 - 10 6.433 A
control 7.2 - 14.24 10.353 B

 

‘amonfield types, mean field sizes With the same letter are not signficantly
different (KW multiple comparisons test, Miller 1980).

EXPERIMENTAL DESIGN

Eighteen fields were involved in this study, divided into 3 treatments, with
6 replications per treatment. The treatments were:
1) annually completely mowing an entire field (in late July). Two fields used in
1994 were flooded in July 1994 and had to be replaced for the 1995 season.
2) strip mowing one-third of a field (with a different third mowed each year of the
study, and taking place at the same time as the annual mowing); and
3) no manipulation (control). These fields were part of a related study that began

in 1991, providing vegetation composition and avian use data for 5 years.

METHODS

Vegetation Characteristics

This study took place from March 1993 through August 1995. Two
sampling periods were delineated to measure vegetation characteristics. These
data were used to determine the vegetation composition and structure that
maintained the greatest diversity, relative abundance, and productivity of avian
species. The first period was 1 May - 31 May, during peak avian breeding
season. The second period occurred 1 July - 31 July, during the maximum
vegetative growth season. One sample was taken on each field per period.

Data were collected from sample points on 6 randomly located 100m
transects per field. Six sample points were spaced at 20m intervals on each
transect. Horizontal cover was measured 4m from the sample point with a Robel
pole (Robel et al. 1970). At each point, height of live and standing dead
vegetation in dm was also assessed with a Robel pole. Percent of canopy cover
made of live, dead, grass, forb, and woody vegetation; percent of bare ground;
and percent litter cover were measured with a 50 x 50 cm frame (Daubenmire
1959). Litter depth in cm was recorded with a meter stick. All plant species
within the frame were recorded to determine frequencies, and the dominant
species on each field was noted.

A post-manipulation vegetation sample was taken on the manipulated
fields to determine the amount of canopy cover remaining after mowing. This

sample consisted of horizontal cover, live vegetation height, proportion of live

10

1 1
canopy cover, and litter depth. Measurements were taken as described above.
This post-manipulation sample was timed to occur when the unmanipulated
control fields were measured in July. Therefore, manipulated fields were each
measured three times and unmanipulated fields were each measured twice. In
analysis, values from the July sample taken on control fields were duplicated for a
post-manipulation control set.

In addition to 2 flooded fields which had to be replaced in 1995, 1 whole
mowed field in 1994 and another whole mowed field in 1995 were mowed ahead
of schedule. Therefore, post-manipulation measurements were not taken on the
2 flooded fields, and July vegetation measurements were not taken on the 2 fields

that were mowed early.

Avian Diversity and Relative Abundance

Three censuses were conducted per year to determine the relative
abundance and diversity of avian species utilizing fields for feeding and breeding
purposes. The census periods ran from 15 May - 15 June, 16 June - 15 July, and
16 July - 15 August. Each field was censused once during each period.

The variable-strip survey method was used along randomly located
transects. Each field first had a 25m buffer delineated around the perimeter. A
corner of each field then was chosen at random, with the first transect starting
25m along the long edge from this comer inside the perimeter. Each transect
thereafter was spaced at 50m intervals. The surveys began at dawn and were
completed by 3 hours after dawn. This allowed for observations to be made
during peak activity times. For each bird seen, researchers noted the
perpendicular distance of the bird from the transect line in meters, the distance of
the bird to the nearest edge in meters, the species, and the gender. The position

of the bird was also recorded on a map. If a flock was seen at a particular

12
location, the center of the flock was used as the location from which to make
measurements. Researchers would also then record the number of birds seen
and the number of birds of each gender.
Censuses were not taken if there was any precipitation (rain, fog) as this
hindered sight and accuracy. Censuses were also not done if the wind speed

was greater than 16kph.

 

Avian Productivity

Researchers looked for nests in 2 censuses to determine productivity and
nesting success. The first census ran from 1 May to 31 May and the second from
1 June to 30 June, with 1 census taken on each field in each period. These time
frames covered the main nesting season and any repeat nesting that occurred.
Researchers walked 1 - 2m abreast across a field until it was completely
traversed. Each nest found was revisited every 2 - 3 days until the young fledged
or the nest was abandoned or destroyed. Species, number of eggs, number of
young, and nest fate were recorded. Any nests found at any other time were also

observed.

ANALYSIS

Vegetation characteristics were compared among periods, among

 

treatments and between years using the Kruskal-Wallis 1-way analysis of
variance (Siegel 1956). Alpha was set at 0.1. Dominant vegetation
characteristics were summarized by treatment. These were then compared to
avian relative abundance, diversity, and productivity by treatment.

Avian diversity was analyzed among periods by comparing Shannon-
Weaver diversity index values (Shannon-Weaver 1949) using the Kruskal-Wallis
1-way analysis of variance (ANOVA). Comparisons among treatments and
between years were also done with Kruskal-Wallis 1-way ANOVA. Alpha was set
at 0.1.

Avian relative abundance was compared among periods, among
treatments, and between years using the Kruskal-Wallis 1-way ANOVA. Alpha
was set at 0.1.

Avian productivity was evaluated as the number of active nests versus the
number of successful nests over both periods within each year. Comparisons
among treatments and between years were done with KruskaI-Wallis 1-way
ANOVA. Alpha was set at 0.1.

Over the 2 year period of the manipulation study, whole mowed fields
averaged 4-5 years in age and strip mowed fields averaged 5-6 years in age. As
found in previous research (Millenbah 1993), age has been found to influence

avian diversity, relative abundance, and productivity by impacting vegetation

13

14
characteristics. It was necessary to match control fields by age to manipulated
fields for analysis. As previously mentioned, the control fields in this study were
part of a regional project started in 1991, and historical data for these fields were
available. Therefore, data collected in 1991 when control fields were 4 years of
age and 1992 when control fields were 5 years in age, were used for
comparisons to whole mowed fields. Data collected in 1992 when control fields
were 5 years of age and 1993 when control fields were 6 years in age, were used
for comparisons to strip mowed fields. As the year the data were collected was
not found to be a factor, this eliminated a confounding influence from analysis.

Statistical evidence for age influence is presented in Appendix I.

RESULTS

Vegetation Characteristics

In 1994 on whole mowed fields, dead vegetation height, dead canopy
cover and forb canopy cover changed significantly between May and July (KW, P
< 0.07). Horizontal cover, live vegetation height, live canopy cover, and litter
depth changed significantly among May, July and post-manipulation (KW, P <
0.02). In 1995, dead vegetation height, percent bare ground, and litter cover
changed significantly between May and July (KW, P < 0.06). Horizontal cover,
live vegetation height, live canopy cover, and litter depth changed significantly
among May, July, and post-manipulation (KW, P < 0.08). Horizontal cover, live
vegetation height, and live canopy cover increased between May and July, then
decreased post-manipulation. Dead vegetation height, dead canopy cover, and
percent bare ground decreased from May to July, while forb canopy cover and
litter cover increased. In 1994, litter depth increased from May to July to post-
manipulation. In 1995, litter depth decreased from May to July, then increased
post-manipulation (Table 2).

15

16

Table 2. Mean values of vegetation characteristics on whole mowed fieldsin
May, July, and post-manipulation in 1994 and 1995 in Gratiot County, Michigan.

 

 

 

 

1994 1995
May July Post- P‘al May July Post- Pa

manipula manipula

tion tion
HCb (dm) 1.83 5.25 2.37 0.01 1.15 5.83 0.79 0.004
LHb (dm) 2.36 7.89 4.63 0.003 2.21 7.9 1.51 0.004
DH” (dm) 1.67 0.22 0.07 0.97 0 0.004
"/0 TC” 95.8 100 0.17 97.3 99.5 0.13
% LCb 67.2 94 63.8 0.01 66.2 98.9 49.2 0.01
% DCb 2.81 0 0.07 1.85 0 0.17
% GC" 37.7 46.6 0.71 32.3 45 0.72
% FCb 27.8 47.8 0.07 31 54 0.2
% WC” 0 0 1 0 0 1
% BGb 4.2 0 0.17 2.75 0.14 0.07
% LtCb 95.8 100 0.17 29.5 98.7 0.01
LtD” (cm) 3.16 3.55 8.81 0.02 3.92 3.79 9.13 0.05

 

'Kruskal-Wallis one-way ANOVA, May vs July

Macy vs July vs post - mani ulation for HC, LH, % LC, and LtD

ll’H (horizontal cover), LH( ive he'ght), DH(dead height), %TC(total canopy
cover), %LC(live canopy cover), °oDC dead canopy cover), %GC(grass canopy
cover), %FC(forb canopy cover), %W (woody canopy cover), %BG(bare
ground), %LtC(Iitter cover), LtD(litter depth)

1994 May values on whole mowed fields were significantly greater than
1995 May values for horizontal cover and litter cover (KW, P < 0.03). 1995 July
values were significantly greater than 1994 July values for live canopy cover
(KW, P < 0.04). 1994 post-manipulation values were significantly greater than
1995 post-manipulation values for horizontal cover and live vegetation height
(KW, P < 0.02) (Table 3).

17

Table 3. Mean values of vegetation characteristics on whole mowed fields in
1994 and 1995 in Gratiot County, Michigan.

 

May July Post-manipulation

 

1994 1995 Pa 1994 1995 P“ 1994 1995 P‘

 

HC" 1.83 1.15 0.025 5.25 5.83 0.462 2.37 0.79 0.011
LH" 2.36 2.21 0.873 7.89 7.9 0.624 4.63 1.51 0.011
DHb 1.67 0.97 0.749 0.22 0 0.18

%TCb 95.8 97.3 0.406 100 99.5 0.264

%LCb 67.2 66.2 0.575 94 98.9 0.032 63.8 49.2 0.394
%DCb 2.81 1.85 0.522 0 0 1

%GCb 37.7 32.3 0.522 46.6 45 0.806

%FC” 27.8 31 0.873 47.8 54 0.806

%WC" 0 0 1 0 0 1

%BGb 4.2 2.75 0.406 0 0.14 1

%Lth 95. 8 29. 5 0. 004 100 98.7 1

LtDb 3.16 3.92 0.337 3.55 3.79 0712 8.81 9.13 1

 

 

'Kruskal-Wallis one-way ANOVA,1994 vs 1995 for May and July

May, July, and post-manipulation for HC, LH, %LC, and LtD

bHC(horizontal cover), LH(live height), DH(dead height), %TC(total canopy
cover), %LC(live canopy cover), %DC(dead canopy cover), %GCE grass canopy
cover), %FC garb canop py cover, %WC(woody canopy cover), % G(bare
ground), %Lt (litter cover), LtD( itter depth)

In 1994 on strip mowed fields, dead vegetation height, dead canopy cover,
and forb canopy cover changed significantly between May and July (KW, P <
0.02). Horizontal cover, live vegetation height, live canopy cover, and litter depth
changed significantly among May, July, and post-manipulation (KW, P < 0.06). In
1995, dead vegetation height, dead canopy cover, forb canopy cover, and litter
cover changed significantly between May and July (KW, P < 0.06). Horizontal
cover, live vegetation height, live canopy cover, and litter depth changed

significantly among May, July, and post-manipulation (KW, P < 0.06). Horizontal

1 8
cover, live vegetation height, and live canopy cover increased from May to July,
then decreased post-manipulation. Dead vegetation height and dead canopy
cover decreased from May to July, while forb canopy cover and litter cover
increased. Litter depth decreased from May to July, then increased post

manipulation (Table 4).

Table 4. Mean values of vegetation characteristics on strip mowed fields in May,
July, and post-manipulation in 1994 and 1995 in Gratiot County, Michigan.

 

 

 

 

1994 1995
May July Post- P‘- May July Post- P‘

manipula manipula

tion tion
HC" (dm) 2.18 4.75 2.28 0.003 1.07 4.34 1.51 0.002
LH" (dm) 2.44 7.16 3.4 0.002 2.11 8.27 2.8 0.003
DH“ (dm) 2.35 0.72 0.01 2.04 0 0.002
% TC" 98.1 96.3 0.52 99.9 100 0.14
% LCID 67.2 94 65.8 0.01 56.9 96.8 80.3 0.02
% DC” 7.34 1.32 0.06 2.18 0 0.06
% GCb 51.7 55.7 0.87 44.7 62.3 0.11
% FC" 15.3 36.2 0.06 11.7 34.7 0.06
% WCb 0.02 0 0.63 0 0 1
% BGb 1.9 0.12 0.63 1.53 0.42 0.9
% LtCb 97.7 99.5 0.63 39.6 96.1 0.004
LtDb (cm) 4.59 4.35 8.85 0.02 5.08 2.6 4.34 0.06

 

‘Rruskal-Wallis one-way ANWWay vs July

Mag vs July vs post - manipulation for HC, LH, % LC, and LtD

bH (horizontal cover), LH(live hei ht), DH(dead height), %TC(totaI canopy
cover), %LC(live canopy cover), °oDC dead canopy cover), %GC( rass canopy
cover , %FC(forb canopy cover), %W (woody canopy cover), %B (bare
ground), %LtC(litter cover), LtD(litter depth)

1 9
1994 May values for strip mowed fields were significantly greater than
1995 May values for horizontal cover, dead canopy cover, and litter cover (KW, P
< 0.05). 1994 July values were significantly greater than 1995 July values for
dead vegetation height, dead canopy cover, litter cover, and litter depth (KW, P <
0.09). 1995 July values were significantly greater for total canopy cover (KW, P
< 0.06). Post-manipulation values for horizontal cover and litter depth were

significantly greater in 1994 than in 1995 (KW, P < 0.04), while live canopy cover

 

was significantly greater in 1995 post-manipulation (KW, P < 0.06) (Table 5).

Table 5. Mean values of v etation characteristics on strip mowed fields in 1994
and 1995 in Gratiot County ichigan.

 

 

 

May July Post-manipulation
1994 1995 P‘I 1994 1995 P‘i 1994 1995 P'
HC" 2.18 1.07 0.004 4.75 4.34 1 2.28 1.51 0.037
LHb 2.44 2.11 0.631 7.16 8.27 0.262 3.4 2.8 0.261
DHb 2.35 2.04 1 0.72 0 ‘ 0.007

%ch 93.1 99.9 0.703 96.3 100 0.059

%ch 67.2 56.9 0.109 94 96.8 0.372 65.3 30.3 0.053
%ch 7.34 2.13 0.042 1.32 0 0.022

"/ocscb 51.7 44.7 0.262 55.7 62.3 0.749

%FC” 15.3 11.7 0.423 36.2 34.7 0.373

%wc" 0.02 0 0.317 0 0 1

%BG" 1.9 1.53 0.523 0.12 0.42 0.902

%LtC" 97.7 39.6 0.003 99.5 96.1 0.039

LtD” 4.59 5.08 0.873 4.35 2.6 0.055 8.85 4.34 0.025

'Kruskal-Wallis one-way ANOVAT'I'Q'STV's—I'995 for May and July
Mag, July, and post-ma n|ipulation for HC, LH, %LC, and LtD

(horizontal cover), L (live height), DH(dead height), %TC(t0tal canopy
cover), %LC(live canopy cover) %DC(dead canopy cover), %GC grass canopy
cover %FC forb canopy cover), %WC(woody canopy cover), % G(bare
ground), %Lt (litter cover), LtD(litter depth)

 

 

20

When differences were significant between whole mowed fields and
control fields for horizontal cover, live and dead vegetation heights, dead canopy
cover, and percent bare ground, control field values were greater (KW, P < 0.04).
When differences were significant for total canopy cover, litter cover, and litter
depth, whole mowed field values were higher (KW, P < 0.1). Live canopy cover
was significantly greater on whole mowed fields in July, 1995 (KW, P = 0.005),
and on control fields in post-manipulation, 1995 (KW, P = 0.019) (Table 6).

When differences were significant between control fields and strip mowed
fields for horizontal cover, live and dead vegetation heights, dead canopy cover,
and percent bare ground, control field values were higher (KW, P < 0.07). When
differences were significant for total canopy cover, live canopy cover, and litter
cover, strip mowed field values were higher (KW, P < 0.1). However, in post-
manipulation, 1994, control fields had significantly higher levels for live canopy
cover (KW, P = 0.031). Litter depth was significantly greater on strip mowed
fields in post-manipulation, 1994 (KW, P = 0.011) and on control fields in July,
1995 (KW, P = 0.055) (Table 7).

Significant differences were seen between whole mowed fields and strip
mowed fields in dead canopy cover in May and July, 1994 (KW, P < 0.07), in total
canopy cover in May, 1995 (KW, P < 0.09), and in all post-manipulation variables
(KW, P < 0.07). Strip mowed field values were higher in all instances (Table 8).

A list of the top 5 plant species found on each manipulation type by year is
shown in Appendix II.

21

.586 6:5 s 52,8 to; .03....

.6520 otmnvwme .098 .628 5 >00 35 .098 39.8 2.0.0.5 .098 30:8 390 00.x. .098 858
383095 69,8 No.88 32.63.. . .98 s 28 .926»... .c; 6; 88.10 .393; 95:: 69,8 5836501..
mEmEoLEoo m> «200 00268 0.9.; .<>Dz< 503.05 Emaimxmabr

00.0 0...0 00.0 00.0 05.0 00.0 00.0 00.0 00.0 ..0.0 00.0 «.0 305
00.0 5.00 05 ..0.0 0.00 0 00.0 00.. 50 3.0V 00 ON 30565.
#00 3.0 5.0 05.0 05.0 N ..0.0 0 0.0 ...0 Ni 5V0 300$
00.0 0 00.0 00.0 0 5.0 05.0 0 0.0 3.0 0 0.0 603$
00 V0 00 05.0 P0 00 05.0 0.5V 5* 0.0 00 05 60065.
00.0 0v 00 00.0 0.00 00 «.0 0.0V 00 0.0 00 00 30035.
..0.0v 0 . «v.0 00.0 00.5 00.5 P0.0v 0 0.5 ..0.0 0.0 0.. 30065.
00.0 0.0V 00 ..0.0v 0.00 00 F 0.00 #0 0...0 v0 00 0.0 V0 00 05.0 50 .5 60.15.
9.0 . 0.00 ..0 00.0 0.50 55 00.0 00.. 00 N00 00 00 305$
5.0V 0 9.0 500V 50.0 0.0.. r0.0v 00.0 0 No.0 5.5 0.0 :10

_.0.0 «0... 5.0.. 00.0 0.5 5.05 3.0 ..N.N 01¢ 90.0 00.4 05 3.0 00.5 or 3.0V v.0 5.5 61..
50.0 0.0 5.5 0.0 00.0 5.5 ..0.0v 0w... 5.0 ..0.0 50.0 : 50.0v 00.0 : ..0.0v 0.5 05.0 no...

 

60...; :00 an._...>> :00 5:5) :00 8. _...>> :00 an. E; :00 an. _...>> :00

. «000
000.20." 00302 3.333% «00.3.35 «0105.02

 

3-32 .5953
.3550 .380 5 .50. 5.0... .228 new .23. ago... 0026.: 29.3 to. $2? 8.65.0220 8.8.00? :85. .0 058.

22

.568 .00... s 09,8 0050......
.0590 0:00.006» 00>8 30:00 5 osvwgo5 .00>8 500:8 8.00.5 .e0>8 500:8 000:0 00.x. ..:0>8 30:00

08300.... 59,8 8:8 9500s . .98 6:8 0.2.0:... .9. .0: 80310 .0299. 02:1. 09,8 038.650....
00.050050 m> 00.0.. 0026:: 050 .<>Dz< 503.08 0503.00.86...

00.0 V0.V 00.0 00.0 0.0 00.0 00.0 00.0 05.0 _.0.0 00.0 00.0 00.0 00V 00.0 0.0 00V 00.0 69...

_.0.0 ...00 5.. F0 0.00 00 ..0.0 0.00 05 50.0 5.50 P0 60335.

V0 0V0 50.0 00.0 00... 50. .. 00.0 0? .0 ...0 0.20 0.9 0... 9000..
V...0 0 V0 V5.0 0 0.0 00.0 0 00.0 0.0 00.0 _..0 30365.
00.0 5.V0 00 V0.0 5. E. VP . 05.0 0.00 00 0.0 0.05 00 6000..
50.0 0.00 V0 00.0 5.VV 50 50.0 5.00 00 V0.0 5. *0 0V 30065.
90.0 0 0? 3.0 9.0 00 3.0 00.5 0V0 00.0 V0.5 00.5 3.000.:

V0.0 0.00 05 50.0 0.00 05 P 0.00 50 00.0 0.00 00 00.0 V0 00 00.0 0.50 V0 304$
00.0 00? 5.00 00.0 0.00 00 00.0 0.00 F0 ..0.0 5.00 ..5 30.585.
..0.0 0 v.0 ..0.0 V0.0 0.? ..0.0 05.0 V0 ..0.0 00.0 0.0? DID
..0.0 0.0 05 ..0.0 50.0 0.. 50.0 .....0 V0.0 500 V0 5.0? 00.0 0.. .5 5.0.. 00.0 VV.0 0V.V 6......
_.0.0 0.? 0.0 5.0 V0.V 0.0 ..0.0 50... 0.0 ..0.0 00.0 5.5 00.0 05V 5.5 50.0 0.. .0 5.0 no...

 

a 9.0 :60 a 9.0 :60 a 9.0 :60 a 9.0 :60 a 9.0 :60

 
 

8. a m. .2 la: «0.0.302

 

0-32 58.6.5.
5:300 .0005 :. .50. 020... 60:8 0:0 .000. 00.0.. 0026:. 0.0.0 to. 00:.0> 00.000.00.050 :o..0.000> :005. .5 0.00...

23

.58: 0:5 0.. .098 00503.:
.8505 000.00..» .098 30:00 > a...» o: .098 30:00 0000.3 .098 30:00 000:0 00.x. .098 0:00

00:300.: .098 8:8 03:00:10.3 0:8 asswbx. .0: 0: 3:31: 0:00: 02:1: :96: 0.805501.

 

00.0.. 0030.: 50 9 020... 0026.: 0.0:; (>020. 33-0:0 0___0B-_0w.0mb..

8.: 3.: 9.: :.: 0.: ::.: :.: ::.0 ::.: F 0:: 3.: 3.: 0:: 00.: :.: :0... :.: .0:
0..: 0:: 0:: :0: 0:: 0:: 0:: 0:: :2 :0: E: 0: .05.:

r :1: 3.: 3.: :0: 0:: 0:: :0: : :0: :e :0 .00.:

F : : P : : F : : :0: 8.: : .03.:

::.: :3 3 0..: 0..: F: :.: :0: :5. :.: :0. 0: .00:

::.: :.:: 0: :.: Z... :.:: 0:: :0: :0: :.: 5: 0: .00.:

0 : : 3.: :0: 00.. B: :0 : 8.: 3.: 3.: .000:

0:: :8 :.:: 2.: :0: :0: :0: :0: :0: . :0: :.:: 00.: 3 3 8.: :.:0 :0 .04.:
v: :2 0:: 0:: :2 :.:: 2.: 0: 8. :0: :0: :: .00.:

r : : :.: 3.: ::.: v: :0: ::.: :0 0:: E .00

50.0.0.0 :0... 05.0 50.0 0.5 05.0 5.0 0.0 0.0 V0 00.V 05.0 0:.5 00.5 0.0 VV.0 V0 :7...
00.0 0.: 05.0 :0 V0.V 00.0 50.0 50.: 0:... 50.0 00.0 50.0 00.0 05.V 00.0 V0.0 0...0 0.: :0...

 

a .50 :8 a e0 :8 a. :50 :8 a. :50 :oo .0 :50 :8

 
 

003 A...» “00.3.0.2 «0.04%.... «3.302

 

0-32 3:222 0:80
000:0 :_ 300. 0200 09.68 050 0:0 2:3. 020... 0030:: 00:3 0.. 000.9 000000000020 800.009 :005. .0 050...

24

Avian Diversity

Comparisons between periods showed that avian diversity on whole
mowed fields was significantly greater the second period (16 June - 15 July ) than
the third period (6 July - 15 August) (KW, P < 0.06) in both 1994 and 1995.
Comparisons between treatments showed a significant difference between whole
mowed fields and control fields in the second period (16 July - 15 August) (KW, P
< 0.06) and third period (16 July - 15 August) (KW, P < 0.01), with greater
diversities occurring on control fields in 1994. A significant difference was found
between whole mowed fields and control fields in all 3 periods (KW, P < 0.03) in
1995, with control fields having more diversity. Comparisons between years
found 1994 values to be significantly greater than 1995 in the third period (16 July
- 15 August) (KW, P = 0.06) (Table 9).

Table 9. Mean avian diversities Shannon-Weaver Index) on whole mowed fields
and control fields in 1994 and 19 5 in Gratiot County, Michigan.

 

Field type
Year Birding period Whole mowed‘ Controla Probabilityb
1994 15 May - 15 June 0.867 (0.3) ABc Ed 1.34 (0.1) 0.15
16 June - 15 July 1.008 (0.14) Ac 6" 1.576 (0.16) 0.055
16 July - 15 August 0.485 (0.21) 3° H“ 1.562 (0.19) 0.01
1995 15 May - 15 June 0.465 (0.28) Acc F“ 1.248 (0.05) 0.025
16 June - 15 July 0.688 (0.23) A° cs.d 1.318 (0.03) 0.01

16 July-15 August 0.063 (0.02) 0° Hd 1.223 (0.18) 0.003

“sample variance in parenthesis.

bKruskal-Wallis one-way ANOVA, whole mowed fields vs control fields.

cwithin whole mowed fields within years, birding periods with the same letter are
not significantly different KW multiple comparisons test, Miller 1980).

dwithin whole mowed fuel 5 between years, birding periods with the same letter
are not significantly different (KW multiple comparisons test, Miller 1980).

 

 

 

25

Avian diversity on strip mowed fields was not significantly different
between periods (KW, P > 0.2) in 1994 or 1995. A significant difference was
found in 1994 between strip mowed fields and control fields in the first period (15
May - 15 June) (KW, P = 0.025) and second period (16 June - 15 July) (KW, P =
0.025), with control fields showing more diversity. A significant difference was
found in 1995 between strip mowed fields and control fields in the second period
(16 June - 15 July) (KW, P < 0.06) and third period (16 July - 15 August) (KW, P
= 0.004), with control fields showing more diversity. Comparisons between years
did not show any significant differences (KW, P > 0.1) (Table 10).

Table 10. Mean avian diversities (Shannon-Weaver Index) on strip mowed fields
and control fields in 1994 and 1995 in Gratiot County, Michigan.

 

Field type
Year Birding period Strip mowed‘ Control‘ Probabilityb

1994 15 May - 15 June 0.797 (0.08) Ac 13d 1.248 (0.05) 0.025
16 June - 15 July 0.803 (0.15) A° CC! 1.318 (0.03) 0.025
16 July - 15 August 0.585 (0.4) A° rsd 1.223 (0.18) 0.149
1995 15 May - 15 June 0.74 (0.11) Ac 8“ 1.213 (0.29) 0.109
16 June - 15 July 0.46 (0.12) A" 0" 1.137 (0.32) 0.055

16 July-15August 0.595 (0.12)A° E“ 1.528 (0.04) 0.004

“sample vanance in parenthesis.

bKruskaI-Wallis one-way ANOVA, strip mowed fields vs control fields.

cwithin strip mowed fields within ears, birding periods with the same letter are not
significant y different (KW multip e comparisons test, Miller 1980).

dwithin strip mowed fields between years, birding periods with the same letter are
not significantly different (KW multiple comparisons test, Miller 1980).

 

 

 

No significant difference was found in 1994 in avian diversity between strip
mowed fields and whole mowed fields in any period (KW, P > 0.2). In 1995,
avian diversity was significantly greater on strip mowed fields than whole mowed
fields in the third period (16 July - 15 August) (KW, P = 0.013) (Table 11).

26

Table 11. Mean avian diversities (Shannon-Weaver Index) on strip mowed fields
and whole mowed fields in 1994 and 1995 in Gratiot County, Michigan.

 

 

 

Field type

Year Birding period Whole moweda Strip mowed‘ Probabilityb
1994 15 May - 15 June 0.867 (0.3) 0.797 (0.08) 0.522

16 June - 15 July 1.008 (0.15) 0.803 (0.15) 0.298

16 July - 15 August 0.485 (0.21) 0.585 (0.4) 0.81
1995 15 May - 15 June 0.465 (0.28) 0.74 (0.11) 0.296

16 June - 15 July 0.688 (0.23) 0.46 (0.12) 0.296

16 July - 15 August 0.063 (0.02) 0.595 (0.12) 0.013

 

'sample variance in parenthesis.
bKruskal-Wallis one-way ANOVA, whole mowed fields vs strip mowed fields.

Fifty-three percent of the birds seen were red-winged blackbirds. When
this species was removed from the whole mowed field census counts and
diversity was recalculated, significantly greater diversity was found in the second
period (16 June - 15 July) than the third period (16 July - 15 August) in both 1994
and 1995 (KW, P < 0.05). Control fields had significantly more diversity than
whole mowed fields in all periods in both years (KW, P < 0.1). Comparisons
between years found significantly greater diversity in the third period (16 July - 15
August) in 1994 than 1995 (KW, P < 0.06) (Table 12).

When the same was done for strip mowed fields, no significant differences
were found between periods (KW, P > 0.1). Control fields had significantly more
diversity than strip mowed fields in 1995 (KW, P < 0.04). Comparisons between
years found significantly greater diversity in the second period (16 June - 15 July)
in 1994 than 1995 (KW, P < 0.03) (Table 13).

27

Table 12. 'Mean avian diversities (Shannon-Weaver Index) excludin red-winged
blackbirds on whole mowed fields and control fields in 1994 and 199 in Gratiot
County, Michigan.

 

Field type
Year Birding period Whole mowed“ Control“ Probability”

 

 

1994 15 May - 15 June 0.682 (0.21) Ac Cd 1.172 (0.09) 0.092
16 June - 15 July 0.772 (0.2) 3° 0“ 1.668 (0.09) 0.005
16 July - 15 August 0.292 (0.11) Ac F“ 1.445 (0.14) 0.004
1995 15 May - 15 June 0.283 (0.21) Ac 0“ 1.05 (0.16) 0.022
16 June - 15 July 0.445 (0.14) 3° E“ 1.38 (0.03) 0.004
16 July - 15 August 0 (0) A° Fd 1.222 (0.16) 0.002

 

‘sample variance W1 pareflhefis.

bKruskal-Wallis one-way ANOVA, whole mowed fields vs control fields.

cwithin whole mowed fields, birding periods with the same letter are not
significantly different (KW multiple comparisons test, Miller 1980).

dwithin whole mowed fields between years, birding periods with the same letter
are not significantly different (KW multiple comparisons test, Miller 1980).

Table 13. Mean avian diversities (Shannon-Weaver Index) excluding red-winged
blackbirds on strip mowed fields and control fields in 1994 and 1995 in Gratiot
County, Michigan.

 

Field type
Year Birding period f Strip moweda Control‘ Probability”
1994 15 May - 15 June 0.923 (0.07) Ac 8“ 1.05 (0.16) 0.574
16 June - 15 July 1.983 (6.18) Ac 0“ 1.38 (0.04) 0.199
16 July - 15 August 0.585 (0.49) A° E“ 1.222 (0.16) 0.147
1995 15 May - 15 June 0.755 (0.19) A° B“ 1.23 (0.14) 0.037
16 June - 15 July 0.457 (0.13) A° Dd 1.165 (0.11) 0.019

16 July-15August 0.477 (0.1)A° E“ 1.435 (0.04) 0.004

rsample variance in parenthesis.

”Kruskal-Wallis one-way ANOVA, strip mowed fields vs control fields.

cwithin strip mowed fields, birding penods with the same letter are not significantly
different (KW multi Ie comparisons test, Miller 1980).

dwithin strip mowe fields between years, birding periods with the same letter are
not significantly different (KW multiple comparisons test, Miller 1980).

 

 

28

Comparisons between strip mowed fields and whole mowed fields showed

significantly greater diversity on strip mowed fields in the first (15 May - 15 June)
and third (16 July - 15 August) periods in 1995 (KW, P < 0.1) (Table 14).

Table 14. Mean avian diversities (Shannon-Weaver Index) excluding red-winged
blackbirds on whole mowed fields and strip mowed fields in 1994 and 1995 in

Gratiot County, Michigan.

 

 

 

Field type

Year Birding period Whole mowed‘I Strip mowed‘ Probabilityh
1994 15 May - 15 June 0.682 (0.21) 0.923 (0.07) 0.336

16 June - 15 July 0.772 (0.2) 1.983 (6.18) 0.2

16 July - 15 August 0.292 (0.11) 0.585 (0.49) 0.442
1995 15 May - 15 June 0.283 (0.21) 0.755 (0.19) 0.096

16 June - 15 July 0.445 (0.14) 0.457 (0.13) 0.565

16 July - 15 August 0 (0) 0.477 (0.1) 0.007

 

'§ample variance in parenthesis.
bKruskal-Wallis one-way ANOVA, whole mowed fields vs strip mowed fields.

Avian Relative Abundance

Comparisons between periods on whole mowed fields in 1994 showed
significantly greater avian relative abundance in the second period (16 June - 15
July) than the third period (16 July - 15 August) (KW, P < 0.03). In 1995, avian
relative abundance on whole mowed fields was significantly different between the
second and third periods (16 June - 15 July and 16 July - 15 August) (KW, P <
0.08), and the first and third periods (15 May - 15 June and 16 July - 15 August)
(KW, P < 0.03), with the third period showing the least abundance. Abundances
were significantly greater on control fields than whole mowed fields in all periods
(KW, P < 0.04) in 1994. Control fields had significantly greater abundances than
whole mowed fields in the second period (15 May - 15 June) (KW, P < 0.03) and

29
third period (16 July - 15 August) (KW, P < 0.04) in 1995. Comparisons between
years showed no significant differences (KW, P > 0.1) (Table 15).

Table 15. Mean avian relative abundances (birds/ha) on whole mowed fields and
control fields in 1994 and 1995 in Gratiot County, Michigan.

 

Field type
Year Birding period Whole moweda Control” Probability”

 

 

1994 15 May - 15 June 1.838 (1.02) AB° C“ 4.135 (2.67) 0.037
16 June - 15 July 2.565 (1.66) 8° C“ 5.701 (3.72) 0.006
16 July - 15 August 1.055 (0.52) N C“ 4.471 (3.51) 0.006
1995 15 May - 15 June 2.575 (2.91) A° c“ 2.508 (2.69) 0.873
16 June - 15 July 1.442 (0.65) A° C“ 3.552 (4.35) 0.025
16 July - 15 August 0.653 (1.02) 8° C“ 3.594 (18.5) 0.036

 

'sample variance in parenthesis.

bKruskal-Wallis one-way ANOVA, whole mowed fields vs control fields.

cwithin whole mowed fields within years, birding periods with the same letter are
not significantly different KW multiple comparisons test, Miller 1980).

“within whole mowed fiel s between years, birding periods with the same letter
are not significantly different (KW multiple comparisons test, Miller 1980).

Comparisons between periods in 1994 on strip mowed fields found
significant differences in avian relative abundance between the first and third
periods (15 May - 15 June and 16 July - 15 August) and the second and third
periods (16 June - 15 July and 16 July - 15 August) (KW, P < 0.04), with the
lowest abundance in the third period. Avian relative abundance on strip mowed
fields in 1995 was not significantly different between periods (KW, P > 0.3).
Control fields had significantly greater abundances than strip mowed fields in the
first period (15 May - 15 June) (KW, P < 0.04) in 1994. Strip mowed fields were
not significantly different from control fields in any period (KW, P > 0.1) in 1995.

Comparisons between years found abundances to be significantly greater in 1994

30
than 1995 in the first (15 May - 16 June) and second (16 June - 15 July) periods
(KW, P < 0.08) (Table 16).

Table 16. Mean avian relative abundances (birds/ha) on strip mowed fields and
control fields in 1994 and 1995 in Gratiot County, Michigan.

 

Field type
Year Birding period Strip mowed“ Control“ Probabilityb
1994 15 May - 15 June 4.653 (4.02) A° C“ 2.508 (2.69) 0.037
16 June - 15 July 5.57 (6.83) N Ed 3.552 (4.35) 0.150
16 July - 15 August 1.767 (4.99) 8° C“ 3.594 (18.5) 0.200
1995 15 May - 15 June 2.722 (5.16) A° o“ 3.659 (1.38) 0.109
16 June - 15 July 2.447 (0.3) A° Fd 3.365 (1.62) 0.200

16 July - 15 August 2.333 (4.4) A° Gd 3.251 (3.88) 0.423

rsample variance in parenthesis.

bKruskal-Wallis one-way ANOVA, strip mowed fields vs control fields.

cwithin strip mowed fields within ears, birding periods with the same letter are not
significant y different (KW multip e comparisons test, Miller 1980).

“within strip mowed fields between years, birding periods with the same letter are
not significantly different (KW multiple comparisons test, Miller 1980).

 

 

 

In 1994, strip mowed fields had significantly greater abundances than
whole mowed fields in the first period (15 May - 15 June) (KW, P < 0.01) and
second period (16 June - 15 July) (KW, P < 0.04). In 1995, strip mowed fields
had significantly greater abundances than whole mowed fields in the second
period (16 June - 15 July) (KW, P < 0.05) and third period (16 July - 15 August)
(KW, P < 0.09) (Table 17).

31

Table 17. Mean avian relative abundances (birds/ha) on strip mowed fields and
whole mowed fields in 1994 and 1995 in Gratiot County, Michigan.

 

 

 

Field type

Year Birding period Whole mowed“ Strip mowed“ Probability”
1994 15 May - 15 June 1.838 (1.02) 4.653 (4.02) 0.01

16 June - 15 July 2.565 (1.66) 5.57 (6.83) 0.037

16 July - 15 August 1.055 (0.52) 1.767 (4.99) 0.749
1995 15 May - 15 June 2.575 (2.91) 2.722 (5.16) 0.873

16 June - 15 July 1.442 (0.65) 2.447 (0.3) 0.045

16 July - 15 August 0.653 (1 .02) 2.333 (4.4) 0.087

 

“sample variance in parenthesis.
bKruskal-Wallis one-way ANOVA, whole mowed fields vs strip mowed fields.

When red-winged blackbirds were removed from whole mowed field
census counts and avian relative abundance was recalculated, significantly
greater abundances were found on whole mowed fields in 1995 in the first period
(15 May - 15 June) than the third period (16 July - 15 August) (KW, P < 0.06).
Control fields had significantly higher relative abundance than whole mowed
fields in the second period (16 June - 15 July) and third period (16 July - 15
August) in 1994 and in the third period (16 July - 15 August) in 1995 (KW, P <
0.02). Comparisons between years showed no significant differences (KW, P >
0.1) (Table 18).

32

Table 18. Mean avian relative abundance (birds/ha) excluding red-winged
blackbirds on whole mowed fields and control fields in 1994 and 1995 in Gratiot
County, Michigan.

 

Field type
Year Birding period Whole mowed“ Control“ Probability“

1994 15 May - 15 June 1.07 (0.45) A° C“ 2.383 (1.64) 0.109
16 June - 15 July 1.208 (0.83) A“ C“ 3.183 (1.08) 0.016
16 July - 15 August 0.637 (0.09) A° C“ 4.038 (2.79) 0.004
1995 15 May - 15 June 1.062 (0.51) A° C“ 1.213 (0.32) 0.575
16 June - 15 July 1.042 (0.92) A° C“ 1.682 (0.47) 0.127

16 July-15 August 0.362 (0.21)8° C“ 2.595 (3.59) 0.006

“sample variance in parenthesis.

l’Knlskal-Wallis one-way ANOVA, whole mowed fields vs control fields.

“within whole mowed fields within years, birding periods with the same letter are
not significantly different KW multiple comparisons test, Miller 1980).

“within whole mowed fiel s between years, birding periods with the same letter
are not significantly different (KW multiple comparisons test, Miller 1980).

 

 

When the same was done for strip mowed fields, significantly greater
abundances were seen in the first (15 May - 15 June) and the second (16 June -
15 July) periods than in the third period (16 July - 15 August) in 1994 (KW, P <
0.03). Control fields had significantly higher abundances in the third period (16
July - 15 August) than strip mowed fields in 1994 (KW, P < 0.01), and in the first
(15 May - 15 June) and second (16 June - 15 July) periods in 1995 (KW, P <
0.04). Comparisons between years found significantly higher abundances in the
first (15 May - 15 June) and second (16 June - 15 July) periods in 1994 than 1995
(KW, P < 0.05) (Table 19).

33

Table 19. Mean avian relative abundance (birds/ha) excludin red-winged
blackbirds on strip mowed fields and control fields In 1994 an 1995 in Gratiot
County, Michigan.

 

Field type
Year Birding period Strip mowed“ Control“I Probability“

1994 15 May - 15 June 1.115 (0.17) A“ C“ 1.213 (0.32) 0.873
16 June - 15 July 1.225 (0.31) A“ E“ 1.682 (0.47) 0.262

16 July - 15 August 0.438 (0.19) 8° G“ 2.595 (3.59) 0.010

1995 15 May - 15 June 0.65 (0.08) A“ o“ 1.692 (1.27) 0.037
' 16 June - 15 July 0.348 (0.08) A“ F“ 2.017 (2.19) 0.016

16 July 15 August 1673 (2.08) A“ c“ 2.55 (2.38) 0.262

“sample variance in parenthesis.
“Kruskal-Wallis one-way ANOVA, strip mowed fields vs control fields.
“within strip mowed fields within ears, birding periods with the same letter are not
significant y different (KW multip e comparisons test, Miller 1980).
“within strip mowed fields between years, birding periods with the same letter are
not significantly different (KW multiple comparisons test, Miller 1980).

 

 

 

When comparisons were done between whole mowed fields and strip
mowed fields, excluding red-winged blackbirds, significantly greater abundances
were found on whole mowed fields in the second period (16 June - 15 July) (KW,
P = 0.05), and on strip mowed fields in the third period (16 July - 15 August) (KW,
P = 0.05) in 1995 (Table 20).

34

Table 20. Mean avian relative abundance (birds/ha) excludin red-winged .
blackbirds on whole mowed fields and strip mowed fields in 1 94 and 1995 In

Gratiot County, Michigan.

 

 

 

Field type

Year Birding period Whole mowed“ Strip mowed“ Probability“
1994 15 May-15 June 1.07 (0.45) 1.115(0.17) 1

16 June - 15 July 1.208 (0.83) 1.225 (0.31) 0.522

16 July - 15 August 0.637 (0.09) 0.438 (0.19) 0.260
1995 15 May - 15 June 1.062 (0.51) 0.65 (0.08) 0.521

'16 June - 15 July 1042 (0.92) 0.348 (0.08) 0.054

16 July - 15 August 0.362 (0.21) 1.673 (2.08) 0.053

 

“sample variance in parenthesis.
“KruskaI-Wallis one-way ANOVA, whole mowed fields vs strip mowed fields.

A list of the avian species found on CRP fields, with scientific

nomenclature, is included in Appendix III. The number of avian species found per

treatment per period is in Appendix lV.

Avian Productivity

In 1994, 19 nests were found on whole mowed fields, all active (produced
at least 1 egg), and 7 successful (fledged at least 1 young) (36.8%); 43 nests

were found on control fields, 40 active, and 24 successful (60%). On strip
mowed fields, 129 nests were found, 107 active, and 37 successful (34.6%);

control fields had 181 nests, 132 active,-and 55 successful (41.7%).

In 1995, 12 nests were found on whole mowed fields, all active, and 5

successful (41.7%); 181 were on control fields, 132 active, and 55 successful
(41.7%). Strip mowed fields had 97 nests, 71 active, and 13 successful (18.3%);

control fields had 116 nests, 100 active, and 44 successful (44%).

Control fields had significantly greater success than whole mowed fields

35
(KW, P < 0.03) in 1994, and than strip mowed fields (KW, P < 0.1) in 1995. The
most common species found nesting in 1994 were red-winged blackbird, song
sparrow, ring-necked pheasant, mallard, and blue winged teal; in 1995 they were
red-winged blackbird, song sparrow, mallard, blue winged teal, eastern

meadowlark, and vesper sparrow (Table 21).

Table 21. Number of active nests, successful nests, and percent of successful
nests found on strip mowed fields, whole mowed fields, and control fields in 1994
and 1995 in Gratiot County, Michigan.

 

 

Year Field type # of active # of successful % successful“
nests nests

1994 whole mowed 19 7 36.8 AD
control 40 24 60 B
strip mowed 107 37 34.6 CD
control 132 55 41.7 C

1995 whole mowed 12 5 41.7 AD
control 132 55 41 .7 A
strip mowed 71 13 18.3 BD
control 100 44 44 C

 

W een field types wrthin a year, W success results with the same letter are ncfi
Significantly different (KW multiple comparisons test, Miller 1980).

When nest success data were combined by year and compared between
treatments, control fields had significantly greater success than whole mowed
fields (KW, P < 0.02) (Table 22).

36

Table 22. Mean nest success per field pe combined by year on whole mowed
fields, strip mowed fields, and control fie ds in Gratiot County, Mlchlgan.

 

 

Field type Mean nest success“b
whole mowed 0.206 (0.08) A '
control 0.489 (0.03) B

strip mowed 0.294 (0.09) AC
control 0.419 (0.02) C

 

“between field types, nest success means wflh the same letter are not
significantly different (KW multiple comparisons test, Miller 1980).
“sample variances in parenthesis.

When red-winged blackbird counts were removed fiom the nesting data, 2
nests were found on whole mowed fields in 1994, both active and neither
successful (0%); control fields had 13 nests, 6 active, and 3 successful (50%).
Strip mowed fields had 5 nests, all active, and 2 successful (40%); control fields
had 18 nests, 16 active, and 6 successful (37.5%).

In 1995, when red-winged blackbird counts were removed from the nesting
data, no nests were found on whole mowed fields; control fields had 18 nests, 16
active, and 6 successful (37.5%). Strip mowed fields had 6 nests, 5 active, and 1
successful (20%); control fields had 18 nests, 16 active, and 1 successful (6%).

Control fields had significantly greater success than whole mowed fields in
1995 (KW, P < 0.01) without red-winged blackbirds (Table 23).

When nest success data without red-winged blackbirds were combined by
year and compared between treatments, control fields had significantly greater
success than whole mowed fields (KW, P < 0.01) (Table 24).

A list of the number of nests found per species per manipulation type is
included in Appendix V.

37

Table 23. Number of active nests, successful nests, and percent of successful
nests, excluding red-winged blackbirds, found on strip mowed fields, whole
mowed fields, and control fields in 1994 and 1995 in Gratiot County, Michigan.

 

 

Year Field type # of active # of successful % successful“
nests nests
1994 whole mowed 2 0 0 AC
control 6 3 50 A
strip mowed 5 2 40 AC
control 16 6 37.5 A
1995 whole mowed 0 0 0 AD
control 16 6 37.5 B
strip mowed 5 1 20 AD
control 16 1 6 A

 

W types witlfin a year, % success results with the sametétter are not
significantly different (KW multiple comparisons test, Miller 1980).

Table 24. Mean nest success excludin red-winged blackbirds per field ty
combined by year on whole mowed fiel s, strip mowed fields, and control elds ln

Gratiot County, Michigan.

 

 

Field type Mean nest success“b
whole mowed 0 (0) A

control 0.36 (0.17) B

strip mowed 0.167 (0.15) A
control 0.193 (0.1) A

 

“EtWeen field types, nest success means with the same letter are not
significantly different (KW multiple comparisons test, Miller 1980).

I’sample variances in parenthesis.

DISCUSSION

Vegetation results will be discussed first, to establish field composition and
structure by manipulation type. Avian diversity will then be tied to vegetation

characteristics by manipulation, as will avian relative abundance and productivity.

Vegetation Characteristics

Whole mowed fields and strip mowed fields had more total canopy cover
and litter cover than control fields; while control fields had greater proportions of
horizontal cover, live and dead vegetation heights, dead canopy cover, and bare
ground than whole mowed fields or strip mowed fields (Tables 6 and 7).
Manipulated fields can therefore be described as having low, live vegetation, and
unmanipulated fields as having tall, dead vegetation.

Whole mowed fields were annually mowed on average for 4 years prior to
the initiation of the study, whereas strip mowed fields had no previous
manipulation history. Over the 2 years of the study, whole mowed fields showed
a decrease in horizontal cover and litter cover, and an increase in live canopy
cover (Table 3). In the same time period, strip mowed fields had a decrease in
horizontal cover, dead vegetation height, dead canopy cover, litter cover, and
litter depth; and an increase in total canopy cover (Table 5).

These results indicate that mowing as a manipulation treatment directly
affects the amount of litter and standing dead vegetation that remains on a field

from year to year. By cutting down the standing dead vegetation, the horizontal

38

39
structure of the field is reduced, and the amount of litter accumulating on the
ground is decreased. The mowing changes the structure of the field by reducing
the amount of leftover standing vegetation present at the beginning of the
growing season, decreasing competition for space among the new growth and
thereby increasing canopy cover. This supports conclusions made by earlier
studies (Holecheck et al. 1982, Comely et al. 1983, US. Dept. of Agriculture
1991).

Vegetation Characteristics and Avian Diversity

Diversity on whole mowed fields followed a consistent pattern in both
years, with the greatest diversities in the middle of the field season and the
lowest at the end of the year (Table 9). This corresponded with the peak
breeding period, which occurred in the middle of the field season; and with young
of the year fiedging, which took place at the end of the season. Removing red-
winged blackbirds from census totals reduced the diversity index on whole
mowed fields, but analysis did not reveal different results (Tables 12).

Control fields were significantly more diverse in both years (Table 9).
Ring-necked pheasants, savannah sparrows, common yellowthroats, and eastern
meadowlarks were consistently found on control fields, but were absent fiom
whole mowed fields (Appendix IV). Vegetation structure on control fields was
taller, and composed of more dead vegetation and less litter cover than whole
mowed fields (Table 6). These Characteristics better fulfilled habitat requirements
for these species than did whole mowed fields (Brewer et al. 1991).

Diversity decreased by half on whole mowed fields between 1994 and
1995, while staying constant on control fields. However, this may reflect the
changes in fields sampled between 1994 and 1995 and not reflect manipulation
effects, because 2 whole mowed fields were replaced between years.

40

Strip mowed fields had relatively constant levels of diversity (Table 10).
When red-winged blackbirds were removed from census counts, diversity indices
increased, and strip mowed fields were found to have greater diversity in 1994
than 1995 (Table 13). Total canopy cover was also greater in July 1995 (Table
5), which may have influenced the increase in diversity.

Control fields had higher levels of diversity than strip mowed fields with or
without red-winged blackbirds, particularly in 1995 after the strip mowed fields
had been manipulated for the first time (Tables 10 and 13). Ring-necked
pheasants, common yellowthroats and northern harriers preferred unmanipulated
fields, while waterfowl such as mallards, Canada geese, and blue-winged teal
were more prevalent on strip mowed fields (Appendix IV). The greater proportion
of litter cover on strip mowed fields would deter pheasants, yellowthroats, and
harriers from utilizing these fields, while the reduced vegetation height on strip
mowed fields would make them more attractive to waterfowl as feeding areas
(Table 7) (Brewer et al. 1991).

Strip mowed fields had higher diversity levels than whole mowed fields in
1995, both with and without red-winged blackbirds (Tables 11 and 14). This was
mainly due to the presence of waterfowl and ring-necked pheasants on strip
mowed fields (Appendix IV). Post-manipulation variables were also higher on
strip mowed fields in both 1994 and 1995 (Table 8). The greater availability of
ground cover on strip mowed fields left on the field from the summer before would
make them more attractive for these ground nesters than whole mowed fields.

Overall, control fields had greater diversity than either manipulation, and
there was no consistent difference between the 2 types of mowed fields. Control
fields had 5-8 unique species present. Whole mowed fields attracted 1-3 unique
species compared to control fields; however strip mowed fields had 3-4 unique

species, mostly waterfowl (Appendix IV). Diversity differences between whole

41
mowed fields and strip mowed fields were due to northern cardinals, field
sparrows, American woodcocks, and American robins on whole mowed fields,
and the presence of mallards, Canada geese, blue-winged teals, tree and barn

swallows, and eastern meadowlarks on strip mowed fields (Appendix IV).

Vegetation Characteristics and Avian Relative Abundance

Relative abundances on whole mowed fields were always lowest at the
end of the summer (Table 15 and Appendix IV). The dispersing of young during
the third birding period would account for the reduction of birds on the fields.
Removing red-winged blackbirds shifted the period of greater abundance on
whole mowed fields from the second period (16 June - 15 July) to the first period
(15 May - 15 June), but significantly lower abundances were only recorded at the
end of the field season (Table 18).

Control fields had greater abundance in both years (Table 15 and
Appendix IV). Again, this reflects the difference in vegetation structure, as the
whole mowed fields were comparatively deficient in horizontal cover and
vegetation height (Table 6). When red-winged blackbirds were removed from
census totals, the differences between control fields and whole mowed fields
were reduced but not eliminated (Table 18).

The lowest abundances on strip mowed fields in 1994 occurred after
manipulations took place in late July. The effect of the mowing was seen into the
1995 season, with abundance levels significantly lower in the first 2 periods in
1995 than they were in 1994 (Table 16 and Appendix IV). Horizontal cover, dead
vegetation height, dead canopy cover and litter cover were all higher in 1994
- (Table 5), and the change in vegetation characteristics reduced avian relative
abundance. Results did not change when red-winged blackbirds were removed

from census counts (Table 19).

42

Control fields had higher levels of relative abundance than strip mowed
fields in 1994, but not in 1995 (Table 16 and Appendix IV). This could be due to
the large number of red-winged blackbirds seen on strip mowed fields in 1994
(Appendix IV), for when this species is eliminated and relative abundance levels
recalculated, mntrol fields have significantly higher levels of abundance only in
the third period in 1994 (Table 19). Removing red-winged blackbirds from
censuses causes control fields to have greater abundances than strip mowed
fields in the first and second periods of 1995 (Table 19). As control fields had
more horizontal cover and dead canopy cover and taller vegetation height (Table
7), the grassland species prevalent on these fields would have found them more
attractive than strip mowed fields.

Strip mowed fields had greater relative abundance than whole mowed
fields in both 1994 and 1995 (Table 17 and Appendix IV). This is related to the
greater numbers of red-winged blackbirds seen on strip mowed fields in both
years (Appendix IV), for when relative abundance is recalculated without this
species, strip mowed fields have greater abundance than whole mowed fields
only in the third period in 1995, and whole mowed fields have greater abundance
than strip mowed fields in the second period in 1995 (Table 20). Analysis of
vegetation characteristics showed post-manipulation values to be higher on strip
mowed fields in both years (Table 8), which would account for greater
abundances on strip mowed fields in this period. The greater abundances seen
on whole mowed fields compared to strip mowed fields in this one period may be
a factor of the difference in field size (Table 1).

Overall, manipulated fields had less relative abundance than control fields,
and strip mowed fields had greater relative abundance than whole mowed fields.
Species common to manipulated and unmanipulated fields, such as red-winged

blackbirds, sparrows, bobolinks, and sedge wrens, were in greater abundance on

43
control fields, then on strip mowed fields. Whole mowed fields had 6 species
represented by only one individual, whereas strip mowed fields had 4.

Previous reports have suggested that while vegetation manipulation may
benefit wildlife, it is important to maintain enough cover on fields to sustain avian
diversities and relative abundance levels (Schenk and Wllliamson 1991). These
results indicate that whole mowed fields do not have adequate cover, and that
strip mowed fields would be likely to support higher relative abundance levels in

their place.

Vegetation Characteristics and Avian Productivity

Whole mowed fields had the fewest nests present of all fields types in both
years, however nests were more likely to be active on whole mowed fields than
on any other type of field. This could be due to nests being built in growing
vegetation, and concealed better than nests built in dead vegetation. Only 3
species were observed nesting on whole mowed fields - red-winged blackbirds,
vesper sparrows, and song sparrows (Appendix V).

Control fields had greater success than whole mowed fields in 1994 and in
both years combined (Tables 21 and 22). Control fields had more standing dead
vegetation and horizontal cover (Table 6), and provided more structure for nest
building.

Removing red-winged blackbirds from whole mowed field nesting data left
1 vesper sparrow nest and 1 song sparrow nest. Annual removal of standing
dead vegetation would greatly reduce nesting cover for ground nesting species
found on other fields, such as waterfowl, ring-necked pheasants, northern
harriers, bobolinks, and eastern meadowlarks (Appendix V). Control fields had
higher nest success in 1995 and in both years combined (Tables 23 and 24).

Strip mowed fields were as successful as control fields in 1994, however

44
after manipulations took place control fields were more successful (Table 21).
Horizontal cover and standing dead vegetation on strip mowed fields were also
reduced compared to control fields in this time period (Table 8), and these
vegetation characteristics Changes reduced nesting cover on strip mowed fields.
Red-wing blackbirds were most prevalent on strip mowed fields, but song
sparrows, field sparrows, mallards, ring-necked pheasants, blue-winged teals,
and vesper sparrows used strip mowed fields (Appendix V). Species composition
did not change significantly from 1994 to 1995, after manipulations occurred, but
nest numbers were reduced (Table 21 and Appendix V).

Red-winged blackbirds made up all but 5 nests in 1994 and 6 nests in
1995 (Appendix V). Nest success without red-winged blackbirds on strip mowed
fields was equivalent to control fields and whole mowed fields (Tables 23 and 24).

Comparisons between whole mowed fields and strip mowed fields showed '
that nest success was not different between manipulations (I’ ables 21 and 22).
Vegetation characteristics were similar on both field types (Table 8), providing
similar habitat and similar Chance for success. Differences in cover remaining on
a field from one year to the next, as shown by the greater post-manipulation
values on strip mowed fields (I' able 8), would account for the differences in
species composition (Appendix V). Ground nesters would prefer the greater
ground cover available on strip mowed fields left over from the year before.

Previous studies have shown that standing dead vegetation is an element
of preferred nesting habitat, and that fields with high levels of standing dead
vegetation have higher avian productivity levels than fields without (Comely et al.
1983, Millenbah 1993). This study supports these conclusions, and if mowing is
necessary for weed control, strip mowed fields may benefit avian populations

more than whole mowed fields by providing more standing dead vegetation.

RECOMMENDATIONS

Results of this study indicate that mowing a field affects vegetation
Characteristics, avian diversity and relative abundance, and avian productivity.
Manipulated fields have less horizontal cover and standing dead vegetation, more
live vegetation, shorter plant height, reduced avian diversity and relative
abundance, and a decline in the total numbers of nests on a field compared to
unmanipulated fields.

These results show that unmanipulated fields provide vegetation
composition and structure preferred by avian populations utilizing CRP fields in
Gratiot County, Michigan. However, previous research has shown that leaving
fields unmanipulated eventually results in decreased diversity and relative
abundance (Millenbah 1993). To satisfy any requirements for weed control, and
to help prevent this decline in avian communities, strip mowing is recommended
as a manipulation. Strip mowing provides for more post-manipulation ground
cover, greater avian relative abundance, and allows more nests on fields than
whole field mowing. Avian diversity does not seem to be impacted more by one
or the other manipulation.

While strip mowing may be more beneficial in providing habitat for avian
communities initially, the long term impacts of strip mowing are not known, and
the effects of strip mowing may change in a continuous annual program. Further
research is required to determine if strip mowing provides preferred habitat for

avian populations when compared to whole field mowing in the long run.

45

APPENDICES

46

Appendix I. Statistical evidence for the effect of field age on avian diversity and

relative abundance.

 

 

 

 

X
8 r’=o.2036 P=0.0|23
E 2 .____---._-._. _
.. 1.8 . '
§ 1.6 " . . . o .-
9 1.4 v- ° 1 A .
E 1.2 0 \:“\
1 " x ‘ ‘ o
o A
E 0.8 , -. ‘

g 0.6 2 — ~ — g
-(2 0.4 l._.~_-- .— _-_v.. ....;.- .72. 2-. 2
0.2 f ' 1 l a 1 t

g 3 4 5 6 7 8

AgeofField
Fieldnumber= .1 '2 .314 x5 +6

 

 

 

 

 

 

 

 

 

 

 

’3 z
‘E r=0.1659 P=0.0255
5 2.5

> o

8 2 t“ -.

g. 1.5 h t: g 4’
S 1 IL- v \N.
C I .
c V

g 0.5 “'7' " x I l:
c 0 t : i - t -

8 3 4 5 6 7 8
5 Age of Field -

Fieldnumbcr= I1v203g4x546

X

C.)

U 2

5 "0321 P=0.00II
a 2

a l:

o a

3:134“

E 1.2 -

9 1 .

1: 0.8 . . :
5:: 0.6 m» x —- -- , : (I
C 0.4 ‘ ‘ c 1‘ t A

8 3 4 5 6 7 8
5 AgeofField

Fieldnumber=

.1.2.3.4 x5 .6

 

    

P=0.6338

   

 

 

 

 

 

 

 

 

 

 

 

0.8
0.6
4 3'3
,5 . ........
.1; 0
0 ~02
‘r -o.4 4
-0.6
0.8 '
1991 1992 1993 1994 1995
Year Data Recorded
Fieldnumbcr= .1VZ03A485“6
r’==o.022l P.=0.4328
1 " ””""—““*%
.9 0.5 .. “ -
a 4‘ A n
.8 O lb—~__:. ' - 0
'g .. ' " ‘8'
m '0'5 1: . __. .____- . ._---.::
x - z
-1 1 t 4
1991 1992 1993 1994 1995
Year Data Recorded
Fieldnumwzn1v203a4x5+6
r’=0.0004 P=0.9IS
1
.2 0.5 3; § ' n
g 0 ll 0 i E "
m I II:
a) . II
I! -0.5 1‘
-1 at ,4‘ : f =
1991 1992 1993 1994 1995
Year Data Recorded
Fieldnumbcr= .1v2-3a4x5+6

Fi ure 2. Divemi Shannon-Weaver Index) regression analysis comparisons by
fiesld age and the t)tletalr the data were recorded for the first, second, and thlrd

birding periods respectively.

47
Appendix I (cont)

2
r =0.3296 P=0.0009

-—_ --—_ -—._.——q

 

    

 

 

 

Mean Relative Abundance
O N A O) o:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S-----.- ”.1“ _ _.. :2 -4
: 4‘ 4 4 f 9 4 t
3 4 5 6 7 8
AgeofFleld
Fieldmw=I1V203A4x5+6
1 .
§ r=0.4205 P=0-000|
(010
U II
5 8 .. '
n '
g 6 1F\'\"
32‘ 4 J. x\\ . 0
£9 ‘ .' .
g 2 4» - ‘\’
§ 0 t c 4. : t c t 0
2 3 4 5 6 7 8
AgeofField
Ficldnumbcr= .1v2.3.4x5.5
Q) 2
2 r=0.4157 P=0.0001
g 14
g 12 . .
2 1o .. .
0 8 0. - —
.2. 61 . - .
E 4 » , . - - o
a 2 ‘* 2 '— "’0‘ x" A It
(3 O 1 9 9 4 1 -
5 3 4 5 6 7 8
Agede

Ficldnumbcr= .1 v2 .3 .4 x5 *6

 

 

 

 

 

 

 

 

 

 

6 rz=0.0168 P=0.4952
2 4 v C '—
‘é 2 ~ 'v I -- -; - ha.
....‘ i' ‘
3 0 I 2 i 0
(K _2 ,_ I m. .4- ._ -ll
.4 0 ‘r f fi‘ 4 f
1991 1992 1993 1994 1995
YearDataReoorded
Fieldnumbcr- -1vz-3_44XS*6
r'=o.0017 P=0.828I
5 .
.12 4 t; "
g 2 " ' . z
:2 x L
§ ° :: 3 ' 1:
-2 .9 . -
.4 : f t .
1991 1992 19.93 1994 1995
YearDataReoorded
Ficldnumber= u1v20344t‘5“6
r2=0.0243 P=0.4I l2
-8 v
6 4
U)
".33 4 0 v
2 2 .. - 2
m ‘ _.1
g 0 “I ! : 9 K
'2 ll 3 ‘ ““7"“
.4 J. V 9 t
1991 1992 1993 1994 1995
YearDataReoorded

Ficldnumbcr= .1 v2 .3 .4 x5 #6

Figure 3. Relative abundance (birds/ha) regression analysis com risons by field
age and the year the data were recorded for the first, second, an third birding

periods respectively.

48

2S 4ndix "' TOP 5 vegetation SpeCies per treatment in Gratiot County, Michigan,

 

 

-1995.
1994 1995
whole mowed orchard grass orchard grass
(Dactlyis glomerata) dandelion
alfalfa alfalfa
(Medicago sativa) oldenrod
dandelion Solidago spp.)
(Taraxacum officinale) red clover
red clover
( Tn‘folium pretense)
uack grass
Agropyron repens)
control alafalfa orchard grass
for whole mowed fields dandelion alfalfa
orchard grass dandelion
timothy grass timoth grass
(Phleum pretense) Cana ian thistle
goldenrod (Cirsium arvense)
strip mowed quack grass orchard grass
alfalfa alfalfa
timothy grass dandelion
orchard grass Queen Anne's lace
dandelion (Daucus carota)
timothy grass
control alafalfa orchard grass
for strip mowed fields dandelion alfalfa
orchard grass dandelion
timothy grass timoth grass

goldenrod

Cana ian thistle

 

49

Appendix III. Bird species observed on Conservation Reserve Program fields in
Gratiot County, Michigan, 1991-1995.

 

 

Species Code Species Scientific Name
AMGO American Goldfinch Carduelis tn'stis
AMRO American Robin Turdus migraton'us
AMWO American Woodcock Philohela minor
BARS Barn Swallow Himndo mstica
BOBO Bobolink* Dolichonyx oryzivoms
BWTE Blue-win ed Teal“ Anas discors

CAGO Canada oose Branta canadensis
COBO Common Bobwhite Colinus virginianus
COGR Common Grackle Quiscalus quiscula
COYE Common Yellowthroat Geothlypis tn'chas
EAKI Eastern Kingbird Tyrannus tyrannus
EAME Eastern Meadowlark Stumella ma na
FISP Field Sparrow“ S izella pusi Ia

HOLA Horned Lark remophila alpestn's
INBU Indigo Bunting Passen'na cyanea
KILL Killdeer Charadrius vocifems
MALL Mallard“ Anas platyrhynchos
MODO Mourning Dove Zenaida macroura
NOCA Northern Cardinal Cardinalis cardinalis
NOHA Northern Harrier“ Circus cyaneus
RNPH Ring-necked Pheasant“ Phasianus colchicus
RWBL Red-winged Blackbird“ Agelaius phoeniceus
SASP Savannah Sparrow“ Passerculus sandwichensis
SEWR Sedge Wren“ Cistothorus platensis
SOSP Song Sparrow“ Melospiza melodia
TRES Tree Swallow Tachycineta bicolor
UNID Unidentified

UNSP Unidentified Sparrow“

VESP Vesper Sparrow“ Pooecetes gramineus

 

“nesting species

Appendix IV. Number of each avian species seen
period in Gratiot County, Michigan, 1994 and 1995 .

50

Per treatment per census

 

 

 

1994 1995
Field 15 May- 16 June- 16 July-15 15 May- 16 June- 16 July-15
type 15 June 15 July August 15 June 15 July August
Whole 27 rwbl“ 49 rwbl“ 15 sewr 34 rwbl“ 17 rwbl“ 7 Mbl
mowed 12 bobo 19 unsp 11 rwbl 12 sosp 10 bobo 7sewr
10 unsp 10 bobo 4 vesp 9 bobo 3 unsp 1 sosp
9 sosp“ 8 sosp“ 1 unsp 2 unsp 2 sewr
1 amro 2 sewr 1 fisp 2 amgo
1 amwo 1 mph 2 eakl
1 unid 1 noca
1 fisp
Control 135 rwbl“ 194 nlvbl“ 84 bobo 123 rwbl“ 178 rwbl“ 68 sosp
for 83 sosp“ 46 sosp“ 52 sasp 35 bobo 38 sosp 56 bobo
whole 31 sewr“ 28 unsp 50 mph 33 sosp 32 unsp“ 38 I'Wbl
mowed 17 sasp 28 sewr“ 44 sewr 10 unsp“ 16 bobo 22 sasp
14 bobo 27 bobo 39 sosp 4 sewr“ 14 sewr“ 16 unsp
6 mph 11 am 0 30 unsp 2 sasp 13 sasp“ 13 mph
4 unid 11 mp 5 unid 2 chsp 6 coye 11 coye
2 amgo 9 unid 3 coye 2 coye 5 mph 10sewr
2 eame 5 came 3 amgo 1 eame 2 vesp 6 hola
1 coye 2 inbu 1 unid 2 eakl 3 amgo
1 mall 1 amgo 2 vesp
1 amro 1 earne 1 noha
1 noha“ 1 unid
Strip 155 rwbl“ 187 rwbl“ 64 rwbl 101 rwbl“ 86 rwbl“ 29 bobo
mowed 14 unsp 17 sewr“ 6 sewr 8 fisp 4 sosp“ 22 rwbl
12 sos “ 8 amgo 5 unsp 4 bwte 2 sewr 10 bwte
5 mal“ 7 bobo 3 sosp 4 bobo 1 mph 10 mall
5 bobo 7 unsp 2 bobo 4 sosp“ 1 fisp 6 sewr
5 sewr 3 sosp“ 2 mph 2 sewr 1 amgo 3 cago
2 mph“ 3 mph“ 2 tres 2 mall“ 1 eame 3 amgo
2 bars 1 amgo 1 unsp 2 sosp
1 tres 1 vesp 1 bars 1 unid
1 fisp
1 mph

 

Appendix IV (con't.)

51

 

 

 

1994 1995
Field 15 May- 16 June- 16 July-15 15 May- 16 June- 16 July-15
type 15 June 15 July August 15 June 15 July August
Control 123 rwbl“ 178 rwbl“ 68 sosp 182 rwbl“ 118 rwbl“ 69 bobo
for strip 35 bobo 38 sosp 56 bobo 42 sosp“ 24 sewr“ 55 rwbl
mowed 33 sosp 32 unsp“ 38 rwbl 24 bobo“ 23 mph“ 38 sosp
10 unsp“ 16 bobo 22 sasp 22 sewr“ 22 tres 31 sewr
4 sewr“ 14 sewr“ 16 unsp 5 mall“ 21 bobo 16 mph
2 sasp 13 sasp“ 13 mph 5 unsp 8 unsp 14 unsp
2 chsp 6 coye 11 coye 2 bwte 8 sosp 10 amgo
2 coye 5 mph 10 sewr 1 cobo 5 coye 3 tres
1 eame 2 vesp 6 hole 1 noha“ 2 am 0 2 coye
1 unid 2 eakl 3 amgo 1 amgo 1 uni 2 same
1 amgo 2 vesp 1 kill 1 sasp
1 eame 1 noha 1 vesp
1 unid 1 cogr
1 coye
1 mph

 

“see Appendix III for species codes.
“nesting species.

appendix V. Number of nests by species found per treatment in Gratiot County,
IC I

gan, 1994 and 1995“.

52

 

number found and species

 

 

Field type 1994 1995
whole mowed 17 rwbl 19 rwbl
1 vesp
1 sosp
control 30 rwbl 163 rwbl
for whole mowed 9 sosp 6 unsg
2 sewr 4 mp
1noha 3 mall
1 unid 2 sewr
2 noha
1 sasp
strip mowed 124 rwbl 91 rwbl
2 sosp 2 sosp
1 fisp 1 vesp
1 mall 1 unsp
1 mph 1 bwte
1 mall
control 163 rwbl 92 rwbl
for strip mowed 6 unsp 8 mall
4 mph 3 mph
3 mall 2 sosp
2 sewr 1 sewr
2 noha 1 unsp
1 sasp 1 bobo
1 noha
1 bwte

 

“see Appentfix III for species codes.

LITERATURE CITED

LITERATURE CITED

Bemer, A. H. 1988. The 1985 farm act and its implications for wildlife. Pa es
13:?ng in Chandler, W. J., and L. Labate, eds. Audubon Society, ew
o , .

Berthelsen, P. S., L. M. Smith, and C. L. Coffman. 1989. CRP land and game
Egg production in Texas High Plains. J. Soil and Water Cons. 44:504-

Brewer, R., G. A. Mc Peek, R. J. Adams Jr. 1991. The atlas of breeding birds in
Michigan. Michigan State University Press, E. Lansing, Michigan. 594pp.

Burger, L. W., E. W. Kurzejeski, T. V. Daile , and M. R. Ryan. 1990. Structural
characteristics of vegetation in CRP elds in N. Missouri and their
ggitgfilaitg as bobwhite habitat. Trans. of N. Am. and Nat. Res. Conf.

Campa, H. III, S. R. Wlnterstein, K. F. Millenbah, and R. B. Minnis. 1991.
Wildlife and vegetative response to diverted agricultural land in Gratiot
County, Michigan. Annual Report. On file at Mich. Dept. Nat. Res. East
Lansing, Mich. 17pp.

, , . , and L. T. Furrow. 1992.
Wllall'fe and vegetative response to al'verta agricultural land in Gratiot
County, Michigan. Annual Report. On file at Mich. Dept. Nat. Res. East
Lansing, Mich. 26pp.

 

. . . . . and A. J.
Pearks. 1993. Wildlife and vegetative res nse to diverted agricultural
land in Gratiot County, Michi an. Annual eport. On file at Mlch. Dept.
Nat. Res. East Lansrng, Mi . 40pp.

 

 

 

. . 1994. Wl'lall'té and vegetative response to diverted
a rlcultural land in Gratiot County, Michigan. Annual Report. On file at
tch. Dept. Nat. Res. East Lansing, Mich. 25pp.

. . R. Minnis. and A. J. Pearks. 1995. Wlldlife and

vegetative response to diverted agricultural land in Gratiot County,

Micnigaaq. Annual Report. On file at Mich. Dept. Nat. Res. East Lansing,
IC . pp.

 

53

54

Comely, J. E., C. M. Britton, and F. A. Sneva. 1983. Manipulation of flood
meadow vegetation and observations on small mammal populations.
Prairie Nat. 15(1): 16-22. WR 190.

Cutler, M. R. 1984. Integrating wildlife habitat features in agricultural programs.
Trans. N. Am. Wildl. Nat. Res. Conf. 49:132-140.

Daubenmire, R. F. 1959. A canopy coverage method of vegetational analysis.
Northwest Sci. 33:43-64.

Edwards, W.R. 1984. Early ACP and pheasant boom or bust! - a historical
rspective with rationale. Pages 71-83 in Dumke, R. B. Stiehl, and R. B.
hl, eds. Proc. Perdix Ill/Greyhfartridge and ring-necked pheasant

workshop, WI Dept. Nat. Res., adison. 203pp.

Erickson, R. E. and J. E. Wlebe. 1973. Pheasants, economics, and land
retirement in South Dakota. VVlldl. Soc. Bull. 1:22-27.

Furrow, Ly T. 1994. Influence of field age on mammalian relative abundance,
diversity, and distribution on Conservation Reserve Program lands in
Michigan. MS. Thesis.

Fee, Rich. 1995. Weed control in CRP plots. Successful Farming. Mid-
March95. 93(5):42-44.

Herkert, J. R. 1991. Prairie birds of Illinois: Population res nse to two centuries
of habitat Change. Illinois Natural History Survey Bu letin 34:393-399.

Holecheck, Je L., Raul Valdez, Sanford D. Schemnitz, Rex D. Pieper, and
Charles . Davis. 1982. Manipulation of grazing to improve or maintain
wildlife habitat. Vlfildl. Soc. Bull. 10(3):204-210. WR 187.

Miller, R. P. 1980. Simultaneous statistical inferences. Second ed. Springer-
Verlag, New York, NY. 299pp.

Millenbah, Kelly F. 1993. The effects of different age classes of fields enrolled in
the Conservation Reserve Program in Michigan on avian diversity, density,
and productivity. MS. Thesis.

National Research Council, Committee on Impacts of Emergin Agricultural
Trends on Fish and Wlldlife Habitat. 1982. Impacts of merging
A ricultural Trends on Fish and Wildlife Habitat. National Academy Press.
ashington, D. C.

Noss, R. 1991. Landscape Connectivity: Different Functions at Different Scales.
Pages 29-39 in Hudson, W. E., eds. Landscape linkages and biodiversity.
Washington, D. 0., Island Press.

Osborn, T. C. 1993. Expiration of Conservation Reserve Program contracts.
USDA, Economic Research Service, Washington, D. C

Outdoo1r6L'rfe. 1994. CRP—-preserving the national habitat. Oct94. 194(4):14-

55

Robel, R. J., J. N. Bri gs, A. D. Dayton, and L. C. Hulbert. 1970. Relationships
between visua obstmction measurements and weight of grassland
vegetation. J. Range Manage. 22:295-298.

Schenck, E. W. and L. L. Williamson. 1991. Conservation Reserve Program
effects on wildlife and recreation. Pages 37-42 in Joyce, L. A., Mitchell, J.
E., and M. D. Skold, eds. The Conservation Reserve - yesterday, today,
and tomorrow. USDA For. Serv. Gen. Tech. Rep. RM-203. 65pp.

Shannon, C. E., and Weaver, W. 1949. The mathematical theory of
communication. Univ. Illinois Press, Urbana. 177pp.

Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-
Hill Book Co. New York, NY. 312pp.

Sousa, P. J. 1987. Habitat management models for selected wildlife
management practices in the Northern Great Plains. Engineering and
léfgesearctclfiienter, US. Dept of the Interior, Bureau of Reclamation.

enver, o o.

Stouffer, D. F. 1990. Evaluating potential effects of CRP on bobwhite quail in
Piedmont, VA.

Taylor, T. T. 1980. Nesting of rin -necked pheasants in Texas Panhandle.
Lubbock, TX: Texas Tech niv. 35pp. MS thesis.

US. Dept. of Agriculture. 1979. Soil survey of Gratiot County, Michigan. Soil
Conservation Service. 142pp.

U.S. Degt. of Agriculture. 1990. Conservation in the 1990 farm bill. USDA.
PP-

U.S. Dept. of Agriculture. 1991. Wlldlife and fisheries habitat improvement
handboo . USDA, Forest Service, Wlldlife and Fisheries.

US. General Accounting Office. 1989. Farm programs: conservation reserve
program could be less costly and more effective: report to the Chairman,
Committee on Agriculture, Nutrition, and Forestry, US. Senate.

GRN STATE UNI V.

lllllllllzlll 0Ill lllLllllzllllllllllall