ANALYSIS OF AMERICAN WOODCOCK
NEST AND BROOD HABITAT
IN NORTHERN LOWER MICHlGAN

Thesis for the Degree of-M. S.
MICHIGAN STATE UNIVERSTTY
ALBERT BOURGEOTS
1975

 

ABSTRACT

ANALYSIS OF AMERICAN wooococx NEST AND
BROOD HABITAT IN NORTHERN LowER MICHIGAN
By

Albert Bourgeois

Habitat characteristics of T6 nest and T9 brood sites of the Ameri-

can woodcock (Philohela minor, Gmelin) were examined in northern Lower

 

Michigan between l5 April and 15 June 1974. Components measured were:
species composition; dominance of the overstory, understory and ground
strata; tree density; diameter at breast height (dbh) and total height
of trees; ground vegetation height; distance to the nearest standing
water and nearest opening; soil characteristics; and earthworm abundance.
Woodcock hens utilized young, second-growth forest stands which
were similar in species composition for both nesting and brood rearing.
A multi-variate discriminant analysis, used to evaluate dissimilarities
between nest and brood sites, revealed a significant discrimination,
however, in habitat structure. .Nest habitat was distinguished mainly
by lower tree density (2l76 trees/ha) and basal area (8.6 mz/ha), by
being close to forest openings (7 m) and by being situated on dry,
relatively well-drained sites. In contrast, woodcock brood habitat had
nearly twice the tree density (3934 trees/ha) and basal area (16.5 mZ/ha),

was located over twice as far from openings (18 m) and characteristically

Albert Bourgeois
occurred on damp sites, near (8 m) standing water. Mean tree dbh was
significantly less in nest habitat, but understory and ground cover
dominance, soil parameters and earthworm abundance were similar in the

two habitats.

ANALYSIS OF AMERICAN WOODCOCK NEST AND
BROOD HABITAT IN NORTHERN LONER MICHIGAN

By

Albert Bourgeois

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of
MASTER OF SCIENCE
Department of Fisheries and Wildlife

l976

ACKNOWLEDGMENTS

Funding was provided by the United States Fish and Wildlife Service
through the accelerated research program, contract no. l4-l6-0008-628,
and administered by the Michigan Department of Natural Resources in
cooperation with Michigan State University.

I am indebted to Dr. Harold H. Prince, committee member, and Mr.
Carl L. Bennett, Jr., wildlife research administrator with the Michigan
DNR, for their guidance and concern throughout this entire study. I
also extend my sincere appreciation to Drs. George A. Petrides and
Donald L. Beaver, committee members, for their assistance in developing
a sound research plan. I thank my entire graduate committee and Mr.
Bennett for critically reviewing and editing previous drafts of this
thesis. Dr. Walt Conley lent valuable assistance in programming the
discriminant function analysis and Dr. Henry Foth kindly offered me use
of the MSU soils laboratory facilities. Dr. John Gill provided valuable
statistical advise.

Personnel from the Wildlife Division of the Michigan DNR were help-
ful in many facets of the study. I particularly thank Andy Ammann, Tom
Prawdzik, Doug Whitcomb, and the staff of the Houghton Lake Wildlife
Research Station for their help in locating woodcock nests and broods.
Andy and Doug also critically critiqued the manuscript. Mr. Dean
Armstrong, Engineering Division, prepared all the figures for this thesis.

I thank Bruce Batt, Rick Kaminski, Ed Langeneau, and Dale Rabe for

reviewing earlier drafts of the thesis.

ii

I owe a debt of gratitude to Molly, my English setter, without
whose help this project would have been impossible.

Andy and Ellen Ammann deserve my sincerest thanks for their encour-
agement and concern througout the study and for being my second family
away from home and making life more pleasant.

Lastly, I am forever grateful to my Mom and brother, Roger, for

their love, understanding and continual encouragement throughout my life.

TABLE OF CONTENTS

 

 

 

 

 

Page
LIST OF TABLES ...................... . . . v
LIST OF FIGURES ......................... vi
INTRODUCTION ............. . . . .......... l
STUDY AREA ........................... 4
METHODS ............................. 7
Locating Nests and Broods . ............... . 7
Habitat Evaluation ........ . . . ......... . 9
Statistical Analysis ................... . ll
RESULTS ............................ . l3
Vegetational Composition ................. . l3
Habitat Characteristics .................. l7
DISCUSSION ........................... 22
RECOMMENDATIONS. . ....................... 26
Appendix A. Percent occurrence of tree species at American
woodcock nest and brood sites in central northern
lower Michigan, 1974 ................ 28
Appendix B. Percent occurrence of understory species at
American woodcock nest and brood sites in central
northern lower Michigan, l974 ........... 29
Appendix C. Percent occurrence of ground cover species at l6
American woodcock nest sites and 19 brood sites
in central northern lower Michigan, l974 ...... 30
LITERATURE CITED ...................... . . 3l

iv

LIST OF TABLES

Table Page

l Searching effort, sites located per habitat type, and
locations found per unit effort for the woodcock banding
season 15 April to l5 June l974, in northern Lower Mich-

igan ........................... 8
2 Importance values for the major tree species found at

American woodcock nest and brood sites in northern Lower

Michigan, l974 ...................... l4
3 Percentages of l6 American woodcock nest sites and l9

brood sites in northern Lower Michigan at which major tree l6

species occurred together .................
4 Importance values for the major understory species found

at American woodcock nest and brood sites in northern

Lower Michigan, l974 ................... l6

5 Mean (: 95% CI) and scaled eigenvector coefficients of UP
I for parameters measured at American woodcock nest and
brood sites in northern Lower Michigan, l974 ....... l8

LIST OF FIGURES

Figure Page
1 Study area in northern lower Michigan .......... 5
2 Frequency distribution of discriminant function scores
for parameters measured at woodcock nest and brood
Sites .......................... 20
3 A graphical representation of woodcock nest and brood
habitat ......................... 23

vi

INTRODUCTION

Habitat manipulation is an important technique used in wildlife
management to encourage higher population densities of particular spe-
cies. For habitat manipulation to be effective the habitat requirements
of the target species must be understood. Most early studies concerned
with the habitat needs of bird species primarily stressed the vegetative
composition of the habitat. During the past 20 to 30 years researchers
investigating habitat selection and bird species diversity have become
aware of a correlation between bird species and specific structural
characteristics of the environment, at least during breeding.

Lack (1933, 1944, 1949) first pioneered this idea and concluded
that interspecific competition has led to adaptations that allow the
individual to detect the most favorable environment for survival (Hilden,
1965). Baker (1938) used the terms ultimate and proximate to delineate
two particular types of environmental factors which influence habitat
selection. This has led to the current view of habitat selection as
described by Hilden (1965): that choice of breeding habitat is released
by certain characteristics of the habitat (proximate factors) which
guide the birds to an environment meeting their ecological requirements
(ultimate factors). Hilden (1965) defined some possible ultimate factors
as being 1) food, 2) structural and functional characteristics of the
Species, and 3) shelter, while proximate factors included 1) landscape;
2) terrain; 3) nest, song, lookout, feeding and drinking sites; 4) other

animals; 5) food; and 6) internal motivation.

2

Recent studies of avian habitat selection have focused attention on
identifying habitat preferences among and between species. Studies by
MacArthur and MacArthur (1961), MacArthur £E.El: (1966), Klopfer (1963,
l965), Sturman (1968), Cody (1968), James (1971), Whitmore (1975) and
others have shown that many passerine species are capable of selecting
habitats or micro-habitats within the environment based on specific cues
in the habitat. The environmental characteristics useful in selecting
specific habitats are related to the life form of the vegetation and the
characteristic structural profile of the habitat, and not the species
composition of the vegetation. In view of these findings, the habitat
requirements of many species, as determined by earlier workers not aware
of these relationships, need to be re-examined.
.)

The habitat requirements for breeding for the American woodcock

(Philohela minor), as determined by earlier studies indicated that wood-

 

cock utilize a variety of cover types throughout their breeding range.
However, these studies have emphasized descriptions of vegetative compo-
sition and little attention has been given to specific habitat structure.
It was generally agreed that woodcock nested near the edges of openings
in young, open second-growth hardwoods, mixed hardwoods and brushland
cover and that broods utilized similar habitats (Blankenship 1957,
Liscinsky 1972, Mendall and Aldous 1943, Pettingill 1936, Sheldon 1967,
Weeden 1955, and Wenstrom l974). Vegetative composition within these
habitats can be alder (Alflu§_spp.), gray birch (Betula populifolia),
white birch (B, papyrifera), aspen (Populus Spp.), and balsam fir (Ab1g§_
balsamea) in the Northeast (Mendall and Aldous 1943, Sheldon 1967) or

aspen, red maple (Acer rubrum), young pines (Pinus spp.) and dogwood

 

(Cornus spp.) in up-state New York (Pettingill 1936), or sweet gum

3

(Liquidambar styraciflua), greenbriar (Smilax spp.), oaks (Quercus spp.)
and pines in Alabama (Causgy__t_al, l974). Liscinsky (1972) noted that
woodcock nesting habitat in Pennsylvania was dominated by hawthorn
(Crataegus spp.), crab apple (Malu§_spp.), elm (Ulmu§_spp.), alder,
aspen, and dogwood. In Michigan, Blankenship (1957) found woodcock nests

in stands dominated by aspen, white birch, juneberry (Amelanchier laevis),

balsam fir, gray dogwood (Cornus racemosa), tamarack (Larix laricina),

 

and hawthorn. Recently, Wenstrom (1974) reported that five radio-equipped
woodcock broods in Minnesota exhibited a preferential use of aspen, birch,
mixed deciduous and mixed conifer-deciduous cover types throughout the +
breeding season.

These studies indicate that the vegetational species composition of
woodcock breeding habitat is highly variable and that more basic, and
perhaps less variable parameters must be considered in order to delineate
specific habitat requirements. However, there is a paucity of data on
the structure of the habitat within the various cover types utilized by
woodcock. Assuming woodcock recognize particular habitats within the
environment, then it may be possible to identify the habitat needs of
the species. The purpose of this study was to: 1) determine the environ-
mental configuration of woodcock nest and brood habitat, and 2) determine

whether habitat structure differs between woodcock nest and brood sites.

STUDY AREA

The study was conducted in the northern portion of Michigan's Lower
Peninsular, in Kalkaska, Missaukee, Roscommon, and Gladwin Counties and
on High and Beaver Islands which are part of Charlevoix County (Figure
l). The study was confined to state-owned lands within these counties.

Logging was the dominant industry in the mainland counties from
about 1870 until the late 1920's. Much of the area was covered with

white pine (Pinus strobus) and red pine (P, resinosa) forests and pine

 

logging dominated from 1870 to about 1895.' As the pine forests were
depleted logging activity declined until about 1905 when hardwood log-
ging gained prominence and continued to about 1925. Limited clearing
of forest land for agriculture also occurred during this time. As the
logging era ended these areas made a slow transition to resort and
tourist trade (Carlson 1973). Today these counties are primarily resort
areas with outdoor recreational opportunities that attract many people.
Pulp logging occurs to some extent throughout these mainland counties
and much of the land is in aspen, with cut-over forests, burned areas
and abandoned farmland supporting young second-growth stands. Mature
oaks, jack pine (P, banksiana), and white pine stands occur on the drier
uplands. Soil types are primarily well-drained and consist of loamy
sands, sandy loams and sand. Swamps and wet swales are found in the
lowlands.

Land use practices on High and Beaver Islands have followed a se-
quence of logging, burning, clearing, farming and abandonment (Whitcomb

4

Lake Super/or

8' A.

UPPER PENINSULA

Lake

   

' RAW
Huron

 

 

    
  

 

ISLAND g BEAVER /

I /
SLAND/

 

PENINSULA

Michigan

 

Figure 1. Study area in northern lower Michigan

6
1974). Portions of Beaver Island are still farmed and limited logging
occurs, but the major industry of the island is tourism. High Island,
which is owned by the state, has not been permanently inhabited since
the mid 1950's. Much of the land on these islands is reverting back to
mature hardwood forests of sugar maple, beech (fagu§_spp.), aspen, red

maple and white birch.

METHODS

Locating_Nests and Broods

 

The vicinities of active singing grounds, previous banding locations
and favorite fall hunting coverts were searched between 15 April and 15
June 1974 with the aid of pointing dogs, following methods described by
Ammann (1973). Searching was concentrated within 50 m of the edge of an
opening, since previous studies indicated that woodcock nests and broods
are most frequently found along the edges of forest clearings (Gregg
l974; Mendall and Aldous 1943; Sheldon 1967). Greatest searching effort
occurred during the first three weeks of May, the normal peak of the
woodcock hatch in central Michigan (Ammann 1967). Aspen habitats were
searched extensively, since aspen was the predominant forest type on
the study area. Mixed hardwoods (maple, oak, white birch, and aspen),
coniferous swamps, alder coverts and clearings (associated with aspen)
were also investigated (Table l).

A location was considered to be a nest site if a woodcock hen was
observed nesting there or if after locating a newly hatched brood (1 day
old chicks) a nest with hatched eggs was found nearby. In this event
the brood location was disregarded and the nest site was used in the
analysis of nest habitat. All other brood locations were analyzed as

brood habitat.

Table 1. Searching

8

effort, Sites located per habitat type, and

 

 

 

 

locations found per unit effort for the woodcock banding
season 15 April to 15 June 1974, in northern Lower Michigan.
Hours Sites located Locations
Habitat type searched Nest Brood per hour
Aspen '58 10 9 0.33
Alder 4 2 3 1.25
Clearings 8 3 O 0.38
Conifer swamp 2 l 2 1.50
Mixed hardwoods 20 0 5 0.25

 

Habitat Evaluation

 

After a nest or brood site was located information on vegetational
composition, habitat parameters and soil characteristics was obtained.
Each site was divided altitudinally into three habitat strata: overstory,
understory, and ground cover. These were sampled using a nested quadrat
procedure (Phillips 1959). Square quadrats of 300 m2, 50 m2, and 10 m2
were employed to sample the strata, respectively. The location of each
nest or brood site served aS the center of each quadrat. Quadrats were
centered by measuring half the diagonal distance in each cardinal direc-
tion from the nest or brood contact point. The overstory included all
woody species greater than 3.0 m in height, while the understory con-
tained all plant species, alive or dead, ranging in height from 0.3
to 3.0 m. Ground cover comprised any vegetation alive or dead, ranging
up to 0.29 m above the forest floor. I

Parameters measured at each site included: vegetational species;
dominance of the overstory, understory and ground cover; diameter at
breast height (dbh) and total height of trees; overstory stem density;
height of the ground vegetation; distance to the nearest opening and
the nearest standing water; earthworm abundance; soil texture and acid-
ity. Overstory dominance was expressed as basal area and was computed
using tree diameters as described by Phillips (1959). Total height and
dbh were recorded by species for all trees over 3.0 m tall using a Haga
altimeter and a metal diameter tape. Understory and ground cover
dominance were determined by recording a visual estimate of the project-
ed ground cover created by stems, leaves, twigs, etc. for each species.
Mean ground cover height was determined by measuring the vegetation

height with a meter stick at 15 randomly selected points within the

10
ground cover quadrat. Distance to the nearest opening and the nearest
standing water were determined by pacing. Any break in the forest canopy
0.1 ha or greater was considered an opening, while standing water ranged
anywhere from a stream to a small puddle. A general description of the
topography and the leaf litter was also noted for each site.

Importance values were computed for all overstory and_understory
species according to methods described by Curtis and McIntosh (1951).

The method involves calculating relative measures of density, dominance,
and frequency for each species and combining them into one "importance
value” for that species. Such values were then used to rank the dif-
ferent species as to their importance within each habitat. The maximum
possible importance value a species can have is 300, since relative
density, relative dominance, and relative frequency each have a maximum
value of 100. species having importance values less than 30 are general-
ly considered of minor importance within the habitat.

To investigate soil characteristics five samples were taken at each
site using a No. 10 tin can depressed to a depth of 7.6 cm--the maximum
depth it is reported that a woodcock can probe (Krohn 1969). One sample
was taken at the center of the quadrats and one in each of the four car-
dinal directions 1.5 m from the center. The numbers of earthworms were
counted after screening with a quarter-inch wire mesh shifter. Soil
texture and acidity were determined in the laboratory using the Bouyucous
method and a Ph meter, respectively (Foth and Jacobs 1964). Vegetation
analysis and soil sampling were accomplished as soon as possible after

each brood was contacted or each nest hatched.

11

Statistical Analysis

 

Standard t-test and chi-square procedures, as described by Sokal
and Rohlf (1969), were used to evaluate differences in individual para-
meters between woodcock nest and brood sites. All percent data were
transformed for statistical analysis using the arcsin transformation.
Significance of parameters was tested at the 0.05 probability level un-
less otherwise indicated.

Ten parameters thought to be factors that might influence nest and
brood habitat selection were collectively subjected to a multivariate
discriminant function analysis to evaluate the dissimilarity between
nest and brood habitat. The objective of discriminant function analysis
is to assign individuals to a group on the basis of data that are related
to the group and to determine which parameters best reveal dissimilari-
ties between groups (Lachenbruch 1975). The analysis develops an equa-
tion which is constructed in such a way that it defines a linear axis
through the data sets (nest and brood site characteristics) which maxi-
mizes the between-group variation. It provides an optimum procedure for
separating the habitats mathematically while permitting a linear order-
ing of the groups such that their separation on the discriminant func-
tion axis is a function of their differences in habitat characteristics
(James 1971). The maximum number of discriminant functions computed
depends on the relative sizes of G (the number of groups) and P (the
number of variables measured). When G - l is less than P, then G - 1
is the maximum possible number of discriminant functions (Cooley and
Lohnes 1971). In the present analysis G = 2, and P = 10, therefore

only one discriminant function (OF I) was computed. The discriminant

12
function program was modified from Cooley and Lohnes (1971) and was run

on the CDC 6500 computer at the Michigan State University computer center.

RESULTS

During the study 92 hours were spent searching for woodcock nests
and broods. Sixteen nest sites and nineteen brood sites were located.
Although the greatest searching success occurred in the coniferous
swamp and alder habitats, there was no significance difference (X2 = 4.86,
df = 4, P >0.05) between the number of hours searched in each habitat
and the number of woodcock nests or broods contacted by habitat type

(Table l).

Vegetational Composition

 

Woodcock hens utilized several early-successional forest stands
both for nesting and brood rearing. These habitats were characterized
by large numbers of saplings (2.5 cm 3_dbh >10.2 cm) and seedlings
(dbh >2.5 cm), the majority of trees ranging from 1.3 to 8.9 cm dbh and
generally less than 9.0 m in height. Aspen was the most important tree
species in both nest and brood habitat (Table 2). Alder, red maple,
balsam fir and dead stems were commonly associated with aspen and im-
portance values for these species indicate that they were dominant spe-
cies along with aspen in both habitats. A complete list of tree species
found at nest and brood sites and their percent occurrence iS presented
in Appendix A.

Comparison of importance values between nest and brood habitats

suggests a shift in importance values between species. Aspen and alder

13

14

Table 2. Importance values for the major tree Species found at American
woodcock nest and brood sites in northern Lower Michigan,

 

 

 

1974.
Species Nest habitat Brood habitat
(n = 16) (n = 19)
Aspen 120 87
Alder 37 28
Dead 21 39
Red maple 19 34
Balsam fir 19 22
White birch 8 13
Black cherry 8 ll
Juneberry 7 14

Other species 61 (n = 15) 52 (n = 12)

 

15
increased in importance in nest habitat while red maple, balsam fir, and
dead stems were less important there.

Comparisons of the major tree species (Table 3) revealed Similar
associations at woodcock nest and brood Sites. Aspen was associated
similarly with both alder and balsam fir. However, there was a closer
association between aspen and red maple at nest sites than at brood
sites. Alder was found with red maple and red maple with balsam fir,
yet alder did not occur with balsam fir.

Importance values for all understory species were low, with only

bracken fern (Pteridium aquilinum) and goldenrod (Solidago spp.) exceed-

 

ing a value of 30 (Table 4). Bracken fern, aspen, brambles (Rubu§_spp.),
goldenrod and blueberry (Vaccinium spp.) were the most important species
in both nest and brood habitats among all the species considered. A
complete list of the understory species found at nest and brood sites

and their percent occurrence are presented in Appendix 8.

There was a shift in the importance of some plant species between
nest and brood habitat. Bracken fern, aspen, sweet fern (Comptonia
peregrina), meadowsweet (Spjr§a_spp.), willow (Saljx_spp.) and red
maple appeared to increase in importance in nest habitat while brambles,

goldenrod, blueberry, alder, black cherry (Prunus serotina), red-osier

 

dogwood (Cornus stolonifera), bush honeysuckle (Diervilla lonicera) and

 

 

beaked hazelnut (Corylus cornuta) decreased in significance.

 

Typical plants found in the ground layer were grasses, dewberry

(Rubus spp.), wintergreen (Gaultheria spp.), wild strawberry (Fragaria

 

spp.), mayflower (Maianthemum canadenses), blueberry and bracken fern.

 

A complete list of ground cover species and their percent occurrence is

presented in Appendix C.

16

Table 3. Percentages of 16 American woodcock nest sites and 19 brood
sites in northern Lower Michigan at which major tree species
occurred together.

 

 

 

 

 

 

Alder Red maple Balsam fir
Species Nest Brood Nest Brood Nest Brood
Aspen 21.4 26.3 50.0 31.6 28.6 26.3
Alder 7.1 10.5 0.0 0.0
Red maple 21.4 26.3

 

Table 4. Importance values for the major understory species found at
American woodcock nest and brood sites in northern Lower
Michigan, 1974.

 

 

 

Species Nest habitat Brood habitat
(n = 16) (n = 19)
Bracken fern 40 22
Aspen 21 16
Brambles 19 29
Goldenrod 19 36
Sweet fern l8 5
Blueberry l4 l7
Meadowsweet 12 9
Willow 12 7
Red maple 10 9
Alder 10 16

Black cherry 9
Red-osier dogwood 7
Bush honeysuckle 6 15
Beaked hazelnut 6
7

Other species 9 (n = 23) 75 (n = 15)

 

17

Habitat Characteristics

 

Topography at all sites was relatively flat, but with small hummocks
and shallow depressions occurring throughout. Low areas were usually
damp or held standing water. All nests were located on dry sites and
Situated at the base of or within 0.3 m of a small tree or other type
of vertical cover. Leaf litter comprised much of the ground cover in
both habitats and consisted primarily of deciduous leaves and to a less-
er extent dead grass and bracken fern.

Discriminant function analysis revealed a significant discrimina-
tion (P >0.5) between woodcock nest and brood habitat. Parameters having
the highest values for the scaled eigenvector coefficients contributed
most to the dissimilarity between habitats.

Basal area around brood sites was nearly double that at nest sites
(Table 5). This was the single most influential parameter separating
the two habitats. Next in importance were the number of trees over 9.0 m
in height and the distance that woodcock nests or broods were found from
water. Woodcock brood habitat was characterized by a greater number of
trees over 9.0 m tall, while nest sites were found significantly farther
from standing water. The correlation matrix computed by the discriminant
analysis revealed a close correlation (r = 0.81) between basal area and
the density of trees over 9.0 m in height.

Distance to the nearest opening, the number of trees 6.1 to 8.8 m
in height, and the mean ground cover height all differed significantly
between nest and brood habitat, but contributed to a lesser extent than
the previous variables in separating the two habitats (Table 5). Wood-
cock nests were located near the edges of old fields and clearings,

whereas broods were over twice as far from openings. Brood habitat had

18

co_pucz$ ucccwewgommv mo ppzmwaurmo.ov a

.mwmxficcm

u AIR .mwmapmca “may-“ we mppammm--Fo.ov a u I, .mo.ov a u I
.mwmxpmcm pcmcwswcum_v mgu cu wFwaLm> mcu yo cowuanwgucou m>wumpmmm

 

arkmmmm.m u u

N —

.vm n mu .o_ u mu

 

F.N Am.m_ - 0.0V N.P_ Am.m_ - o.mv P.0F me Lag mocmucznm ELOZESEHU gem:
N.@ Am? - omv mm AFE - omv m4 La>oo scoumtauca Scootaa
N.m Ame - _NV _m Awm - mmv mm La>oo ucaoea “Emotaa
P.__ A¢.FP - m.mv 0.0. Am.m~ - m.__v m.m_ I AEUV Lm>ou nczoem any co Seawa; cam:
m.¢_ Amm - NV m_ A__ - my A R, AEV mcwcaao “magma: as» ou mocmpmwo
o.qm Am_ 1 av m Aom I NFV _m « Asv Loam: mcwucmum
ummgwm: ms“ ow museumwo
N.o Am_om - N_m_v oomm AQNFN - wmmv Ame, u;m_a; c? a m.m - o.m
m; can moms» eo Lansaz
©.N_ Amm¢_ - __Nv NADA Aomm - ¢N_v “mm Iagmwm; cw E m.m - _.o
a: gag mmmcu $0 Lamazz
m.om Amom - Nomv mmm Aomm - wm_v Rom “swam; cw E o.mx
a: emu women we consaz
o.N¢ Am.om - A.N_V m.o_ A_.m_ - P.4v o.w «,Am;\msv mate mewm
mpcmwuwmwmou AmP u cv Am” n :V memEmcmm
Louum>cmm_m wpwm uoocm mpwm ummz
mafiaom

 

 

snowcmE< um umczmmme

.qum— .cwmwcuwz LmZOJ :szpLO: c_. mmuwm vOOLn Ucm “mm: xUOUUOOZ

mcmumEmcma LOU H do we mpcmwowewmoo Logom>cmmwm cmpmum wcm aHo xmm AV cow:

.m wFQmH

19
almost three times the number of trees in the intermediate height class
as nest habitat, while ground vegetation surrounding nest Sites was
slightly taller than in brood habitat. Percent understory cover, percent
ground cover, earthworm abundance, and the number of trees in the height
range 3.0 to 5.8 m did not differ statistically between nest and brood
habitats and offered little discriminatory power (Table 5).

Discriminant function scores for woodcock nest and brood sites were
computed using a standardized grand mean of 50 and a standard deviation
of 10 (Cooley and Lohnes 1971) in order to allow comparison of the var-
ious habitat characters measured. A frequency distribution of these
scores shows the dissimilarity between habitats (Figure 2). Sites with-
in the 46 - 55 range (37 percent of the brood sites and 19 percent of
the nest Sites) could not be assigned to either habitat classification
with much confidence.

Previous studies have reported stem density to be an important hab-
itat component in diurnal woodcock covers (Liscinsky 1972, Wenstrom
1973). This parameter, however, was highly correlated with several of
the parameters used in the discriminant analysis and was therefore ana-
lyzed separately. Tree density was significantly lower in nest habitat
(2176 trees/ha) than in brood habitat (3934 trees/ha). In addition,
mean tree diameter differed significantly between habitats, with trees
around nest Sites averaging 5.6 cm dbh and those at brood sites averag-
ing 6.7 cm dbh.

The mean number of trees in the three height categories (Table 5)
was tested to determine if the proportion of trees within each class

was similar between woodcock nest and brood habitat. A disproportionate

FREQUENCY

 

 

20

  

NEST (N: I6)

BROOD (N = l9)

  

 

26-35 36 -45 46-55 56-65 56-75
DISCRIMINANT FUNCTION SCORES

 

 

 

Figure 2. Frequency distribution of discriminant function scores for
parameters measured at woodcock nest and broad sites.

21

increase in trees occurred in broad habitat, primarily in the intermedi-

ate height category (X2 45, df = 2, P <0.05).
Soil properties were Similar in nest and broad habitat. Sandy loam
occurred at 92 percent of the nest sites and 77 percent of the broad
Sites, while loamy sand was found at 8 percent of the nest sites and 15
percent of the broad sites. Clay loam comprised the soil type at one

site. Mean soil ph was 5.2 at nest sites and 4.9 at brood sites and did

not differ significantly.

DISCUSSION

Previous studies have reported that woodcock utilize similar habi-
tats as nest and broad cover (Blankenship 1957, Mendall and Aldous 1943,
and Sheldon 1967). However, few studies have been concerned with
structural aspects of the habitat and little quantitative data is avail-
able on woodcock habitat characteristics. It was apparent in this study
that the habitat requirements of the American woodcock are more complicated
than once believed and involve many environmental components and behavioral
traits.

Woodcock hens evidently select somewhat different habitats for nest-
ing and brood rearing. While similar in species composition, these hab-
itats are structurally different (Figure 3). In brood habitats, tree
density and basal area was almost double that of nest sites. Although
this increase may be due in part to nests being located close to the
edges of forest openings, there was a change in habitat structure be-
tween sites. Tree density in the lowest and highest height classes
increased approximately 60 percent from nest to broad habitat, while
trees in the intermediate class increased 200 percent. Woodcock hens
appear to select sites with greater tree densities in the 6.1 to 8.8 m
height class to raise their broods. A shift to denser habitats by wood-
cock broods may have two benefits. First, reduced ground cover under
a dense overstory may allow easier movements of young chicks. Percent

ground cover in fact, was slightly less in broad habitat, although

22

HEIGHT ABOVE GROUND (m)

23

 
  
  

1' b" WOODCOCK NEST
HABITAT

 

 

 

 

 

 

 

 

2?: wooacacx BROOD
HABITAT

 

 

 

 

 

I L I I I I
5 K) I5 20 25 30 35 4O 45 50
DISTANCE T0 WATER (m)

Figure 3. A graphical representation at woodcock nest and broad habitat.

24
highly variable. Secondly, greater tree density may offer greater pro-
tection against avian predation as broods forage on the forest floor.

In addition to differences in tree distribution and density woodcock
brood habitat was located closer to standing water than was nest habitat.
Woodcock need dry sites to nest. Thus, hens would be expected to select
sites in the spring that are first to be free of snow and well drained.
The site characteristics plus the use of openings by singing males make
the edge of a forest a likely site. The average 7 m distance of a nest
from a forest Opening is similar to the distances reported by Blankenship
(1957), Mendall and Aldous (1943), Sheldon (1967) and Wenstrom (1973).

In contrast, broods averaged over twice as far from openings as nests

and were found relatively close to standing water. Sheldon (1967) stated
that woodcock hens usually "lead their broods to poorly-drained areas
where the chicks can probe for worms and other invertebrates." Wenstrom
(1974) believed that litter insects may serve as a primary food source
through the first three weeks of the woodcock's life. Wenstrom also
noted that radio-equipped woodcock broods in Minnesota showed little
affinity for habitat edges. Movements which result in broods avoiding
openings may relate to increased protection in denser habitat and better
feeding opportunities for woodcock chicks.

Height of the ground cover also contributed to the dissimilarity
between nest and broad habitats, although less important than the pre-
viously discussed parameters. Taller ground cover in nest habitat affords
the hen more protective cover as she Sits on her nest but may limit
movement by young woodcock chicks.

Differences in structure between nest and broad habitats indicates

that the habitat needs of woodcock hens, at least during the breeding

25
season, are complex. Future management decisions must consider these
habitat differences if management is to be effective. Tree density and
basal area, tree distribution, distance to the nearest opening and the
nearest standing water, and height of the ground vegetation were apparently
important distinguishing characteristics of woodcock nest and brood hab-
itats. Other factors, not measured in the present study, may also contri-
bute significantly to the habitat needs of the woodcock. Further detailed
analysis of woodcock habitat structure and vegetative characteristics is
needed in order to delineate the habitat requirements of the American
woodcock and identify important habitat parameters for sound management
of the species. In addition, quantitative data is needed on brood move-
ments and habitat utilization, predation, food habits of young woodcock,
food supply and those factors affecting it, and optimum nest and broad

densities.

RECOMMENDATIONS

In good reproductive years, when woodcock are abundant, they may
nest and raise broods in sub-optimal as well as Optimal habitat. When
papulations are low, however, only optimal habitats may be occupied. It
is difficult to determine in a one-year study whether the habitats sampled
represented prime habitat or were complicated by sub-optimal habitat
characteristics. Where possible, therefore, habitat studies should be
of long duration.

Biologists interested in creating or improving woodcock habitat need
data on management techniques and woodcock response to these techniques.
The present study was a quantitative description of woodcock habitat and
did not test management techniques. The data do, however, suggest some
possible recommendations which may be applicable to management of woodcock
breeding habitat.

Young, second—growth hardwoods, particularly aspen, were utilized
extensively by woodcock hens during the breeding season. Similar habitat
should be selected for management. Forest Openings should be interspersed
throughout this habitat, since the majority of woodcock nests and broods
are found within 50 m of the edge of an opening (Gregg 1974). These open-
ings also serve as singing grounds for males (Sheldon 1967). If openings
do not exist in the habitat, small clearings should be created. These
Openings should be located on dry sites, but close to low, poorly-drained

areas .

26

27
The data also indicate that manipulation of tree densities may
improve habitats for nesting and broad rearing. Within 10 m of an open-
ing the tree density should be approximately 2100 trees/ha and consist
primarily of seedlings and saplings less than 6.0 m in height. From 10
to 30 m of the edge of an opening tree density should approximate 4000

trees/ha and consist of mainly trees less than 9.0 m in height.‘

APPENDICES

28

Appendix A. Percent occurrence of tree species at American woodcock
nest and broad sites in central northern lower Michigan,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1974.
Species Nest Site Brood Site
I" = 15) (n = 19)

Aspen (Populus spp.) 81 79
Dead trees 56 89
Red maple (Acer rubrum) 44 42
White birch (Betula papyrifera) 25 26
Alder (Alnus spp.) ’ 25 37
Balsam fir (Abies balsamea) 25 42
Black cherry (Prunus serotina) 25 . 16
Willow (Salix spp.) 25 21
Juneberry (Amelanchier laevis) 19 32
White pine (Pinus strobuS) 19 10
White cedar (Thuja occidentalis) 19 10
Jack pine (Pinus banksiana) 12 5
Yellow birch (Betula alleghaniensis) 6 5
Spruce (Picea spp.) 6 10
Oak ( uercus spp.) 6 16
White ash Fraxinus americana) 6 16
Apple (Malus spp.) 6 5
Sugar maple (Acer saccharum) 6 21
Red pine (Pinus resinosa) 6

Hornbeam (Carpinus caroliniana) 6

Hobblebush (ViBurnum alnifOTium) 6

Elm (Ulmus spp.) 6

Witch-hazel (Hamamelis virginiana) 6

 

Beaked hazelnut (Corylus cornuta) 10
Hemlock (Tsuga canadensis) * 5

 

 

 

29

Appendix 8. Percent occurrence of understory Species at American wood-
cock nest and broad sites in central northern lower Michi-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

gan, 1974.
Species Nest Site Brood Site
(n = 16) (n = 19)

Aspen (Po ulus spp.) 56 37
Red maple (Acer rubrum) 50 32
Brambles (Rubus spp.) 50 42
Juneberry (Amelanchier laevis) 44 37
Willow (Salix spp.) 44 26
Rose (Rosa spp.) 38 10
Black cherry (Prunus serotina) 33 47
Bracken fern (Pteridium aquilinum) 31 47
Beaked hazelnut (Corylus cornuta) 31 21
Grass (Gramineae) 25 16
Bush honeysuckle (Diervilla lonicera) 25 32
Honeysuckle (Lonicera sppf) 25 5
Sweet fern((Comptonia peregrina) 25 21
Goldenrod Solida o spp.) 25 26
Meadowsweet (S irea spp.) 25 21
Red-osier dogwoo Cornus stolonifera) 19 21
Alder (Alnus spp.) 19 32
Blueberry (Vaccinium spp.) 19 26
Viburnum (Viburnum spp.) 19 10
Sugar maple (Acer saccharum) 19 10
Balsam fir (Abies balsamea) 12 26
White ash (Fraxinus americana) 12 5
Chokeberry (Pyrus spp.) 12 10
Jack pine (Pinus banksiana) 6 5
False solomons seal (Smilacina spp.) 19

Oak (Quercus spp.) 12

White birch (Betula papyrifera) 12

Witch-hazel 0~I_amamel is virginiana) 12

White cedar (Thuja occidentaliS) 6

Apple (Malus spp.) 6

 

Larch (Larix laricina) 6
Bitter Nightshade (Solanum dulcamara) 6
Wild sarsaparilla (Aralia nudTCaulis) 6
Fire cherry (Prunus pennsylvanica) 6
6
6
6

 

 

 

 

Cinquefoil (Potentilla spp.)
Sumac (Rhus sppCTS
Burdock (Arctium spp.)

 

 

 

Dogwood (Cornus spp.) 10
Ground hemlock (Taxus canadensis) 10
White pine (Pinus strobus) 5
Sedge (Carex spp.) 5

Horsetail (Eguisetum spp.) 5

 

30

Appendix C. Percent occurrence of ground cover Species at 16 American
woodcock nest Sites and 19 brood sites in central northern
lower Michigan, 1974.

 

 

 

 

 

 

 

 

 

 

 

Species Nest Site Brood Site
(n = 16) (n = 19)

Grass (Gramineae) 94 63
Dewberry (Rubu§_spp.) 31 26
Wintergreen (Gaultheria spp.) 31 21
Strawberry (Fragaria spp.) 31 21
Goldenrod (Solidago spp.) 31 21
Blueberry (Vaccinium spp.) 19 26
Bracken fern (Pteridium aquilinum) 12 21
Clubmoss (Lycopodium spp.) 12 21
Aster (Astgr_spp.) 12 5
Bush-honeysuckle (Diervilla lonicera) l2 5
Mayflower (Maianthemum canadenses) 6 21
Sweetfern (Comptonia peregrina) 6

Horsetail (Eguisetum Spp.) 6 5
Poison ivy (Rhus radicans) 6 5
Violets (Viola_spp.) 10
Trillium (Trillium spp.) 5
Prince's pine (Chimaphila umbellata) 5
Dandelion (Taraxacum officinale) 5

 

False solomons seal (Smilacina spp.) 5

 

LITERATURE CITED

31

LITERATURE CITED

Ammann, G. A. 1967. Woodcock banders newsletter, No. 2. Mich. Dept.
Conserv. Information Circular No. 145. 6 pp.

1973. Methods of capturing American woodcock broods. XI
International Congress of Game Biologists, Stockholm, National
Swedish Environment Protection Board. Editor, Sven Lundstrom.
631 pp.

Baker, J. R. 1938. The evolution of breeding seasons. 6. R. deBeer:
Evolution. Essays on aspects of evolutionary biology presented to
E. S. Goodrich. 161-177. Oxford.

Blankenship, L. H. 1957. Investigations of the American woodcock in
Michigan. Mich. Dept. Conserv. Rep. 2123. 217 pp.

Carlson, B. W. 1973. The Houghton Lake Area. Pendell Publishing Co.,
Midland, Mich. 47 pp.

Causey, K., J. Roboski, and G. Horton. l974. Nesting activities of
the American woodcock (Philohela minor Gmelin) in Alabama. Pre-
sented at the Fifth Woodcock WorkShop, Georgia. 13 pp.

 

Cody, M. L. 1968. On the methods of resource division in grassland
bird communities. Amer. Naturalist 102:107-147.

Cooley, W. W., and P. R. Lohnes. 1971. Multivariate data analysis.
Wiley, New York. 364 pp.

Curtis, J. T., and R. P. McIntosh. 1951. An upland forest continuum
in the prairie—forest border region of Wisconsin. Ecology 32:
476-496.

Foth, H. D., and H. S. Jacobs. 1964. Laboratory manual for introductory
soil science, Wm. C. Brown CO. Publishers, Dubuque, Iowa. 82 pp.

Gregg, L. E. 1974. Woodcock nesting habitat in northern Wisconsin.
Presented at the Fifth Woodcock Workshop, Georgia. 5 pp.

Hilden, 0. 1965. Habitat selection in birds. Annales Zoologici
Fennici Vol. 2:53-75.

James, F. C. 1971. Ordinations of habitat relationships among breeding
birds. Wilson Bull. 83:215-236.

32

Klopfer, P. H. 1963. Behavioral aspects of habitat selection: the
role of early experience. Wilson Bull. 75:15-22.

1965. Behavioral aspects of habitat selection: a preliminary
report on stereotypy in foliage preferences of birds. Wilson Bull.
77:376-381.

Krohn, W. B. 1969. Feeding habits of woodcock using summer fields in
central Maine. Presented at the Third Woodcock Workshop, Maine.

15 pp.

Lachenbruch, P. A. 1975. Discriminant analysis. Hatner Press, New
York. 128 pp.

Lack, D. 1933. Habitat selection in birds, with special reference to
the effects of afforestation on the Breckland avifauna. J. Anim.
Ecol. 2:239-262.

. 1944. Ecological aspects of species formation in passerine
birds. Ibis 86:260-286.

1949. The significance of ecological isolation. G. L.
Jepsen, E. Mayr and G. G. Simpson. Genetics, paleontology and
evolution. Princeton. 299-308.

Liscinsky, S. A. 1972. The Pennsylvania woodcock management study.
Pa. Game Comm. Res. Bull. 171. 95 pp.

MacArthur, R. H., and J. W. MacArthur. 1961. On bird species diversity.
Ecology 42(3):594-598.

, H. Recher, and M. Cody. 1966. On the relation between habi-
tat selection and species diversity. Amer. Naturalist 100:319-332.

Mendall, H. L., and C. L. Aldous. 1943. The ecology and management of
the American woodcock. Maine Coop. Wildl. Res. Unit, Univ. Maine,
0rono. 201 pp.

Pettingill, 0. S. 1936. The American woodcock, Philohela minor (Gmelin)
Mem. Boston Soc. Natur. Hist. 9:169-391.

 

Phillips, E. A. 1959. Methods of vegetation study. Henry Holt & Co.,
Inc., New York, 107 pp.

Sheldon, W. G. 1967. The book of the American woodcock. Univ. of
Massachusetts Press, Amherst. 227 pp.

Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co.,
San Francisco. 776 pp.

Sturman, W. A. 1968. Description and analysis of breeding habitats
of the chickadees Parus atricapillus and P, rufescens. Ecology
49(3):418-431.

 

33

Weeden, R. B. 1955. Cover requirements of breeding woodcock in central
Maine. M.S. Thesis, Univ. Maine.

Wenstrom, W. P. 1973. Habitat utilization and activities of female
American woodcock (Philohela minor Gmelin) in northeastern Minne-
sota during spring and summer, Ph.D. Thesis, Univ. Minn., St. Paul.

203 pp.

 

Whitcomb, D. A. 1974. Characteristics of an insular woodcock population.
Mich. Dept. Nat. Res., Wildl. Div. Rept. No. 2720. 78 pp.

Whitmore, R. C. 1975. Habitat ordination of passerine birds of the
Virgin River Valley, southwestern Utah. Wilson Bull. 87:65-74.

~.Mh, MW

‘-..AJ_ __1, A ‘2‘. -'

HICHIGQN STATE UNIV. LIBRARIES

 

31293182159989