.A STUDY OF WATERFOWL RI LOGY AT i’ONDS

BLASTED IN MAWITOBA’S DEH’R MARSH

Thesis fur. the Degree of M. S.
mzcmem STATE UNVERSITY
RONALD H. HGFFMAN
1967

..vvx‘gz[i‘“‘:‘ ‘

‘ MIN!!! ILIIQIIJWI LIN (m m HM unwilling I}; ll ;

 
  

 

Midi?“ C‘7.'ce.te.';.-

Universgcy

 

‘1

ABSTRACT

A STUDY OF WATERFOWL ECOLOGY AT PONDS BLASTED
IN MANITOBA'S DELTA MARSH

by Ronald H. Hoffman

Forty—one ponds were blasted with ammonium nitrate~
in a prairie marsh. During 33 weeks of field work (fall
of 1964 through 1966) quantitative measurements were made
of pond morphometry, soils, waters, macrophytes, inverte-
brates, and waterfowl use.

Changes in morphometry indicated a minimum life of
30 years for the 5 x 26 x 52-foot ponds. High water loss
contributed to increased salinization of pond soils and
waters. A few large storms added more water to the ponds
than the same amount falling in many small showers. Ponds
within 300 feet of the bays received seepage water. The

macrophyte, Potomogeton pectinatus, and the invertebrates,

 

Daphnia and Tendipedidae, dominated the aquatic community.
Plants common to the surrounding meadow invaded the pond
shoulders. During the first two years, the quality and
quantity of macrophytes and invertebrates increased in
the ponds.

The waterfowl species composition of the ponds was

characterized by few diving ducks and a large percentage

Ronald H. Hoffman

of blue-winged teal (Anas discors). Use by ducks followed

 

a pattern of highest abundance early in the morning during
spring. Duck selection of certain ponds was not signifi-
cantly influenced by measured variations in pond age,
food, cover, water levels, or bay-to-pond distance. The
pattern of pond location in relation to other features in
the marsh seemed to influence waterfowl distribution at
the ponds.

Duck nests were concentrated along the bays before
pond construction. After blasting the ponds in a uniform
pattern, duck nests were more evenly distributed. Both
pond proximity and vegetation seemed to affect nest density.
Severe nest predation resulted from predators being at—
tracted to the ponds.

The 25 ponds appeared to increase the carrying
capacity by 35 breeding pairs. Waterfowl abundance and
available habitat probably modified the degree of response.
Actual production was far less than expected from the
breeding population because of severe nest predation. The
ponds chiefly functioned as isolation areas for breeding
pairs, so other production requirements must be offered

in nearby areas.

A STUDY OF WATERFOWL ECOLOGY AT PONDS BLASTED

IN MANITOBA'S DELTA MARSH

By

Ronald H. Hoffman

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

1967

ACKNOWLEDGMENTS

I was indeed fortunate to have the counsel and ex-
perience of Dr. Miles D. Pirnie of Michigan State Uni-
versity provided in supervising this study.

Financial assistance was furnished by the Wildlife
Management Institute, Ducks Unlimited (Canada), Delta
Waterfowl Research Station, and Michigan State University.
I am sincerely grateful to each of these organizations.

Dr. H. Albert Hochbaum and staff of the Delta Water-
fowl Research Station made helpful suggestions and offered
a stimulating atmosphere for conducting the study. Peter
Ward, manager of the Bell Estate, afforded access to the
private land in the Delta Marsh. Dwight D. Moore and
William R. Miller of the Delta Marsh Development Committee
supplied data concerning waterfowl populations in the East
Marsh. James Rutledge assisted with field work in 1966.
Early spring duck censuses were conducted in 1965 by David
P. Mossop and Dwight D. Moore. They also helped blast
ponds and aided on numerous other occasions. William G.
Leitch and Charles H. Lacy of Ducks Unlimited (Canada)
visited the study area several times and offered recom-
mendations. Theodore Laidlaw, University of Manitoba,

helped with soil and water analyses. Much of the nesting

ii

data was collected in cooperation with Gerry M. Lynch of
the University of Wisconsin. To each I am obligated.

I am indebted for advice offered by Dr. Kenneth J.
Linton in statistics, Dr. Jethro O. Veatch in soils, and
Dr. Robert C. Ball and Dr. Niles R. Kevern in limnology
of Michigan State University.

I would like to thank Dr. Rollin H. Baker, Dr. John
E. Cantlon, and Dr. Leslie W. Gysel of Michigan State Uni-
versity for their critical reading of the manuscript and
their comments. Dr. H. Albert Hochbaum of the Delta Water-
fowl Research Station also examined sections of the manu—

script.

iii

TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS . . . . . . . . . . . . ii

LIST OF TABLES . . . . . . . . . . . . vi

LIST OF FIGURES . . . . . . . . . . . . viii-
Chapter

INTRODUCTION . . . . . . . . . . . . 1

History of Subject . . . . . . . . 1

Nature Of St Udy o a o o o 0 o o o 3

Study Area . . . . . . . 3

Waterfowl Population Levels . . . . 9

METHODS O O O O O I O O O O 0 0 O 0 ll

Pond Construction. . . . . . . . . ll

Abiotic Measurements. . . . . . . . 12

Sampling of Vegetation . . . . . . . 13

Sampling of Invertebrates . . . . 14

Waterfowl Investigations . . . . l6

FINDINGS AND ANALYSES . . . . . . . . . 18

Changes in Pond Morphometry . . . . . 18

Construction Costs . . . . . . . . 22

Water-level Fluctuations . . . . . . 22

Soils and Waters of Study Area . . . . 23

Vegetation . . . . . . . . 29

Invertebrates of l— and 2-year Ponds . . 36

Waterfowl Populations . . . . . . . A3

Duck Behavior at the Ponds. . . . . . 69

Nesting . . . . . . . . . . . 72

Duck Production . . . . . . . . . 79

iv

Chapter
DISCUSSION . . . .

Census Methods .
Factors Limiting

MANAGEMENT . . . .
SUMMARY .
LITERATURE CITED .
APPENDIX A . . . . .

APPENDIX B . . . . .

Duck Production

Page
814

8A
86

91
95
100
108

109

Table

10.

11.

12.

13.

1“.

LIST OF TABLES

Results of Pond Morphometrical Measurements

Results of Pond Soil Analyses. . .

Results of August Pond and Bay Water
Analyses. . . . . . . .

Plant Species Composition of the Meadow
Pond Shoulders. . . . . . . .

Average Stem Length of Meadow and Pond
Shoulders . . . . . . . . .

Aquatic Plant Species Composition at l-
2-year Ponds o o o o o o o 0

Comparison of l- and 2-year Pond Bottom
ples for Invertebrates . . . . .

and

Sam-

Comparison of 1- and 2-year Pond Net Tows

for Invertebrates. . . . . . .

Comparison of the Duck Species Compositions
of the East Marsh, the Study Area Bays,

and 27 Blasted Ponds. . . . . .

Comparison of June 1965 and 1966 Dabbler Duck

Numbers in the East Marsh and at 1?
Artificial Ponds . . . . . . .

The Blue-Winged Teal Populations Composition

for 25 Artificial Ponds. . .
Second Lead Breeding Pair Population .

Comparison of Duck Use at l- and 2-year
Ponds O 0 O O O O 0 0 0 O 0

Summary of a Comparison of l- and 2—year

Ponds. . . . . . . . .

vi

Page

21
26

27

30

32

3A

39

A0

A8

51

58

59

61

Table Page

15. Comparison of Pond Locations and Breeding
Pair Use . . . . . . . . . . . 65

16. Time Ducks Spent Per Pond-Visit, Feeding,
and Loafing While at the Ponds. . . . 68

17. Comparison of the Distance from Duck Nests
to the Bays Before and After Pond Con-
struction. . . . . . . . . . . 73

18. Comparison of Duck Nest Densities at an
Experimental Area with Artificial Ponds

and at a Check Area Without Ponds. . . 7A
19. Comparison of Nest Fates at Experimental

and Check Areas. . . . . . . . . 77
20. Potential Duck Production of 25 Ponds . . 82

vii

Figure

l.

10.

11.

l2.

13.

1A.

15.

LIST OF FIGURES

Map of a Portion of the East Marsh, Includ-
ing the Study Area . . . . . . .

Map of Second Lead Study Area Showing Pond
Locations . . . . . . . . . .

Map of Study Area Showing Distribution of
Four Primary Cover Types . . . . .

Yearly Abundance of Ponds and Breeding
Waterfowl in Southern Manitoba . . . .

Changes in Pond Morphometry After Two Years.
Photographs of Ponds . . . . . . .

Relationship of Rainfall to Bay and Pond
Water Levels . . . . . . . . .

Effect of Bay-to—Pond Distance on Pond-
Water Loss. . . . . . . . . . .

Comparison of l- and 2-year Pond Inverte-
brate Biomass. . . . . . . . . .

Species Composition of Ducks in the East
Marsh . . . . . . . . . . .

Species Composition of Ducks at Study Area
Bays 0 O O O O O I O O O 0

Species Composition of Ducks at Artificial
Ponds . . . . . . . . . .

Seasonal Abundance of Ducks . . . .

Graph Showing Poor Association Between
Breeding Duck Use and Invertebrate Bio-
mass at the Ponds . . . . . . . .

Diurnal Duck Activity Pattern at the Ponds

viii

Page

lO

19
20

2A

25

38

”5

A6

“7

5A

6A
70

INTRODUCTION
History of Subject

A major concern of North American conservationists
is the loss of waterfowl habitat. Recurring droughts and
expanded drainage programs have reduced our wetlands
(Sayler, 1962). In addition, the quality of many wetlands
is threatened by the increasing demands of our society
(Jahn and Hunt, 196A).

Steps are being taken to ameliorate this threat to
waterfowl habitat.‘ Vast programs of wetland acquisition,
leasing, and development are being tried. Presently ac-
quisition and leasing are stressed. However, as the price
of land increases, wetland development will receive addi-
tional attention (Uhler, 1956).

Pond construction, one form of wetland development,
has been based on the findings of various workers. Leopold
(1933:131) described game as a product of edge. Hochbaum
(191M), Sowls (1955:53), Evans and Black (1956), McKinney
(1965), and others have suggested territorialism as limit-
ing the carrying capacity for breeding pairs of some local
areas. Studies of natural potholes by Evans g£_al. (1952),

Evans and Black (1956), and Glover (1956) have showed an

increase in breeding ducks after additional precipitation
made more water areas available.

A variety of methods have been used to construct
waterfowl ponds. Pirnie (1935:292) mentions broods using
small open water areas resulting from peat fires. Mechani-
cal methods include--draglines (Mathiak and Linde, 1956;
Lacy, 1959; Hammond, 196A; Mathiak, 196A), bulldozers
(Mathiak, 196A), and backhoes (Mathiak, 196A). Scott and
Dever (19AO), Provost (19A8), Mendall (19A9), and Mathiak
(196A) used dynamite to blast small ponds. Recently,
ammonium nitrate has been used instead of dynamite (Mathisen
gt_al., 196A; Mathisen, 196A and 1965).

Most of these studies have been concerned with methods
of construction and not with waterfowl use. For example, of
25 papers I reviewed on blasting, only six dealt.with bio-
logical studies. Provost (19A8) concluded that dynamited
ponds in an Iowa sedge marsh were attractive to migrant
blue-winged teal, but had dubious value as an inducement to
nest. In Maine, Mendall (19A9) reported black ducks in-
creased from four to 1A pairs after dynamiting ponds. Lacy
(1959) and Hammond (196A) found about one pair of ducks per
pond in North Dakota. Workers in Minnesota (Mathisen, 196A
and 1965) and Wisconsin (Mathiak, 196A) report duck use of
ammonium nitrate blasted ponds in areas having little use
previous to blasting.

Currently ammonium nitrate-blasted ponds are being

emphasized. Several marsh development plans include large

appropriations for blasting (Westerberg, 1965; Byrant,
1965). In addition, exhibitions of blasting are creating

widespread public interest.

Nature of the Stqu

 

This study was initiated to gain a clearer under-
standing of blasting as a habitat development technique.
My objectives were to examine waterfowl use of ammonium
nitrate—blasted ponds and to compare use with food and
cover differences.

The study was a cooperative two-year project of the
Delta Waterfowl Research Station and Michigan State Uni-
versity. The Delta Marsh, its previous and current water-
fowl studies, financial support, and the station's labora-
tory facilities provided an excellent opportunity to carry
out the field work. Planning and writing the manuscript

were conducted at the University.

Study Area
Along Lake Manitoba's south shore lies the Delta
Marsh (Fig. 1). Extending as far as five miles inland it
borders the Lake for nearly 2A miles. The Delta Marsh has
been separated into three parts; West Marsh, East Marsh,
and Lake Francis Marshes. This study was undertaken in
the East Marsh which extends from the towns of St. Ambroise

on the East to Delta on the West; a distance of 11 miles.

.moouficwz wpaom mmma .mohw mUSpm omoq ocooom on» wcfiosaoca swam: ummm
on» mo coHuLOQ m cam swam: maaoQ on» mo cofipmooa on» wcfizonm dwzln. H maswfim

 

 

 

. (wad >03hm
/
D<MJ 0200mm

  

 

 

<m0.....2<_>_ m¥<4 430.5222).

 

 

Today Lake Manitoba is but a small remnant of glacial
Lake Agassiz. Formed by the retreating Wisconsin glacier,
this ancient lake's lacustrine sediments cover Devonian
limestones (Ellis, 1938; Johnston, 193A). The Delta Marsh
lies upon a onetime lakebed, called the First Prairie
Steppe (Ellis, 1938).

During recent times ice and wind have formed a series
of sandy ridges along Lake Manitoba's south shore. In
addition, the ancient Assiniboine River and several smaller
streams, especially the Portage River, have deposited their
loads of alluvium in~a meandering pattern of natural levees.
The result which one sees today is a land mass with a com-
plex of shallow bays, sloughs, and ponds less than six feet
deep. I

Delta has a humid continental climate with cool sum—
mers. Winnipeg (60 miles southeast) records a mean annual
temperature of A5.3 degrees Fahrenheit and 20.1 inches
mean annual precipitation (Dominion Government Department
of Transport).

As Hochbaum (19AAz6) has pointed out, the Delta Marsh
has a history of fluctuating water levels. During periods
of drought as much as half of the marsh may become dry,
but during wet years most of it has been under water.

Starting at the lakeshore and moving southward one
encounters a series of vegetative types; a narrow strip
of ridge forest, a belt of phragmites encompassing most of

the marsh, next come haylands, and finally grainfields of

the Portage Plains (Hochbaum, 19AA; and Love and Love,
195A). Although the marsh is covered chiefly by phragmites,
many water areas are bordered by bulrushes; and the inter-
mittently flooded areas usually are whitetop meadows. The
East Marsh thus conforms to Reid's (1961:13) definition
of a marsh as "a broad wet area on which abundant grasses
and sedges grow."

Most of this study was conducted on a 78—acre penin-
sula seven miles east of Delta, (T.1A N, R.6 WPM, sec. 25);
an area often referred to as the Second Lead. A sandy
road enters the area from the North. Here ponds S-l thru
8-28 were blasted (Fig. 2) and shall be referred to as an
experimental nest study area. The intrazonal soil (Finch
g§_§1., 1957) had a profile of a thin humus layer covering
strata that ranged from a texture of sandy loam to clay loam.

A whitetop meadow covered 60 percent of the study area.
The ground water table was near the surface only in bulrush
and mixed emergent vegetation cover types (Fig. 3).

A 20-acre nesting check area was located to the east
of First Bay (Fig. 2). During the spring the check area
was separated from the rest of the marsh by flooded shallows.
The soil and vegetation of the check area resembled those of
the Second Lead study area.

Ponds were blasted in several other locations. Ponds
T-l thru T-8 were blasted three miles east of Delta, B-l

1-1/2 miles east of Delta, and R—l thru R-A 1-1/2 miles

 

 

.CLAIR
LAKE

EBCIIRI.EE

J db5¢i F=1r.

 

 

Figure 2.--Map of Second Lead study area showing pond
locations, 1966, Delta, Manitoba.

 

 

   

o 650 Ft
FOREST AND SHRUB-

WHITETOP

V
MIXED EMERGENTS
BULRUSH -

 

 

 

 

Figure 3.—-Map of study area showing distribution of four
primary cover types, 1966, Delta, Manitoba.

south of Delta. (Additional information is given in

Appendix A).

Waterfowl Population Levels

 

The abundance of water areas and waterfowl in south-
ern Manitoba has fluctuated in the past (Fig. A). Aerial
censuses during May, covering nearly 39,000 square miles
of southern Manitoba (Delta Marsh included), give a yearly
index of pond and duck abundance (Smith and Droll, 1966:66).
From 1956 through 1966, the pintail, (scientific names in
Appendix B), blue—winged teal, lesser scaup, shoveller, and
the mallard have fluctuated more than the gadwall and green—
winged teal (Smith and Droll, 1966). During 1965, the total
number of ducks was 27 percent below the l2-year average,
and the number of natural ponds was 23 percent above the
average (Smith and Droll, 1965). In 1966, the total number
of ducks was 20 percent below the 13-year average and ponds
36 percent above (Smith and Droll, 1966). Blue—wings were
60 percent below the long—term average (op. cit.). Thus,
this study was undertaken during a period of above-average

water conditions and below-average duck abundance.

10

 

   
  
    
 

d
hJ

DUCKINDEX

l J L A J

 

Lug - \ (1x 100000)
2 P /’/
<I ' ///I 1
QB. /
g r.POND INDEX L — I

L
m3_ (1x 100) 1
< .

ad

 

 

CD

 

1956 1958 1960 1962 1964 1966
Figure A.--The yearly abundance of ponds and breeding water-
fowl in southern Manitoba (Smith and Droll, 1966).

METHODS

Pond Construction

 

All of the study ponds were blasted with Amex, a
commercial mixture of ammonium nitrate and fuel oil. For
the remainder of this paper they will be referred to as,
"the ponds."

Pond construction followed the procedures and safety
recommendations of Mathisen gt_al. (196A), Mathiak (1965),
Radtke and Byelich (n.d.), and Spencer Chemical Co. (1963).
Three holes, 16 feet apart and three to four feet deep,
were dug in a straight line. Each end of a length of
detonating cord (about 20 feet) was attached to a half
stick of 50 percent ditching dynamite and then placed in
the center of a 50-lb. bag of Amex. One bag was lowered
into each hole and packed with dirt. Thus each bag of
Amex was connected by detonating cord. Then one blasting
cap, either fuse-type or electrical, was attached to the
detonating cord. The material was then ready for deton—
ation.

Ponds S-l through 8—17 were blasted in August 196A
and ponds S-18 through 8-25 in August 1965. Ponds S-l

through 8-17 were called 1-year ponds in 1965 and 2-year

11

12

ponds in 1966. Ponds S—l8 through 8-25 were considered

l—year ponds in 1966.

Abiotic Measurements

Ponds S—l through 8-25 were measured for width,
length and depth. A line was stretched between two perma-
nent markers across the longitudinal axis of the ponds.

By lowering a ruler at 2—foot intervals along this line,
pond depth was measured to the nearest inch. During 1965
and 1966 morphometrical measurements were made in July of
2—year ponds and in August of l-year ponds. Weekly water-
level records were kept for two ponds in 1965, six ponds
in 1966, and for First Bay during both years.

Water tests were made of First Bay, Second Bay, and
of randomly selected ponds. A Hach colorimeter (Hach, 5th
ed.) was used to test for nitrite, nitrate, ammonium,
ortho and meta phosphate, turbidity, and calcium during
August 1966. In August of 1965 and 1966, specific con-
ductance was measured with a Radiometer conductivity meter
and pH with a Beckman model N meter. The limit of visi-
bility in pond waters was measured with a Secchi disk
'(Welch, 1958:159-160) every month during the summer.

Aquatic soil samples were collected from 10 randomly
selected ponds during August 1966. The samples were air—
dried and then stored in plastic bags until January. The
Soil Testing Laboratory at Michigan State University

analyzed the samples for various nutrients and pH with

13

the Spurway technique (Spurway and Lowton, 19A9); and for
percent organic matter with the Walkely-Black method

(Jackson, 1960:219—221)

Sampling of Vegetation

 

Aquatic Vegetation

Pond plants are described in terms of percent fre-
quency and weight. All plants growing within the exca-
vations were considered aquatic, even though water levels
declined drastically during the summer. Frequency refers
to presence in a sampling unit (Brown, 195A:2A).

The aquatic vegetation in eight l-year and eight 2-
year ponds was sampled during July and again in August.
Two 6- x 2A-inch plots were randomly located in each pond,
one in the middle and one near the edge. All plants in
these plots were collected with an Ekman dredge, placed in
plastic bags, and returned to the laboratory where they
were sorted to species, air-dried, and then weighed on a
trip-balance scale to the nearest 0.1 gram. Frequency was
based on these plots in 1965; but in 1966, the frequency
sampling unit included the entire excavations of ponds S-l

through S-25.

Terrestrial Vegetation

 

Meadow and pond "shoulder" vegetation is described
in terms of percent frequency, percent coverage, and stem
length. Pond "shoulder" or edge arbitrarily included a

6-foot wide area surrounding the excavation. Coverage

1A

was defined as "the amount of ground covered by an indivi-
dual species" (Brown, 195A:A2). Length refers to the above—
ground stem projection and in most cases was equivalent to
plant.height.

Terrestrial vegetation was measured with the line-
intercept method (Canfield, 19Al) using 2-foot plots.
Forty-eight randomly located plots were sampled during July
and again in August on the shoulders of eight 1-year and.
eight 2-year ponds. Twenty plots were randomly selected
from a 950-foot line bisecting the Second Lead meadow near
ponds 8-10 and 8-7. Twenty plots were similarly chosen
from a 700—foot line in the check nesting area. Meadow
vegetation was sampled only during July.

Plant identification was based on Gleason (1952),
Scoggan (1957), Fassett (1960), and comparisons with
identified specimens in the Delta Waterfowl Research Station

herbarium.

Sampling of Invertebrates
Invertebrates were sampled with a tow net and an
Ekman dredge. A 3A-meSh per inch net with a 9 l/2-inch
diameter was pulled for a distance of 6.9 feet (approx.
200 liters) through the waters of four l-year and four 2-
year randomly selected ponds. A bottom sample was col—
lected from the edge and middle of the same ponds with a

l/A square-foot Ekman dredge. Dredge contents were

l5

strained through a 30-mesh per inch screen and the residue
placed in quart jars along with a few milliters of undiluted
formalin.

Permanent sampling sites were randomly located in
the ponds. Net tows were always taken between the same
two markers, whereas a particular bottom site was never
sampled more than once. Net tows were taken May 12, and
then every two weeks from June through August of 1966.
Bottom samples were collected every two weeks from July 13
through August 11 of 1965 and June through August 10 of
1966.

In the laboratory, invertebrates were analyzed ac-
cording to the procedure described by Welch (19A8z303-306).
Animals were sorted from the remaining detritus in a white
porcelain pan, volumetric measurements made with a gradu-
ated centrifuge tube, and then separated into taxonomic
groups. If a sample contained less than an estimated 500
individuals, direct counts were made. If the sample con-
tained more than an estimated 500 individuals, volumetric
measurements were made of the taxa and the number of indi—
viduals estimated from the volume. Pennak (1953), Usinger
(1963), and personnel of the Delta Waterfowl Research
Station aided identification. Once processed, all samples

were placed in labeled vials with a preservative.

l6

Waterfowl Investigations

 

Each week ducks were counted on the study area. The
counts were made during the periods of May A through
August 31 in 1965 and April 23 through August 25, 1966.
The usual procedure was to count ducks on the bays first
(which seldom flew) and then to systematically check each
pond from which ducks almost always flushed. This minimized
duplication of counts. The censuses were made between 8:00
and 10:00 A.M. The species, sex, number, and location of
all ducks seen were recorded on field forms.

Each week, personnel of the Delta Marsh Development
Committee made low-altitude aerial duck censuses of the
marsh. By flying 60 miles an hour over half-mile east-
west transects, two observers were able to give complete
coverage. Counts began on June 9 in 1965 and on May 10 in
1966 and continued through the study period. All counts
were completed before 11:00 A.M. Except for green-winged
teal and blue-winged teal which were grouped into "teal,"
ducks were recorded by species, or when identification was
uncertain, as "undetermined."

A 15-foot observation tower was erected on the study
area. This concealment made it possible to watch several
ponds from one location. Records were kept of the time
undisturbed ducks spent on top of the pond shoulders, at
the water's edge, in the water, or feeding. In addition,

general behavior associated with pond use was noted.

17

Various techniques were used to investigate the
effects of pond construction on duck nesting. The Second
Lead area was burned in August 196A and again in 1966.

The check area was also burned only in 1966. After burn-
ing, the number of old duck nests and their locations
were recorded for the periods before and after pond con-
struction.

A chain, with tin cans attached, was dragged between
two vehicles in a systematic search for active nests (Sowls,
1950). The study areas were dragged twice during 1966,

May 27-30 and June 20—21. Some.nests were discovered inci-
dental to other investigations. All nests were marked by
placing consecutively numbered lath 5-paces south of the
nest. Species were determined by identification of flushed
hens, egg size and shape, and nest down (Broley, 1950).
Weekly checks of the nests revealed clutch size and nest—
ing success. "Hair catchers" (single wire with barbs)

were placed near each nest to help determine predators
visiting those nests destroyed. The condition of egg-
shells and the nest were used to ascertain nesting success
or the cause of nest loss (Kalmback, 1937; Sowls, l9A8;

Rearden, 1951)-

FINDINGS AND ANALYSES

Changes in Pond Morphometry

 

Morphometrical measurements were taken to determine
the rate of siltation in newly created ponds. One month
after blasting, the ponds averaged five feet deep, 25.6
feet wide, and 52.3 feet long. During the next two years
the steep-sided shoulders eroded into the ponds (Figure 5).
After two years the mean pond length had increased A.3
feet, width increased 0.7 foot, and depth had decreased
four inches. Most of the dimensions overlap at the 95
percent confidence interval (Table l).‘ The ponds increased
in length more than in width because most of the spoil was
thrown parallel to the longitudinal axis of the ponds.

The rates of siltation can be affected by soil type,
pond shape, water level, vegetation, muskrats, etc. (Pro-
vost, l9A8, and-others). At Delta, muskrats seem to have
the greatest potential for increasing siltation. Muskrats
did not use any of the ponds measured, but elsewhere their
use substantially decreased depth (Figure 6). High muskrat
use could drastically reduce the duration of the ponds.

From my findings it is difficult to predict accu-

rately the life expectancy of the ponds. Provost (19A8:

l8

19'

    

 

 

 

0P\ / '1
A ONE-MONTH POND /
1::3 20 L ‘
p. .
~ I: 40 ’
‘0
ll ,_° ‘.
a E 9"
v l l L L l
10 20 30. 40 50

LENGTH (FEE T)

Figure 5.--Changes in pond morphometry after two years.
Spacing of ammonium nitrate charges caused the two
mounds in the pond bottom, 1966, Delta, Manitoba.

20

pm

.BQOpficmz .mpHmQ .mwma .mBCOQ

.Ucoq m pm poHHm>onm
oco cam Ham» commazumSHn @0929 .Q

 

3m3< mama Ca ocoo smmmlm HmOfidze .m

one mo chMLwOBOLmII.w opsmam

.mpmnxmse ho commobocfi coapmpafim ocom

 

.zmz mapmm CH ocoo Lacuna Hmofidze

.0

.¢

 

21

TABLE 1.--Results of pond morphometrical measurements.
Confidence limits are at the 95 percent level, 1966,
Delta, Manitoba.

 

 

 

 

 

Pond Age (Months) 1 ll 23
Number of Samples 8 1A 1A
Length (feet) 52.6 1 1.A 5A.6 1 2.0 56.6 1 1.3
Width (feet) 25.6 1 1.7 26.6 1 1.8 26.3 1 1.5
Distance from Edge Depth (inches)
2 ft. interval 16 1 2 16 1 2 21 1 2
A " " 35 i 31 i 3 33 f 2
6 " " 50 i “3 i 3 A5 1 2
8 " " 60 1 3 51 1 3 52 1 2
10 " " 68 1 2 61 1 5 60 1 3
12 " " 7A 1 69 1 A 65 1 3
1A " " 75 1 71 1 3 69 1 A
16 " " 71 1 3 69 1 3 66 1 3
18 " " 65 1 7 6A 1 5 62 1 3
2o " " 58 1 3 6o 1 5 61 1 A
22 " " 67 1 3 61 1 6 62 1 A
2A " " 72 1 3 66 1 5 65 1 A
26 " " 73 i 5 67 i A 67 j 3
Mean 60. 56.1 56.0
% Mean Loss -- 7.0 7.3

 

22

36A) found that the average depth of ponds dynamited in
peat decreased 35 percent during the first five years with
a decelerating rate of siltation in later years. The
average depth of ponds at Delta decreased 7.0 percent after
one year and showed no substantial decrease the following
year (Table 1). Strohmeyer (1963) studying Provost's

ponds 16 years later also found a deceleration in the rate
of siltation and believed that they might persist for "25
to 30 years or more." Since the Delta ponds had a lower

initial rate of siltation, they may well last 30 years.

Construction Costs

 

The average pond in this study cost $20.00. Of this
amount ammonium nitrate accounted for $13.00, other materials
$A.00, and labor $3.00. If the ponds persist for 30 years
the prorated construction cost would range between $0.60

and $0.70 per pond per year.

Water-level Fluctuations

In contrast to bays, ponds usually receive most of
their water from precipitation (Hochbaum, 19AA; Evans
g£_a1., (1952). From June 16 through August 25, Delta had
3.63 inches of rain in 1965 and A.25 inches in 1966. Dur-
ing the same periods pond S—3 lost 27 inches of water in
1965 and 52 inches in 1966. First Bay lost only four to
five inches both years. In 1965, there was less total

rainfall, but more storms with rain of a half-inch or

23

more than in 1966 (Figure 7). Therefore, it appears that
many small rain showers supplied less water to the ponds
than the same amount falling in fewer, but more severe
storms.

Although the ponds obtained most of their water from
precipitation, bay-pond distance also influenced water
levels (Figure 8). Ponds within 300 feet of the bays re-
ceived seepage inversely proportional to their distance
from the bays. Beyond 300 feet the bays contributed little

seepage water to the ponds.

Soils and Waters of Study Area

Soil samples were collected to give an index of
chemical composition of the ponds. The Spurway technique
does not indicate the availability of nutrients. Further-
more, Kadlec (1960 50) justly criticized air drying of
aquatic soil samples. Therefore, results of soil analysis
(Table 2) are offered chiefly as an index to the general
level of nutrients.

Soil tests showed little difference in the chemical
composition of l- and 2-year ponds (Table 2). The 95 per-
cent confidence limits of chloride, magnesium, nitrogen,
and pH did not overlap, but their differences are not con—
sidered biologically significant to this study.

Tests of bay and pond waters showed differences in
mean turbidity, pH, specific conductance, and calcium con-

tent (Table 3). Turbidity was higher in the bays than in

'2A

 

 

 

 

 

 

m o 1 fl f
4A
1' ID
a:
.4
g; 21)
u:
'2
3: 130
.4
.4
1<
u.
z. 1.0
‘: 0.5
a .
4-1!)
0’
.4
nu
>'
u:
.4
“I
In
F.
'1
3
.4
.4
g 1.0». .
E 005 "
‘1
¢ MAY JUNE 3 JULY AUGUST

Figure 7.-~Relationship of rainfall to water-level fluctua-
tions of First Bay and pond S-3. Data expressed in inches.
Rainfall measurements recorded by the staff of the Delta
Waterfowl Research Station, Delta, Manitoba.

WATER LOSS UNCHES)

25

 

 

 

 

 

‘o‘_\BAY .
\
\
20’ \ ‘
\
30- \ -
40- .
so- .
6 160 260 360 400

BAY-POND DISTANCE (FEET)

Figure 8.-—The effect of bay—to-pond distance on
pond-water loss, August 25, 1966, Delta, Manitoba.

26

TABLE 2.--Results of pond soil analyses, 1966, Delta,
Manitoba. Confidence limits are at the 95 percent level.

 

l—Year Ponds 2-Year Ponds

 

Number of Number of

 

Samples Samples
A 6

% Organic matter 11.8 1 7.2 7.A 1 2.9
pH 8.0 1 0.0 8.2 1 0.0
Nitrogen (ppm) A 1 3 15 1 6
Sulphate (ppm) 1100 1 1500 800 1 500
Sodium (ppm) 1050 1 350 980 1 AAo
Calcium (ppm) 560 1 70 620 1 30
Chloride (ppm) 260 1 130 90 1 10
Potassium (ppm) 101 1 16 76 1 30
Magnesium (ppm) 17 1 0 18 1 0
Manganese (ppm) A 1 3 5 1 0
Phosphorus (ppm) 2 1 0 1 1 0
Iron (Ppm) tr tr

 

27

.homnsdom mHnmCOHpmo:@**

.heaes Heflofioese coaxoo0e

 

 

 

 

I I I Asdqv
0.0 + 0.H III **0.0 + m.m 0.0 + 0.0 III III who: .oeoedhoem
I I I AEQQV
H.0 + m.0 III *e0.0 + H.H 0.0 + H.0 III III oneso .oeoeomoem
0 H em III AH H HNH mm H mHH III III Asoov eeoHHoo
0.0 H 0.H I H 0.0 m.0 H m.m 0.0 H 0.0 0.0 H 0.0 m.0 H 0.0 Amoeesv Hefiefieosoeoo
00.0 III 00.0 00.0 III III Assoc oeaeofiz
0.0 H m.m III 0.0 H H.m 0.0 H H.m III III Asodv operHz
0.0 H 0.0 III 0.0 H 0.0 0.0 H 0.H III III Asoov ssaeosee
H.0 H m.0 I H 2.0 0.0 H H.0 0.0 H 0.0 H.0 H 0.0 0.0 H 0.H mo
Hm H 00 III 0 H 00 am H Hm III III *A090V HeHofiopee
III III 0 H mm m H 0: III III A.eH0 xoflo Heooom
m H 0 0 0H m oNHm oaosom
000A mmmfi 000a 000A m00a mead seem
HmowIm HmowIH HmmeH npcozIH
mmmm mucom
.Ho>oH Hcoonom mm 02H Hm ohm mHHEHH mocooncoo
.moopHcmz .muHmo .momzamcm Hopmz amp cam onca pmsws< Ho muHSmmmII.m mamHB

28

the ponds, while pH, calcium, and specific conductance
were lower in the bays.

Wave action and water levels may explain the major
differences noted between bay and pond waters. Because
of their small size, the ponds received less wave action;
and therefore were less turbid.

The drastic reduction in pond water levels (Figure 7)
during the summer and reflooding the following spring not
only may have concentrated, but also increased the dis-
solved nutrients. Cook and Powers (1956) noted an inverse
relationship between total alkalinity and water levels and
suggested that evaporation may have been responsible.
Kadlec (1960:51) believed that drying reduced the colloidal
fraction of sediments, releasing adsorbed ions which re-
sulted in an increase of dissolved nutrients upon reflood-
ing. Reid (1961:181) and others have stated the inter-
relationship involving water loss, the carbon dioxide-
carbonate system, calcium, and hydrogen ion concentration.

Testing showed differences in l- and 2-year pond
waters (Table 3). Phosphate concentrations and pH were
higher in the 2-year ponds, while specific conductance,
nitrate, ammonium, and visibility (Secchi disk) decreased.
Phosphate readings in the 2-year ponds are of questionable
accuracy because such high concentrations are seldom en-
countered. Except for May 1966, a Secchi disk could be
seen at the bottom of the ponds. Only May disk readings

are shown (Table 3).

29

The influence of water loss on dissolved nutrients
(discussed above) may also be applicable to the comparison
of l-year and 2-year ponds. Welch (1952:372-373) de-
scribed situations where decreasing water volumes were
accompanied by changes in pH and conductivity. He
(op. cit.) suggested that precipitation of calcium and
magnesium carbonate reduced electrolytes and increased
pH. The cycle of repeated drying and reflooding could com-
pound this effect.

Pond soils and waters were saline. The differences
observed were similar to those often encountered, so are
not considered atypical of the region. After two years of
study, it is not clear if an "alkali problem," similar to
that described by Keith (1958 88), will develop. If
salinization does increase with pond aging, pond vege—

tation could be drastically altered in the future.

Vegetation

 

Meadow Community

 

Differences between 1965 and 1966 meadow vegetation
(Table A) probably were due to burning in August of 196A.

Chenopodium rubrum, Erigeron-Solidago-Aster group, and

 

Ranunculus sceleratus especially were more abundant in

 

1965. Burning did not seem to affect the frequency or

cover of Scolochloa festucacea or Atriplex patula. Burn-

 

ing did reduce leaf litter cover, but the meadow still fits

the "Scolochloetum" described by L5ve and Lave (195A).

 

N
H
H

O

30

NOLflOOO
O

IHb~H

Ln CDkDKO

(I)

HOCDOCXDO :I'
O N O \O
H

O\C\J\O
I—‘l

Ln Or—IO

:3”
H

("\JOr-IOCDO
N

mm:
[\2’

O
(I)

m.HH

m.o :

o o

o o

m.m NH
o o

w.o m

H.m mH
w.: mm
m.o m:
m.H mm

Ln
O
H
m
N
H

O

mwmnm>oo nomad Hmpoe

Honuo
mamopusm msmuHom

msEpHme xmssm
momcm>mm mssocom

 

 

 

mspmanoom msHSocscmm
.Qdm HonH

a .owwvHHom .copmeAm

sznnsn ssHooaoCmno

mHspma memme<

 

 

 

 

[\Nm :TOOKOOO
I-I:rm O

mmm Guam—woo
NOON I-lOr—IO

(\l
O
H
\O
(\l

mmoMospmom moanooaoom

AoeeHH Hood

 

Am>oouvopm

Hm>oouuomm

m:
003

mocom

Ho>oouumpm

m:

mmmH
mocom

Ho>oouvopm Ao>oouuohm

mpoaa Ho Amnssz
Ammw
mop<

 

.mnopHcmz .mpHom .mmmH .mwmno>oo cam mosmsdomg owmpcoonmd

CH oommmpmxo .onUHdonm vcoa Ucm zovmos mo QOHHHmomsoo moHooam pcmHmII.: mqm<e

31

Edge Vegetation

 

Pond shoulders varied from areas of little spoil
deposition to areas where soil was piled in mounds 2-foot
high. This variability in substrate undoubtedly affected
plant colonization. The following discussion includes
sample areas of varying depths of spoil.

Eight plant species were commonly recorded on the

pond shoulders. Scolochloa festucacea, Atriplex patula,

 

 

Chenopodium rubrum, and Sonchus arvenses had the highest

 

 

percent frequency and coverage (Table A). When stem length

is considered Scolochloa festucacea, rose two to three

 

times taller than Atriplex patula, Chenopodium rubrum, and

 

Sonchus arvenses (Table 5).

 

One-year pond shoulder vegetation differed from that

of 2-year ponds (Table A). Scolochloa festucacea frequency

 

and Atriplex patula cover increased with pond age, while

 

Chenopodium rubrum decreased. Scirpus paludosus appeared

 

 

after two years, and leaf litter substantially increased
in both frequency and cover.

Provost (19A82372) believed that plants invaded
dynamited potholes from two sources, "undisturbed marsh"
and from plants which established themselves "in the
excavations." The latter group "met the invaders from
the marsh usually at the waterline under the banks"

(op. cit.). After the first year invasion by Chenopodium

 

rubrum and other marsh plants, the pond shoulders resembled

32

TABLE 5.-—Average stem length* of meadow and pond-shoulder
vegetation, Delta, Manitoba.

 

 

 

 

 

 

 

 

 

 

Species 192§ad§g66 1-1966P0nd 21966 Pond
Scolochloa festucacea 29 31 22 35
Atr1plex patula 2 3 9 6
Chenopodium rubrum A 12 9 12
Erigeron, Solidago, &

89223 spp. 6 8 ll 2
Ranunculus sceleratus 2 -- -- --
Sonchus arvenses 6 13 ll 8
Rumex maritmus l2 -— -- --
Scirpus paludosus -- -- -- 12
Other -- —— A5 --

 

*Length in inches.

33»

the surrounding meadow. However, after only two years.
of study, it is not clear whether the vegetation on the

shoulders will become part of the Scolochloa community or

 

if it will become a separate community.

The appearance of Scirpus paludosus on the shoulders

 

of 2-year ponds may indicate a separate shoulder community.

Coupland (1950) noted that S. paludosus can tolerate high

 

concentrations of salts and grows around bare central areas.
If pond salinization increases with age (p. 29) the pond
shoulders would resemble Coupland's (1950) conditions and

thus favor a separate shoulder community.

Aquatic Macrgphytes

 

Plants quickly appeared in the ponds. Zannichellia

 

palustris and Potomogeton pectinatus were among the most

 

 

widespread (high frequency) species and comprised over 90
percent of the total weight (Table 6). Fourteen additional
species, some of which had a high frequency, were found in

the ponds. On the basis of frequency and weight, Zannichellia

 

palustris and Potomogeton pectinatus dominated the plant

 

 

community of the ponds.

Reported studies offer possible explanations for
sources of plants in newly denuded areas. Shull (191A:
337), Penfound(l953:569), and others noted that submerged
seeds will germinate after many years of dormancy and.thus
are an important origin of pioneering plants. Birds have

been known to transport aquatic plants either by adherence

.eemHoe $0.0 some omoH I ee*

.memC<

oCm HHCh mCHCCU mmmH CH mpOHC mm CCm mmmH CH mHOHQ mH Co Comma prHmz .pmswz< oCm
HHCh mCHACU mmmH CH mUCOC mCHpCo mm CCm mmmH CH mpOHC mH Co woman moCoCuome

 

3A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=.mHH m.m m.mm Amsmmwv HCmHmz Hmpoe
OOH mm mm ommHCoonmC HmHoB

m mHmeHC> mHCCQmHm

m mspmmeoo 0HomoCom

m HoCHE MCEoH

m mCHCom mCmnHom

m mCsHHCmE meCm

Hm m msmovsHmm mCQmHom

m mH mHCCEEoo mopHEwmACm

m mm mH mCoomoCmeo ECHHmCmoHCmS

0 mm eHHoCHeoH shame

H a» no moomozpmom mOHCoOHoom
0 H 0 0 NH oHCmmHae oHCoHsoHCpp

0H 0H 0 NH .om memeo

H mm om p mH mzmeonom mCHCoCCCmm

mm m mm exp mm mCoHHm> msmuHom

mm :m mm mm Hm mm memCHpooC Coumm080pom

Hm 0H H0 H0 R00 00H mHeemeHoo oHHHoeoHeeoN

 

prHoz moCmsomCm quHmz moCvaoCm prHoz zoCoCdmam
momH mme mmmH mmHooCm

 

mUCom CmowIm moCom HmmeH

 

s.mCOHHsz .mHHoQ .Amsmmwv prHmz Hmpop Ho owmquoHoC CCm zoCozumCH
CH oommmpoxo nmoCoa HmozIm CCm IH pm COHpHmomsoo mmHomCm pCmHQ OHHmCd<II.m mqm<9

35

of various plant parts to their bodies or passage through
the digestive tract (Arber, 1920:301; Schlichting, 1958).
Undoubtedly wind is also responsible for short-range dis-
persal of plants. Certainly differences in 1— and 2-year
pond flora were affected by the available sources of vege—
tative parts and seeds, and by the means of dispersal of
each species. Probably most of the pond macrophytes were
carried by ducks or came from dormant seeds already pre-

sent in the soil.

Several pond species flowered and a few produced seeds.

Ranunculus sceleratus, Myriophyllum exalbescens, Scolochloa

 

festucacea, and Sencecio congestus flowered, but I found no

 

 

fruits or seeds. Practically every clump of Potomogeton

 

pectinatus formed fruit, but few if any mature seeds.

 

Scirpus validus, S, acutus, and S. paludosus produced seeds.

 

 

Other species (Table 6) did not flower or produce fruits.

It appears that once introduced to the ponds, only the

Scirpus §223 reproduced by sexual means; and the other species
seem to have increased by vegetative means only.

Data from all aquatic plots and sampling dates were
pooled into two groups, 1- and 2-year ponds. The 1965 total
macrophyte weight of l-year ponds was nearly seven times the
1966 l-year pond weight (Table 6). This difference may have
been due to yearly variations in growth and dissimilar grow-
ing sites (l-year ponds were S-l through S—l7 in 1965 and
S—l8 through S-25 in 1966, Figure 2). The higher frequencies

noted in 1966 l-year ponds as compared to 1965 ones, were

36

mainly due to a change in the frequency sampling unit.
After two years, the ponds contained five more species and
three times the weight of the l—year ponds. In spite of
the high degree of variability during the early stage of
pond succession, the trend was towards increasing number
of plant species and quantity with pond age.

Future changes in the aquatic plant community are
difficult to predict. Bird (1930:Al2) states that suc-
cession in aspen parkland lakes progresses from various

species of Potomogeton to a stage of Scirpus, Typha, and

 

 

Carex. Natural pond vegetation in the Delta Marsh (Lave

and L8ve, 195A) indicates that a sere of Utricularia,

 

Myriophyllum, and Hippuris will follow the Potomogeton

 

 

stage. The same investigators (op. cit.) found Lemna spp.
common, but it was absent in the ponds even though the ponds

seemingly offered a possible habitat.

Invertebrates of l- and 2-Year Ponds

 

The quantity of invertebrates in the ponds was sparse
in comparison with that of the marsh. Collias and Collias
(1963), making total counts, found as many as 1,000
Daphnia per liter in a small bay near Delta. Using a
macroplankton net, the highest 1 ever found was an average
of 18 per liter.

A Chi-square two-sample case test (Siegel, 1956:
lOA-lll) showed a statistically significant difference

(p < 0.001) in the proportion of invertebrate biomass in

37

l- and 2-year ponds (Figure 9). Older ponds contained a
greater biomass of invertebrates during most of the summer.
They were more abundant in l-year ponds only during late
June and early July, when zooplankton and nekton reached

a peak. '

In contrast to biomass, the total number of inverte-
brates during the summer was similar in the two groups of
ponds, although their seasonal abundance differed (Table 7,
and 8). In general, invertebrates in 2-year ponds reached
a peak earlier in the summer. This was especially true of
benthic organisms. The 1-year ponds had a slightly higher
total number of invertebrates during the summer.

A qualitative comparison of the two pond groups offers
an explanation to the contrasting biomaSs and total number
estimates of invertebrate abundance. In net tows Daphnia
composed 97 percent of the l-year pond invertebrates, while

five taxonomic groups--Daphnia, Lymneae, Chaoborus, Physidae,

 

and Coroxidae—-comprised the same percent in 2-year ponds.
Similarly, in bottom samples only three taxa (Tendipedidae,
Daphnia, and Oligachaeta) comprised 95 percent of the younger
ponds samples, while 95 percent of the 2-year pond samples

consisted of seven taxa (Tendipedidae, Lymneae, Daphnia,

 

Oligochaeta, Spongillidae, Physidae, and Hydrophilidae).
Lymneae and Physidae, snails, made up a large portion of

the 2-year sample biomass.

'38

 

 

 

    

 

 

 

’ 2-year ponds ‘
10", /\\ .1
/ \
F / \\ . 1-year ponds -
81- ‘ -
6- -
4- .1
m h _
m .
P12.
.“ t -
m:
a:
w I
p. 8?
a:
III /\ 1
>. '71 / \
z / \
' Z'year ponds/I
IL 6*- \ .1
O / \
J St / \ -
E / \
\
4~ / 1
3b \\ / c1
\ /
2r \/ .-
1-year ponds
1' .
l l 4
MAY JUNE -JULY AUGUST
Figure 9.--Comparison of l- and 2-year pond invertebrate

biomass, 1966, Delta, Manitoba. Expressed as volume per
200—liter net tows and per one—square—foot bottom samples.

39

TABLE 7.--Comparison of l- and 2—year pond bottom samples
for invertebrates, 1966, Delta, Manitoba. Expressed in
numbers per square foot.

 

 

June July August Per-
1 15 30 13 27 10 Total cent
Tendipedidae
l—yr. 1965 -- —- -- 117 256 169 5A2 62.A
l-yr. 1966 3 68 200 161 238 50 720 A0.1
2-yr. 1966 138 151 150 9A 61 7A 668 A3.A
Daphnia ppp,
l-yr. 1965 -- -- -- Al 3A 23 98 11.3
l-yr. 1966 20 158 212 1A0 106 21 657 36.6
2-yr. 1966 29 102 16 Al 22 9 219 1A.2
Oligochaeta
l-yr. 1965 -- -- -— 0 0
l-yr. 1966 5 1A 31 56 122 102 330 18.A
2-yr. 1966 28 111 13 5 l3 1 171 11.2
Lymneae spp.
l-yr. 1965 -- -- -- 0 0
1-yr. 1966 l A 3 8 0.3
2-yr. 1966 8 8A 96 52 2A0 15.6
Physidae
l-yr. 1965 -- -— -- A 17 25 A6 5.3
l-yr. 1966 0 0
2-yr. 1966 l 3 1 22 15 A2 2.7
Spongillidae
l-yr. 1965 -- —- -- 5 30 A 39 A.5
l-yr. 1966 6 10 16 0.9
2-yr. 1966 A 16 15 9 1A 1A 72 A.7
Hydrophilidae
l-yr. 1965 -- -- -- 1A 5 l 20 2.3
l-yr. 1966 l 3 15 A 23 1.2
2-yr. 1966 7 A 21 32 2.1
Other
l-yr. 1965 -- -- -- 15 Al 68 12A 1A.1
1—yr. 1966 2 3 2 5 15 17 AA 2.5
2-yr. 1966 A 8 8 10 3A 31 95 6.2
Total
l-yr. 1965 -— -- -- 196 383 290 869 100.0
l-yr. 1966 30 2AA AA5 366 506 207 1798 100.0
2eyr. 1966 20A 391 211 250 266 217 1539 100.0

 

A0

 

 

 

N.00 00H: m0H 00N 000 00H 000 wemH 0HNH 0N CNoHIN
0.00 00H0 00H 00H HHm Hm0H 0NH HmNH 000 00 CNoHIH
Heeoe
0.H 00 HH 0H 0H 0 N H N N emoHIN
0.0 0N H m N m H 0 m CmoHIH
eoeeo
H.m HHH 00 HH m0 mH H AmoHIN
o o CmomIH
ooonHeN
0.0 00 N NH NN H H H CeoHIN
he N H H sooHIH
ompHOHCmmCooo
H.m 00H 00 0 00 0N 0N meoxIN
N.0 0 H H m m CmmHIH
2mmm mmmCeNH
0.H 00 N 0N Hm H emoHIN
H.0 HN m H 0 H H CNoHIH
mmvonCoo
0.0 NmN H0 30 mm mm N: 0N 0 NH CmomIN
0.0 0N 0 H 0H H N pooHIH
2mmm mzponomno
0.H H: N H N N 0 0N semHIN
N.H mm 00 H H H H H 0 emoAIH
mmoHCCoonpmm
0.H0 H000 HH mmH 0HN 00 000 0HmH mHNH 0H emoHIN
0.H0 H000 HNH 00 00N 0H0H 0HH NHNH 000 00 emoHIH
ammm MHCCme
eeoo Hepoe MN 0. 0N mH. 0N HH H NH
ICmm HmCmC< HHCh oCCh mm:

 

.mCmHHH oom Hod mHoCECC CH commommxm

.oooeHemz .oeHoo
.mmmH .mmmeCoHC0>CH Com m30p HoC UCOQ HmomIm CCm IH Ho ComHCmQEOOII.m mqm<e

Al

In both groups of ponds, Daphnia and Tendipedidae,
respectively, were the most numerous zooplankton and
benthic organisms. A Chi-square two-sample case test
(Siegel, 1956) showed a statistically significant difference
(p < 0.001) in the numerical proportion of Daphnia and
Tendipedidae in the two groups of ponds (Tables 7 and 8).
Both of them were more abundant in the 2-year ponds early
in the summer. However only the total number of Tendipedidae
was similar in both ages of pond while Daphnia were more
abundant in l-year ponds. Thus, during the summer, in-
creased diversity of invertebrates in 2-year ponds was
accompanied by an increase in biomass, but not in the total
number of organisms.

The means of dispersal of each taXa probably affected
their abundance in the ponds. Various animals have been
found transporting viable aquatic organisms (Maquire, 1959).
Water birds have been known to transport snails (Malone,
1965) and crustacean eggs (Proctor, 196A). Pennak (1953:
15) believed wind could carry plankton and benthos. Flying
insects easily could have reached the ponds.

Upon reaching the ponds, environmental requirements
and biotic interactions affect each organism's survival.
These limiting factors are more easily recognized at the
species level than at the taxa levels used in this study,
so only broad generalizations are indicated.

Daphnia, Tendipedidae, and Oligocheata were the most

common invertebrates in l-year ponds. In general, they

A2

are primary consumers feeding on organic matter, algae,
and macrophytes.(Dineen, 1953; Pennak, 1953). Their
"life form" (Odum, 1959:295) can be classified as plankton
and benthos. Since the l-year ponds did contain organic
matter (Table 2), vegetation, and few predators; Daphnia
and Oligochaeta were more numerous in them. The early
season low number of Tendipedidae and Oligochaeta was
probably because few individuals had dispersed to the
younger ponds.

The most common invertebrates of the 2-year ponds
not only included the three taxa of the 1-year ponds, but

in addition Lymnaea, Physidae, Chaoborus, Coroxidae,

 

Spongillidae, and Hydrophilidae. Chaoborus is a plankton

 

predator while the others are primary consumers (Dineen,
1953; Pennak, 1953)., Periphyton, a new "life form" (Odum,
1959), was added to the list of common invertebrates with
the inclusion of Lymnaea, Physidae, and Spongillidae.

Chaoborus probably reduced Daphnia, but this reduction in

 

numbers was replaced with the addition of periphyton.
Although there was little measurable increase in organic
matter (Table 2), the macrophytes did increase (Table 6)
with pond aging. Undoubtedly, producer and consumer organ-
isms found in the 2-year ponds increased the available
microhabitats and niches which resulted in a greater

diversity and biomass of invertebrates.

A3

Waterfowl Pppulations
Ducks were counted for four main reasons: (1) to
compare pond species composition with that of the marsh,
(2) to compare pond seasonal abundance with that of the
marsh, (3) to determine the size and distribution of the
breeding population, and (A) to investigate the factors
of pond age, food, cover, water levels, and bay-to-pond

distance on duck distribution at the ponds.

Species Composition

 

The Wilcoxon matched-pairs signed-ranks test (Siegel,
1956:75-83) was used to compare the duck species at the
ponds with those of the East Marsh and the bays surround—
ing the study area. The weekly percent composition of each
species in each area was compared with the same species in
other areas. The comparison was limited to the months of
May and June because few ducks used the ponds at other»
times.

Several sources of bias may have been introduced by
comparing the species composition of the three areas.
Counts of the East Marsh were taken from the air while
those of the bays and ponds were taken from the ground.
Smith (1958:56) found that if 50 percent of a species were
seen from the air, its species composition would be similar
to that recorded on the ground. Probably additional bias

was introduced because different observers made the aerial

AA

and ground counts. Neither of these factors was considered
a large source of error for comparing species compositions.

This statistical method of comparison contains cer—
tain limitations. Because of above mentioned bias, a com—
parison of duck density would be meaningless. Therefore
the percent composition of each species was compared. The
percentage of the total population that each species com-
prised was an average proportion; and should not be confused
with density.

The duck species composition of the Second Lead bays
(Figure 11) was compared with that of the East Marsh (Figure
10, Table 9). Mallards, gadwalls, shovellers, redheads, and
ruddy ducks were found in relatively the same proportion in
both areas. The Second Lead bays contained a higher per-
centage of pintails and teal while baldpates, canvasbacks,
and scaup comprised a smaller proportion. Evidently the
Second Lead bays had a more restricted species composition
than the surrounding marsh.

The duck species composition at the ponds (Figure 13)
was compared to that of the East Marsh and the Second Lead
bays (Table 9). Only the percentage composition of gadwalls
and shovellers were similar in the three areas. Teal made
up a larger percentage of the pond composition than they
did in the marsh or bays, even though the bays contained
higher percentages of teal than the East Marsh. Mallards,

pintails, baldpates, canvasbacks, redheads, scaup, and

A5

.mnouHsz .muHmQ aoopHHEEoo pCoECOHo>oQ swam: mpHoQ 0:» an
mHCzoo me003 CO oommm .Cmpmz Hmmm oCH CH mxoso no CoHHHmOCEoo moHomdmII.OH oCCmHm

 

 

 

 

 

 

 

 

 

 

 

     

 

     
  

 

06— $0.: coo- o\co— oc— A«cc— $00—
0 0 . s . N ntho
a n 2030 >DDDE
— Hlol n3<0m
Ill—llllhl D<WIDWE
«—
e u ¥U<mm<>2<u
o WF<QDJ<U
J<WIP
nu n— 0—
EW JJW>UIW
m— m n—
JJ<30<0
0— w—
.IIMhWIIIIIIIIIIIIIIIII .J_4\1Pa‘—nu
h . o. o—
| DE<I_.H<_>_
a. a. o. m a. a. n.
OonNh— 05050 mm—ou en—m— 2350— 330— —Qon— WN.m ”JR-24‘":
~m:m:< 3:3 0:3... as! «mamznx 3:... 023—.
000p moa—

 

 

 

A6

.mnouHCm= .mHHon .muCCoo

meooz Co Comwm .mzmn Comm mospm on» up mxosc mo COHHHmOQEoo moHooCmII.HH ohstm

 

$60—$06— #60— flcfi— .809— RGG— 806—

 

». m _

 

 

 

    

     
     

 

   

EmIkD
XUDD_>DDDE

 

 

n3<Um
D<mIDmfl
¥U<mm<>2<u
Uh<n04<m
4<mh
EM44m>DIm
«—
44<>>D<U
.« o o m -
.IIIIl-IIIIII 42E;2n
o. a. n. o. n. nu «— .
T
a. a. «— 1'1 0 n A . Dfl<44<§
«an 2: Wm» nun oo— n: ma. m..__N...m..i|Im4n.>.<m
awaua< :21. 0...... an! «nan:< 3:... 0......
000— . noo—

 

 

 

A7

.muHoa .mpCCoo meooz Co comma

.mnouHsz

.mCCoo 02» um mxoCC mo CoHpHmanoo moHoodmII..mH opstm

 

890— 800— .800— $60—

 

 

 

 

 

flca— $OG— 806.— 896.— J<PDF

gnu-JD ¥U<Jm

£3400 :—
I'vl. J<WF..2’IU

I. .2... .3-..
ON an an

  

 

 

Ill ENJJW>OIW
o— 0—
.ll 4.32330
nu ma _
lull 4.4.5,..."—
.. o o .
’ lull DE<44<E
.oo. «5 . Q
— on Nov «nu a— — 0' an MN.” WAR-21%“
«aaaaa‘ at... «can. an! gnaw-0‘ at... 0:3... :0!
ooo__ . moo—

 

 

A8

TABLE 9.——Comparison of the species compositions of the

East Marsh, the study area bays, and 27 blasted ponds.

(The Wilcoxon matched-pairs test was used to compare the

weekly percent composition of each species during June
1965 and May through June 1966.)

 

 

Comparison of: Bays to Ponds to Ponds to
Species East Marsh East Marsh Bays
Mallard 0* ___ -
Pintail + _- _-_
Gadwall 0 0 0
Shoveller 0 0 0
Teal +++ +++ +++
Baldpate -- _-- _-__
Canvasback - ___ ___
Redhead 0 ___ -_-
Scaup - _-- __-
Ruddy Duck 0 --_ ___

 

*Level of significance:

99 percent +++ higher percent, --- lower percent
98 percent ++ higher percent, -- lower percent
95 percent + higher percent, - lower percent

Less than 95 percent 0.

A9

ruddy ducks were either completely absent in the ponds or
found in lower proportions than at the surrounding areas.
This suggests that of the 11 species most commonly found
in the area, only two, Shoveller and gadwall, used the
ponds as much as would have been expected and that teal
used them more than was expected.

The differences between species compositions at the
ponds and the surrounding areas could be due to specific
habitat preferences, to overcrowding, or a combination of
the two. There can be little doubt that some species
prefer certain habitats because of structural adaptations,
physiological responses, and specific behavior (Odum, 1959:
27). In addition, the number of individuals of a species
is affected by the carrying capacity of a particular time
and place, and the extent to which a specific population
fills that capacity (Odum, 1959:183).

In case of the low pond use by diving ducks, it
appears that they respond mostly to their habitat prefer-
ences. Evans and Black (1956:36) found that diving ducks
usually remained on larger waters than dabbling ducks. Most
diving ducks patter along the surface when rising from the
water, and thus require longer "take-offs." Although the
50-foot ponds offered ample distance for take-off, the
small size probably limited diving duck use.

In contrast to divers, dabbling ducks (as a group)

showed no clear preferences for any of the three areas.

50

In Iowa, Provost (19A8) noted larger numbers of blue-winged
teal, smaller numbers of mallards and shovellers, a few

wood ducks and green—winged teal, and no gadwalls or bald-
pates using blasted ponds; even though all frequented the
marsh. Hochbaum (19AA:78), Smith (1958:33), and Drewien
(1966) found specific habitat preferences among dabbler
ducks while Keith (l96l:A6-A7) reported that they frequented
ponds and lakes to nearly the same extent. Similarly, my
findings showed that dabbling ducks as a group, did not
prefer any one of the three areas compared, although indi-
vidual species differences did exist. Therefore, each species
must be examined separately.

Some species of dabbling ducks may have been affected
more by overcrowding than by habitat preference. Evans and
Black (1956:38) postulated that the most abundant breeding
ducks dispersed into small, less-desirable wetlands as
their population density increased. However, gadwall seem
to have less dispersal urge than other dabbling ducks (Gates,
1962; Drewien, 1966).

A The yearly abundance of each species in the marsh and
ponds offers a partial explanation to the differences in
species composition of the three areas. In the East Marsh
during June, all species of dabbling duck increased in
1966 over 1965, while only shovellers and teal increased
in the ponds (Table 10). This may indicate that teal and

shovellers may have been close to the carrying capacity of

51

TABLE 10.--Comparison of June 1965 and 1966 dabbler duck
numbers in the East Marsh and at 17 artificial ponds,
Delta, Manitoba.

 

 

 

 

 

East Marshl Ponds
Species % 7?—
1965 1966 Change 1965 1966 Change
Mallard 2,225 A,8A0 +118 2 0 —
Pintail 1,515 2,030 +3A 3 2 -
Gadwall 2,655 A,220 +58 12 10 -17
Shoveller 1,930 3,920 +103 5 3A +580
Teal 2,015 3,025 +50 27 1AA +A33
Baldpate ___Sg; _g11g; 1121 _g __g -
Total 11,165 20,160 +80% A9 190 +380%
1

Data from Delta Marsh Development Committee.

52

the adjacent area and therefore increased their use of the
newly created ponds. In contrast, other dabbling duck
species showed no increase at the ponds because their
populations possibly were below the carrying capacity of
the local area.

Mallard and pintail use of the ponds probably was
"low" because these early nesters had completed much of
their breeding activities before the ponds were free of ice.
Mallards return to Delta in early April (Hochbaum, 19AA)
yet the ponds were frozen until mid-April.

Current information--including this study--does not
provide a positive answer to the observed differences in
dabbler duck species composition. Circumstantial evidence
indicates that both habitat preferences and overcrowding
in the study area may have resulted in a pond species

composition dominated by teal.

Seggonal Abundance

 

Ice probably restricted early season duck use of

the ponds. In general, spring break-up of the bays occurred
several days to over a week earlier than in the ponds. In
1965 on May 6 (the first census) the ponds were free of

ice and ducks were present. In 1966, the ponds were first
clear of ice by April 22, but were frozen again from April
28 to May 2. The first ducks were seen at the ponds May A
even though they had been using the surrounding bays for

several weeks.

53

From early May until mid-June, pond utilization
followed the trends of waterfowl abundance in the marsh
(Figure 13). Dillion (1955) described a similar Delta
Marsh trend as a decrease from a prebreeding high to a
breeding low. Sowls (1955:13) believed that during early
spring, Delta's population was comprised of transient and
resident ducks.

In contrast to the marsh, duck abundance at the ponds
sharply increased in late June (Figure 13). This is a
period when many ducks move into the marsh to moult and
pass their flightless period (Hochbaum, 19AA:119). Some.
ducks using the ponds were still paired while others were
gatherings of postbreeding drakes. Paired ducks most likely
were renesters. I

The ponds received little use after mid-July even
though ducks increased in the surrounding areas (Figure 13).
Only two newly-hatched broods and a few adults were seen on
the ponds during this period. Mathisen gp_g1. (l96A) also
reported little brood—use of blasted ponds. Broods appar-
ently prefer larger water areas which offer better escape
from possible predators (Evans and Black, 1956:A0). Once
ducks are able to fly they prefer large open potholes
(Evans-gp_g1., 1952:Al). Flightless ducks were never seen
on the ponds, but they did frequent the surrounding bays.
Hochbaum (19AA:122) similarly never encountered flightless

ducks at natural potholes in the Delta Marsh.

5A

 

I I I 1

     
    

 

 

 

 

    

East Marsh
lxloflflfl *
21 . H. 1
' . - """°'°'l"9..6 5'
...16- ' ' ' zoo -
= .
0. 1x10
(Laltz.b d
210* “
m .
GI. lir' '
V, 6t 1966 -
a: . I" '
a 4- z . 1
= . . ‘2
a 2 ' I965 .-".. ..' ° . 3"; ‘
I...., 1 'I'
_ Ponds * -
20 . o ... -I

 

 

 

IIIIFMII'III III'I'IIII IIIIIIIsI

Figure 13.-—Seasonal abundance of adult and flying juvenile

, ducks at the East Marsh, study area bays, and 17 ponds. East

Marsh data from the Delta Marsh Development Committee, Delta,
Manitoba.

55

Ducks utilized the ponds during the time of year
commonly associated with spring migration, breeding, and
postbreeding activities (Figure 13). The population
composition of the most common species (Table 11), blue-
winged teal, reflects these activities. From April to
mid-June breeding pairs comprised the bulk of the blue-
wings present, indicating little use for migration pur—
poses. After mid-June breeding pairs decreased and non-
breeders, mostly gatherings of postbreeding drakes, in-
creased in abundance. The population composition of
species other than blue-wings showed the same general

pattern.

BreedingrPopulation

 

The breeding duck population was determined by a
method described by Evans and Black (1956:5A).' Ducks were
recorded as lone pairs, single males, single females,
groups of pairs, groups of males, one male with a pair,
and unidentified. Lone pairs, lone drakes, and lone hens
were assumed to represent nesting pairs. It also was
assumed that half of the nesting pairs seen on the First
and Second Bays were in the study area, while all breeders
seen on the Clair Lake transect were included (Figure 2).
The peak number of breeding pairs for each species was
taken as the breeding population for that species. For
example, 12 pairs of blue-winged teal, the peak number,

were seen on May 26 (Table 11), so the breeding population

56

.OOOH Com mOCOC popmmHo mm Ho COHHNHCCQC CHNC mCHOmmAn OmmeHOCH*

 

 

: mm ON 5 O H m 0\0
O mm m mm a m O om\O
OH 0m O a w a O mm\o
OH mH o o o m m mH\m
HH 0N 0 e 0 H H 0\0
m mm o a a m m m\w
*mH 0m O m H H m mm\m
O OH O o O m w mH\m
m wm o o m o m MH\m
H m o o O o H ©\m

mCHmm .HOCO. mmHmz mCHmm onz CHmm

mCH mxoza oCOH puma

Ivmmhm Hmuoe

 

masopo wCHOmoAnCoz

mxoso mCHOmmCm

 

.oooeHesz .NoHoo .000H .meeoo
HeHoHHHees 0N eoC eoHpHmooeoo eoHomHsooo Heoo eomcHsIosHo oeeII.HH mHmHH

57

of teal was estimated at 12 pairs for the ponds. The
breeding population of all species was the sum of the indi-
vidual peak numbers of breeders for each species, regard-
less of the peak date (Table 12).

In 1966, the Second Lead area contained a calculated
6A breeding pairs (Table 12). Only parts of the bays were
censused in 1965 so no calculation was made of the whole
Second Lead area. On the average, an artificial pond had
0.8 pairs in 1965 and 1.A pairs in 1966. For every mile
of edge there were 15 pairs along the bays and 27 pairs
along the ponds in 1965, while in 1966 there were 16 pairs

per mile of bay edge and 51 pairs per mile of pond edge.

Factors Affecting Pond Selection

 

The weekly duck-count total ranged from only one duck
on pond S-2A to 7A on S—3. Of the ducks seen on pond S—3,
25 were a flock of drakes! Flocks of this size were en-
countered only a few times, so it was not possible to deter-
mine from the total number of ducks if certain ponds were
being favored.

By grouping the ponds into 1- and 2-year classes, it
was possible to minimize the bias of flocking to particular
ponds. Arrangement of the data in this manner fulfilled
the assumptions of the Student t-test (Li, 1957). A com-
parison of the two age groups showed no statistically signi-

ficant difference (p > 0.20) in total duck use (Table 13).

58

TABLE l2.--Second Lead breeding pair population based on
weekly counts, Delta, Manitoba.

 

 

 

 

 

Peak Flooded
Species Date Ponds Areas Bays* Total
1965 17 Ponds
Pintail 5/12 0 0 3 3
G-W. Teal 5/12 1 0 l 2
Shoveller 5/28 2 0 0 2
Gadwall 5/28 3 0 0 3
Mallard 6/9 1 0 2 3
Canvasback 6/16 0 0 l l
Redhead 6/16 0 0 1 1
B-W. Teal 6/23 _6 __g _a __8_
Total. 13 0 10 23
1966 25 Ponds
Baldpate 5/5 0 O 2 2
Mallard 5/6 1 0 A 5
Pintail 5/6 2 l 3 6
Shoveller 5/13 11 2 1 1A
Gadwall 5/13 A 0 2 6
G-W. Teal 5/13 3 0 O 3
L. Scaup 5/19 2 l 2 5
Canvasback 5/19 0 0 l 1
B.-W. Teal 5/26 12 3 3 18
Redhead 6/2 0 0 3 3
Ruddy Duck 6/9 _9 _Q _1 _1
Total 35 7 22 6A

 

*Only

a portion of the bays censused in 1965.

59

TABLE l3.--Comparison of duck use at l- and 2-year ponds,
1966, Delta, Manitoba.

 

 

Total Ducks Breeding Pairs
Pond Age* l-year 2-year l—year 2-year
B—W. Teal 86 99 A A.O
G-W. Teal 33 21 l 1.0
Shoveller 15 35 3 A.0
Gadwall 10 12 0 2.0
Pintail 6 A 1 0.5
Mallard 3 1 0 0.5
Lesser Scaup 0 A 0 1.0
Black Duck 2 0 0 0
Total 155** 176** 9 12.0

 

*Based on eight l-year ponds and a weighted sample
of 17 2-year ponds.

**No statistically significant difference (Student
t-test p > 0.20).

60

Even though duck use was similar at the two age
classes of pond, two species did seem to indicate a
preference (Table 13). Shovellers appeared to select
2-year ponds and green-winged teal the l—year class. It
is not clear whether such a difference did exist because
of the small sample size.

It is of particular interest that ducks used both
age groups of pond with nearly equal intensity. Certainly
macrophytes and invertebrates increased qualitatively and
quantitatively (Table 1A). Based on Welch's (1952:379)
description of pond-age classes, the ponds passed from a
young stage (little or no vegetation) to an adolescent
stage of invading plants. Apparently, changes in early
stages of succession did not significantly affect water-
fowl use of the ponds.

A Chi-square one-sample test (Seigel, 1956) was used
to ascertain if breeding pairs were uniformly distributed
on the ponds. Flocking did not bias the distribution be—
cause only lone males, lone females, or lone pairs were
considered as breeding pairs. The total number of breeding
pairs seen on each pond served as an index of breeding duck
use. The pair index ranged from one on pond S-2A to 15 on
S-8, with a mean of 7.2 for ponds S-l through S-25. A
statistically significant difference (p < 0.001) indicated

that breeding pairs were not uniformly distributed.

61

.30C 08mm oCH CH moCmCmHHHO pCmOHHHCwHw HHHNOHpmHHmpm oz**
.30C 05mm 0C9 CH ooCmOCCCm HmCommmm CH ooCmCmHHHO pCmOHHHCMHm mHHmOHpmHHmHm*

 

**ONH
05mm

*.00.om\He N.N
*meoeHH 00H\He H.0

.00.00\0NH
meoeHH 00H\N0N

mxmp m+

.0
same 0+ .0

$0.0H I
moHomdm m+ .m

.HH.Om\mEmAw O.HH
moHoodm mH+ .m

CmHHEHm

HmHHEHm
HmHHEHm
CmHHEHm

**mmH
moHooCm m+ Hmoe 3Im

*.00.oo\Hs 0.H
emsooHH 00H\Hs 0.0

.00.om\00H
meoeHH 00H\0Nm

0x00 N+ oeeHoooHoeoe
mHCcme

RO.HH
mmHomOm 3+ NOHCOOHoom

.pH.Om\msmCm m.O

 

mmHooCm OH + CoumwdEOHom

mCHHmm

eooH 00
Home 0N
poem 0

HHmHOHV HHHHCmCO
HHHHNCO
Hzommopmz

moCuCon
CouHCmHC
mmmEOHC Cams
mOCpCmn
CoumeHC
HomECC Cams
szquCO
monCon
COpmeHC
HHHHmCO
mopmaoopC0>CH

Ano>oov HHHHCNCO
HHHHOCO
owvo

ACmoEV mHHHCmCO
HHHHmCO
OHHmCOm
momeCopomz

HCHmHEoCO Cmpmz OCm HHom

prCoH Cams

CHOHC Cmoe

Canon Cams
HCmeOCQCoz 6Com

 

mOCom CmmmIm

mOCom HmoHIH

 

pooumm poC OHO monCCoHC0>CH OCm mmmedonomE Ho mpHquCO

A. mmd H30hfimpm3

OCm HHHHmCO CH moowCov

.mnouHsz .mpHmQ .OOOH .mOCOC HmoHIm OCm IH Ho ComHCmCEoo m Ho hCmssCmII.:H mHmde

62

Breeding pair distribution can be affected by a
variety of environmental factors. Studies by Furniss
(1938) and Rogers (l96A) have demonstrated that water
levels influence breeding pair use. Girard (19Al, Mendall
(19A9), and Keith (1961:77-78) believed that food was im-
portant. However, Ignatoski (1966) and Evans and Black
(1956:38) found that breeding ducks did not seem to be
attracted by food. Cover has been reported to affect pair
use (Saugstad, 1939; Smith, 1953; Smith, 1958). In con—
trast, Evans and Black (1956:38) and Keith 1961:77) found
variations in cover had little effect on breeding-duck
usage. These are not the only factors affecting pair distri-
bution on a local area, but they are the ones commonly sup-
posed to affect such use. Water levels, bay—to-pond dis—
tance, food, and cover were examined in this study.

The combined effect of increased food and cover was
previously examined when total duck use at l- and 2-year
ponds was compared (Table 13). This did not specifically
compare breeding pair use; but since there was no statisti-
cally significant difference in total use, it seemed in-
advisable to compare statistically one type of use, that by
pairs. Both the quality and quantity of food and-cover
increased in the older ponds (Table 1A), yet this change
does not appear to have significantly affected waterfowl

use .

63

I

The available biomass of invertebrates in individual
ponds was compared with breeding pair use (Figure 1A). A
Spearman rankcorrelation test (Siegel, 1956) showed no
statistically significant association (rs = 0.15A p > 0.05)
between breeding pair use and the invertebrate biomass of
the ponds.

Invertebrates were not the only food source in the
ponds during the spring. Sago pondweed occurred in most of
the ponds, but was more abundant in the 2-year group (Table
6). Rhizomes were present, but no tubers were found in the
ponds. Since duck use at the l- and 2-year ponds was
similar and since little plant material was available in
the spring, it_is doubtful that aquatic vegetation played
a major role in affecting pair use of the ponds.

A Chi-square 2 x 2 contingency test (Siegel, 1956) was
used to discover if the distance from the ponds to the bays
influenced pair distribution. During May and June the
number of pairs using the ponds nearest the bays was com—
pared with those further away (Table 15). No statistically
significant difference (p > 0.35) indicated that bay-to-pond
distance had little affect on pair use.

These results can be applied also to the influence of
pond water levels on duck use, because water levels were
inversely proportional to bay-to-pond distance (Figure 8).
At the end of June the water levels of three ponds nearest

the bays had lost an average of 11 inches and two ponds

6A

 

 

 

 

. 7. -
In . .
III
F' :3.(3 t a
I? . -
m In F‘ ‘
“III _ .' 7. 5
F E goo '- . . “
I: 3 _ .
m o
>.CJ - -
E D " . -1

r
u. ‘1.l3 r '1 ‘
D E I- . 1
Lu 0
. D. I C1 .1 ‘
.I . 0 H
E O
D 5 ‘ID ’I5

INDEX DF BREEDING DUCKS

Figure lA.-—Graph showing poor association between breed-

ing duck use and invertebrate biomass at the ponds, Delta,
Manitoba. Each value is the mean of two net tow samples
for invertebrates and the total of four counts of breed-
ing pairs of ducks during May 13 through June 30, 1966.

65

TABLE 15.-—Comparison of pond locations and breeding pair
use, Delta, Manitoba, 1966.

 

 

Bay-Pond Distance Central Location*
<325 Ft. >325 Ft. End Ponds Center Ponds
May Al A2 27 66
June 52 A6 35 53
Total 93** 88** 62*** 119***

 

*See Figure 2 End Ponds S-l through S—A, S-6, and
S-l9 through S-25 and center ponds S-7 through S—l8.

**Chi-square p > 0.35 < 0.A0.

***Chi—square p > 0.05 < 0.10-

66

further away averaged 22 inches below the edge of the ponds.
This difference in water levels did not cause a statisti-
cally significant difference in waterfowl use of the two
pond groups.

Breeding ducks did use certain ponds more than others,
but pond age (food and cover), invertebrate abundance,
water levels, and distances from bay to pond (as encountered
and measured in this study) did not cause this selection.
Admittedly these factors, either singly or combined may
affect pair use in other situations.

Ponds S—7, S-8, S-l3 and S—l8 had the highest breeding
pair use, and ponds S-2, S-23, and S-2A had the lowest.
Most of the high-use ponds were centrally located while
the low-use ponds were at either end of the pond chain
(Table 15,,Chi—square p > 0.05 < 0.10). This would sug-
gest that pairs reacted more to the ponds as a complex, as
opposed, to a factor or factors within a particular pond.

Man and wildfowl view their environment differently.
Because of this difference it is hard to understand why
ducks use certain ponds more than others if it was not due
to some environmental factor within the ponds. Therefore
the following explanation based on duck response to the
macrohabitat (in contrast to microhabitat--that with a
pond) is speculation on my part.

Not all of the needs of breeding waterfowl were

satisfied in the ponds. Observations indicated that the

67

ponds served mainly as isolation and loafing areas. Ducks
spent only a short time per pond visit and did little feed-
ing (Table 16). Hochbaum (1955), Sowls (1955), and others
found that loafing, feeding, nest site, graveling areas,
etc. may be widely separated. Thus, breeding ducks made
frequent trips to other areas of the marsh because the
ponds were only part of their home range.

Movement from one area of the home range to another
may explain the distribution of the breeding pairs.
Hochbaum (1955) states:

These ducks of the Delta Marsh use their environment

in an orderly fashion; there is a pattern to their

travels. . . ., the marsh itself is a pattern of
aerial lanes as well marked by the flights of water-
fowl as the roads on a highway map.

Furthermore, Peter Ward (1p_litt.) states:

There is a well developed east to west flight line

that passes just north of the mid-point of your pot-

hole complex. While not as easily seen now, it waS~

very obvious during the high water period, 1955-57.
I do not know if the increased flight activity above the
most-used ponds was due to the ponds themselves or to the
presence of a "flight path."

If’a "flight path" did exist over the centrally
located ponds in 1966, ducks would most likely use the
ponds in an inverse proportion to the distance from ponds
to "flight path." Perhaps ponds S-7, S-8, S—13, and S-l8
received the highest use because they were under the

"flight path," and the end ponds had fewer-breeding ducks

because they were farther away. Only circumstantial

68

.mopCCHs mo ConesC Coos

.AOACo Onmvcmpm

.oNHm mHosmm u C

N
"ID

IIH

 

 

 

 

 

H
am N0 a00H 0H «a. “00 00 . 0.H 0.0H mHN Hoooe
nu nu ”HI nu nu qu nu 04m wdmm HHI 05000 »H
II II II II II II II H.m m.NH 0 omoneom
II II II II II II II II 0.0 m Heme .3IO
H Nm 00H 0H OH 00 N0 H.H 0.NH 00H Home .3Im
m 0 00H 0 0 H0 HH H.N H.mH 00 AoHHoeoCm
m m OOH : N mm m m.m 0.0m am HHmzva
0 H «00H “0 R0 HOOH 0 0.0 0.0H 0N HHopCHm
II II II II II II II 0.N H.0H 0H oeeHHez
a C C mm .x C moHooCm
wCHOoom Hmpoe Cmmmz hwmmm HHoCm HCCom mlCo. "HCMQW oEHB

 

.eooeHeoz .meHoO .000H see 000H oeeHIsoz

.mOCoa oCu um oHHC3 wCHHmOH pCm .wCHOoom .pHmH>IOCoq Add uCoCm mxosc oEHBII.OH mqm¢e

69
evidence of this study supports such a theory. Certainly

it is a topic which deserves future consideration.

Duck Behavior at the Ponds.

 

Observations were made from a tower overlooking the
ponds for two principal reasons: (1) to determine if duck
use was higher at any particular time of day, and (2) to
ascertain the average time spent at the ponds, feeding,

loafing, etc.

Diurnal Activity Pattern

Observations were made of the number of ducks using
the ponds at all times of day to determine if a diurnal
activity pattern existed during the spring. Based on 16
different periods (total of 31 hours of observation in
1966) the pattern appeared to be one of highest use early
in the morning (from sunrise to two hours after sunrise)
and then declining utilization for the remainder of the day
(Figure 15).

Time Spent at Ponds

 

Ducks constantly moved from one pond to another and
to surrounding areas. On the average, a duck spent 15 to
16 minutes per pond visit (Table 16). Gadwalls spent the
longest time per visit--27 minutes, and green-winged teal
the shortest--9 minutes. Based on 87 observations, ducks
undisturbed by man moved 6A percent of the time from one

pond to another and 36 percent of the flights were to

70

 

 

 

 

 

 

a) , . . . .
5
:3 25* ‘
o

20- :
U.
0 I51 . . 1
a: . ' .
w '0 O
In
2 5- '
___) .
Z 6 s 10 I2 I4 16 Is 20 22

TIME OF DAY

Figure 15.--Diurna1 duck activity pattern at the ponds
during the spring. The curve smoothed by eye is based on
31 hours of observation in 16 days during 1965 and 1966,
Delta, Manitoba.

71

adjacent bays. Similarly, Evans and Black (1956:AA)
found that pairs seldom remained on one pothole for any
length of time. A

The diurnal activity pattern and the constant move-
ment of the birds affected the accuracy of censuses. Most
of the censuses were taken between 8:00 A.M. and 10:00 A.M.
so neither the maximum nor the minimum number of ducks was
present on the ponds for that particular day. Also be-
cause the ducks were constantly moving from place to place,
it was not possible to predict consistently the number of
ducks using a particular pond at a certain time of the day.

During May and June, ducks spent most of their time
while at the ponds on top of the shoulders, instead of at
the water's edge or in the water (Table 16). This was
especially true when the tops of the Spoil banks were bare
of vegetation and thus offered a loafing place with a good
elevated view of the surrounding area. Provost (19A8) also
believed that ducks were attracted by the naked shoulders
of blasted ponds. Ducks spent less than 10 percent of their
time in the water and less than half of that time feeding
(Table 16). Either these ponds were not used as feeding
areas or food was so abundant that little time was needed

to meet their food requirements.

72

Nesting

Ponds and Nest Locations

Old nests were located before and after pond con—
struction to determine if the ponds affected the distance
from nests to the bays. Before the ponds were blasted,
73 percent of the nests were within A00 feet of the bays
while only AA percent were within the same distance after
pond construction (Table 17).

Hens often used the ponds as jumping-off places en-
route to their nests. During 31 hours of observation in
1966, hens were seen on A5 occasions leaving the ponds
for nearby nests. The jumping-off ponds were not always
those closest to the hen's nest, but most were within 200

feet.

Ponds and Nest Density

 

In 1966, a 78-acre experimental area containing 25
artificial ponds and a 21-acre check area without ponds
(Figure 2) were searched for duck nests to evaluate the
effect of artificial ponds on nest density. Blue-winged
teal nests composed A2 percent of the 108 nests found in
the two areas (Table 18). Shoveller and pintail nests were
the next most common. The check area contained 2.1 nests
per acre. The experimental area only had 0.3 nests per
acre where there were 5.1 acres per pond, and l.A nests

per acre where there were 2.2 acres per pond.

73

TABLE l7.--Comparison of the distance from duck nests to
the bays before and after pond construction, 196A and
1966, Delta, Manitoba.

 

Bay—Pond Distance Cumulative Percent

 

 

(feet) Before After

0—100 6 0
101—200 26 9
201-300 A6 1“
301—A00 72 A“
A01-500 80 8A
501-600 100 100

Number of Nests 99 A9

 

7A

TABLE 18.--Comparison of duck nest densities at an experi-
mental area with artificial ponds and at a check area with-
out ponds, 1966, Delta, Manitoba.

 

 

Check Experimental
Number of Ponds 0 8 l7
Blue-winged Teal l8 3 25
Shoveller 6 9
Pintail 10 O 5
Mallard A 3 l
Gadwall l 1 l
Redhead 2 0 O
Lesser Scaup 0 0 1
Unknown* _1 _S _1
Total A5 1A A9
Acres 21 Al 37
Acres/pond -- 5.1 2.2
Nests/acre 2.1 0.3 l.A

 

*Old nests found chiefly after an August burning of
cover.

75

Other studies have shown that vegetation and proximity
to water may affect nest location. Williams and Marshall
(1938), Bue gp_S1. (1952), Sowls (1955:67), Evans and
Black (1956:A9), Glover (1956), Keith (1961:55), and others
have stressed the importance of cover to nest densities.
Some investigators (Bennett, 1938:106; Steel gp_g1., 1956)
stressed proximity of water. When nesting cover was nearby,
Evans gp_g1. (l952:A0—Al) and Evans and Black (1956:A8)
found little relationship between nest location and potholes
used by breeding ducks.

Quantitative sampling of the experimental and check
area vegetation revealed differences in their flora, yet
both could be classified as whitetop meadows. In the check
area the percent cover was--leaf litter 93.A percent,

Scolochloa festucacea 0.8 percent, Sonchus arvenses 1A.8

 

 

percent, and Cirsuim 5p. 7.A percent. In comparison with
the experimental area (Table A), leaf litter, S. arvenses,
and Cirsium 92: were more abundant in the check area. The
frequency and average height of plants in the two areas
were similar. Since plants had not begun to grow when most
ducks were selecting their nest sites, the higher percent
cover of leaf litter at the check area was the most obvious
difference during the spring.

Mathiak and Linde (1956) reported a higher concentration
of nests in an area with level ditches than areas without
ditches. Provost (19A8) found that blasted ponds induced

few ducks to nest. At the experimental area, as the number

76

of ponds per acre increased, so did the number of nests.
At a nearby area without artificial ponds there were more
nests than at the experimental area, suggesting that
creating ponds in itself does not always influence nest
density. Although as Dzubin (1955) has pointed out, the
nest site must have some relationship to the water areas
used by pairs, it does not necessarily have to be close
to the water. The lower density at the experimental area
was probably due to the differences in vegetative cover

in the two areas.

Fapes of Nests

 

A larger percent of the nests were unsuccessful in
the experimental area, where 80 percent of the nests were
known to be destroyed while only 60 percent were lost in
the check area (Table 19). Sowls (19A8:130) found that 35
percent of the nests hatched in the area he studied at the
Delta Marsh. Mathiak and Linde (1956) reported very high
nest loss near level ditches in Wisconsin.

Predators were the chief cause of nest loss (Table 19).

The striped skunk (Mephitis mephitis), Franklin ground

 

squirrel (Citellus franklinii), and the raccoon (Procyon

 

19233) destroyed most of the unsuccessful nests.

Several factors may have contributed to higher pred—
ation in the experimental area. Early in the spring the
check area was separated from the adjacent dry marsh by

a flooded shallow zone. Although the check area was not

77

TABLE l9.--Comparison of nest fates at experimental and
check areas, 1966, Delta, Manitoba.

 

 

Number Percent
Check Exp. Check Exp.
Successful 2A A 32 6
Incomplete Data* 6 9 8 1A
Unsuccessful AA 50 60 80
Skunk 15 20 20 32
Squirrel l3 9 18 1A
Raccoon A 5 5 8
Deserted 2 5 3 8
Human 0 2 0 3
Unknown 1S ._2 ‘ 1A 1A
Total 7A** 3

 

*Nests chiefly found after August burning.

**Also includes nests found in general vicinity of
check area, so the total is larger than the A5 nests in
Table 18.

78

free of skunks, it did not lose as many nests to skunks as
did the experimental area (Table 19). A road leading to
the experimental area, plus the fact that more man-hours
were spent around the ponds may have attracted the larger
predators. Kalmback (1938) and Keith (1961:78) found

more nests destroyed in dense vegetation than in sparse
cover. Because a large number of leopard frogs (5223

pipiens) and Dakota toads (Bufo hemlophrys) were found

 

around the ponds, skunks may have been attracted to the
ponds and consequently covered areas they normally would
not have searched for nests. All of these factors--flood-
ing, the road, activity by humans, differences in cover,
and the ponds—~may have resulted in higher predation at

the experimental area.

Displaced Eggs and Eggshells

 

Duck eggs and eggshells were found at many of the ponds.
On August 11, 1965, eight ponds were checked and four con-
tained either whole eggs or eggshells. Three eggs, one
each of blue-winged teal, shoveller, and gadwall were
located in or near the ponds. Also six eggshells cracked
by predators were found.

On June 21, 1966 I saw a female redhead fly away from
her nest with a cracked egg containing a partially developed
embryo. Examination of the nest revealed that a tractor

had run over it, cracking at least three eggs.

79

Sowls (1955) was able to induce a Shoveller hen to
make 81 trips from nest to pond each time carrying a shell.
Probably nesting hens carried most of the cracked eggs to
the ponds.

More eggs and eggshells were found near the ponds than
in other areas of the marsh. However, they only may have
been easier to locate at the ponds. Bennett (1938:36),
Hochbaum (19AA:88), Sowls (1955:82), and Glover (1956)
have also noticed displaced eggs. Mathiak and Linde (1956)
found eggshells in level ditches. The role which this
phenomenon plays in the breeding biology of ducks is un-

known and deserves further study.'

Duck Production

 

Duck production is commonly based on direct counts
of broods near the flying age. However, broods in the
experimental area did not use the ponds, but rather dis-
persed to surrounding bays. Therefore my production esti-
mates of pond production are based on (1) nesting data

and (2) the breeding population.

From Nesting Data

 

Production estimates can be made from nesting studies
if human disturbance does not affect nest predation and if
all nests are found. When proper care has been taken,
human disturbance does not increase nest predation (Hammond
and Foward, 1956; Keith, 1961:70). Not all of the nests

were found during the spring. A partial search of the

80

experimental area after fall burning revealed eight
formerly missed nests. Bennett (1938:122), Glover
(1956), and Steel gp_g1. (1956) concluded they found
about 70 percent of the nests present (approx. same per—
cent I found in the spring).

If the above assumptions are correct, the actual
production at the experimental area was far less than its
potential. Forty-eight nests were found during the spring
and if this represented 70 percent of the nests, then the
experimental area contained an estimated 69 nests. Since
only 6 percent of the nests were successful, only 3A eggs
would have hatched (average of 8.2 eggs per nest) compared
with the 32 known to have hatched.

Keith (1961:69) summarized six nesting studies con—
ducted on the Canadian prairies, and found that 39 percent
of the dabbler duck nests were successful. Under average
nest-loss conditions described by Keith (1961), a potential
of 220 eggs might have hatched from 27 successful nests in
the experimental area--far exceeding the known hatch.

Nests do not give a clear evaluation of pond production.
Although nests probably-did offer an accurate estimate of
production at the study area, my estimates may have in-
cluded pairs that never used the ponds. An area without
ponds was compared with the experimental area, but an addi-
tional variable——cover--complicated analysis. Therefore
the following is a forecast of production based on pair

counts.

81

From Breeding Population

 

Production can be based on the breeding population
when the percentage of hens producing broods and the
average brood size at flight age are known. Abnormally
few hens produced broods in the study area (Table 19) so
calculations were based on other prairie studies reporting
the percentage of hens successful. Brood data also was
taken from these reported studies. The following estimate
should be recognized as a hypothetical production--an
average expected for prairie conditions--and not an ob-
served production because only pair data came from my
study.

The ponds theoretically could have produced 113 fly-
ing ducks in 1966 (Table 20) and the entire experimental
area (including all water areas), approximately 200 ducks.
Brood mortality and inaccuracy of this estimate could ac-
count for the difference between the theoretical estimate
of 220 eggs hatching and the 200 flying ducks produced.

The usefulness of predicting production from the breed—
ing population depends on the accuracy of three factors:
(1) pair counts, (2) percentage of the hens successful,
and (3) brood size.

Pair counts are evaluated below in the discussion.

The percentage of successful hens and the average
brood size vary between areas and years. Such variations
were minimized by selecting data from ten different

studies (Table 20).

82

TABLE 20.--Potentia1 duck production of 25 ponds, 1966,
Delta, Manitoba.

 

 

Approximate Average Class

No. of Pairs* Percent Hens III Brood Production
at 25 Ponds Successful Size Ducks
B-W. Teal 12 501‘9 7.81:“’9 A6
Shoveller 11 AAl‘Ll 6.21:” 32
Gadwall A All‘“ 6.01:” 11
G-W. Teal 3 861‘l 5.0” 13
Pintail 2 571’2’u A.61’” 5
L. Scaup 2 332 “.910 3
Mallard _1 All"9 6.01"“,9 __3
Total 32 113

 

*From Table 12.

lBue, et a1. (1952), S. Dakota.

2Keith, (1961), SE Alberta.

3Evans and Black, (1956), S. Dakota.

“Evans et a1. (1952), S. Manitoba.

5Dzubin, (1956), S. Manitoba.

6Leitch, (1956), SW Saskatchewan.

7Reeves et a1. (1956), SW Saskatchewan.

8Sterling, (1956), S. Saskatchewan.

9Stoudt and Yeager, (1956), sw Saskatchewan.

loDillion, (1955), Delta, Manitoba.

83

Two factors affect brood sizes: (1) initial nests
have larger clutch sizes than renests, and (2) ducks near
flight age often form aggregations (Keith, 1961:71).

The high rate of predation in the study area could have
decreased clutch size. Unfortunately, no procedure is
presently available to evaluate these factors (Jahn and
Hunt, 196A:A5).

This method admittedly has many short—comings, but
it does give an indication of expected production. Many
areas of the marsh other than the ponds were used to pro-
duce each flying duck. Therefore pond production depends
also on these areas. However, this potential easily might
have been realized if predators had destroyed only an

average or normal number of nests.

DISCUSSION

An evaluation of breeding duck use of the ponds

depends on two factors: (1) the accuracy with which the

breeding population is estimated and (2) a knowledge of

factors limiting duck production.

Census Method

 

Breeding population estimates usually are based on

pair counts. Computed breeding densities may vary from

the actual population because of the following sources of

error:

Error may result from some pair bonds breaking
before the peak in breeding numbers. McKinney
(1965) has shown that both species and individual
variations affect the stage in the breeding cycle
when the male breaks contact with the hen.
Usually mallards, pintails, and gadwalls have
weaker pair bonds than shovellers and blue-wings
(Sowls, 1955:96).

Mallards and pintails probably were under-
estimated because many of the hens were already
nesting before the earliest censuses. Since the

ponds were frozen during this early period,

8A

85

censuses of them would not be affected as greatly
as counts of pairs using the bays.

Some pairs may move into the area after the

peak in breeding numbers. If "late pairs" were
renesters, probably they would have been counted
earlier (Sowls, 1955:138). If the "late breeders"
are nesting for the first time, the breeding
population might be underestimated. Because of
the large nest losses in the area (Table 19), it
can be assumed most of the late breeders were
renesters and did not represent new pairs.

Low (19A7) pointed out that lone drakes should

be used cautiously as indicating a breeding pair;
but Bennett (1938:123), Dzubin (1955), Glover
(1956), and Smith (1958) have shown that lone
drakes at waiting sites usually do indicate nest—
ing pairs.

Undoubtedly two drakes accompanying a hen repre-
sent a breeding pair. However, no clear evidence
supports this hypothesis, so they were not in-
cluded as breeding pairs.

There is considerable turn-over in ducks using

an area during a day (Table 16, Evans and Black,
1956:AA). Also ducks have specific differences
in their mobility (Dzubin, 1955). Censuses were

taken as quickly as possible once a week to

86

minimize the influence of this factor. It was
assumed that the number of ducks moving onto the
area was equal to those moving off.

6. The mid-morning counts probably were conserva-
tive because duck use of the ponds was highest
early in the morning (Figure 1A).

7. The intensity of sampling can affect the accuracy
of an estimate. More censuses might have shown
higher use, but greater intrusion also might
have affected duck use. Weekly censuses were
considered a compromise between an adequate
sample and duck tolerance to repeated distur-
bance by investigators.

Most of these sources of error tend toward a low estimate
of the population. Therefore I believe that this estimate
is conservative but fairly close to the actual breeding

population of the study area.

Factors Limiting Duck Production

 

Wetland improvement is aimed at attracting waterfowl.
Blasting ponds is no exception. An effective program of
habitat manipulations demands a knowledge of limiting
factors. Likewise attempts to increase waterfowl pro—
duction should be aimed at conditions which most directly
affect the carrying capacity of a breeding area.

Hochbaum (19AA) and others have suggested that dur-

ing a brief period in the spring, the carrying capacity of

87

an area may be limited by drake intolerance to other pairs
of the same species. Dzubin (1955) used the words "amoeb-
oid" and "moving territory" to describe the defended por-
tion of a pair's home range. Some species such as
shovellers are more territorialistic (McKinney, 1965) than
the gadwall and certain other species (Hammond and Mann,
1965; Gates, 1962). McKinney (1965) concluded that "chas-
ing" tended to disperse pairs and that it could signifi-
cantly effect breeding densities.

Based on these and other studies it is often assumed
that creating potholes in the breeding range of ducks will
allow more pairs to use an area and thereby incread duck-
ling production. However, many suitable ponds lack ducks,
and unless a reservoir of nonbreeding birds exists because
of insufficient water areas, creating new ponds may just
dilute the existing breeding populations.

Lack of water may limit the number of breeding ducks
in local areas. For a long time drought and extensive
drainage programs have continued to reduce breeding habitat.
Studies by Evans gp_g1. (1952), Evans and Black (1956),
Glover (1956) and data from Smith and Droll (1966, Figure
5 prior to 1963) show a close correlation between duck
abundance and available water areas.

With the return of water, ducks are expected to in-
crease. Since 1963, an increase in total water areas in

southern Manitoba has not been accompanied by an increase

88

in ducks (Figure 5). Apparently scarcity of water areas
was not the only factor limiting duck abundance in
Manitoba's prairies in recent years. Consequently, wide-
spread pond construction might not have resulted in a
greater production of ducks.

The number of ducks attracted to any area may vary
from year to year. Factors limiting duck abundance are
constantly changing. Under one set of conditions a popu—
lation's tolerances may be entirely different from those
under another set of conditions. Thus, there are as many
variations of duck use as there are combinations of factors
limiting duck production. This is especially true for
highly mobile populations like waterfowl. They are
opportunists, inevitability responding to nearly ideal
conditions and avoiding less favorable areas. This causes
shifts that are unexplainable from data collected at small
areas.

Long-term information on duck abundance or production
is lacking for my study area. However, I believe it safe
to assume that trends in waterfowl abundance at the study
area generally followed those of southern Manitoba. If
so, then the lack of water areas probably did not limit
breeding populations during my study.

Nevertheless I believe that blasting ponds did in-
crease the local carrying capacity for breeding pairs. I
base this conclusion on four points: (1) pond use was

highest during the breeding season, (2) the ponds contained

89

more pairs per mile of edge than the bays, (3) the number
of pairs at the ponds increased in 1966 from 1965 while
the Bay's population was similar during both years, and
(u) reported studies of breeding pair behavior. Mendall
(19U9:62), Hochbaum (1944), Evans and Black (1956:45), and
Lacy (1959:76) also believed that small water areas
functioned chiefly for breeding pairs.

In 1966, the carrying capacity of the study area was
increased to accommodate 35 additional breeding pairs
(Table 12). This does not mean that these 35 pairs would
have failed to nest without the ponds. Certainly the con—
stant movement between the marsh and ponds (Table 16) indi-
cates that the ponds were only part of the complex used
by ducks. Nevertheless, an opportunity was available for
other pairs to fill the vacuum left by the 35 pairs at-
tracted to the ponds. Figure 5 indicates that this vacuum
was not filled in 1966, so attracting ducks to the ponds in
1966 probably diluted the population rather than increasing
it.

If new ponds increased the carrying capacity by 35
additional breeding pairs, then presumably production
should have increased. An additional 113 ducks (Table 20)
theoretically should have been produced, in part, by the
ponds (under an average rate of predation) if water areas
had been limiting production. However, this potential was

not achieved because predators destroyed most of the nests

90

near the ponds, and water did not appear to limit duck
abundance.

In general, species which have strong territorial-
istic behavior and whose population density had reached
the carrying capacity of the area should have benefited
most from the ponds. Probably other species benefited
less depending on their density and tolerance to other

species.

MANAGEMENT

Objectives are basic to selection of management
techniques. Blasting small water areas for waterfowl may
achieve four major aims: (1) increase the carrying
capacity for breeding pairs, (2) attract ducks for pro—
tection, (3) attract ducks for recreation, and (u) stimu-
late public interest in waterfowl management.

The relationship between pond construction and in-
creases in the carrying capacity was appraised in the
discussion.

It may be possible to increase duck production by
attracting pairs away from areas of low nesting success.
Although vegetative cover also affects nest site selection,
the site must bear some relationship to the water area
used by pairs (Dzubin, 1955). Pond construction adjacent
to good nesting cover may concentrate nests and thus
facilitate control of predators, flooding, fire, and human
disturbance.

It is also possible to attract ducks to areas of
low nesting success. Nests located near the test ponds
were far less successful than those in areas without ponds.

Undoubtedly, predators also are attracted to the ponds.

91

92

An effective predator control program may be justified
under such conditions.

Ponds must be constructed sufficiently close to
brood areas so newly hatched ducklings can safely traverse
the distance. Evans EE_§l- (1952) believed that broods
could safely travel up to two miles overland. However,
the possible hazards of such a long trip would be reduced
by inducing hens to nest closer to brood areas.

Aside from attracting ducks to increase production,
pond construction simply may be justified by duck use.

The addition of small water areas may increase the attrac-
tiveness of many refuges. Furthermore, artificial ponds
offer excellent opportunities for bird watching, photography,
hunting, and other forms of recreation.

There can be little doubt that pond construction, and
blasting in particular, stimulates public interest. Many
other useful techniques of wildlife management fail to
attract attention. Blasting ponds is visual proof of ef—
fort expended and thereby, in itself, may vindicate its
use because of the positive attitude created with the public.
However, such a philosophy produces few additional ducks.

The use of ammonium nitrate to create pond ponds has
distinct advantages and disadvantages in comparison with
other means of construction (Mathisen gt_al., l96u: Mathiak,
l96u and 1965; Radtke and Byelich, n.d.). Each project

proposal should consider cost, safety, pond size and shape,

93

accessibility, available equipment and personnel, etc. be-
fore selecting the appropriate method.

Pond cost in terms of flying ducks produced appears
to be relatively cheap. For example, if the prorated con-
struction cost of a pond is $0.70 per year (p. 22) and if
113 flying ducks theoretically could have been produced,
in part, on 25 ponds; then a flying duck would have cost
$0.15 in 1966. At best this is a crude estimate which
covers only construction costs. Such a cost is economically
feasible if the lack of ponds limits duck production. How-
ever, production is often limited by several factors so
the actual production may be far less than forecast, as
it was in this study. Under such conditions, pond cost can
not be justified by duck production. I

Findings of this study indicate that during the first
few years after construction, food and cover in the ponds
does not affect pair use. Certainly there is no need to
plant aquatics where they pioneer quickly.

Constructing ponds, by itself, does not guarantee
more ducks. Creating ponds may increase ducks when and
where a lack of water is limiting local production or when
pairs are attracted to areas of higher nesting success.
Recreation and stimulation of public interest help to off-
set occasions when blasting ponds fail to increase duck
production.

In a final analysis, wise management of a resource

necessitates a broad base. A wide variety of wildlife

9“

management tools are available; no one of which by itself
will increase waterfowl. One tool, habitat improvement,
can be accomplished by several techniques. Some of the
characteristics and limitations of blasted ponds are
offered in this paper as a guide for placing pond con-

struction in perspective with the total spectrum of water-

fowl management.

SUMMARY

Pond construction has long been suggested as a
means of improving waterfowl habitat. Ammonium
nitrate now is being used as an economical means
means of blasting ponds. Because few studies
have dealt with biological evaluation of such
habitat improvement, this study was initiated to
help fill this gap in understanding. Thirty-
three weeks were spent in the field from the fall
of 196A through 1966.

Twenty-five ponds, l7 blasted in 196” and eight
in 1965, were studied intensively while less in-
tensive work was conducted on 16 other ponds.
Located in Manitoba's Delta Marsh, the primary
study area consisted of a 78-acre whitetop meadow
surrounded on three sides by bays. The study
period was characterized by above-average water
conditions and below-average duck abundance in
southern Manitoba.

Three 50-pound bags of ammonium nitrate placed

16 feet apart blasted a hole 5 x 26 x 52 feet.

Two years after construction pond depth decreased

95

96

7 percent. The ponds may persist for 30 years.
If so, a pond would have a prorated yearly cost
of between $0.60 and $0.70.

The ponds received most of their water from preci-
pitation. A few large storms added more water to
the ponds than the same amount falling in many
small showers. Within 300 feet of the bays, the
ponds received seepage water at an inverse pro-
portion to their distance from the bays.

Due to large water loss, soil and water nutrients
increased, resulting in saline conditions in the
ponds.

Plants common in the surrounding whitetop meadow
were first to pioneer onto the ponds shoulders.

Scholochloa festucacea, Atriplex patula, Cheno-

 

podium rubrum, and Sonchus arvense had the highest

 

 

frequency and cover. Scirpus paludosus seemed to

 

be favored by increased salinization that accom-
panied pond aging.

Potomogeton pectinatus and Zannichellia palustris

 

 

began growing in the ponds soon after blasting.
During the first two years, the ponds were domi-

nated by these two species, but Utricularia,

 

Myriophyllum, and Hippuris may increase.

 

A significantly larger invertebrate biomass occurred
in the 2-year ponds than in l-year ponds. Daphnia

and Tendipedidae, the most numerous invertebrates,

10.

ll.

12.

13.

97

each had a significantly different seasonal
abundance in the two pond age classes. In
general, the quality and quantity of inverte-
brates increased with pond age.

The number of ducks in the East Marsh increased

as the season advanced from spring to fall. How-
ever, ducks used the pond most during the spring
and early summer, very little during mid-summer,
and to a limited amount in late summer.

Few diving ducks used the ponds even though they
frequented the marsh. Dabbler duck species used
the ponds to varying degrees, depending on their
habitat preferences and abundance. Blue-winged
teal dominated the species composition.

Ducks used the ponds in a diurnal activity pattern
of highest use early in the morning and declining
use as the time of day advanced.

Ducks spent an average of 15 minutes per pond visit.
Most of their time was spent loafing and little
time feeding.

In 1966, an estimated 6A breeding pairs used the
Second Lead area, of which 35 pairs were at 25
ponds. In 1965 and 1966, respectively, the bays
had 15 and 16 pairs per mile of shoreline, and
the ponds 27 and 51 pairs. Blasting ponds in-
creased the carrying capacity of the Second Lead

area for 35 additional breeding pairs.

14.

15.

l6.

17.

18.

98

Breeding ducks used certain ponds more than
others. Their selection was not significantly
affected by the measured variations in pond

food, cover, water levels, or bay-to—pond dis-
tance. Therefore it was concluded that ducks
responded more to the ponds as a whole, rather
than to some factor(s) within particular ponds.
Displaced eggs and eggshells were found near

many of the ponds. The possible role which this
phenomenon plays in the breeding biology of ducks
is unknown.

Nesting hens used the potholes as jumping-off
places enroute to their nests. Since the ponds
were evenly distributed, neSts also were distri-
buted more evenly after pond construction than
before.

In an area with artificial ponds the number of
nests increased as the number of ponds increased,
but there were more nests in another similar area
without ponds. Apparently hens located their
nests in response to pond proximity but differ-
ences in vegetative cover prevented a clear
examination.

Predators destroyed a high percentage of the nests
around the ponds. Only 6 percent of the nests
were known to be successful in the experimental

area as compared to 32 percent in the check areas.

19.

20.

21.

99

A theoretical production of 113 flying ducks
was calculated for 25 ponds. Since pairs and
broods used many areas other than the ponds,
and since the breeding population did not seem
to be limited by water areas, this estimate is
a theoretical potential and not an observed pro-
duction.

Based on the theoretical estimate of production
and the prorated cost of pond construction, a
flying duck would have cost $0.15 in 1966.
Blasting ponds potentially fulfills three ob-
jectives; to increasing the carrying capacity
for breeding pairs, to attracting ducks for
recreation and protection, and to stimulate
public interest. Whether one or all of these
aims are accomplished depends on individual

situations.

LITERATURE CITED

American Ornithologists' Union. 1957. Check-list of
North American birds. Lord Baltimore Press, Inc.,
Baltimore. 691 pp.

Anonymous. 1963. Blasting with ammonium nitrate—fuel
oil mixtures. Spencer Chemical Co., Kansas City.

25 PP-

. Hach direct reading colorimeter methods manual.
5th ed. Hach Chemical Co., Ames. 91 pp.

Arber, A. 1920. Water plants a study of aquatic angio-
sperms. Cambridge Univ. Press. London. 436 pp.

Bennett, L. J. 1938. The blue-winged teal its ecology
and management. Collegiate Press, Inc. Ames. IAN pp.

Bird, R. H. 1930. Biotic communities of the aspen parkland
of central Canada. Ecol., 11:356—443.

Broley, J. 1950. Identifying nests of the Anatidae of the
Canadian prairies. J. Wildl. Mgmt., 14(4):“52-456.

Brown, D. 195A. Methods of surveying and measuring vege—
tation. Bull. U2, Crown Press. Reading, England.

223 pp.

Bryant, J. E. 1965. Private marshes in southwestern
Ontario. Progress Report Project 02-5-1. Can.
Wildl. Serv. 57 pp.

Bue, I. G., L. H. Blakenship, and W. H. Marshall. 1952.
The relationship of grazing practices to waterfowl
breeding populations and production on stock ponds
in western South Dakota. Trans. N. Amer. Wildl.
Conf., 17:396-A1A.

Canfield, R. H. 1941. Application of the line inter-

ception method in sampling range vegetation. J.
For., 39:388-394.

100

lOl

Collias, N. E. and E. C. Collias. 1963. Selective feeding
by wild ducklings of different species. Wilson Bu11.,
75(1)6-1A.

Cook, A. H. and C. F. Powers. 1958. Early biochemical
changes in the soils and waters of artificially
created marshes in New York. N. Y. Fish and Game J.,
5(l)=9-65.

Coupland, R. T. 1950. Ecology of mixed prairie in Canada.
Ecol. Mono. 20(4):271-315.

Dillion, S. T. 1954. Waterfowl breeding grounds surveys,
on the Delta Marshes of south-central Manitoba--195A.
In Waterfowl populations and breeding conditions—-
summer 1954. U. S. Fish and Wildl. Serv., and Can.
Wildl. Serv., Spec. Sci. Report-~Wild1., No. 27, pp.
92-95.

Dineen, C. F. 1953. An ecological study of a Minnesota
pond. Amer. Midland Nat., 50(2):3A9-376.

Drewien, R. C. 1966. Ecological relationships of breeding
waterfowl to prairie potholes. S. Dakota Dept. Game,
Fish, and Parks. Progress Report 1964-65. 9 pp.
(mimeo).

Dzubin, A. 1955. Some evidences of home ranges in water-
fowl. Trans. N. Amer. Wildl. Conf., 20:278-297.

1956. Waterfowl populations and production on
the Roseneath study area, southern Manitoba, 1955.
In Waterfowl populations and breeding conditions--
summer 1955. U. S. Fish and Wildl. Serv., and Can.
Wildl. Serv., Spec. Sci. Report-—Wild1., No. 30,
pp. 86—88.

Ellerbrock. W. J., Jr. 1956. Waterfowl surveys, Chippewa
National Forest, Minnesota 1937-1955. In Waterfowl
populations and breeding conditions--summer 1955.

U. S. Fish and Wildl. Serv., and Can. Wildl. Serv.,
Spec. Sci. Report--Wild1. No. 30, pp. 202-206.

Ellis, J. H. 1938. The soils of Manitoba. Winnipeg.

Evans, C. D. and K. E. Black. 1956. Duck production studies
on the prairie potholes of South Dakota. U. S. Fish
and Wildl. Serv., Spec. Sci. Report-—Wildl. No. 32.

52 pp-

102

Evans, C. D., A. S. Hawkins, and W. H. Marshall. 1952.
Movements of waterfowl broods in Manitoba. U. S.
Fish and Wildl. Serv., Spec. Sci. Report--Wild1.
No. 16, 47 pp.

Fassett, N. C. 1960. A manual of aquatic plants. Revised.
Univ. of Wis. Press, Madison. 405 pp.

Finch, V. C., G. T. Trewortha, A. H. Robinson and E. H.
Hammond. 1957. Elements of geography physical and
cultural. McGraw-Hill Book Co., Inc., New York.

693 pp-

Furniss, O. C. 1938. The 1937 waterfowl season in Prince
Albert District, central Saskatchewan. Wilson Bu11.,

Gates, J. M. 1962. Breeding biology of the gadwall in
northern Utah. Wilson Bu11., 74(1):43-67.

Girard, G. L. 1941. The mallard: its management in western
Montana. J. Wildl. Mgmt., 5(3):233—259.

Gleason, H. A. 1952. The new Britton and Brown illustrated
flora of the northeastern United States and adjacent
Canada. New York Bot. Garden. New York.

Glover, F. A. 1956. Nesting and production of blue-winged
teal (Anas discors Linnaeus) in northwestern Iowa.

 

Hammond, M. C. 1964. Artificial potholes of Lower Souris.
The Delta seminar on the breeding biology of water-
fowl. 3 pp. (mimeo).

Hammond, M. C. and W. R. Forward. 1956. Experiments on
causes of duck nest predation. J. Wildl. Mgmt.,
20(3):243-247.

Hammond, M. C. and G. E. Mann. 1956. Waterfowl nesting
islands. J. Wildl. Mgmt., 20(4):345-352.

Hochbaum, H. A. 1944. The canvasback on a prairie marsh.
Amer. Wildl. Inst., Washington. 201 pp.

. 1955. Travels and traditions of waterfowl.
Univ. of Minn. Press., Minneapolis. 301 pp.

Ignatoski, F. J. 1966. The ecology of wild ducks in two
central Michigan impoundment marshes. MS. thesis,
Mich. State Univ. East Lansing. 37 pp.

103

Jackson, M. L. 1958. Soil chemical analysis. Prentice—
Hall, Inc., Englewood Cliffs, N. J. 498 pp.

Jahn, L. R. and R. A. Hunt, 1964. Duck and coot ecology
and management in Wisconsin. Tech. Bull. No. 33.
Wis. Cons. Dept. 212 pp.

Johnston, W. A. 1934. Surface deposits and ground-water
supply of Winnepeg map-area, Manitoba. Can. Geol.
Survey. 174:1-110. (mimeo).

Kadlec, J. A. 1960. The effect of a drawdown on the
ecology of a waterfowl impoundment. Game Div. Report
2276. Mich. Dept. Cons., Lansing. 181 pp.

Kalmback. E. R. 1937. Crow-waterfowl relationship in the
Prairie Provinces. Trans. N. Amer. Wildl. Conf.,
2:380-392.

1938. A comparative study of nesting waterfowl
on the Lower Souris Refuge: 1936-1937. Trans. N.
Amer. Wildl. confo’ 3:610—6230

Keith, L. B. 1958. Some effects of increasing soil
salinity on plant communities. Can. J. Bot.,
36(1):79—89.

1961. A study of waterfowl ecology on small
impoundments in southeastern Alberta. Wildl. Mono.
No. 6. 88 pp.

Lacy, C. H. 1959. Artificial potholes and level ditch
development as a means of increasing waterfowl pro-
duction. MS. thesis. Utah State Univ., Logan.

Leitch, W. G. 1956. Waterfowl breeding ground surveys of
Ducks Unlimited (Canada) in Alberta, Saskatchewan and
Manitoba (Caron Pothole study area-~Saskatchewan).

In Waterfowl populations and breeding conditions--
summer 1955. U. S. Fish and Wildl. Serv., and Can.
Wildl. Serv., Spec. Sci. Report--Wild1., No. 30,
pp. 102—104.

Leopold, A. 1933. Game management. Charles Scribner's
Sons, New York. 481 pp.

Li, J. C. R. 1957. Introduction to statistical inference.
Edwards Bros., Inc., Ann Arbor. 568 pp.

L6ve, A. and D. Lave. 1954. Vegetation of a prairie
marsh. Bull. Tor. Bot. Club, 81:16-34.

104

Low, J. B. 1947. Review of new techniques—-waterfow1.
Trans. N. Amer. Wildl. Conf., 12:339-343.

Malone, C. R. 1965. Dispersal of aquatic gastropods via
the intestinal tract of water birds. Nautilus,
78(4):l35-139-

Maquire, B., Jr. 1959. Passive overland transport of
small aquatic organisms. Ecol., 40:312.

Mathiak, H. A. 1964. Testing small pothole development
for wetland wildlife. Wis. Cons. Dept. Annual
Progress Report II-A. Madison. 7 pp.

1965. Pothole blasting for wildlife. Publ. 352.
Wis. Cons. Dept., Madison. 31 pp.

'Mathiak, H. A. and A. F. Linde. 1956. Studies of level
ditching for marsh management. Tech. Wildl. Bull.
No. 12. Wis. Cons. Dept., Madison.

Mathisen, J. 1964. An evaluation of reclaimed woodland
potholes for waterfowl use, production and harvest.
Progress Report No. 1. Chippewa Nat'l. Forest, Cass
Lake, Minn. 4 pp.

. 1965. An evaluation of reclaimed woodland pot-
holes for waterfowl use, production, and harvest.
Progress Report No. 2. Chippewa Nat'l. Forest, Cass
Lake, Minn. 4 pp.

Mathisen, J, J. Byelich, and R. Radtke. 1964. The use of
ammonium nitrate for marsh blasting. Trans. N. Amer.
Wildl. Conf. 29:143—149.

McKinney, F. 1965. Spacing and chasing in breeding ducks.
Wildfowl Trust Annual Report. 16:92-106.

Mendall, H. L. 1949. Breeding ground improvements for
waterfowl in Maine. Trans. N. Amer. Wildl. Conf.
14:58-63.

Odum, E. P. 1959. Fundamentals of ecology. W. B. Saunders
Co., Philadelphia. 546 pp.

Penfound, Wm. T. 1953. Plant communities of Oklahoma lakes.
Ecol., 34(3):561-583.

Pennak, R. W. 1953. Freshwater invertebrates of the United
States. Ronald Press Co., New York. 769 pp.

105

Pirnie. M. D. 1935. Michigan waterfowl management.
Mich. Dept. Cons., Game Div., Lansing. 328 pp.

Proctor, V. M. 1964. Vialibity of crustacean eggs re-
covered from ducks. Ecol., 45(3):656-658.

-.Provost, M. W. 1948. Marsh-blasting as a wildlife manage-
ment technique. J. Wildl. Mgmt., 12(4):350-387.

Radtke, R. and J. Byelich. undated. Marsh blasting with
ammonium nitrate. Mich. Dept. Cons., Lansing.

13 pp.

Rearden, J. D. 1951. Identification of waterfowl nest
predators. J. Wildl. Mgmt., 15(4):386-395.

Reeves, H., M. Lundy, and F. Kreller. 1956. Waterfowl
breeding ground survey-—Success study area--
Saskatchewan, 1955. In Waterfowl populations and
breeding conditions--summer 1955. U. S. Fish and
Wildl. Serv., and Can. Wildl. Serv., Spec. Sci.
Report-—Wild1., No. 30, pp. 54-62.

Reid, G. K. 1961. Ecology of inland waters and estuaries.
Reinhold Publ. Corp., Pittsburgh. 375 pp.

Rogers, J. P. 1964. Effects of drought on reproduction
of the lesser scoup. J. Wildl. Mgmt. 28(2):213-222.

Saugstad. 1939. Effects of artificial ponds on our duck
populations. Bimonthly Bull. 2(2):6-8.

Sayler, J. W. 1962. Effects of drought and land use on
prairie nesting ducks. Trans. N. Amer. Wildl. Conf.
27:69-79.

Schlichting, H. E. 1958. The role of waterfowl in the
dispersal of algae. Ph.D. Dissertation. Michigan
State Univ., East Lansing. 259 pp.

Scoggan, H. J. 1957. Flora of Manitoba. Nat'l. Museum
of Can. Bull. No. 140.

Scott, T. G. and W. L. Dever. 1940. Blasting to improve
wildlife environment in marshes. J. Wildl. Mgmt.
4(4):373-374.

Shull, C. A. 1914. The longevity of submerged seeds.
Plant World. 17:329-337.

Siegel, S. 1956. Nonparametric statistics for the be-
havioral sciences. McGraw-Hill Book Co., Inc.,
New York. 312 pp.

1

f-.‘

 

J

3 ‘ .
y.’-h-‘l_-J

 

106

Smith, A. G. 1958. A progress report on the Lousana
waterfowl study area 1953-1957. Lousana, Alberta.
95 pp. (mimeo.)

Smith, M. M. and R. Droll. 1965. Breeding ground surveys——
Southern Manitoba, 1965. In Waterfowl status report
1965. U. S. Fish and Wildl. Serv., Spec. Sci. Report--
Wildl., No. 90, pp. 22-24.

. 1966. Breeding ground surveys--southern Manitoba,
1966. In Waterfowl status report 1966. U. S. Fish
and Wildl. Serv., Spec. Sci. Report-—Wildl., No. 99,
pp. 14—15.

Smith, R. H. 1953. A study of waterfowl production on
artificial reservoirs in eastern Montana. J. Wildl.
Mgmt. 17(3):276-291.

Sowls, L. K. 1948. The Franklin ground squirrel, Citellus
franklinii (Sabine), and its relationship to nesting
dUCkS. Jo Mammal. 29(2):ll3-l370

 

. 1950. Techniques for waterfowl-nesting studies.
Trans. N. Amer. Wildl. Conf. 15:478-487.

. 1955. Prairie ducks. Stackpole Co., Harrisburg.
193 pp-

Spurway, C. H. and K. Lowton. 1949. Soil testing a practical
system of soil fertility diagnosis. Mich. State Univ.
Agri. Tech. Bul. 132, East Lansing.

Steel, P. E., P. D. Dalke and E. G. Bizeau. 1956. Duck
production at Gray's Lake, Idaho, 1949-1951. J. Wildl.
Mgmt. 20(3) 279-285.

Sterling, R. T. 1956. Waterfowl breeding ground surveys of
Ducks Unlimited (Canada) in Alberta, Saskatchewan and
Manitoba (Southey Area--Saskatchewan). In Waterfowl
populations and breeding conditions-—summer 1955.

U. S. Fish and Wildl. Serv., and Can. Wildl. Serv.,
Spec. Sci. Report--Wild1. No. 30, pp. 104-105.

Stoudt, J. H. and B. A. Yeager. 1956. Waterfowl breeding
ground survey of Redvers Area, Saskatchewan, 1952-
1955. In Waterfowl populations and breeding conditions
--summer 1955. U. S. Fish and Wildl. Serv., and Can.
Wildl. Serv., Spec. Sci. Report--Wild1., No. 30,
pp. 63—72.

Strohmeyer, D. L. 1963. An evaluation of artificial
potholes in marsh management. MS. Thesis. Univ.
of Iowa, Ames.

107

Uhler, F. M. 1956. New habitats for waterfowl. Trans.
N. Amer. Wildl. Conf. 21:453-469.

Usinger, R. L. Aquatic insects of California with keys to
North American genera and California species. Univ.
of Calif. Press, Berkley. 408 pp.

Welch, P. S. 1948. Limnological methods. McGraw-Hill
Book Co., Inc., New York. 381 pp.

. 1952. Limnology. McGraw-Hill Book Co., Inc.,
New York. 538 pp.

Westerberg, D. D. 1965. Wildlife management plan/wetlands
of the Chippewa National Forest, Minnesota. U. S.
Forest Service, Cass Lake. 17 pp.

Williams, C. S. and W. H. Marshall. 1938. Duck nesting
studies, Bear River Migratory Bird Refuge, Utah,

Ponds mentioned

following locations.

Ponds
S-l thru S-28
T-l thru T-8
B-l

R—l thru R-4

108

APPENDIX A

in the text were blasted

T.14 N,
T.14 N,
T.14 N,

Location
R.6 WPM. sec.
R.6 WPM. sec.
R.6 WPM. sec.

R.7 WPM.

sec.

in the

25
20, s. 1/2
18, NW, 1/4

12, sw, 1/4

109

APPENDIX B

COMMON AND SCIENTIFIC NAMES OF

DUCKS MENTIONED IN THE TEXT

The scientific names of ducks were taken from the

American Ornithologists' Union Check-list (1957).

 

 

 

 

 

 

 

 

 

 

Mallard Anas platyrhynchos
Black Duck Anas rubripes
Gadwall Anas strepera
Pintail Anas acuta
Blue-winged Teal Anas discors
Green-winged Teal Anas carolinensis
Shoveller Spatula c1ypeata
Baldpate Mareca americana
Canvasback Aythya valisineria
Redhead Aythya americana
Scaup (Lesser Scaup) Aythya affinis

 

Ruddy Duck Oxyura jamaicensis

 

"7'1111111111111111“