INFORMATION TO USERS This re p ro d u c tio n was m ad e from a copy o f a d o c u m e n t sent to us fo r microfilming. While the m o st advanced tech no log y has been used to p h o to g ra p h and repro d u ce this d o c u m e n t, the q uality o f the re p ro d u c tio n is heavily d e p e n d e n t u p o n the qu ality o f th e m aterial s u b m itte d . T h e following e x p lan a tio n o f tech niq ues is p rovided to help clarify m arkings or n o ta tio n s w hich m ay ap p e a r o n this re p ro d u ctio n . 1 .T h e sign or “ t a rg e t” fo r pages ap parently lacking from the d o c u m e n t p h o to g ra p h e d is “ Missing Page(s)” . I f it was possible to o b tain the missing page(s) or section, they are spliced into th e film along w ith adjac en t pages. This may have necessitated c u ttin g thro ug h an image and du plicating adjacent pages to assure co m p lete c o n tin u ity . 2. When an image on th e film is o blite rated with a ro u n d black m ark , it is an indication o f e ith e r blurred copy because o f m o v e m e n t during ex po sure, duplicate co p y , o r co p y rig h te d m aterials t h a t should n o t have been filmed. F o r blurred pages, a good image o f the page can be fo u n d in the ad jac en t fram e. If co p y rig h te d m aterials were deleted, a target n o te will ap p e ar listing th e pages in th e a d jac en t fram e. 3. When a m ap, draw ing o r ch a rt, etc., is p a rt o f th e m aterial being p h o to g ra p h e d , a definite m e th o d o f “ sectioning” th e m aterial has been followed. It is c u sto m ary to begin filming at the u p p e r left h an d co rn er o f a large sheet and to co n tin u e from left to right in equal sections w ith small overlaps. If necessary, sectioning is co n tin u e d ag ain—beginning belo w the first ro w and co n tin u in g on until co m p lete. 4. F o r illustrations th a t c a n n o t be satisfactorily re p ro d u c e d by xerographic m eans, p h o to g ra p h ic p rin ts can be purch ased at additio nal cost and inserted into y o u r x erographic copy. These p rin ts are available u p o n request from the D issertations C u s to m e r Services D e p a rtm e n t. 5. Some pages in any d o c u m e n t may have indistinct p rint. In all cases the best available c o p y has been filmed. University Microfilms International 300 N. Z eeb R oad Ann Arbor, Ml 48106 8520555 R u sz , P a tric k J a m e s WATERBIRD RESPONSES TO HABITAT CHANGES ON AN OPEN WATER SYSTEM IN CENTRAL MICHIGAN M ic h ig a n State U n iv e rs ity University Microfilms International 300 N. Zeeb Road, Ann Arbor, Ml 48106 Ph.D. 1985 PLEASE NOTE: In all c a se s this material h as b een filmed in th e best possible w ay from the available copy. Problem s encountered with this d o cum ent have been identified here with a ch eck mark V . 1. Glossy photographs or p a g e s ______ 2. Colored illustrations, paper or print 3. Photographs with dark b a c k g ro u n d ______ 4. Illustrations a re p o o r c o p y _______ 5. P ages with black marks, not original c o p y _______ 6. Print shows through as there is text on both sides of p a g e _______ 7. Indistinct, broken or small print on several pages 8. Print exceeds m argin req u irem en ts_______ 9. Tightly bound co p y with print lost in s p in e ________ 10. Computer printout pages with indistinct print_______ 11. P a g e (s)_____________lacking w hen material received, a n d not available from school or author. 12. Page(s) seem to b e missing in numbering only as text follows. 13. Two pages num bered 14. Curling and wrinkled pages . Text follows. 15. Dissertation co n tain s pages with print at a slant, filmed a s received 16. Other University Microfilms International WATERBIRD RESPONSES TO HABITAT CHANGES ON AN OPEN WATER SYSTEM IN CENTRAL MICHIGAN By Patrick James Rusz A DISSERTATION Submitted to Michigan State University in partial fulfullment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife 1985 ABSTRACT WATERBIRD RESPONSES TO HABITAT CHANGES ON AN OPEN WATER SYSTEM IN CENTRAL MICHIGAN By Patrick James Rusz Waterbird use of a newly-created 350 ha industrial cooling pond in central Michigan was examined over a 6year period, 1979-1984. Numbers, distributions, and ac­ tivities of waterbirds were recorded. Water depth, aquatic macrophytes, and benthos were sampled, and data on the fish community were obtained from secondary sources. A chi-square test was used to determine if dis­ tributions of waterbirds and macrophytes coincided. The water level was lowered by about 1.3 m for the entire year in 1980, and again in fall of 1983. The cool­ ing pond was gradually drained in fall of 19 84. There was only 1 macrophyte species present, and the distribution of plants was similar each year. The benthic community was poorly developed throughout this study. The reduced water level in 1980 resulted in a much greater density of plant material near the surface than in any other year. Small fish were most numerous in 1979, 1982, and 1983. Waterbird use, especially by American wigeon, redhead, and American coot, was much greater in 1980 than in any other year. Common loon and horned grebe had more use-days Patrick James Rusz in 1979, double-crested cormorant and red-breasted mergan­ ser were most abundant in 19 82, and common merganser use was greatest in 19 83. Canada goose use was high in all 3 years in which the water level was reduced, and greatest during the progressive drawdown in 1984. The number of species seen declined each year, but the mean weekly species diversity index was highest in 19 80. Coincidence of birds and vegetation was particularly evident for American wigeon, redhead, and American coot in 19 80, and these species were most numerous where plant growth was most extensive. Plant-eating and piscivorous species did not use the cooling pond intensively except in years when the levels of plants and small fish exceeded that found in most open water systems. Canada goose use seemed influenced by the amount of dry ground loafing sites. ACKNOWLEDGEMENTS Funding for this study was provided by Consumers Power Company. I owe special thanks to Gary Dawson, of Consumers Power Company's Environmental Department, for his valuable suggestions and technical assistance. I am indebted to Harold Prince, my research advisor, for his guidance and editing of the manuscript. I thank Jonathan Haufler, Donald Beaver, and Leighton Leighty for serving on my research committee. Stanley Zarnoch and Carl Ramm provided valuable advice on data analysis. I am particularly grateful to Richard Rusz for his help in collecting and synthesizing the field data. ii TABLE OF CONTENTS Page LIST OF TABLES...................................... iv LIST OF FIGURES..................................... vii INTRODUCTION......................................... 1 DESCRIPTIONS OF STUDY A R E A S........................ 5 Location And General Setting.................. Cooling P o n d ................................... Dow Tertiary Treatment P o n d ................... 5 5 9 METHODS AND MATERIALS............................... 11 Waterbird Observations........................ Measurement Of Habitat Variables............. Data Analysis.................................. 11 12 14 RESULTS.............................................. 17 Habitat Conditions In The Cooling P o n d ....... Vegetation................................ Benthos................................... Fish....................................... Waterbird Use, 1979-1983...................... Waterbird Activities.......................... Relationships Between Waterbird Distri­ butions And Habitat Variables.............. Waterbird Response To Fall Drawdown in 1984.. 17 17 23 23 26 42 DISCUSSION........................................... 55 BIBLIOGRAPHY......................................... 71 APPENDIX............................................. 75 iii 44 50 LIST OF TABLES Page Table 1 2 3 4 5 Relative water depths and plant densities in the cooling pond during 1979-1983 by season and year. Water depths and plant densities varied in fall of 1984 as water was drained from the cooling p o n d ............. 20 Estimated use-days by year for the 17 most abundant waterbird species on the cooling pond. Estimates are to the nearest 10 use-days. The highest number of use-days recorded in a year for each species is underlined...................................... 28 Estimated use-days by season, summed for 5 years (1979-1983), for the 17 most abun­ dant waterbird species on the cooling pond. Estimates are to the nearest 10 use-days. The highest number of use-days by season is underlined for each species.................... 30 Estimated use-days by year for the 13 most abundant waterbird species on the Dow ter­ tiary treatment pond. Estimates are to the nearest 10 use-days. The highest number of use-days recorded in a year for each species is underlined.......................... 32 Estimated use-days by season, summed for 5 years (1979-1983), for the 13 most abun­ dant waterbird species on the Dow tertiary treatment pond. Estimates are to the near­ est 10 use-days. The highest number of use-days by season is underlined for each species.......................................... 33 Estimated use-days for waterbird species observed on the cooling pond in at least 1 year but not in all 5 years, and/or for which a low number (5-year total <650) usedays were recorded. Numbers in parentheses are total observations; dashes indicate species not seen in that y e a r ................. 35 iv LIST OF TABLES (cont'd): Table 7 8 9 10 11 Page Estimated use-days for waterbird species observed on the Dow tertiary treatment pond in at least 1 year but not in all 5 years, and/or for which a low number (5year total <130) use-days were recorded. Numbers in parentheses are total observa­ tions; dashes indicate species not seen that y e a r .......... ............................ 37 Mean 24-hour species diversity indices (x) and standard deviations (SD) by sea­ son and year for waterbirds on the cool­ ing pond. Numbers in parentheses are total 24-hour observation periods in a season or ye a r ................................. 41 Percentages of birds of the 17 most abun­ dant waterbird species on the cooling pond observed feeding. Numbers in paren­ theses are total 15-second observations of individuals of a species.................. 43 Number of grid squares (N=440) occupied by both birds and vegetation (C) and to­ tal grid squares occupied by birds (n) of the 17 most abundant species on the cool­ ing pond. Species-year combinations are ranked based on the difference in the number of grid squares occupied by birds with and without vegetation. Higher ranks generally indicate better coinci­ dence of birds and vegetation, but small sample sizes for certain species-year combinations prohibit some comparisons....... 45 Occurrence of birds of 3 species in 4 vegetated locations in the cooling pond in 1980. Total areas (in ha) of loca­ tions are: 1 - 3.0, 2 - 2.5, 3 - 11.0, 4 - 68.5. Locations 1, 2, and 3 con­ sisted of continuous 0.5 ha grid squares with at least 10 percent of each grid square covered by macrophytes. Location 4 consisted of the remaining vegetated grid squares in the cooling pond (see Figure 3 ) ....................................... 49 v LIST OF TABLES (cont'd): Page Table 12 A1 Estimated use-days, by species, in fall of 6 years (1979-1984) for waterbirds on the cooling pond. Dashes indicate species not seen that fall.................. 51 Common and scientific names of waterbird species observed on the cooling pond and Dow tertiary treatment pond during 1979-1983. Numerals indicate number of years in which a species was seen. Letters indicate relative abun­ dance: VA = very abundant (>1,000 usedays on cooling pond, >200 use-days on Dow tertiary treatment pond) each year; A = Abundant (>1,000 use-days (>200 for Dow tertiary treatment pond) in at least 1 but not in all 5 years); C = common (>650 use-days for 5-year total, >130 use-days on Dow tertiary treatment pond, but not meeting criterion for abundant); U = uncommon (<650 use-days, < 1 3 0 on Dow tertiary treatment pond for 5-year total)....................................... 75 vi LIST OF FIGURES Page Figure 1 2 3 4 5 6 Location of the study area in central Michigan ................................ 6 Areas of different depths (in meters) and general locations of soil undis­ turbed during excavation of the cooling pond. Within the areas delineated, 9 0 percent of measured depths differed by less than 0.3 m. Depths were 1.3 m lower in 19 80 and from late July to December of 1983. A progressive draw­ down of the cooling pond occurred in fall of 1984.............................. 7 Locations of vegetated grid squares, in the cooling pond. Darkened grid squares indicate at least 10 percent of the grid square was covered by macrophytes. Num­ erals (1-3) indicate locations where there were contiguous grid squares in which vegetation covered at least 10 percent of each grid square............. 18 Dense growth of macrophytes on the cool­ ing pond in early fall of 1980. The upper photo shows macrophytes draping boat oar; the lower photo shows 1 m^ wooden frame supported by macrophytes growing to water surface................ 21 Relative estimated total waterbird usedays on the cooling pond and Dow ter­ tiary treatment pond by y e a r ............ 27 Number of species observed on the cool­ ing pond and Dow tertiary treatment pond by y e a r ................................... 39 vii INTRODUCTION While freshwater wetlands are diminishing at an alarm­ ing rate (Shaw and Fredine 1956, Whitesell 1970) open water systems such as lakes, recreational impoundments, and in­ dustrial ponds for cooling or wastewater treatment are in­ creasing in number (Street 1982, Uhler 1956). The morpho­ metry, water chemistry, and plant and animal communities of open waters differ markedly from those of wetlands. Open water systems are typically deep, lack substantial emergent vegetation, and are often subject to considerable wave ac­ tion. Diversity of waterbird microhabitats within most open water systems seems low in comparison with wetlands (Weller and Fredrickson 1974) . Despite their seemingly low diversity of microhabitats, many open water systems are used by significant numbers of a wide variety of waterbirds during specific times (Prince et al. 1984, Reeves 1980, Thornburg 1973, Bellrose 1976). How­ ever, there is a dearth of information on the nature of waterbird use of open waters because such habitats are usually not managed for waterbird use by industries or gov­ ernmental agencies. There are a few detailed descriptions of waterbird use of certain open waters such as wastewater treatment lagoons (Willson 1975, Dornbush and Anderson 19 64, 1 Medema 19 80), and there have been several recent attempts to correlate waterbird use with selected habitat variables on open waters. Teer White and James (197 8) and Hobaugh and (1981) used multivariate methods to characterize waterfowl use of open water systems in Texas, but the for­ mer study was primarily focused on the marshy fringes of lakes rather than the lake proper. Reeves (19 80) used multiple regression analysis to compare levels of certain habitat variables with distributions of waterbirds (by species) (1982) on 3 lakes in central Michigan. Blomberg used regression analysis to determine habitat characteris­ tics which influenced duck use of gravel pits in Colorado. Such correlation studies have led to management re­ commendations by the respective authors, but verification of waterbird responses to habitat manipulations on open waters is virtually absent. Street An exception is the study of (19 82) who reported increased waterfowl use of a gravel pit in England after construction of sheltered, shore-based loafing sites, introduction of a variety of vegetational species, and addition of organic matter to stimulate production of invertebrates. The complexity of interrelated factors which might in­ fluence habitat selection in open water systems makes it difficult to determine specific cause-effect relationships. Habitat manipulations usually affect many variables such as the density, relative abundance, diversity, availability, and species composition of both benthic and surface 3 dwelling plants and animals. Hence, measured responses of birds can seldom be directly linked to changes in a single parameter or even set of parameters. Costs and other con­ siderations make strict replications or use of control areas nearly impossible in studies of large open water systems. In addition, year-to-year differences in migration chronology and/or population numbers due to weather or con­ ditions on the breeding or wintering grounds may further confound interpretation of results of field studies involv­ ing migrating birds. Thus, there is need for long term field studies in open water systems in which such problems are minimized. The primary purpose of this study was to determine, over a 6-year period, the species assemblages, and levels and types of waterbird use associated with a newly-created open water system lacking the complexity of habitat compo­ nents usually found in naturally-formed open waters. The water depth in this system was manipulated during my study, so an adjacent open water system was also studied during the same period to facilitate interpretation of results from the primary study area. Implicit in my research is the hypothesis that open water systems are used by discrete waterbird communities which can be predicted on the basis of relatively few proximate factors. For purposes of this report, the term "waterbird(s)" refers to all avian species, except herring gull and ring­ billed gull, which used the off-shore areas of the open water systems I studied. Herring gull and ring-billed gull were the most frequently observed species on the primary study area, and a large breeding colony of ring-billed gull was present there each year of my study. These 2 species are excluded from my study since the data on their use of the study area are presented by Rusz (19 85). Various species of small, non-swimming shorebirds were also present on the shores of the open water systems I studied, but are not referred to in this report. Scientific names of bird species included in this study are in Appendix A. DESCRIPTIONS OF STUDY AREAS Location And General Setting The open water systems I studied are a 350 ha indus­ trial cooling pond and a 150 ha wastewater treatment pond located in central Michigan near major migration corridors (Bellrose 1976) and concentration areas of Canada geese and both dabbling and diving ducks. The cooling pond, which served as the primary study area, and the wastewater treat­ ment pond are in an industrial complex on the southern edge of the City of Midland (Figure 1). The industrial complex includes Consumers Power Company's Midland Energy Center and Dow Chemical Company facilities. Land use near the complex is industrial and commercial on the north, residen­ tial and light commercial on the west, and rural residen­ tial and agricultural on the south and east. Cooling Pond The cooling pond was constructed to serve Consumers Power Company's Midland Energy Center and associated nuclear power plant. Soil was excavated to the level of a lowland drained by a small creek. land was left undisturbed Soil in most of the low­ (Figure 2), however additional earth was excavated to provide a deeper supplemental water River R E S ID E N T IA L L IG H T N U C L EA R POND Industrial City Agriculture Rural 22 CTl Study Area Figure 1. Location of the study area in central Michigan. Figure 2. Areas of different depths (in meters) and gen­ eral locations of soil undisturbed during ex­ cavation of the cooling pond. Within the areas delineated, 90 percent of measured depths diff­ ered by less than 0.3 m. Depths were 1.3 m lower in 19 80 and from late July to December of 1983. A progressive drawdown of the cooling pond occurred in fall of 19 84. Figure 2 -{—»\ F— o Io \ o Undisturbed Soil lit# o 75 percent open water) spring, to 15 June; summer, 16 June to 14 September; fall, 15 September to ice-in (<50 percent open w a ter). Diversity indices for each week's sampling period in 197 9-1983 for the cooling pond were calculated by the Shannon-Wiener formula (Wilson and Bossert 1971): s Hs = - 2 Pj^loggPi i=l where H s is the diversity index, s is the number of species in a group, and pj_ is the proportion of all birds in the ith group. In the calculation of diversity indices and in all statistical analyses, actual count data, not estimated use-days, were used. I performed one-way analyses of variance to determine if differences in the mean weekly diversity indices among seasons and among years were statistically significant. Bartlett's test of heterogeneity of variances and the Kolmogorov-Smirnov test of non-normality were significant for the untransformed data, so the subsequent ANOVA and a priori contrasts were made using transformed data (log x + 1 ). A chi-square goodness-of-fit test was used to deter­ mine if the observed number of grid squares occupied by both plants and birds on the cooling pond was significantly different from the expected number based on the proportion 16 of the total grid squares which contained plants. The same test was used to determine if the observed number of grid squares with birds but without plants was significantly different from the expected value. Waterbird distribution data for 17 species on the cooling pond in each of 5 years (1979-1983) were subjected to this chi-square test. Further analysis of distribution data for 3 species of birds for which exceptionally high use and coincidence with plants was recorded in 1980 was performed. Vegetated grid squares in the cooling pond were considered to be in 4 lo­ cations. Locations 1, 2, and 3 consisted of contiguous 0.5 ha grid squares with at least 10 percent of each grid square in the location covered by macrophytes. Location 4 consisted of the remaining vegetated grid squares not in a single location) on the cooling pond. (actually I adopted the null hypothesis that the numbers of birds of each species seen in a particular location would be proportional to the total area of that location. Only birds which occupied vegetated grid squares were considered in the analysis. A chi-square test was used to determine the va­ lidity of the null hypothesis, and an a posterior test em­ ploying the Bonferroni jz statistic was used to identify significant components of the chi-square analysis. al. Neu et (1974) suggested this 2-step method for evaluating preference or avoidance by animals of specific habitats. RESULTS Habitat Conditions In The Cooling Pond Vegetation Two species of macrophytes were present the first year (1979) after the cooling pond was filled. species was spiked water milfoil A small bed The predominant (Myriophyllum spicatum) . (<0.2 ha) of floatingleaf pondweed (Potamogeton natans) was found in the northwest part of the cooling pond. From early summer of 19 80 through 19 84, only spiked water milfoil was found. Visual observations and SONAR and grab samples indicat­ ed the macrophytes grew only where the soil (cooling pond bottom) was undisturbed during construction (Figure 2). Colonization of these areas of inundated organic soil was extensive. Because of this bottom type - plant relation­ ship, the distribution of macrophytes was similar each year. Analysis of the false color infrared aerial photographs taken in 19 81 revealed that about 40 percent of the grid squares had some detectable macrophytes within about 1 m of the surface (Figure 3). The percentage of each grid square covered by macrophytes ranged from 1 to 90 percent. were 3 locations There (Figure 3) with contiguous grid squares in which vegetation covered at least 10 percent of each grid 17 Figure 3. Locations of vegetated grid squares in the cooling pond. Darkened grid squares indi­ cate at least 10 percent of the grid square was covered by macrophytes. Numerals (1-3) indicate locations where there were con­ tiguous grid squares in which vegetation covered at least 10 percent of each grid square. Figure 3. □ g 360m I 1-- 1-- 1 1=1 * rS_ ^ 1 Powerblock 19 square. About 39 percent of location 1, 17 percent of lo­ cation 2, and 29 percent of location 3 were covered by macrophytes. The average percent coverage by plants of the remaining vegetated grid squares was about 4 percent. The density of plant materials near the surface was much greater in 19 80 than in any other year (Table 1). In late summer and fall of 19 80, the spiked water milfoil formed extensive mats on the surface (Figure 4). Until late October, these mats were so dense that waterbirds (in­ cluding great blue heron) could stand on them, and the plant material supported a i m strips. 2 wooden frame made of 2.5 cm square The frame was placed in 125 randomly selected loca­ tions within plant beds in early September, and in each lo­ cation vegetation covered 100 percent of the frame's inter­ ior. Primarily as a result of post-growing season sene­ scence, the macrophyte density gradually decreased visibly in November. Macrophytes were deposited by waves in piles up to 0.5 m deep and 4 m wide along the east shore in late November, but the plant density throughout the vegetated areas remained many times higher than in any other year. The density of plants in 1980 was associated with a lowered water level throughout the year (Table 1). The water level was lowered to a similar level in late July of 1983 (after much of the growing season was over), but the plant density did not noticeably increase and appeared simi­ lar to that in 1979, 1981, and 1982. The mean dry weight of plant material within 1 m of the surface was 12.9 g/m (n=9, 20 Table 1. Relative water depths and plant densities in the cooling pond during 19 79-19 83 by season and year. Water depths and plant densities varied in fall of 19 84 as water was drained from the cooling pond. Year Season Water Depth Plant Density 1979 Spring Summer Fall High3High High Low Low Low 19 80 Spring Summer Fall Low^ Low Low Moderate-3 High4 High 19 81 Spring Summer Fall High High High Low Low Low 19 82 Spring Summer Fall High High High Low Low Low 1983 Spring Summer Fall High High Low Low Low Low3 Mean depth 5.4 m. 2 Mean depth 4.1 m. 3 Intermediate between low and high densities in late spring; lower during most of spring waterbird migration. 4 Very dense mats of plant material at surface (see Figure 4). 5 Mean dry weight of plant material within 1 m of sur­ face was 12.9 g/m3 (n = 9, SD = 3.1) in late summer of 1983. Figure 4. Dense growth of macrophytes on the cooling pond in early fall of 1980. The upper photo shows macrophytes draping boat oar; the lower photo shows 1 m 2 wooden frame supported by macrophytes growing to water surface. 21 22 SD=3.1) in late August of 19 83. Thereafter, the plant den­ sity appeared to decrease steadily during the fall. the mean sample dry weight of 12.9 g/m Thus, was indicative of the maximum amount of plant material near the surface available to migrating waterbirds in fall of 1983. The distribution and density of plant material in late summer of 1984 appeared similar to that in 1979, 1981, 1982, and 19 83. However, from 20 September to 27 September (prior to drawdown) macrophytes were cut and removed from the upper 1.5 m of about 7 5 percent of the cooling pond surface. Approximately 250 m^ of plant material moved (non-dried) was re­ (about one-third of the estimated total amount). Plants were cut in 2 of the 3 locations (locations 1 and 2) where macrophytes were most extensive, and in most grid squares north of location 3 (Figure 3). The water level in 19 84 was lowered by about 15 cm each day from 1 October to 18 October, and by about 20 cm each day from 19 October to 1 November. Thereafter, water was mostly confined to isolated depressions and the deep area in the northeast part of the cooling pond (Figure 2). From about 10 October to 29 October, the density of plant mater­ ials at the surface increased steadily (as stem length ex­ ceeded water depth) until there was essentially no standing water in the plant beds on 29 October. Dense mats of macro­ phytes were present in water less than 30 cm deep from 22 October to 2 8 October in the unharvested southeast part of the cooling pond. 23 Benthos Sampling with an Eckman dredge in 40 randomly selected grid squares in late May and June of 19 81 revealed a very sparse and poorly developed benthic community. Only 9 of 85 samples had invertebrates. The most specimens captured in any sample was 6 chironomid (midge) larvae. The other 8 samples which contained invertebrates had 4 or less chiro­ nomid larvae. Two samples had a few snail and clam shell fragments, and 1 sample had a single mayfly larvae. (Hexaqenia s p .) Each of the samples which contained chironomid larvae were from within the macrophyte beds and also con­ tained organic material from the unexcavated area of the cooling pond bottom. Hence, it was concluded that chirono­ mid larvae were sparse and generally confined to the macro­ phyte beds and organic bottom type. ous rocks Examination of numer­ (rip-rap) along the shore both in 19 81 and in 19 84 after much of the pond was drained, revealed only a few mayflies (Heptageniidae) and sow bugs (Isopoda). After most of the cooling pond was drained in early December of 19 84, a sparse population of large clams ha) was visible. (density <10 per Because of the low numbers of specimens in my samples and previous (1979) samples of Lawler, Matusky & Skelly Engineers, Inc. (1980), no further samp­ ling of benthos was conducted. Fish No systematic attempt to periodically sample fish pop­ ulations in the cooling pond was made. However, 24 electrofishing and occasional netting by Lawler, Matusky & Skelly Engineers, Inc. (1980 and unpublished data), more intensive netting by Consumers Power Company data) (unpublished in fall of 1984, my field observations, and the hist­ ory of colonization of the cooling pond by fish all suggest the fish community was numerous and diverse during my study. Fish entered the cooling pond via intake pipes during filling in spring (7 April to 4 May) and fall to late December) of 197 8. (8 November Screens prevented entry of most fish larger than 15 cm in length; hence, mostly minnows and young of other species entered the cooling pond (Mr. P. Bradley Latvaitis, Environmental Department, Consumers Power Company, personal communication). large and dense schools of minnows In 1979, extremely (Cyprinidae) could be seen from shore, and electrofishing provided further evi­ dence of the abundance of minnows communication). (Mr. Latvaitis, personal In 19 80 and 19 81, my observations and occasional netting by Lawler, Matusky & Skelly Engineers, Inc. indicated that yellow perch rock bass (Perea flavescens) and (Ambloplites rupestris) more than 10 cm in length were extremely abundant. In 19 82 and 19 83, extremely dense and extensive schools of small (less than 4 cm in length) gizzard shad (Dorosoma cepedianum) were visible near the cooling pond's surface on numerous occasions. These schools were so dense that it was possible to scoop them from the water with 25 hand-held nets. ponds, Gizzard shad often spawn in sloughs, lakes, and large rivers (Becker 1976), so the gizzard shad I observed and captured could have originated from either spawning in the cooling pond or entry through the intake pipes in late 1980. Water from the Tittaba- wassee River, which has very heavy fall runs of gizzard shad, was added to the cooling pond in late November and December of 19 80 to bring its level back to normal after drawdown. From 14 September to 19 December of 1984, Consumers Power Company conducted an intensive trap netting program to transfer game fish from the cooling pond to the Tittabawassee River prior to completion of the drawdown. Results of the trap netting indicated gizzard shad and carp nis carpio) had the highest standing crops among species in the cooling pond. (Cypri- (biomass) It was estimated that at least 16 metric tons of gizzard shad and a similar bio­ mass of carp were in the cooling pond in fall of 19 84. Among fishes less than 12 cm in length, black crappie xis nigromaculatus) was most numerous, followed (Pomo- (in order of numerical abundance) by gizzard shad, various sunfishes (Centrarchidae), and yellow perch communication). (Mr. Latvaitis, personal Small minnows, which electrofishing re­ sults and observations indicated were extremely abundant in 1979, were not numerous in fall of 1984. 26 Waterbird Use, 1979-1983 The total number of waterbird use-days on the cooling pond ranged from a high of about 147,000 in 1980 to a low of approximately 20,700 in 1982 (Figure 5). The number of use-days recorded in 19 80 was much higher than the com­ bined total of 117,700 use-days for the other 4 years for which yearly use was estimated, and was 17 8 percent higher than the 5-year average for the cooling pond. Use-days estimated for the Dow tertiary treatment pond (Figure 5) ranged from a high of about 29,40 0 in 197 9 to a low of about 7,800 in 1983. The 1979 total was 72 percent higher than the 5-year average; the 19 80 total of 18,20 0 was only 6.7 percent higher than the 5-year average for the Dow tertiary treatment pond. All of the 17 most abundant species using the cooling pond exhibited year-to-year differences in amount of use (Table 2). 1980. Peak use by 12 of the 17 species occurred in The largest absolute yearly differences in use were recorded for American wigeon, redhead, American coot, and scaup which were all abundant that year. Two species— common loon and horned grebe— had more use-days recorded in 197 9 than in any other year. Double-crested cormorant and red-breasted merganser were most abundant in 19 82, but the difference between the 1982 and 1983 totals for red-breast­ ed merganser was only about 14 percent. ganser was greatest in 1983. Use by common mer­ 160- 144- 128- Use-Days (Thousands) 112 - Cooling Pond □ 96 - Dow Tertiary Treatment Pond 80 64 48 32 16 1979 1980 1981 1982 1983 Year Figure 5. Relative estimated total waterbird use-days on the cooling pond and Dow tertiary treat­ ment pond by year. 27 Table 2. Estimated use-days by year for the 17 most abundant waterbird species on the cooling pond. Estimates are to the nearest 10 use-days. The highest number of use-days recorded in a year for each species is underlined. Species Horned grebe Pied-billed grebe Common loon Double-crested cormorant Common merganser Red-breasted merganser Mallard Gadwall American wigeon Redhead Canvasback Scaup Ring-necked duck American goldeneye Bufflehead Canada goose American coot 1979 1,160 110 830 20 40 2,540 2,660 60 150 500 50 2,910 100 1,160 980 6,580 4,820 19 80 530 960 340 150 3,050 2,380 14,360 1,170 17,360 25,470 2,990 19,020 2,360 3,000 1,050 13,060 38,260 Year 1981 1982 500 80 320 370 1,500 2, 810 5,000 30 330 1,440 310 4,750 40 1,950 90 7,980 5,380 330 50 190 970 1,030 9,240 2,720 10 10 60 90 1,130 10 370 220 3,940 100 1983 140 10 60 100 8,110 8,130 7,070 10 10 240 40 970 20 120 10 11,890 40 29 Among the 17 most abundant species on the cooling pond, common loon, horned grebe, double-crested cormorant, scaup, common merganser, and red-breasted merganser had more use-days recorded in spring than in summer or fall (Table 3). Mallard use was greatest in summer, and use by Canada goose, American wigeon, redhead, American goldeneye, and American coot was greatest in fall. There were only slight seasonal differences in use by pied-billed grebe and gadwall, and between spring and fall use by ring-necked duck, canvasback, and bufflehead. High use was recorded for each of the 17 most abundant species on the cooling pond in particular season-year com­ binations. Common loon and horned grebe had 25 percent and 37 percent, respectively, of their total 5-year use-days recorded in spring of 1979 (Table 3). Pied-billed grebe, Canada goose, gadwall, American wigeon, redhead, canvasback, American goldeneye, and American coot had 2 0 to 77 percent of their total use-days recorded in fall of 1980. Ring­ necked duck, scaup, and American goldeneye had 2 3 to 51 percent of use in spring of 1980 (Table 3), and 32 percent of all mallard use was recorded in summer of 1980. The highest use by double-crested cormorant and red-breasted merganser was in spring of 1982, while 48 percent of common merganser use occurred in spring of 1983. High use recorded in particular season-year combina­ tions was the result of both more birds and longer duration of use. (larger flock sizes) An exception was the high use recorded for scaup in spring of 1980 which resulted from Table 3. Estimated use-days by season, summed for 5 years (1979-1983), for the 17 most abundant waterbird species on the cooling pond. Estimates are to the nearest 10 use-days. The highest number of use-days by season is underlined for each species. Species Horned grebe Pied-billed grebe Common loon Double-crested cormorant Common merganser Red-breasted merganser Mallard Gadwall American wigeon Redhead Canvasback Scaup Ring-necked duck American goldeneye Bufflehead Canada goose American coot Spring Season Summer 1,910 23 0 1,090 1,210 9,550 21,560 8,510 160 370 4,880 1,640 22,280 1,280 1,650 1,160 8,540 8,3 80 30 210 370 280 0 10 16,650 440 3,680 4,990 10 20 100 0 70 6,460 6,590 Fall Period Of Highest Use 620 Spring 1979 Fall 1980 670 280 Spring 1979 120 Spring 1982 4,180 Spring 1983 3,530 Spring 1982 6,650 Summer 1980 Fall 1980 670 Fall 1980 13,790 17,840 Fall 1980 Fall 1980 1,840 6,480 Spring 1980 1,150 Spring 1980 4,950 Spr,Fall 1980 1,120 Fall 1979 Fall 1980 28,450 Fall 1980 33,640 Percent Of Total In Highest-Use Period (tie) 37 48 25 42 48 36 32 52 77 63 50 51 47 23 27 20 59 31 increased flock sizes only. ing pond for about 2 weeks season. Up to 991 scaup used the cool­ (6 April to 19 April) in that Intense, but brief, scaup use also occurred in fall of 19 80 when a relatively large number on 19 October) (peak of 662 of scaup was seen on the cooling pond, but only for a few days. Among the 13 most abundant species on the Dow ter­ tiary treatment pond, substantial year-to-year differences in use-days were recorded for mallard, American wigeon, blue-winged teal, redhead, goose, and American coot scaup, ring-necked ducks, Canada (Table 4). However, both the ab­ solute and relative year-to-year differences were generally much smaller than those recorded for the most abundant species on the cooling pond. In addition, the years of peak abundance were not the same as for the corresponding species on the cooling pond, and 1979 was the year of the greatest total number of use-days. As on the cooling pond, there were major differences between years in the relative abundance of species on the Dow tertiary treatment pond, but scaup was the most abundant species each year. The general chronology of waterbird use of the Dow tertiary treatment pond was also similar each year of my study. Waterbird use in spring was essentially limited to brief use by large American goldeneye in April. (>100) numbers of scaup and Those two species contribut­ ed 85 percent of the 5-year total of use-days recorded in spring (Table 5). From late April to mid-June, total Table 4. Estimated use-days by year for the 13 most abundant waterbird species on the Dow tertiary treatment pond. Estimates are to the nearest 10 use-days. The highest number of use-days recorded in a year for each species is underlined. Species Mallard Black duck American wigeon Blue-winged teal Northern shoveler Redhead Scaup Ring-necked duck American goldeneye Bufflehead Ruddy duck Canada goose American coot 1979 5,070a 150 1,030 5,870 20 2,520 9,340 250 1,610 19 0 320 1,210 1,390 1980 Year 1981 1982 1983 1,680 90 130 1,730 20 2,320 8,860 40 1,490 740 590 300 140 290 195 10 460 20 630 7,380 50 2,900 37 0 150 30 10 2,740 150 350 1,920 30 2,040 6,130 50 2,950 550 140 30 10 1,660 180 10 320 10 590 3, 390 10 1,020 130 90 410 10 a Higher estimated use-days for mallard in 197 9 is likely due in part to more com­ plete censusing of broods in summer. In 197 9, several censuses were made from the Dow tertiary treatment pond's dikes. In the other years, censusing of broods was done from less advantageous locations. Table 5. Estimated use-days by season, summed for 5 years (1979-1983), for the 13 most abundant waterbird species on the Dow tertiary treatment pond. Estimates are to the nearest 10 use-days. The highest number of use-days by season is underlined for each species. Spring Season Summer Fall Period Of Highest Use Mallard 560 8,560 2,320 Summer 197 9 35 Black duck 100 10 660 Fall 1981 25 20 260 1,230 Fall 1979 59 330 8,210 1,850 Summer 1979 49 40 20 30 Spring 1982 28 170 850 7,070 Fall 1980 26 8,810 10 26,290 Fall 1979 27 20 10 370 Fall 1979 64 2,180 0 7,800 Fall 1981 29 Bufflehead 620 0 3,070 Fall 1982 14 Ruddy duck 140 10 1,180 Fall 1980 34 Canada goose 150 930 900 Summer 1979 41 American coot 140 10 1,410 Fall 1979 90 Species American wigeon Blue-winged teal Northern shoveler Redhead Scaup Ring-necked duck American goldeneye Percent Of Total In Highest-Use Period 34 waterbird use was negligible (<400 use-days) each year. The 5-year total summer use was about 1.4x that of spring on the Dow tertiary treatment pond. Nearly 9 0 percent of the summer use was by mallard and blue-winged teal, and an additional 5 percent was by Canada goose. Substantial summer use by mallard and blue-winged teal, including from 5 to 2 0 broods, occurred each year except 1981 and 1983. Total fall use was greater than in the other seasons. The 5-year total for fall was nearly 3x that of summer and about 4x that of spring (Table 5). Nine of the 13 most abundant species on the Dow tertiary treatment pond had their highest number of use-days recorded in fall. Ninety percent of the total use of the Dow tertiary treatment pond by American coot occurred in fall of 1979, and 59 percent of American wigeon use and 64 percent of ring­ necked duck use also occurred in fall of 1979 (Table 5). Eleven species of waterbirds were seen on the cooling pond each year from 1979 to 1983, but were not abundant in any year (Table 6). An additional 13 species were seen in at least one year, but not in all 5 years. Year-to-year differences in use were pronounced for several of these species. Use by wood duck, blue-winged teal, and pintail was greater in 19 80 than in the other 4 years, use by common tern and green heron was greatest in 1979, use by snow goose, Bonaparte's gull, and ruddy duck was greater in 19 80, and whistling swan use was higher in 19 83. Table 6. Estimated use-days for waterbird species observed on the cooling pond in at least 1 year but not in all 5 years, and/or for which a low number (5-year total <650) use-days were recorded. Numbers in parentheses are total observa­ tions; dashes indicate species not seen in that year. Species Western grebe Eared grebe Black tern White pelican Common egret Wood duck Red-necked grebe Green-winged teal Oldsquaw Common tern Northern shoveler Snow goose Black-crowned night heron Bonaparte1s gull Caspian tern Hooded merganser Black duck Blue-winged teal Pintail White-winged scoter Ruddy duck Whistling swan Great blue heron Green heron 1979 1 (1) 2 (9) 12 (21) 1 (4) 1 (2) 1 (1) 14 (11) 2 (4) 32 (11) 555 (226) 10 (28) 3 (9) 10 (28) 1 (2) 53 (49) 5 (8) 11 (24) 64 (30) 3 (8) 2 (4) 61 (48) 14 (23) 11 (9) 192 (156) 1980 -----457 (281) 1 (2) 24 (58) 6 (16) 61 (29) 9 (12) 1 (3) 7 (8) 10 (29) 197 (127) 6 (14) (68) 76 143 (75) 252 (143) 32 (12) (92) 88 37 (16) 45 (48) (80) 139 Year 1981 -----(4) 5 2 (6) 1 (1) 22 (24) 1 (5) 370 (214) 7 (12) 116 (73) 87 (63) 4 (6) 89 (88) 42 (57) 6 (16) 1 (1) 125 (199) 70 (157) 1 (4) 17 (37) - 1982 — — — — — — — — — 2 (7) 2 (4) 8 (16) 1 (1) 23 (37) 116 (97) 1 (2) (38) 53 3 (5) 3 (17) 4 (12) (57) 13 11 (18) 27 (19) 3 (8) 1983 --------2 (2) ----(68) 5 165 (105) 6 (24) 66 (55) 2 (4) 2 (5) 1 (2) 3 (27) 170 (138) 23 (13) 16 (8) 36 Five species were present in low numbers on the Dow tertiary treatment pond each year, and an additional 16 species were observed in at least 1 year, but not in all 5 years (Table 7). With exception of common tern, species found on both the cooling pond and the Dow tertiary treat­ ment pond in low numbers had their peak use in different years. Among these species, year-to-year differences in use of the Dow tertiary treatment pond were largest for gadwall, common tern, and canvasback (Table 7). Forty-one and 34 species of waterbirds were observed on the cooling pond and Dow tertiary treatment pond, re­ spectively, from 1979-1983 (Appendix A ) . Although the species which occupied the Dow tertiary treatment pond, except Wilson's phalarope, over the course of a year were also observed on the cooling pond, the species compositions of the 2 open water systems were often different in parti­ cular 24-hour observation periods. Species seen on the cooling pond only were Western grebe, eared grebe, common loon, black tern, white pelican, whistling swan, common egret, and green heron. Ten species— horned grebe, common loon, double-crested cormorant, common merganser, hooded merganser, gadwall, pintail, white-winged scoter, great blue heron, and green heron— were observed on the cooling pond in all 5 years, but in 4 or fewer years on the Dow tertiary treatment pond. In contrast, only northern shov- eler and Wilson's phalarope were seen in more years on the Dow tertiary treatment pond than on the cooling pond. Table 7. Estimated use-days for waterbird species observed on the Dow tertiary treat­ ment pond in at least 1 year but not in all 5 years, and/or for which a low number (5-year total <130) use-days were recorded. Numbers in parentheses are total observations; dashes indicate species not seen in that year. Species Snow goose Black-crowned night heron Red-necked grebe Double-crested cormorant Pied-billed grebe Pintail Wood duck Hooded merganser Great blue heron Horned grebe Gadwall Oldsquaw White-winged scoter Common tern Green-winged teal Common merganser Bonaparte's gull Caspian tern Red-breasted merganser Canvasback Wilson's phalarope ______________________________ Year____________________________ 1979 1980 1981 1982 1983 16 1 (13) (2) — " __ 1 6 — 17 (24) --1 (1) 2 (5) 147 (148) 3 (12) (3) 1 142 (69) (66) 55 -1 (1) 9 (13) 3 (6) 7 (30) 3 (17) 1 1 1 1 — 1 5 1 1 3 2 4 19 6 1 1 1 (2) (4) (2) (3) (2) (35) (2) (2) (22) (13) (10) (46) (24) (4) (4) (1) 1 (1) 2 (8) 1 (2) 2 1 (8) (2) 1 1 2 1 1 75 3 (2) (2) (9) (3) (1) (60) (14) (1) (9) ---- 2 1 (5) (1) — — 7 (20) 1 4 (1) (15) — — 1 (1) 7 (30) 1 (2) 4 (16) 8 (13) 3 (10) 16 (34) 22 (100) — — 1 (2) — 2 (5) 6 (31) 1 (2) 1 (2) 2 (9) 2 (8) 38 Sixteen species were observed on both open water sys­ tems in all 5 years. They were: Bonaparte's gull, Caspian tern, red-breasted merganser, mallard, black duck, American wigeon, blue-winged teal, redhead, canvasback, scaup, ring­ necked duck, American goldeneye, bufflehead, ruddy duck, Canada goose, and American coot. However, the only species relatively abundant on both open water systems were mallard, American wigeon, redhead, scaup, American goldeneye, buffle­ head, Canada goose, and American coot. These same 8 species and Caspian tern and common tern were the only species for which movements of individual birds or flocks between the 2 systems within a 24-hour observation period were recorded. Such movements were infrequent and involved few birds, ex­ cept in the case of fall-migrating American goldeneye which often used the cooling pond as a night roost after spending most of the day on the Dow tertiary treatment pond. Most birds of all other species appeared to select 1 of the 2 open water systems and remain there as long as they stayed in the immediate area. The number of species seen on the cooling pond declined steadily each year from a high of 41 in 1979 to 29 in 1983 (Figure 6). An overall declining trend in the number of species seen also occurred on the Dow tertiary treatment pond, but the pattern was not as consistent. There were 28 species seen on the Dow tertiary treatment pond in 1979, 2 9 in 1980, 25 in 1982, and 22 in 1983 (Figure 6). 50 m Cooling Pond □ Dow Tertiary Treatment Pond 30 - 20 - 10 - 0 - Number Of Species 40 1979 1980 19 81 1982 1983 Year Figure 6. Number of species observed on the cooling pond and Dow tertiary treatment pond by year. 40 Five species— Western grebe, eared grebe, black tern, white pelican, and common egret— were seen on the cooling pond in 1979, but not in subsequent years (Table 6). Wood duck was present in 1979 and 1980, but not observed in the next 3 y e a r s . Red-necked grebe and green-winged teal were not seen in 1982 or 1983 (Table 6). Four additional spe­ cies— common tern, northern shoveler, snow goose, and black-crowned night heron— were not seen in 1983 after having been observed in the 4 previous years. Oldsquaw was seen on the cooling pond each year except 1982. This gen­ eral pattern of species appearing to stop using the cooling pond after previous use was not observed for the Dow ter­ tiary treatment pond (Table 7). Only 6 of 16 species seen in 1 year but not in all 5 years on the Dow tertiary treat­ ment pond exhibited this use pattern. The mean weekly species diversity index for the cool­ ing pond was significantly higher in 1980 than in the other 4 years for which year-around data were recorded (Table 8). Mean weekly diversity in each season was higher than for the respective seasons in the other 4 years. test Bartlett's (Snedecor and Cochran 1967) indicated the variances in mean weekly species diversity were lower in 19 80 than in the other years in each season. Inspection of the data used to calculate the diversity indices revealed that the higher diversity and lower sample variances in 19 80 were due to better numerical balance among species occupying the cooling pond and more persistent use by several species Table 8. Mean 24-hour species diversity indices (x) and standard deviations (SD) by sea­ son and year for waterbirds on the cooling pond. Numbers in parentheses are total 24-hour observation periods in a season or year. Summer Spring Year X SD 1979 1.23^ (9) 0.31 0. 1980 1.68b (12) 0.24 1981 1.19a (11) 1982 1.17a 1983 1.17a Fall SD X X Year Total SD X SD (13) 0.25 1.21ab (12) 0.41 1.01a (34) 0.42 1.43b (12) 0.18 1.47a (10) 0.29 1.53b (34) 0.25 0.45 0.72a (12) 0.22 l.llabc (12) 0.41 0.81a (38) 0.42 (13) 0.31 0.46a (13) 0.27 0.85bc (12) 0.41 0.81a (38) 0.43 (14) 0.48 0.53a (12) 0.28 0.69bc (10) 0.64 0.79a (36) 0.55 6 8 a a-c Column means with unlike superscripts differ and Cochran 1967). (P<0.05) by Scheffe's test (Snedecor 42 including redhead, American wigeon, American coot, mallard, and Canada goose. In the other years, the depressing effect on species diversity of differing migration chrono­ logy among species was greater in spring and fall. Waterbird Activities There were significant year-to-year differences in the percentages of birds observed feeding for 12 of the 17 most abundant species on the cooling pond (Table 9). Years of greatest percentages observed feeding coincided with the years of highest use-days for 6 of those species. The percentages of Canada geese, mallards, gadwall, American wigeon, redhead, and American coot seen feeding were significantly greater in 1980, the year of highest use by these species on the cooling pond, than in any other year. The highest percentages of birds seen feeding in a particular year were 67 percent for American coot, 60 per­ cent for gadwall, and 55 percent for American wigeon in 1980). (all In contrast, double-crested cormorant, common merganser, red-breasted merganser, canvasback, scaup, ring­ necked duck, American goldeneye, and bufflehead had rela­ tively low percentages of birds feeding in their high-use years on the cooling pond (Tables 2 and 9). Significantly higher percentages of birds feeding were recorded for horned grebe in 197 9 and 1983, for common loon in 1980, for red-breasted merganser and scaup in 1979, and for double-crested cormorant in 19 83 than in the other years. There were no significant differences among years Table 9. Percentages of b irds of the 17 most abundant w a t e rbird species on tho cooling pond observed in parentheses are total 15-second observations of individuals of a species. Yoar 1981 1980 1979 Numbers 1982 1983 50.7* (675) 17. 9^ (408) 8.8° (2S0) 2 7 . 015 (241) 42.4® (66) 4.5* (199) 23. 7b (440) 14. 3b (84) 31. 3b (16) 11. lb (36) Common loon 28.9* (5,739) 34 .7^ (2,520) 29.4s (1,844) 23.9* (3,340) 23.5* (925) Double-crested cormorant 12.7® (165) 6.9® (1,540) 9.5s (1,264) 6.3* (554) 6.4® (76) 7.8® (1,092) 1.6a (367) 4.8® (1,031) 5.2* (3,386) 37.4® (944) 22. 3b (1,340) 25.4b (1,495) 8. 6C (6,712) 5.^ (3,310) M allard 6.8® (2,478) (P (5,013) 0. 9C (982) 0. 3C (703) 8.8* (1,756) Gadwall 0.0® (37) 59.8b (570) ll.la (18) 0.0 (3) 0.0* (16) 23.6® (165) 5 4.6b (7.8R8) 0.0s (104) 0.0 (5) 0.0 R edhead 9.0* (378) 2 5 . 3b (13,687) 15.3s (894) 0.0* (33) 16.5* (236) Canvasback 6.9* (29) 4.2* (1,488) 0. 9s (340) 0.0* (26) 1.9* (212) Scaup 9.3* (589) 4 .4b (8,855) 1.5C (2,445) 0. 4 C (1,021) 0.0C (489) Ring-necked duck 0.0* (39) 5.8* (1,320) o Q Species feeding. (32) 0.0* (12) 0.0* (65) American goldeneye 2.8® (249) 1.9® (213) 0.0* (15) 3.8* (26) 0.8* (125) Bufflehead 5.6® (594) 6.1* (446) 8.7* (69) 4.4* (158) 10.0* (30) Canada goose 7.0® (2,566) 27.7b (6,644) 2. Rc (3,136) 8.3* (1,477) 5.1* (3,800) 47.0* (2,450) 6 7 . 0b (17,497) 40. 3C (4,010) 6. 3d (80) 17.5d (67) Homed grebe Pied-billed grebe Common merganser R ed-breasted merganser American w i geon American coot ® “d 25. Row percentages w i t h unlike superscripts are s i gnificantly different distribution. Percentages w i t h o u t superscripts indicate sample si~c (ncO.os) in. based on binomial 30. (P (100) (2) 44 in the percentages of canvasbacks, ring-necked ducks, Amer­ ican goldeneyes, buffleheads, and common mergansers seen feeding. Relationships Between Waterbird Distributions And Habitat Variables Chi-square goodness-of-fit comparisons indicated that the observed number of waterbird occupied grid squares with and without macrophytes differed significantly from the ex­ pected number based on proportion of occurrence for 30 of 85 species-year combinations tested for the cooling pond (Table 10). American coot, American wigeon, redhead, scaup, ring-necked duck, pied-billed grebe, canvasback, mallard, gadwall, and Canada goose occupied significantly more grid squares with vegetation than expected in 19 80, and the coincidence of birds with vegetation was more evi­ dent for those species in 19 80 than in the other years (Table 10). Redhead occupied significantly more vegetated grid squares than expected in 4 years 19 83) . (1979, 19 80, 19 81, and American coot, American wigeon, and pied-billed grebe occupied more vegetated grid squares in 3 years (1979, 1980, and 1981). Scaup and canvasback were similar­ ly associated with vegetation in 1979 and 19 80, and ring­ necked duck (in 1980 and 1981), American goldeneye (1979 and 19 82), common loon (19 81), and double-crested cormo­ rant (19 82) also occupied more vegetated grid squares than expected (Table 10). Species which occupied significantly 45 Table 10. Number of grid squares (N=440) occupied by both birds and vegetation (C) and total grid squares occupied by birds (n) of the 17 most abundant species on the cooling pond. Species-year combinations are ranked based on the difference in the number of grid squares occupied by birds with and without vegetation. Higher ranks gen­ erally indicate better coincidence of birds and vegetation, but small sample sizes for certain species-year combinations prohibit some compar­ isons . Rank Species Year 1 2 3 4 5 6 7 8 9 10 11 12 13 American coot American wigeon Redhead Scaup Ring-necked duck Pied-billed grebe Common loon Canvasback Redhead Mallard American wigeon Redhead Redhead Gadwall Pied-billed grebe American coot Ring-necked duck Pied-billed grebe American coot American goldeneye Canvasback American wigeon Double-crested cormorant Scaup American goldeneye Horned grebe Ring-necked duck Gadwall Double-crested cormorant American wigeon Common merganser Common loon Gadwall Ring-necked duck Pied-billed grebe American coot 1980 1980 1980 1980 1980 1980 1981 1980 1979 1980 1981 1983 1981 1980 1979 1979 1981 1981 1981 1979 1979 1979 1982 1979 1982 1980 1979 1979 1983 1983 1979 1979 1982 1982 1983 1983 15 18 21 22 23 24 26 28 30 33 36 C 118** 88** 123** 88** 35** 40** 55** 31** 27** 109** 24** 27** 57** 27* * 24** 45** 22** 24** 63** 16** 10** 15* 21* 26* 11* 19 7 5 7 3 2 102** 1 4 7 4 n 165 108 181 139 37 47 83 40 34 203 33 40 44 21 37 78 11 38 116 22 11 23 36 47 17 34 10 7 11 4 2 202 1 7 13 8 46 Table 10 (cont'd): 3ank 38 42 44 48 50 56 58 60 61 62 63 65 66 67 71 72 73 74 76 77 79 80 81 Pied-billed grebe Gadwall Scaup Ring-necked duck American goldeneye Canvasback Double-crested cormorant Common merganser Gadwall Double-crested cormorant Canvasback Canada goose American goldeneye American wigeon American goldeneye Bufflehead Bufflehead Scaup Common loon Double-crested cormorant Canvasback Redhead Bufflehead Bufflehead American coot Canada goose Scaup Bufflehead Common loon Red-breasted merganser Common merganser Horned grebe Mallard Mallard Canada goose Mallard Canada goose Horned grebe Common loon Canada goose Red-breasted merganser Horned grebe Mallard Horned grebe Red-breasted merganser Year C 1982 1981 1981 1983 1983 1983 1980 1981 1983 1979 1982 1980 1981 1982 1980 1983 1981 1983 1983 1981 1981 1982 1979 1980 1982 1983 1982 1982 1980 1980 1980 1983 1979 1981 1979 1983 1981 1981 1982 1982 1979 1979 1982 1982 1983 5 6 61 3 6 5 2 12 2 2 4 98* 4 0 61 3 14 24 17 6 11 3 38 13 5 28 20 15 22 37 31 8 47 53 57 24 34 37 49 29 83 77 27 19 * * 48 n 10 13 123 7 13 12 6 27 7 7 11 200 12 5 127 11 33 53 39 18 28 13 83 35 20 67 52 42 58 90 80 34 112 124 133 69 90 97 121 85 194 182 84 83 143 47 Table 10 Rank 82 83 84 85 (cont'd): Species Red-breasted merganser Common merganser Common merganser Red-breasted merganser Year C 1981 1982 1983 1982 48 29** 68* 67** * Chi-square test significant, P<0.05. ** Chi-square test significant, P<0.01. n 143 101 201 231 48 more grid squares without vegetation than expected were horned grebe (in 1982), red-breasted merganser common merganser (1982 and 1983). (1982), and In 1982 and 1983, rela­ tively large numbers of horned grebes, red-breasted mer­ gansers, and common mergansers were observed in the deeper water area in the northeast part of the cooling pond (Figure 2). Coincidence of birds and vegetation was particularly evident for American coot, American wigeon, and redhead in 19 80 (Table 10), and the abundance and incidence of feed­ ing by those species were also much higher in 1980 than in any other year. Chi-square analyses of the distributions of American coot, American wigeon, and redhead in 19 8 0 in­ dicated that birds of those species.were not evenly dis­ tributed among the vegetated grid squares. species seemed to prefer locations Each of the 3 (Figure 3) of the cool­ ing pond where macrophytes were most extensive (Table 11). More American wigeon occupied all 3 locations where there were contiguous grid squares with at least 10 percent of each grid square covered by macrophytes than expected based on proportion of occurrence tat) . (availability of habi­ American coot appeared to prefer only locations 2 and 3 (both in the southeast part of the cooling p o n d ) , and redhead were more numerous than expected in locations 1 and 3. The relative difference between observed and ex­ pected numbers of birds of all 3 species in 19 80 was greatest for location 3, the largest of the 3 locations Table 11. Location Occurrence of birds of 3 species in 4 vegetated locations in the cooling pond in 1980. Total areas (in ha) of locations are: 1 - 3.0, 2 - 2.5, 3 - 11.0, 4 - 68.5. Locations 1, 2, and 3 consisted of contiguous 0.5 ha grid squares with at least 10 percent of each grid square covered by macrophytes. Location 4 consisted of the remaining vegetated grid squares in the cooling pond (see (Figure 3) . Proportion3 Of Total Area Species Number Of Birds Observed Number Of Birds Expected Proportion Observed In Each Loca­ tion (Pi) Family Of 99% Confidence In­ tervals On Pi*5 0.035 American wigeon Redhead American coot 866 1,365 20 432 693 924 0.070** 0.069** 0.001** 0.0631,000 use-days on cooling pond, > 200 use-days on Dow tertiary treatment pond each year); A = abundant (>1,000 use-days (>200 for Dow tertiary treatment pond) in at least 1 but not in all 5 years); C = common (>650 use-days for 5-year total, >130 use-days on Dow tertiary treat­ ment pond,but not meeting criterion for abundant); U = uncommon (<650 usedays, < 1 3 0 on Dow tertiary treatment pond for 5-year total). Common Name Western grebe Red-necked grebe Horned grebe Eared grebe Pied-billed grebe Common loon Bonaparte's gull Common tern Caspian tern Black tern Double-crested cormorant White pelican Common merganser Red-breasted merganser Hooded merganser Mallard Black duck Gadwall Scientific Name Aechmophorous occidentalis Podiceps grisegena P . auritus Colymbus podiceps Podilymbus podiceps Gavia immer Larus Philadelphia Sterna hirundo S. caspia Chlidonias niger Phalacrocorax auritus Pelecanus erythrorhynchos Mergus merganser M. serrator Lophodytes cucullatus Anas platythynchos A. rubripes A. strepera Cooling Pond 1U 3U 5A 1U 5C 5C 5U 4U 5U 1U 5C 1U 5A 5VA 5U 5VA 5U 5A Dow Tertiary Treatment Pond 2U 3U 2U 5U 4U 5U 2U 4U 5U 2U 5VA 5U 3U Table Al (cont'd): Common Name American wigeon Green-winged teal Blue-winged teal Northern shoveler Pintail Wood duck Redhead Canvasback Scaup Ring-necked duck American goldeneye Bufflehead Oldsquaw White-winged scoter Ruddy duck Snow goose Canada goose Whistling swan Great blue heron Common egret Green heron Black-crowned night heron American coot Wilson's phlalrope Scientific Name A. penelope A. crecca A. discors A. clypeata A. acuta Aix sponsa Aythya americana A. valisneria A. marila, A. affinis A. collaris Bucephala clangula B. albeola Clangula hyemalis Melanitta fusca Oxyura jamaicensis Anser caerulescens Branta canadensis Cygnus columbianus Ardea herodias Casmerodius albus Butorides virescens Nycticorax nycticorax Fulica americana Steganopus tricolor Cooling Pond 5A 3U 5U 4U 5U 2U 5A 5A 5A 5A 5A 5A 4U 5U 5U 4U 5VA 5U 5U 1U 5U 4U 5A Dow Tertiary Treatment Pond 5A 4U 5VA 5U 2U 2U 5VA 5U 5VA 5U 5VA 5A 3U 3U 5C 1U 5A 2U 1U 5A 5U