Activity Budgets of Nonbreeding Gadwalls in Louisiana

Activity budgets of nonbreeding gadwalls (Anas strepera) in coastal southwestern Louisiana were studied from October 1977 through April 1978 and related to sex, pair status, and environmental factors. Gadwalls spent 64% of their time feeding, 11% locomoting, 11% resting, 9% alert, 5% preening, and less than 1% each in agonistic and courtship activities, respectively. Gadwalls spent significantly more time feeding during night than day (P 0.05). Time spent feeding increased from 44% in October to 77% in April. Feeding activity decreased as air temperature increased (r = -0.163, P < 0.001) and varied depending upon habitats and foods used. All gadwalls spent similar amounts of time alert, but unpaired males spent more time locomoting than those paired (P < 0.01). Gadwalls spent more time resting during night than day (P < 0.01), and birds spent more time resting and preening when using shallow-water areas on refuges than other habitats. Although males only spent 1.5% of their time courting in fall, by late November 86% of females were paired. Since gadwalls spend most of their time feeding, the primary goal of management should be to ensure that preferred feeding habitats are preserved. Observations also suggested that current land-use modifications along coastal Louisiana may be especially detrimental to gadwalls. J. WILDL. MANAGE. 48(2):371-380 Numerous studies have examined the ecology, taxonomy, and distribution of North American waterfowl, yet of 1,748 studies reviewed by Reinecke (1981), only 8% pertained to wintering waterfowl. Fretwell (1972) emphasized that factors outside the breeding season were crucial in regulating populations of migratory species. Condition of waterfowl leaving the wintering grounds is probably important to reproduction (Paulus 1980, Krapu 1981). Gadwalls have been intensively studied on the breeding grounds (Gates 1958; Oring 1968; Dwyer 1974, 1975; Serie 1974; Blohm 1979) but little is known of their wintering ecology. Over three-fourths of the North American gadwall population, numbering between 500,000 and 1 million, winter in Louisiana (Bellrose 1976). Gadwalls comprise 29% of all ducks in coastal Louisiana (H. A. Bateman, unpubl. reps., La. Wildl. and Fish. Comm., 197778) and 10% of the waterfowl harvest in Louisiana (Novara et al. 1981). Fredrickson and Drobney (1979) and Reinecke (1981) considered analysis of activity budgets a useful tool in determining he needs of wintering waterfowl. Verner (1965) suggested that each species exhibited an optimal time budget for each environmental condition and that selection favored individuals whose time budgets were most adapted. I examined activity budgets of gadwalls in winter and the influence of time of day and year, weather, and social status of individuals on activity budgets. I also examined the influence of habitat and diet on activity budgets and suitability of various management practices in providing for needs of wintering gadwalls. I thank R. D. Crawford, T. Joanen, L. L. McNease, K. Paulus, D. L. Trauger, and W. Wrenn for help during this study. A. Afton, C. D. Ankney, R. J. Blohm, R. D. Sayler, and two anonymous reviewers provided criticisms of earlier versions of the manuscript. Financial support was 'Present address: Department of Zoology-Entomology, Auburn University, Auburn, AL 36849. J. Wildl. Manage. 48(2):1984 371 This content downloaded from 157.55.39.196 on Sat, 24 Sep 2016 04:34:52 UTC All use subject to http://about.jstor.org/terms 372 ACTIVITY BUDGETS OF NONBREEDING GADWALLS * Paulus provided by U.S. Fish and Wildl. Serv., La. Dep. of Wildl. and Fish., Univ. of N.D., and Soc. of Sigma Xi. STUDY AREAS AND METHODS Gadwalls were observed in coastal southwestern Louisiana on Rockefeller and Marsh Island state wildlife refuges, both closed to hunting, and on private lands within 13 km of Rockefeller Refuge on which hunting was allowed. Brackish marsh comprised much of the study areas and several areas have been impounded to stabilize water conditions. I have described the study area in detail elsewhere (Paulus 1980, 1982). Activity budgets of gadwalls were determined by instantaneous sampling procedures (Altmann 1974, Dwyer 1975) from October 1977 through April 1978. Observations were made with a 15-40 x spotting scope and 7 x binoculars. A tape recorder, stopwatch, and super-8 movie camera were used to time and record specific events. Activity of a single individual or both members of a pair were recorded every 20 seconds at the tone of a metronome (Wiens et al. 1970). Activities were divided into seven categories: (1) feeding, (2) locomoting (walking, swimming, and flying not associated with courtship activity), (3) resting (loafing and sleeping), (4) preening (preening and bathing), (5) alert, (6) courting (displays and copulations), and (7) agonistic (bill threats, chasing, and biting). Diurnal activities were recorded during 1-hour randomly selected sampling periods. Nocturnal activities were recorded during nonrandom periods under clear to mostly cloudy skies when the moon was between the first and last quarter stages. Nocturnal observations were limited to birds within 30 m of the blind. Data collected from observation periods of <30 minutes were eliminated from analysis. During observation periods, sex and pair status of the individual observed was determined, and date, ambient temperature (C), wind velocity (using a portable anemometer), cloud cover, precipitation intensity, and water depth were recorded. For analysis, cloud cover was coded as (1) 0-25%, (2) 26-50%, (3) 51-75%, or (4) 76100%; rainfall intensity as (0) none, (1) light, (2) moderate, or (3) heavy; and water depth as (1) 0-15 cm, (2) 16-30 cm, (3) 31-45 cm, (4) 46-50 cm, and (5) 51-75 cm. Foods consumed by gadwalls were d termined from observation, aquatic vegetation analysis, or from gadwalls coll cted for food habits analysis (Paulus 1982). Habitats used were classified based upon vegetation, physiography, and water depth characteristics (Table 1). Vegetation was identified from Radford et al. (1968). Pair status was determined by observing associations of females with males. Chance or temporary associations were a potential source of error, and on many occasions even widely separated (_15 m) gadwalls later resumed close associations and were judged paired. Judgmental error was reduced by considering as paired only those birds (1) mutually avoiding or threatening other birds; (2) exhibiting consistently synchronized activities, especially locomoting; and (3) remaining within 3 m of each other during most of the observation period. The Kruskal-Wallis test and multipleconfidence-interval procedures (Marascuilo and McSweeney 1977:306-319) were used to detect differences among group means with each activity category as dependent variables and date, sex-pair status, or environmental variables as independent variables. Simple correlation (r) analysis was used to examine relationships J. Wildl. Manage. 48(2):1984 This content downloaded from 157.55.39.196 on Sat, 24 Sep 2016 04:34:52 UTC All use subject to http://about.jstor.org/terms ACTIVITY BUDGETS OF NONBREEDING GADWALLS * Paulus 373 Table 1. Characteristics of major habitats used by nonbreeding gadwalls in Louisiana during 1977-78. Water depth Major foods Habitat (cm) consumed Other characteristics Stream 30 Potamogeton pusillus Fresh to intermediate water, slow moving Shallow, natural marsh 30 Leptochloa fascicularis, Seasonally flooded in winter algae Deep, natural marsh 30 Ruppia maritima, algae Under tidal influence, intermediate to brackish Shallow impoundment 15-30 Eleocharis parvula, algae, Seasonally flooded in winter, Leptochloa fascicularis fresh Deep impoundment 30 Ruppia maritima, algae Large bodies of brackish to intermediate water not