OKLAHOMA. Improving Management for Bobwhite Quail: Re-evaluation of Movements, Group Dynamics, Habitat and Feeding Ecology

Transcription

1 W B663q OKLAHOMA o Improving Management for Bobwhite Quail: Re-evaluation of Movements, Group Dynamics, Habitat and Feeding Ecology

2

3 GRANT TITLE: Improving Management For Bobwhite Quail: Re-evaluation of Movements, Group Dynamics, Habitat and Feeding Ecology

4 climatic stress in winter on use of supplemental feeders and their impact on rate of survival in winter we examined 783 crops from harvested northern bobwhite in 5 years. Crops of bobwhites that were harvested from areas with supplemental feeders contained 28.2% supplemental food compared with 5.5% (P < 0.001) for those from areas without supplemental feeders. Low regression coefficients from the relationship of the amount of supplemental food in crops and prevalent climatic conditions on the non-supplemented (R 2 = 0.20) and supplemented areas (R 2 = 0.15) suggested that wimer climate was not a significant predictor of use of supplemental feeders. Rates of winter survival with hunting deaths censored were significantly greater on areas with supplemental feeders compared with non-supplemented areas in winters (P < 0.001) and (P = 0.001) but not (P = 0.131) and (P = 0.365). Results suggested that bobwhite can gain nutritional benefits from supplemental feeders during times of severe winter stress. A. Evaluate "quality" nesting and brood rearing habitat as defined by vegetation characteristics among successful vs. unsuccessful nests and broods of northern bobwhites on the Packsaddle WMA in western Oklahoma. 1. Evaluate feeding habits of chicks to determine insect preferences and compare distribution and abundance of insect species among successful brood rearing habitats vs. unsuccessful brood rearing habitats. B. Determine seasonal changes in home-range size, group associations, and movement patterns of radio-collared bobwhites as influenced by supplemental feeding on the Packsaddle WMA.

5 c. Compare density estimates from line transect flush counts and mark and recapture models. D. Determine annual changes in food habits of hunter-harvested bobwhite (crops) on the Packsaddle WMA to evaluate relative use of supplemental feed vs. native foods, and the impact of climatic conditions on feeding habits. II. INTRODUCTION Ground nesting birds in shrub and grassland habitats suffer greater nesting mortality than other species, and many such species are documented to be in long-term population declines (Martin 1993). Declining populations of the northern bobwhite (Colinus virginianus) are no exception and have been well documented (Klimstra 1982, Church et al. 1993). Oklahoma experienced a 16% decrease from 1961 to 1988 (Brennan 1991). Although the reason for those declines remain unknown, successful reproduction is an important factor of bobwhite ecology that depends on adequate nesting and brood-rearing habitat (Berner and Gysel 1969). Many studies have described the macrohabitat of bobwhite nest sites throughout their range (Klimstra and Roseberry 1975; Lehmann 1984:78-83; Roseberry and Klimstra 1984:18-23; Taylor 1991) However, few have compared vegetative differences between successful and non-successful nest sites and nest sites vs. random non-use sites. Winter survival is another important factor of bobwhite ecology and can be extremely low in many populations (Moseby and Overton 1950; Kabat and Thompson 1963; Robel and Fretwell 1970; Curtis et al. 1988; Robinette and Doerr 1993; Burger et al. 1994, 1995). To improve winter survival of bobwhite, managers of intensely hunted areas have used supplemental feeders with the belief that nutrition and severe climatic stress can be contributing factors to population

6 declines. However, supplemental feeders have not been universally regarded as an effective management tool (Lehmann 1984:276; Guthery 1986:59; Landers and Mueller 1986:19). In addition to a lack of solid evidence that populations of bobwhite are nutritionally stressed during winter, several biologists have suggested that supplemental feeders could concentrate predators and facilitate the spread of disease, thereby having an overall detrimental effect on populations (Lehmann 1984:276; Guthery 1986:56; Landers and Mueller 1986:19). The suggestion that supplemental feeders may help increase survival during especially stressful winters when foods may become limited has never been fully evaluated (Lehmann 1984: 165,273; Guthery 1986:48; Landers and Mueller 1986: 19). Our objectives were: 1) to determine if the choice of nest site by bobwhites is related to site characteristics and if such characteristics influence the likelihood of nest success and 2) to evaluate use of supplemental feeders by northern bobwhite in response to changing climatic conditions in winter and assess their effect on winter survival. We hypothesized that prevalence of supplemental food in crops of harvested bobwhites would be greater on areas with supplemental feeders in winter and that use of supplemental feed would be correlated with climatic winter stresses. We further hypothesized that survival rates would be higher for populations with access to supplemental feeders. m. METHODS A. Approach: 1. Radio-telemetry Birds were trapped on both treatment areas (feeder, control) using modified Stoddard funnel traps (Wilbur 1967) baited with sorghum throughout the year and by nightlighting

7 (Huempfuer 1975) sessions prior to the nesting season (March-April). Captured birds were marked with <7-g radio-transmitters (Holohill Systems Ltd., Ontario, Can. and Wildlife Materials, Inc., Carbondale, II...), sexed, aged, and banded with aluminum leg bands 0Nebb and Guthery 1982). Birds were monitored at least once daily (ideally 3 times/day) throughout the nesting and brood-rearing season (May - October). 2. Habitat Use of Bobwhites a) Nest Site Nest sites were marked, and the surrounding microhabitat characterized after parents permanently left the nest. Success of the nest was compared with structural characteristics of the nest site to elucidate "quality" nesting habitat. Successful nests were defined by a hatch of> 1 chick from each nest. Depredated nests were characterized in 2 categories: (1) predation (mammal or snake) and (2) abandoned. Ten Daubenmire (0.10 m 2 ) quadrats were used to characterize cover within a I_m 2 plot positioned directly over the nest site. Habitat measurements were taken at each of2 plots: a plot centered directly over the nest and a plot 20 m from the nest selected in a random direction (Badyaev 1995). Measurements of percent cover by species and bare ground were recorded consistent with Daubenmire's coverage classifications (Daubenmire Physiographic variables such as aspect of the nest, and aspect downhill (Asnest, Asdwn; Appendices C, D), slope (%) (Sieg and Becker 1990), and distance to nearest stand of shinnery oak (Quercus havardii) or any other noticeable abrupt change in habitat (edge) or major disturbance (roads, burns, food plots, etc.) was recorded. Diameters of the nest at the top (Toplngth, Topwdth; Appendices C, D) and depth of the dome and bowl (Dpthdme, Dpthbwl;

8 Appendices C, D) and thickness of the nest lining (Linthick; Appendices C, D) were recorded (Lehmann 1984). Tradeoffs for nest-site selection between a bobwhite's visibility of its surroundings and its concealment from predators were evaluated (Gotmark et al. 1995). Visual obstruction was evaluated using a vertical profile board placed 3 m from each nest or non-use site (Nudds 1977). Obstruction was recorded in 4 profiles: <0.25 m, m, m, and m (Cba25, CbaSO, CbalO, Cba20, respectively; Appendices C, D). Percent cover of vegetation was differentiated into 6 categories: <2.5%, %, 26-50%, 51-75%, 76-95%, and >95% (Schmutz et al. 1989). Measurements from within each nest site (a bobwhite's view from the nest) were taken in 4 different directions (Angelstam 1986). The first was in a random direction and subsequent measurements were taken at 90 0 intervals. Nest concealment (Nconceal; Appendices C, D) from outside the nest (predator's view) was quantified by 9 points: 8at45 0 compass intervals 1 m from the nest and 1 overhead view taken at 0.5 m from the nest (Gotmark et al. 1995, Holway 1991, Keppie and Herzog 1978, Martin and Roper 1988). Concealment was classified as follows: 0 = 0% visibility; 1 = 20% visibility, 2 = 40% visibility, 3 = 60% visibility, 4 = 80% visibility, 5 = 100% visibility (Holway 1991). Density of little bluestem (Schizachyrium scoparium) patches around the nest and nonuse site within 1_m 2 and at 1-m, 2-m, and 5-m radii was recorded (Lbpm2, Lbp1, Lbp2, Lbp5; Appendices C, D). That density was compared with nest success in relation to predation. Martin and Roper (1988) hypothesized that probability of predation decreased with increases in density of a particular foliage type utilized as nest sites. Shrub densities also were enumerated at 1-m, 2-m, and 5-m radii around each nest and non-use site (Shr1, Shr2, Shr5; Appendices C, D). Shrubs were defined as woody vegetation

9 >0.50 m in height and with a stem diameter <2 em. Density was defined as the number of stems/m 2 (Holway 1991). Effective plant height, length, and width (Totaiht, Clumplth, Clumpwth; Appendices C, D) was measured using a meter tape (Higgins et al. 1994). b) Brood Site Brood habitat and home range were monitored by a collection of radio locations spanning the date of hatching to 2 weeks of age. Ideally, broods were monitored 3 times/day after hatching: morning (first 25% of daylight), midday (middle 50% of daylight), and evening (last 25% of daylight) (Taylor and Guthery 1980). Vegetation in a brood's 2-week home range was sampled with 30 Daubenmire (0.10 m 2 ) quadrats. Five location points within the 2-week home range constituted a starting point. Daubenmire quadrats were placed 20 m from each starting point: N, S, E, W, and in 2 non-cardinal random directions. Vegetation composition estimates of percent cover, percent bare ground, and herbaceous quail foods were sampled in these quadrats. Composition of insect species was sampled by 2 different methods: (1) a sweep-net method (after Hollifield and Dimmick 1995) and (2) a modification of Stewart and Wright's (1995) Bug- Vac. Sweeps were conducted along E (30 sweeps) and W (30 sweeps) transect lines, where 1 sweep consisted of either 1 forehand or 1 backhand motion of the net and 60 sweeps comprised 1 sample (Hollifield and Dimmick 1995). The Bug- Vac is a suction apparatus consisting of a 0.25-m diameter collecting funnel attached to a conventional leaf blower that contains an insect collection bucket. Each time the funnel touched the ground comprised a sample. The apparatus was run along N (30 samples) and S (30 samples) transect lines. Three location points within the 2-week home range comprised a starting point for insect sampling. From each starting point, insects were sampled by sweep net and Bug-Vac methods.

10 Insect orders were identified within a brood's 2-week home range to determine richness and relative abundance of insect orders (Hill 1995). Successful brood rearing habitat was defined as the brood's first 2-week home range if ~ 1 chick within the brood survives to the age of2 weeks. Unsuccessful brood-rearing habitat was represented by no chick survival at 2 weeks of age. 3. Diet Analysis Broods monitored for habitat use were captured for diet analysis. Standard night-lighting procedures (Labisky 1968, Huempfner et al. 1975) were used to capture bobwhite parents and their chicks. Chicks used for diet analysis were night-netted, fitted with radiotransmitters weighing ca g, and marked with patagial wing bands. Diets were sampled using standard neck constriction procedures. Samples were collected using a wire pipe cleaner tied below the crop to prevent chicks from swallowing ingested food items (Wilson 1966, Orians 1966). Chicks were released and allowed to feed for ca. 1-2 hours and recaptured. Upon recapture the wire pipe cleaner was removed, and contents of the crop were to removed with forceps; food obtained was to be largely whole and identifiable, while leaving the chick unharmed (Lockie 1955). Qrians (1966) stated that attempts to obtain samples from chicks <3 days old will result in either slippage of food past the band or strangulation of the young. Therefore, chicks were sampled between the ages of 4 to 14 days old. Samples were collected for 3 different time periods (early morning, ; mid-day, ; and late-afternoon, ) to assess any differences in species composition during feeding periods.

11 4. Additional Analyses Mark-recapture methods incorporating bird banding and hunter harvest data collected from 1990 to 1995 were used to estimate bobwhite density using the Jolly-Seber model (Jolly 1965, Seber 1965, Lefebvre 1982). This open-population model allows immigration, emigration, recruitment, and mortality during the trapping period. This population estimator was compared with density estimates derived from line transect flush counts to assess precision and accuracy (Guthery 1988, Guthery and Shupe 1989). Correlations between these 2 population estimators and subsequent annual harvest were examined (proc CORR, SAS 1989). Food habits were determined from bobwhite crops from hunter-harvested quail to evaluate use of feeders and native forages during the previous 5 years of the study. Quail crops have been collected by biologists on the Packsaddle WMA since the 1990 hunting season. A sub-sample of seed contents from hunter-harvested crops (200/year) was separated into native vs. supplemental feed, dried, and weighed. The ratio of supplemental feed as a percentage of total crop contents was determined (Korschgen 1948). Date of harvest and prevalent climatic conditions at the time of harvest (NOAA weather data) were correlated with intensity of supplemental feeder utilization to test the hypothesis that deteriorating weather conditions result in increased use of feeders bobwhites. III. Results A. Nest Habitat Use vegetative Cover. We compared percent vegetative cover and nest characteristics between successful and non-successful nests and nest sites vs. random non-use sites with analysis of variance (SAS Institute, Inc. 1996). A detrended correspondence analysis (DCA) was used to

12 find patterns in the coverage of vegetative species composition between successful and nonsuccessful nests and nest sites vs. random non-use sites using the program CAt~OCO (ter Braak 1998); data were square root transformed to minimize the effect of potential outliers. The DCA was performed on 85 plant species and 80 nest sites that were analyzed with rare species down weighted (ter Braak 1998). Nesting data from 1996 through 1998 consisted of 42 successful and 38 non-successful nests. Coverage of bare ground within 1 m 2 of nest sites and their respective random sites appear in Figure 1. In 1996 successful nests had less bare ground than non-successful nests (P = 0.013), but they did not differ in 1997 and In 1996 and 1997, coverage of bare ground was greater at corresponding random non-use sites that at successful nest sites (P = and P = 0.036, respectively). Non-successful nest sites also had less coverage of bare ground than their respective random non-use sites in 1996 (P = 0.066) and 1997 (P = 0.001). Results of the detrended correspondence analysis are shown in Figure 2. Due to the small amount of variance explained by axis 3 (eigenvalue = 0.128) and axis 4 (eigenvalue = 0.104), valid biological gradients are difficult to interpret, and as a result, we concentrate only on the first 2 axes. The ordination diagram suggested that low DCA scores of axis 1 consisted of species closely associated with bare ground and high DCA scores of axis 1 consisted of species that were associated with litter. DCA scores of axis 2 were grouped along a disturbance gradient with low disturbance occupying low DCA scores and high disturbance occupying high DCA scores. Species' cover indicated that bobwhites tended to select for less disturbed sites that were associated with an intermediate cover of litter compared with random non-use sites. The DCA revealed no statistical differences between successful and non-successful nest sites. DCA scores

13 11 of axis 1 were greater at both successful (x = 1.48) and non-successful (x= 1.42) nest sites than at random non-use sites (x= 1.18), and DCA scores of axis 2 were lower at successful (x= 0.83) and non-successful (x= 0.77) nest sites than at random non-use sites (x= 0.99; Table 1). DCA scores did not differ between successful and non-successful nest sites (Table 2). vegetative Characteristics. Vegetation cover 3 m from the nest, shrub densities, and density of little bluestem patches around nest sites at I-m, 2-m and 5-m radii, with the exception of I-m shrub stem counts at successful nest sites, was consistently greater than their respective random non-use sites (Table 3). In 1996, little bluestem density ~ I-m 2 of successful (x = 7.09) and non-successful (x= 7.06) nest sites was greater than that of their respective random non-use sites (x= 3.78; P < and x= 4.44; P = 0.012, respectively), but they did not differ in 1997 and We found no differences in the vegetative characteristics between successful and nonsuccessful nest sites (Table 4). However, nest concealment was related to nest success. Successful bobwhite nests were less visible than non-successful nests in 1996 (P = 0.026) and 1997 (P = 0.012), but they did not differ in 1998 (P = 0.536; Figure 3). Other than concealment, we were unable to identify any vegetative characteristics of nest sites that differentiate between successful and non-successful nests. Martin and Roper (1988) hypothesized that increased density of nest-site foliage (within a habitat patch surrounding the nest) decreased a predators chance of finding the nest. However, we found that the mean density of little bluest em patches at successful nest sites did not differ from non-successful nest sites. Estimates of concealment have been documented to be an important component of nest success (Keppie and Herzog 1978, Riley et ai. 1992). In 1996 and 1997 we found that greater

14 12 concealment reduced the chance of nest predation (Figure 3). Bowman and Harris (1980) found that spatial heterogeneity (disturbance) was more important than concealment in reducing nest predation. Bobwhites at PWMA selected sites that were associated with low disturbance (high spatial heterogeneity). Such sites presumably offered superior cover that helped prevent predation by inhibiting chemical, auditory, or visual clues (Martin and Roper 1988) and protect incubating bobwhites from weather and other disturbances (Riley et al. 1992). Although nest-site selection differed from random non-use sites, it did not differentiate between successful and nonsuccessful nests. As a result, we conclude that concealment may be more important to nest success of bobwhites than disturbance because wild bobwhites, regardless of nest success, apparently select nest sites with low disturbance that maximizes their chance for survival. B. Brood Habitat Use Vegetation cover for each brood's habitat was sampled in a manner similar to that for each nest site. There were no differences in vegetative cover between successful and non-successful brood sites in In 1997 and 1998, data to compare non-successful broods were unavailable; i.e., ~1 chick within each brood survived to the age of2 weeks (Table 5). Sweep nets were used to sample insect order abundance within each brood's 2-week home range. Abundances oflarva, Arachnida, Coleoptera, and Diptera were greater for successful than non-successful broods in Abundance of Hornopt era was greater for non-successful than successful broods in No data were available for non-successful broods in 1997 and 1998; i.e. ~ 1 chick within each brood survived to the age of2 weeks (Table 6).

15 C. Additional Analyses Food Habits. We compared percentage of supplemental food in quail crops between control and feeder areas with analysis of variance (SAS Institute, Inc. 1996); data were arcsine transformed prior to analysis. To assess the influence of climatic variables on the intake of supplemental food in quail, we used stepwise regression analysis (proc REG; SAS Institute, Inc. 1996), with the amount of supplemental food as the dependent variable and climatic conditions as independent variables. Variables selected for inclusion into the model were deemed to be significant when p~ 0.15 (Cody and Smith 1997). The prevalent weather conditions at time of harvest (maximum and minimum temperature, precipitation, and snowfall; National Oceanic and Atmospheric Administration 1992, 1993, 1994, 1995, 1996) and ~7-day averages just prior to harvest were included as independent variables. Associations with climatic variables also were analyzed by Pearson correlation analysis (proc CORR; SAS Institute, Inc. 1996). We computed survival rates using a modification of the SAS program by White and Garrot (1990), which uses the staggered-entry design of the Kaplan-Meier procedure or product-limit estimator (Kaplan and Meier 1958). Winter survival was estimated over a 5-month period, which included 1 month prior to and 1 month after crop collections (1 November-31 March). We partitioned our survival estimates into the following categories: a) survival without hunting or predation censored (noncensored); b) survival with hunting losses censored; and c) survival with predation losses censored. We used chi-square analysis to determine differences in survival curves and a Z- statistic to define differences in 5-month survival rates (pollock et al 1989) between feeder and control study areas. Differences were deemed to be significant when P~0.05.

16 Percentage of supplemental food in the winter diet of quail was consistently greater in birds harvested from the feeder area compared with the control area in all years (P < 0.001). Supplemental food in the diet reached a peak in winter , comprising 45.2% of the diet on the feeder area. The amount of native seed was inversely proportional to the amount of supplemental feed (Table 7). Multiple regression models that predicted supplemental food use by quail in winter on feeder (R 2 = 0.15, 7-variable model) and control (R 2 = 0.20, lo-variable model) areas were significant, but they explained only a small percentage of the variation in supplemental food use at the time of harvest (Table 8). The 7-day mean maximum temperature prior to harvest accounted for most of the variation ( %) in supplemental food use on these two areas; when mean maximum temperature declined, supplemental feed became more important (Table 9). Both mean minimum and maximum temperatures during the previous 1-7 days prior to harvest were correlated significantly and negatively with percentage of supplemental food in crops of birds harvested from the feeder area (Table 9). Variables associated with temperature generally were not correlated with use of supplemental food on the control area, except for 6- or 7-day mean maximums (Table 9). The amount of precipitation or snowfall 1-3 days prior to harvest was correlated positively with supplemental food use on the feeder study area. The amount of snowfall 1-7 days prior to harvest was correlated negatively with supplemental food in crops of birds from the control area, indicating that birds did not have access to, or they elect not to use, food plots during snow events. Although many significant correlations existed between climatic variables and percentage of supplemental food in crops of harvested birds, all correlation

17 15 coefficients were small (r < 0.22; Table 8), indicating that combinations of other extrinsic factors influenced supplemental food use by quail in western Oklahoma. Considering all fonns of mortality, the 5-month survival rate over winter was greater in the population from the feeder area compared with the control in 2 of 4 winters (Table 10). Survival rate was 6-fold higher in and 2-fold higher in on the feeder area compared with the control. In winter , the population on the control area had a survival rate 2-fold higher than the population on the feeder area. Because mortality risks may be elevated from hunters and predators concentrating their activities in the vicinity of supplemental feeders, we censored those groups from the data set and recalculated rates of survival for each year (Table 10). Censoring causes of predation from the data set (largely only hunting-related deaths included) revealed that survival rates were about 1.5- fold greater on the control than feeder study areas in 2 of 4 winters surveyed ( ), suggesting that hunters concentrated their activities around feeders. Censoring hunting mortalities from the data set revealed that during 2 winter seasons, survival rate was 2- ( ) to 5-fold ( ) greater in the quail population in the feeder area compared with the control population. The control population did not have survival rates that were significantly greater than the population on the feeder study area with hunting mortalities censored, indicating that feeders did not adversely concentrate predators. Our data demonstrated that bobwhites with access to supplemental feeders use this nutritional resource even when food plots are available. Results suggest, however, that bobwhites on the control were denied access or did not utilize food plots during periods of snow cover. Results further suggest that bobwhites with access to feeders take greater advantage of

18 supplemental feed during precipitation and snowfall events that occurred 1-3 days prior to harvest. Bobwhites progressively increased supplemental feeder use with decreasing ambient temperatures, but low correlation coefficients suggest that winter climatic conditions, alone, were not strong predictors of use of supplemental feeders. We attribute low correlation coefficients to the year-round availability and daily use of feeders by bobwhites, regardless of prevalent weather conditions. Our results support the hypothesis that supplemental feeders can increase survival during winter when climatic conditions are especially stressful (Lehmann 1984:273, Guthery 1986:48, Landers and Mueller 1986:19). Kendeigh (1969, 1970) proposed that daily fluctuations in body weights are greater at low temperatures because of increased energy demands to maintain core body temperatures causing greater depletion offat reserves. Leifand Smith (1993) reported that bobwhites that consumed low-energy foods were unable to accumulate as much body fat as bobwhites that consumed high-energy foods. Supplemental feeding not only provides bobwhites with a high-energy food source that is required to maintain existence energy during low ambient temperatures (Robel et al. 1979) but also may help to preserve energy reserves and increase predator avoidance by decreasing foraging activities. We believe that lower loss to predators on the feeder area in and than on control area supports that hypothesis (Table 10). Raynor (1980) found an inverse correlation between mean population density of bobwhites and days of snow cover (r = -0.51) and total snowfall (r = -0.32) during preceding winters. Roseberry and Klimstra (1984) also documented population declines in winter that varied from 36% to 81% in (x= 63%) with the greatest natural mortality in January-

19 March. Peoples et al. (1994) documented benefits of essential and nonessential amino acids to supplementally fed bobwhites in winter, and Robel (1969) found that bobwhites with access to a supplemental food source had greater body weights, body fai, and food in their crops than those without access to supplemental food. Although climatic conditions were not strong predictors of the prevalence of supplemental food in quail crops in our study area, the combination of decreased native foods and severe winter stresses could have acted together to regulate supplemental food use. Thus, periods of food and climate stress in winter may cause bobwhites to reach a critical nutritional threshold; birds with access to supplemental feeders can receive nutritional benefits that ultimately affect their winter rates of survival. We documented that survival rates, calculated with predation losses censored (largely only hunter-related deaths), were higher on the control area in winters This evidence supports Lehmann's hypothesis (1984:273) that feeders can congregate both birds and hunters. Furthermore, survival rates with censored hunting losses (largely only predator-related deaths) were higher on the feeder area in winter and did not differ in (Table 10). We found no evidence to support the hypotheses of Lehmann (1984:276), Guthery (1986:56), and Landers and Mueller (1986: 19) that supplemental feeders concentrate non-human predators. Our results suggests that supplemental feeders can have a positive impact on winter survival of bobwhite quail in western Oklahoma in some years. Chick Diets. We were unable to conduct diet analyses of chicks due to difficulty in relocating and capturing chicks during the day. Parents would "dump" their chicks when field personnel were present. Only a few feet of separation between parent and chick would be

20 adequate for them to avoid capture. Although we were granted permission to sacrifice chicks in 1998 for diet analysis, sample size from successful nests was very low. As a result, with only 4 successful broods monitored in 1998, we chose to maximize our sample size of radiocollared chicks. Thus, we were unable to gain adequate samples for chick-diet analysis. As a result, the chick-feeding habit d?~:1'was not collected. Insect Ayail-..illi1I. Insects were sampled only by sweep net. The modification of the Bug-vac did not work due to the amount of sand that was pulled into the vac and frequently plugged the hose. Moyement Patterns. Arrangements were made to format data for use in SAS to determine seasonal changes in home range size, group associations, and movement patterns of radio-collared bobwhites. Density Estimates. These data were properly formatted and delivered to the ODWC. Mark-recapture data collected from through bird banding and hunter harvest were prepared for analysis using CAPTURE (White et al. 1982), a computer program that estimates population densities. CAPTURE failed to provide an appropriate model because of an insignificant number of recaptures and excessive trap occasions where no birds were captured. As a result, estimates of density from mark-recapture data were inconclusive. The program DISTANCE was used to analyze line transect flush count data collected during this same time period. Density results were submitted for publication in the Wildlife Society Bulletin; the manuscript is entitled "The Effect of Quail Feeders on Northern Bobwhite Density in Western Oklahoma" by S. DeMaso et al.

21 Darrell Townsend II, R.L. Lochmiller and D.M. Leslie, If. Oklahoma State University

22 Harold Namminga Oklahoma Department of Wildlife Conservation

23 Church, K E., 1. R. Sauer, and S. Droege Population trends of quails in North America. 21 Pages in K. E. Church and T. V. Dailey, eds. Quail Ill: national quail symposium. Kansas Department of Wildlife and Parks, Pratt, Kansas. Cody, R. P., and 1. K Smith Applied statistics and the SAS programming language. Prentice Hall, Upper Saddle River, New Jersey. Curtis, P. D., B. S. Mueller, P. D. Doerr, and C. F. Robinette Seasonal survival of radiomarked northern bobwhite quail from hunted and non-hunted populations. Biotelemetry International Radio-telemetry Symposium 10: Daubenmire, R A canopy-coverage method of vegetational analysis. Northwest Science 33: Gotmark, F., D. Blomqvist, O. C. Johansson, and 1. Bergkvist Nest site selection: a tradeoff between concealment and view of the surroundings. Journal of Avian Biology 26: Guthery, F. S Beef, brush and bobwhites: quail management in cattle country. Caesar Kleberg Wildlife Research Institute Press, Kingsville, Texas. _~, F. S Line transect sampling of bobwhite density on rangeland: evaluation and recommendations. Wildlife Society Bulletin. 16: , F. S. and T. E. Shupe Line transect vs. capture-removal estimates of bobwhite density. Proceedings of the Annual Conference of the Southeast Association offish and Wildlife Agencies. 43 : Higgins, K. F., 1. L. Oldemeyer, KJ. Jenkins, G.K Clambey, and R.F. Harlow Vegetation sampling and measurement. Pages in T.A. Bookhout, ed. Research

24 and management techniques for wildlife and habitats. The Wildlife Society. Bethesda, Maryland. Hill, C. J Linear strips of rainforest vegetation as potential dispersal corridors for rain forest insects. Conservation Biology 9: Hollifield, B. K., and R. W. Dimmick Arthropod abundance relative to forest management practices benefiting ruffed grouse in the southern Appalachians. Wildlife Society Bulletin. 23: Holway, D. A Nest-site selection and the importance of nest concealment in the blackthroated blue warbler. The Condor 93: Huempfuer, R. A., S. J. Maxson, G. J. Erickson, and R. J. Schuster Recapturing radiotagged ruffed grouse by nightlighting and snow-burrow netting. Journal of Wildlife Management. 39: Jolly, G. M Explicit estimates from capture-recapture data with both death and immigration-stochastic model. Biometrika 52: Kabat, c., and D. R. Thompson Wisconsin quail, Population dynamics and habitat management. Wisconsin Conservation Department Technical Bulletin No. 30. Kaplan, E. 1., and P. Meier Nonparametric estimation from incomplete observations. Journal of the American Statistics Association 53: Kendeigh, S. C Energy responses of birds to their thermal environments. Wilson Bulletin 81: Keppie, D. M., and P. W. Herzog Nest site characteristics and nest success of spruce grouse. Journal of Wildlife Management. 42:

25 23 Klirnstra, W. D Bobwhite quail and changing land use. Pages 1-5 in F. Schitoskey, Jr., E. C. Schitoskey, and 1. G. Talent, eds. Proceedings Second National Bobwhite Quail Symposium, Oklahoma State University, Stillwater, Oklahoma., and Roseberry Nesting ecology of the bobwhite in sourthern illinois. WildlifeMonographs 41:1-37. Korschgen, L Late fall and early winter food habits of bobwhite quail in Missouri. Journal of Wildlife Management. 12: Labisky, R. F Nightlighting: its use in capturing pheasants, prairie chickens, bobwhites, and cottontails. illinois Natural History Survey. Biological Notes 62: Landers, 1. L., and B. S. Mueller Bobwhite quail management: a habitat approach. Tall Timbers Research Station, Tallahassee, Florida. Lefebvre, L. W., D. L. Otis, and N. R. Holler Comparison of open and closed models for corton rat population estimates. Journal of Wildlife Management 46: Lehmann, V. W Bobwhites in the Rio Grande Plain of Texas. Texas A & M University Press, College Station. Leif, A. P. and L. M. Smith Winter diet quality, gut morphology and condition of northern bobwhite and scaled quail in west texas. Journal offield Ornithology. 64: Lockie, 1. D The breeding and feeding of Jackdaws and Rooks with notes on carrion Crows and other Corvidae. Ibis 97: Martin, T. E Nest predation among vegetation layers and habitat types: revising the dogmas. The American Naturalist. 141:

26 -----J T. E. and Roper Nest predation and nest-site selection ofa western population of the hermit thrush. The Condor 90: Moseby, H. S., and W. S. Overton Fluctuations in the quail population on the Virginia Polytechnic Institute fanns. Transactions of the North American Wildlife and Natural Resources Conference 15: National Oceanic and Atmospheric Administration Climatological data Oklahoma. National Climatic Data Center, Asheville, North Carolina Climatological data Oklahoma. National Climatic Data Center, Asheville, North Carolina Climatological data Oklahoma. National Climatic Data Center, Asheville, North Carolina Climatological data Oklahoma. National Climatic Data Center, Asheville, North Carolina Climatological data Oklahoma. National Climatic Data Center, Asheville, North Carolina. Nudds, T. D Quantifying the vegetative structure of wildlife cover. 37th Midwest Fish and Wildlife Conference. Toronto, Ontario. Orians, G. H Food of nestling yellow-headed blackbirds, Cariboo Parklands, British Columbia. The Condor 68: Peoples, A. D., R. L. Lochmiller, 1. C. Boren, D. M. Leslie Jr., and D. M. Engle Limitations of amino acids in diets of northern bobwhites (Colinus virginianus). American Midland Naturalist 132:

27 Pollock, K. R., S. R. Winterstein, C. M. Bunck, and P. D. Curtis Survival analysis in telemetry studies: the staggered entry design. Journal of Wildlife Management 53:7-15. Raynor, G. S Bobwhite population changes on Long Island, New York American Birds 34: Riley, T. Z., C. A. Davis, M. Ortiz, and M. J. Wisdom Vegetative characteristics of successful and unsuccessful nests of lesser prairie chickens. Journal of Wildlife Management 56: Robel, R. J Food habits, weight dynamics, and fat content of bobwhites in relation to food plantings in Kansas. Journal ofwi1dlife Management 33: _~, and S. D. Fretwell Winter mortality of bobwhite quail estimated from age ratio data. Transactions Kansas Academy of Science 73: _~, A. R. Bisset, A. D. Dayton, and K. E. Kemp Comparative energetics of bobwhites on six different foods. Journal of Wildlife Management 43: Robinette, C. F., and P. D. Doerr Survival of northern bobwhite on hunted and nonhunted study areas in the North Carolina sandhills. Pages in K. E. Church and T. V. Dailey, eds. Quail III: national quail symposium. Kansas Department ofwtldlife and Parks, Pratt, Kansas. Roseberry, J. L., and W. D. Klimstra Population ecology of the bobwhite. Southern TIlinoisUniversity Press, Carbondale, illinois. SAS Proprietary Software Release 6.08 ts407. SAS Inst. Inc., Cary, North Carolina. Schmutz, J. A., C. E. Braun, and W. F. Andelt Nesting habitat use of Rio Grande Wild turkeys. Wilson Bulletin. 101:

28 Seber, G. A. F A note on the multiple-recapture census. Biometrika 52: Stewart, A. J. A. and A. F. Wright A new inexpensive suction apparatus for sampling arthropods in grassland. Ecological Entomology 20: Taylor, J. S Aspects of northern bobwhite reproductive biology in south Texas. M.S. Thesis, Texas A&I University, Canyon, Texas. Taylor, M. A. and F. S. Guthery Fall-winter movements, ranges, and habitat use oflesser prairie chickens. Journal of Wildlife Management. 44: ter Braak, C. J. F CANOCO--a FORTRAN program for canonical community ordination by [partial] [detrended] [canonical] correlation analysis, principal components and redundancy analysis. Agricultural Mathematics Group, Wageningen, The Netherlands. Webb, W. M., and F. S. Guthery Response of bobwhite to habitat management in northwest Texas. Wildlife Society Bulletin. 10: White, G. C., and R. A. Garrott Analysis of wildlife radio-tracking data. Academic Press, Incorporated, San Diego, California. White, G. C., D. R. Anderson, K. P. Burnham, and D. L. Otis Capture-recapture and removal methods for sampling closed populations. Los Alamos Natl. Lab, Los Alamos New Mexico. Wilbur, S. R Live-trapping North American upland game birds. U.S. Fish and Wildlife Service Special Scientific Report, Wildlife No. 106:1-37. Wilson, M. F Breeding ecology of the Yellow-headed Blackbird. Ecological Monographs. 36:51-77.

29 Table 1. Detrended Correspondence Analysis (DCA) mean sample scores of successful and nonsuccessful nest sites vs. their random non-use sites at PWMA, Ellis County, Oklahoma, (all data pooled). 27 Nest Success Nest Site Random Site DCA Axis Eigenvalue n x SE n x SE Successful Axis Axis Axis Axis Non-Successful Axis Axis Axis Axis adifferences evaluated with analysis of variance.

30 Table 2. Detrended Correspondence Analysis (DCA) mean sample scores of bobwhite successful and non-successful nest sites at PWMA, Ellis County, Oklahoma, (all data pooled). Successful Non-Successful DCA Axis Eigenvalue n X SE n x SE E- Axis Axis Axis Axis anifferences evaluated with analysis of variance.

31 Table 3. Comparison of vegetative characteristics of successful and non-successful bobwhite nest sites vs. their respective random sites on PWMA, Ellis County, Oklahoma, Nest Success Nest Site Random Site Characteristic n X SE n x SE Successful Nests Shrub Stem Count 1m Shrub Stem Count 2m Shrub Stem Count 5m Little Bluestem Patch 1m Little Bluestem Patch 2m Little Bluestem Patch 5m Cover Board m Cover Board m Cover Board m Cover Board m Non-Successful Nests Shrub Stem Count 1m Shrub Stem Count 2m Shrub Stem Count 5m Little B1uestem Patch 1m Little Bluestem Patch 2m Little B1uestem Patch 5m Cover Board m Cover Board m Cover Board m Cover Board m adifferences evaluated with analysis of variance.

32 Table 4. Comparison of vegetative characteristics between successful and non-successful bobwhite nest sites at PWMA, Ellis County, Oklahoma, Successful Non-Successful Characteristic n X SE n x SE P Shrub Stem Count 1m Shrub Stem Count 2m Shrub Stem Count 5m Little Bluestem Patch 1m Little Bluestem Patch 2m Little Bluestem Patch 5m Cover Board m Cover Board m Cover Board m Cover Board m Total Height Clump Width Clump Length Bowl Width Bowl Length Depth Dome Depth Bowl Lining Thickness "Differences evaluated with analysis of variance.

33 31 Table 5. Comparison of percent cover by species of successful and non-successful bobwhite brooding areas on PWMA, Ellis County, Oklahoma Successful Non-successful Year n X SE n X SE Species 1996 Bare Ground Leaf Litter &hizachyrium scoparium II U Andropogon gerrardii Paspa/um sp Panicum virgatum II Ambrosia psilostachya Eriogonum annuum II Boute/oua curtipendula II Crotonsp Cyperussp Plantago patagonica Quercus havardii Artemisia sp Celtis sp Heterotheca sp Conyza Canadensis Bare Ground Leaf Litter &hizachyrium scoparium Andropogon gerrardii Paspalum sp Panicum virgatum Ambrosia psilostachya Eriogonum annuum Boute/oua curtipendula Crotonsp Cyperussp Plantago patagonica Quercus havardii Artemisia sp Celtis sp Heterotheca sp

34 32 Conyza Canadensis Bare Ground Leaf Litter &hizachyrium scoparium Andropogon gerrardii Paspalum sp Panicum virgatum " Ambrosia psilostachya Eriogonum annuum Bouteloua curtipendula Croton sp Cyperussp Plantago patagonica Quercus havardii Artemisia sp Celtis sp Heterotheca sp Conyza Canadensis 'Differences evaluated with analysis of variance.

35 Table 6. Abundance of insect orders collected by sweep net within successful and non-successful brood sites on PWMA, Ellis County, Oklahoma Successful Non-Successful Year Order n X SE n X SE 1996 Larva Arachnida Neuroptera Orthoptera Hemiptera Homoptera Coleoptera Diptera Lepidoptera Hymenoptera Mantodea Larva Arachnida Neuroptera Orthoptera Hemiptera Homoptera Coleoptera Diptera Lepidoptera Hymenoptera Mantodea Larva Arachnida Neuroptera Orthoptera Hemiptera Homoptera Coleoptera Diptera Lepidoptera Hymenoptera Mantodea anifferences evaluated with analysis of variance.

36 Table 7. Relative percentage of food items in the crops of hunter-harvested bobwhites from study areas with (feeder) and without (control) supplemental feeders during the vymters of Control Feeder Winter season n x SE n x SE % Supplemental <0.001 % Native <0.001 % Vegetation % Insects % Miscellaneous % Supplemental <0.001 % Native <0.001 % Vegetation % Insects % Miscellaneous % Supplemental <0.001 % Native % Vegetation % Insects % Miscellaneous % Supplemental <0.001 % Native % Vegetation <0.001 % Insects % Miscellaneous Winters pooled % Supplemental 400 S <0.001 % Native <0.001 % Vegetation <0.001 % Insects % Miscellaneous <0.001 IDifferences evaluated with analysis of variance.

37 35 Table 8. Climatic variables selected by a stepwise multiple regression analysis to predict the prevalence of supplemental foods in the crops of hunter-harvested bobwhite from study areas with (feeder) and without (control) supplemental feeders in winter (data from pooled). Parameter Area Variable' estimate SE PartialR 1 F P Control Intercept daymax dayprec daysnow < dayprec < daymax I-day min day prec daymax day snow dayprec <0.001 Feeder Intercept < daymax < dayprec < day snow < daysnow dayprec daymin daymin a 1- to 7-day (n-day) averages for maximum (max) or minimum (min) daily temperature, precipitation (prec), and snowfall (snow).

38 Table 9. Correlation coefficients for the relationships between prevalence of supplemental food in quail crops and mean prevalent climatic conditions that existed 1-7 days prior to their harvest in Oklahoma, winters (all data pooled). Area Maximum Minimum Mean period Precipitation Temperature Ternperature Snowfall Prior to harvest r pa r pa r pa r pa Control 1 day days days days days days days < Feeder 1 day < days < < days < days days days < days < 'Differences evaluated with correlation analysis.

39 Table 10. Estimated 5-month survival rate in winter with 95% confidence intervals for bobwhite residing in study areas with (feeder) and without (control) supplemental feeders in Oklahoma for the period Differences in survival curves and survival rates between study areas were tested using chi-square analysis and Z-statistics. Group Control Feeder Survival curves Survival rates Winter n S 95% CI n S 95% CI i P Z P Non-censored a , Hunting-censored b < Predation-censored" a Pertains to predation and hunting censors only; birds which fell into the missing and unknown categories were censored. bonly predation deaths. COnlyhunting deaths.

40 Figure 1. Mean coverage of bare ground for successful and non-successful bobwhite nest sites and their respective random non-use sites at PWMA, Ellis County, Oklahoma, Q) C) l'ls 40 Successful Nests "- Q) 1m Successful Random > 30 0 o Non-Successful U ~ Non-Successful Random 0 20

41 Figure 2. Results of a detrended correspondence analysis of 85 plant species from 81 individual bobwhite nest sites and their corresponding random sites at PWMA, Ellis County, Oklahoma, (all data pooled). Species scores have a minimum weight of 1. High Dlsturban~e.. q:y rp. PI pa : : 1;, an : Cr Ip. x ""Am pi.. Nonsuccessful.. Random Sites ~ Low '<7 Disturbance Se Ie Bare Ground AD Be ~ Litter l cellp. QUi ha ~ -0.5

42 Figure 3. Mean estimates of nest concealment for successful and non-successful bobwhite nest sites at PWMA, Ellis County, Oklahoma, (* = P < 0.05 for within-year comparisons). 20 ~ ~-.c 15 fi) :> ~0 10 Successful o Non-Successful 1997* Year