Journal of the Southeastern Association of Fish and Wildlife Agencies

We recorded plant community characteristics in an 18-year-old, white-tailed deer (Odocoileus virginianus) exclosure in the Lower Coastal Plain of Mississippi during summer 1996. The 0.4-ha exclosure was constructed in 1977 within upland pine (Pinus spp.) forest of the Leaf River Wildlife Management Area in Perry County, Mississippi. Surveys at the time of exclosure construction revealed that similar plant cover and species richness existed inside and outside of the exclosure. Surveys were conducted during 1996 along 8 37-m transects located inside and outside the exclosure. Forest canopy did not differ between exclosure and control sites. Exclosure habitat supported 59 plant species and unprotected control sites contained 43 plant species. Density and coverage of midstory vegetation were greater (/• < 0.0001) in exclosed sites than control sites.

We describe a new, non-destructive procedure for visually estimating forage biomass based on volumetric cover. The accuracy of this procedure was tested against actual dry biomass by clipping and weighing 41 plots of wax myrtle (Myrica cerifera). Visual estimates of forage biomass were significantly related (P ≤ 0.001) to actual biomass determine by clipping (r2 = 0.925; y = 16.36 + 2.52 X, where y = dry biomass and X = volumetric cover). We developed this procedure to apply 3 white-tailed deer (Odocoileus virginianus) habitat suitability index (HSI) models to a suburban development. The 3 traditional HSI models evaluated did not include several variables unique to developed areas that could affect deer habitat quality. Therefore, the models may need to be modified before they can be applicable in developed areas.

Recent declines in deer densities in some areas of Eglin Air Force Base (AFB), Florida, have prompted concerns regarding the impact of coyote (Canis latrans) predation. We determined the food habits of coyotes from analysis of 166 scats collected on Eglin AFB from November 1994 to October 1996. We compared the frequency of white-tailed deer (Odocoileus virginianus) remains found in scats collected in high- and low-deer density areas during the deer fawning season to evaluate impacts of coyotes on white-tailed deer. Important coyote foods (by frequency of occurrence) were shrub/vine fruit (80%), beetles (55%), persimmon (27%), and deer (15%). Deer occurred most often (29%) during the fawning season. There was no difference in the frequency of deer remains found in scats collected in high- and low-deer density areas. The dominance of soft mast in the diet illustrates the important role that soft mast can play in the diet of coyotes.

Traditionally, reducing game-bird nest depredation has involved lethal means of predator control. We evaluated a non-lethal alternative, conditioned taste aversion (CTA), in Tom Green County, Texas. Simulated nests were constructed and baited with 3 eggs injected with lithium chloride, an aversive chemical. Simulated nests were constructed along the perimeter of a 40-ha pasture. A 21-day treatment phase was conducted with depredated nests being rebaited daily with treated eggs. A 28-day posttreatment phase involved establishing 24 non-treated nests in both the treated pasture and a control pasture. The study was replicated over 2 sites: the Management, Instruction, and Research Center (MIRC) and Stone Ranch (SR). There was no difference in nest survival between treatment and control pastures at MIRC (F = 5.0; 1, 3 df; P = 0.1). At SR, nest survival was higher in the treated pasture (F = 11.64; 1, 3 dfi P = 0.03).

Detailed knowledge of a species' biology and ecology is required before sitespecific management programs can be implemented. Therefore, we examined seasonal changes in food habits, nutritional status, and reproduction in a sample of 54 feral hogs (Sits scrofa) collected over a 2-year period from Cumberland Island National Seashore, Georgia. Seasonal variations in food habits probably were related to changes in food availability. Differences in seasonal rainfall patterns between years appeared to be related to dietary changes. No significant seasonal changes in body mass, fat indices, or crude protein in stomach contents were evident, which indicated that hogs on this subtropical island may not undergo pronounced seasonal variations in nutritional status.

Existing landcover maps offer an inexpensive opportunity to conduct largescale habitat assessments for black bears (Ursus americanus), but because cover classes used in these maps may have been developed without consideration for bears, inferring bear food and cover distribution from these maps may be difficult. We evaluated the information content of a habitat map that we constructed using National Wetlands Inventory (NWI) data for a composite home range of 21 radio-tagged adult black bears in coastal Louisiana. Habitat types having potentially different food and cover resources for bears and recognizable from NWI data were deciduous broadleaf forest, bald cypress forest, mixed deciduous broadleaf and bald cypress forest, scrub-shrub wetlands, brackish and fresh marsh, deciduous broadleaf forest spoil, upland hardwood forest, and agriculture. We compared measurements taken from 113 plots in 77 stands distributed among 7 habitat types.

We studied use of a new, less expensive design of wildlife crossing to determine the acceptance of the structure by wildlife. We documented wildlife use of 2 precast concrete wildlife crossings from 27 March 1995 to 30 June 1996 on State Road (SR) 29 in southwest Florida. Two additional crossings of a different design were monitored on Interstate (I)-75 for comparison. Over 1,000 photographs were taken of >20 species of wildlife, domestic animals, and humans using those 4 wildlife crossings. The SR 29 structures were utilized by Florida panthers (Felis concolor coryi), black bears (Ursus americanus), bobcats (Felis rufus), white-tailed deer (Odocoileus virginianus), raccoons (Procyon lotor), and 17 other species. Panther use of the 1-75 wildlife crossings increased over time. The new design of wildlife crossing on SR 29 allowed safe passage of many species of wildlife, including panthers.

We evaluated a habitat suitability (HSI) model developed for mallards (Anas platyrhynchos) wintering in the Lower Mississippi Valley by comparing mallard densities obtained from aerial surveys with habitat suitability indices derived from satellite imagery for 25, 256-km2 sampling units. Regression models that related mallard densities to habitat suitability indices accounted for only 29% of the variability in the data and the 95% confidence interval of predicted mallard densities included zero for most habitat suitability indices evaluated. Thus, we conclude that the published HSI model is a poor predictor of wintering mallard density in the Lower Mississippi Valley. We suggest model revision to allow users to remotely obtain model inputs for habitat characteristics at landscape scales. Further, we suggest the model be revised to consider yearly variation in habitat and flood conditions that better reflect the ability of an area to support wintering mallards.

An understanding of the mechanisms by which northern bobwhite (Colinus virginianus) populations respond to old-field habitat management is important to evaluate efficacy of these practices. We examined reproductive strategies and success of 114 radio-marked bobwhite on a managed wildlife area in east-central Mississippi during 1994—1996. Fifteen female and 5 male bobwhite incubated 23 nests. Male-incubated nests, female-incubated first nests, and female-incubated renests contributed 21.7%, 65.2%, and 13.0% to total nesting effort, respectively. Female-incubated first nests and male-incubated nests each accounted for 44% of successful nests. Of birds alive on 15 April (40 female and 74 male), 37.5% of females and 6.8% of males attempted ≥1 nest, whereas 12.5% of females and 5.4% of males were successful. Female nest initiation peaked in mid-May prior to the onset of male nesting.

We estimated interval (Dec-Feb) recovery and survival rates of marked northern bobwhites (Colinus virginianus) to determine effects of radiomarking and supplemental feeding on the Packsaddle Wildlife Management Area (WMA) in western Oklahoma from 1991 to 1996. We also estimated unretrieved harvest (crippling loss) reported by hunters and compared it to unretrieved harvest of radiomarked bobwhites to determined accuracy of hunter information. We banded 308 and radiomarked 296 bobwhites. Interval survival and recovery rates were estimated using the computer program MARK. Recovery rates of banded bobwhites (0.39) differed (χ2 = 5.03, P = 0.03) from radiomarked bobwhites (0.30). Estimated interval survival rates differed (χ2 = 42.1, P < 0.01) between banded (0.19) and radiomarked bobwhites (0.56). In our study, radiomarking bobwhites had an apparent positive influence on survival.

Call counts for a number of gamebirds (e.g., northern bobwhite [Colinus virginianus] and wild turkey [Meleagris gallopavo]) have been used to index population levels and trends and to document species presence or absence. Call counts for wild turkeys have been used for these purposes, but gobbling activity has not been related quantitatively to population size, reproduction, weather, male age structure, or hunting variables. Consequently, we examined these factors as they affected gobbling activity on Tallahala Wildlife Management Area, Bienville National Forest, in central Mississippi, from 1984 to 1995. Using multiple linear regression, we determined that within-year gobbling activity was related to hunter effort, days into call count period, wind velocity, year, and dewpoint. Among years, gobbling activity was related to hunter effort and hunter success.

A persistent shortcoming of wild turkey (Meleagris gallopavo) management programs is the inconsistency in survey techniques. One approach to standardize turkey population monitoring is to use cameras and infrared sensors. The 7 primary assumptions associated with using cameras and infrared sensors to monitor turkey populations can be grouped into those pertaining to baiting and those associated with sampling design. Because none of these assumptions have been tested, our objective is to outline an experimental design appropriate for determining which theoretical assumptions are practically valid. We recommend that testing these assumptions be a priority for additional research on using camera and infrared sensors for monitoring wild turkey population(s).

Most researchers arbitrarily delineate study areas even though a quantitative estimate of study area size can be generated from capture and subsequent locations of radio-equipped animals. Arbitrary delineation may result in biased estimates of density. Density is often determined with capture-mark-recapture designs that do not include locational data from radio-equipped animals. We used logistic regression to determine probability of recapture of radio-equipped wild turkey (Meleagris gallopavo) hens based on pre-sample distances from bait sites for hens using and not using baited sites. We then used the posterior probability function from logistic regression analogously to the detection function from variable circular plot methods.

To gain a better understanding of the roles of public and private hunting areas in the harvest dynamics of an eastern wild turkey (Meleagris gallopava var. sylvestris) population, we conducted a banding study in southeastern Louisiana from 1989-1997. Physical measurements were collected on birds at capture and harvest. During spring gobbler seasons, 137 of 199 banded gobblers were recovered through hunting. For the 1991-1994 seasons, we observed higher hunting pressure and more effort to bag a gobbler on public (0.22 hunter-days/ha and 40.7 hunter-days/bird) than on private land (0.13 hunter-days/ha and 15.6 hunter-days/bird). We captured a higher proportion of juvenile gobblers on a public area than on surrounding private land (P = 0.024), but found only weak evidence (P = 0.073) that hunters on the public land harvested a greater proportion of juveniles. The direct recovery rate for juveniles was 23% (SE = 3.7) and for adults was 70% (SE = 5.5).

Quality management for white-tailed deer (Odocoileus virginianus) is becoming increasingly popular in the southeastern United States, yet surprisingly little information has been published that describes quality or trophy management strategies in detail. The quality deer program at the McAlester Army Ammunition Plant (McAAP) is unique because it maintains high hunter opportunity while producing high-quality white-tailed deer. Several strategies have been incorporated into the management program to help maintain its unique characteristics. The first is a regulation that limits hunters to traditional archery equipment (recurve or longbow), thereby reducing hunter success and providing bucks with greater opportunity to reach maturity. The second is a centrally located, 4,500-ha refuge that receives little hunting pressure and increases the proportion of mature bucks in the population.

We compared antler characteristics and body mass at 4.5 years of age (adult) of 140 white-tailed deer (Odocoileus virginianus) reared in a captive herd at the Kerr Wildlife Management Area (Hunt, Texas) from 1973 to 1990. Each yearling (1.5 years old) was classified as spike- (N = 43) or fork-antlered (N = 97), and its live body mass recorded. Fork-antlered yearlings were further partitioned into 3-5 points (N = 33) and ≥6 points (N = 64) subclasses based on the number of antler points ≥2.54 cm in length. All deer were reared in 1.62-ha enclosures and maintained on a 16% crude protein diet ad libitum. In ensuing years, antlers were removed and live body mass recorded. At 4.5 years, the gross Boone and Crockett (GBC) score of each buck was measured. The average GBC score of adult deer that were fork-antlered yearlings (127.8 ± 2.0 SE) was greater (P <0.001) than those of spike-antlered yearlings (89.9 ± 2.8).

Home range size is an important component of wild turkey (Meleagris gallopavo) ecology. We estimated 95% convex polygon home ranges for gobblers and hens within biological seasons in central Mississippi. Mean home range size of gobblers (N = 97) varied from 607.1 ha (subadults during spring) to 809.9 ha (subadults during fall/winter). Mean home range size of hens (N = 127) varied from 97.2 ha (early brood) to 541.9 ha (fall/winter). Male home range size did not differ among seasons (P >0.05). However, gobblers tended to have larger home range sizes than hens, which likely reflected sexual dimorphism of turkeys and movements of gobblers to associate with hens during spring. Home ranges for hens were smallest during the early brood period (P<0.003). Home ranges on our area were larger than reported from other comparable Southeastern studies.

Wildlife management is the interaction of wildlife populations, habitats, and people. The eastern wild turkey (Meleagris gallopavo silvestris) is an important wildlife resource. However, quantitative data on the human dimensions of wild turkey management are scarce. Therefore, we surveyed 2,143 Mississippi turkey hunters by mail to determine characteristics of this group, examine attitudes towards regulations and management issues, and determine how hunter characteristics influenced attitudes. We received responses from 1,524 participants (71.1%). The average respondent was male, a Mississippi resident, 39 years old, had completed at least some college education, and had hunted turkeys for 13 years. Respondents hunting on private land reported higher harvest rates (0.73) and hunter success (0.44) than public land hunters (0.38 and 0.27, respectively). Attitudes towards 17 questions dealing with turkey hunting regulations or pertinent management issues were reported.

The wild turkey (Meleagris gallopavo) has ecological, aesthetic, recreational, and economic values. Gross expenditures of hunters have been used to estimate economic value of game species. We examined the economic impact and associated values of the wild turkey in Mississippi. Expenditures of wild turkey hunters were obtained from a mail survey (N = 2,143, 69.6% response to economic section) and were used in an input-output model to determine economic impacts for the state. There were 39,775 hunters engaged in 334,856 activity-days in 1993. Wild turkey hunters spent an estimated $14.8 million or $44.27 per hunter day in 1993. Total sale impacts were $16.7 million. The value-added component of the economic impact totaled $10.4 million and supported 385 jobs. We also examined the structure of the economy in relation to the wild turkey.

Past research suggests nesting habitat limits survival and growth of some eastern wild turkey (Meleagris gallopavo silvestris) populations. However, information on nest-site selection is primarily on established populations with limited data on restocked birds. Our objectives were to assess nest-site characteristics of relocated birds and determine causes of nest failure in the Post Oak Savannah of eastern Texas. Radio-tagged wild turkey hens (N = 48) were relocated to 4 areas in winter 1994. We compared understory and ground cover characteristics between 22 nest sites of radio-tagged turkeys and 22 random sites. Furthermore, we measured Euclidean distance to transition zones (“edges”) and presence/absence of protective barriers (“guard object”) at nest and random sites. Important nest-site characteristics were lateral cover, height of vegetation, and protective barriers. Mammalian predation was the major cause of nest failure.

Habitat use of eastern wild turkey (Meleagris gallopavo silvestris) broods (1993: N= 7; 1994: N = 9) in western Tennessee were monitored for up to 2 months following hatching to determine preferred habitats during this critical period. Based on availability within their home ranges, hens and their broods selected bottomland hardwoods over other habitat types during weeks 1-4 (P < 0.001). Older broods (4-6 weeks) selected upland hardwood, upland pine, bottomland hardwood, and open habitats in that order over other habitat types (P < 0.001). Intensive monitoring of 6 broods in 1995 verified that telemetry data accurately reflected habitat use. Invertebrate biomass was lower in bottomland forests than in open habitats (P = 0.02), but apparently was adequate to meet the needs of developing poults.

We investigated attributes of 5 colony sites of endangered least terns (Sterna antillarum athalassos) nesting on the Arkansas River from Tulsa to Muskogee, Oklahoma, in 1992 and 1993. We tested hypotheses that least terns selected areas to nest with the lowest amount and height of vegetation and nested near driftwood or similar debris. River flows prior to the nesting season in May 1993 scoured all sandbars, removing vegetation and increasing abundance of driftwood. Attributes of colony sites of nesting least terns varied substantially from one location to another. Distances from nests and random points to vegetation and driftwood varied among colony sites. Occurrence of driftwood at colony sites increased after floods in 1993, and least tern nests at 4 of 5 colony sites were significantly closer to driftwood than were random points inside a colony.

We field-tested the Reproductive Index (RI) of the great blue heron (Ardea herodias) Habitat Suitability Index (HSI) model in the southcentral Great Plains with the aid of Geographic Information System (GIS) technology. From January 1993 through May 1994, populations of great blue herons in 18 rookeries located throughout Oklahoma were monitored, and GIS was used to evaluate data on rookery habitat structure and surrounding landscape features. Eighteen rookeries were classified as potential nest sites and RI ratings were determined for each rookery according to model criteria. The RI identified only 3 (17%) of the 18 rookeries as suitable habitat for reproduction. After modifying the RI using habitat and landscape data from the 18 rookeries in Oklahoma, the RI more reliably characterized the habitat suitability of rookeries in the southcentral Great Plains.

Vegetation and soils were sampled 1 month prior to and 1 year after a pipeline was constructed in a Texas coastal marsh. Submerged aquatic (SAV) and emergent vegetation (EV) were sampled to detect changes in taxa frequencies and percent cover within 3 pipeline corridor treatments (soil deposit/borrow, pipeline ditch, construction equipment) and a control. Taxon richness was not significantly altered by pipeline construction within EV plots. However, pipeline construction decreased total vegetative coverage of EV plots within all 3 pipeline treatments. A 33% (2.3 ha) decrease in EV coverage was calculated within a 30.4-m strip along the pipeline ditch using pre- and post-construction aerial photographs and Geographic Information System (GIS) technology. Similarly, data from EV quadrats indicated a 49% (2.2 ha) loss within the 19.8-m wide construction corridor. No analyses were performed on SAV data because these plants were only present during pre-construction sampling.

Habitat fragmentation is one of the greatest threats to the conservation of biodiversity and has 3 components: habitat loss, patch isolation, and patch size. We tested the effects of forest-clearing size on small mammal populations in the Upper Coastal Plain of South Carolina. These clearings act as islands for many species of small mammals, particularly old-field mice (Peromyscus polionotus) and cotton rats (Sigmodon hispidus). We live-trapped small mammals in 3 size classes of clearcuts (small <6 ha, medium 10-15 ha, and large >25 ha) on the Savannah River Site and compared relative density and diversity of small mammal species among the patch size classes. Peromyscus polionotus were captured on all grids and were the only species captured on small grids; P. polionotus relative density did not vary significantly among patch sizes. In contrast, S.