Journal of the Southeastern Association of Fish and Wildlife Agencies

We examined food habits of bowfin (Amia calva), in the Black and Lumber rivers, North Carolina. Stomachs from 192 Black River bowfin and 175 Lumber River bowfin were analyzed to determine frequency of occurrence and percent by number and weight of individual food items consumed from 1994 to 1997. Crustaceans [primarily crayfish (Astacidae) and grass shrimp (Palaemonidae)] were the dominant food item consumed by bowfin in both rivers. They occurred, on average, in 79% of Black River bowfin stomachs containing food and accounted for 65% by number and 53% by weight of all food items consumed. In the Lumber River crayfish occurred, on average, in 71 % of bowfin stomachs containing food and accounted for 63% by number and 27% by weight of all food items consumed. Fishes, primarily centrarchids and ictalurids, were of secondary importance in the diet of bowfin.

Food of bluegills (Lepomis macorchirus), redear sunfish (L. microlophus), and largemouth bass (Micropterus salmoides) from Belews Lake was examined during their recovery from selenium (Se) contamination to evaluate the influence of diet on the continued bioaccumulation of Se in their skeletal muscle tissue. The redear sunfish and largemouth bass consumed food items that originated primarily from Belews Lake while bluegills consumed mostly terrestrial insects. Food items originating from the lake generally exhibited higher concentrations of Se and resulted in higher concentrations of Se in the skeletal muscle tissue of the redear sunfish and largemouth bass when compared to bluegills.

In October 1997, apparent spatial heterogeneity in the black crappie (Pomoxis nigromaculatus) age distribution was observed in B.E. Jordan Reservoir. The number of age groups for black crappie increased from downlake to uplake. The objective of this study was to verify this spatial heterogeneity. Trap nets were used to collect black crappie in 4 sections of the reservoir. Black crappie ages ranged from 1 to 17 years, and fish ≥10 years were found only in the upper 2 sections of the reservoir. Significant (P<0.05) differences in the black crappie age distribution were found between most reservoir sections in April 1998. However, only the lower most section and upper most section were significantly different in November 1998. There were no differences between sections in April 1999.

Contest angler exploitation rates for crappie (Pomoxis spp.) ≥254 mm total length (TL) were estimated for Lake Texoma, Texas-Oklahoma, from tags returned by anglers fishing in Crappiethon USA contests in 1994 and 1995. Unadjusted contest angler exploitation was 27% in 1994 and 23% in 1995. Exploitation rates were adjusted for handling and tagging mortality and tag retention. Non-reporting was assumed to be 0%. Handling and tagging mortality rate estimates were 22% in 1994 and 10% in 1995. Tag loss for both years was calculated at 14%. Adjusted contest angler exploitation rate of crappie ≥254 mm TL during Crappiethon was 41% in 1994 and 30% in 1995.

Crappie Pomoxis spp. population structure on Arbuckle Reservoir was monitored from 1984-1997 using fall trap-net samples. A creel survey was conducted from 1985-1995. A 254-mm length limit/15-fish daily creel regulation was placed on the lake 1 January 1993. Prior to the length limit, the population was characterized as fast-growing and short-lived. Angler harvest steadily increased after the length limit and in the final year of the creel survey, yield (kg/ha) was higher than in 5 of the 8 years data were collected prior to the length limit. Angler acceptance of the length limit was high. Growth rates remained unchanged following implementation of the length limit. Although no differences in mortality rates could be demonstrated using the trap-net data, size and age structure of the crappie population improved. Based on the data collected, the length limit regulation met the objectives as stated.

Streams and rivers in the Blue Ridge Mountains of Virginia provide an excellent cold water resource and have historically supported exceptional wild trout populations. In June 1995, a flood of greater than 500-year recurrence interval created a unique opportunity to assess the impact on trout populations within 3 rivers of the Shenandoah National Park (SNP). Debris flows impacted the lower one- to two-thirds of the Rapidan, Staunton, and North Fork Moormans rivers, either extirpating or greatly depressing trout populations. The number of trout collected in debris flow areas were significantly reduced (P≤0.05) in 1996, 1997, and for the post-flood 3-year mean when compared to pre-flood means. Trout populations in the flooded headwater reaches of all 3 rivers were not reduced. Debris flows, in association with severe flooding, greatly depress and even extirpate native brook trout (Salvelinus fontinalis) populations, but flooding alone may have little effect on populations.

Growth and condition of rainbow trout and brown trout in the Caney Fork River below Center Hill Dam in middle Tennessee were investigated for 1 year. Rainbow trout stocked in that spring grew faster (13 mm and 20 g/month) than rainbow trout stocked in June and August 1997. Brown trout grew slower in summer and fall (8 mm and 10 g/month) in summer and fall and faster (17 mm and 61 g/month) in winter. Adjusted mean weight of brown trout dropped from 141 g in May 1997 to 113 g in October 1997. Rainbow trout stocked in summer 1997 lost 14% of their body weight by fall 1997. Condition of both species improved in late fall with a corresponding improvement in tailwater dissolved oxygen concentration.

We used a multiple linear regression approach to develop models predicting water, protein, and lipid content of bluegills (Lepomis macrochirus) under 4 measurement approaches varying in terms of time and money. Inputs were length, weight, relative weight, total body electrical conductivity, and water. Models predicting water and protein weights were very accurate (<5% mean error). No regression predicting lipid weight was accurate enough to be used as a predictor (>37% mean error). We then attempted to reduce inaccuracy by standardizing lipid weight 4 ways. No standardization substantially improved predictive accuracy (>30% mean error). However, our results suggest that increasing the range of values used to fit the regressions may increase precision and accuracy of prediction.

Fishery managers routinely collect scales from black basses (Micropterus spp.) for age determination; however, the validity of these ages is often unknown. The objectives of this study were to evaluate the accuracy and precision of scales and whole otoliths relative to sectioned otoliths for determining ages of largemouth bass (M. salmoides) in North Carolina. Scales and sagittal otoliths were collected from largemouth bass during spring and summer of 1997 from the Chowan river, Neuse River, B. E. Jordan Reservoir, W. Kerr Scott Reservoir, Hyco Reservoir, Tillery Reservoir, and Santeelah Reservoir. Sample sizes varied between 94 and 149 largemouth bass per water body and were pooled for data analysis. Reader-derived ages, mean total length at age capture (TL, mm), and between reader precision for scales and whole otoliths were compared against otolith sections. For comparative purposes in this study, ages determined from otolith sections were considered to be the correct ages.

We surveyed black bass (Micropterus spp.) populations in Baron Fork in northeastern Oklahoma and Glover River in southeastern Oklahoma during 1994 and 1995 to assess population characteristics and management options particularly for smallmouth bass (M. dolomieu) in eastern Oklahoma streams. Smallmouth bass dominated the catch of black bass species in both streams during both years. The fishery potential for smallmouth bass in Glover River was limited in part by low abundance, poor year-class success, and high annual mortality resulting in low recruitment to older ages. In contrast, smallmouth bass in Baron Fork were abundant, exhibited good year-class success, and low annual mortality. Growth of early-age smallmouth bass was similar between streams.

Largemouth bass (Micropterus salmoides) were tagged and released following a series of weekly tournaments at Lake Thunderbird, Oklahoma, in 1993. Forty-nine percent of the 176 fish recaptured in 1993 were caught within 0.8 km of the release sight and 64% were caught within a 1.6-km radius. Dispersal during the second year was slightly greater with 35% of the 56 recaptured bass coming from within 0.8 km and 46% from 1.6 km of the release site. The maximum distance from the release site that a tagged fish was recaptured was 12 km with a median distance of 1.6 km after 18 months. Natural dispersal and the use of live-release boats to redistribute tournament-caught bass during the second year of the study reduced the number of tagged fish that were recaptured; however, the harvest rate of tagged bass caught by non-tournament anglers was similar between 1993 (68%) and 1994 (67%).

Angler exploitation, sex and size selection, temporal and spatial capture patterns, and angling mortality were determined in 1996 and 1997 for largemouth bass (Micropterus salmoides), spotted bass (M. punctulatus) in Norris Reservoir, Tennessee. Adjusted annual angler exploitation was 20% (± 4) for 1996-tagged largemouth bass and 14% (±4) for 1997-tagged largemouth bass. Adjusted annual angler catch was 47% (±8) for 1996-tagged fish and 34% (±7) for 1997-tagged fish. No significant differences (P>0.05) were detected between years for either exploitation or catch. No significant differences (P>0.10) by sex or size were detected for largemouth bass in either year. Anglers caught the majority of largemouth bass in the spring and 86% were captured within the embayment of tagging. Total angling mortality was estimated at 23% (± 1) in 1996 and 16% (± 1) in 1997 for largemouth bass.

From 1987 to 1998, gill netting was conducted annually in Jordan Lake, North Carolina, usually 3 times per year with 9 nets distributed over 3 reservoir sub-basins. White perch (Morone americana) began appearing in gill-net samples in 1988. By 1993 they were the second-most abundant species (11.2/net night), following black crappie (Pomoxis nigromaculatus) (21.2/net night). White perch catch rates in 1994 (16. I/net night) exceeded rates for other fish captured in gill nets. White perch continued to constitute a large percentage (21.1%) of the gill-net catch from 1995-1998. Analysis of variance of log-transformed white perch catch data detected significant year effects (P=0.0001). No significant season or sub-basin effects were detected. Neither interaction between sub-basin and year nor interaction between sub-basin and season was significant. The 3-way interaction between year, sub-basin, and season was not significant.

Striped bass have been stocked into the Upper Coosa River Basin (Alabama and Georgia) since the 1970s. The increased presence of small striped bass within these waters in recent years suggested the possibility of a resident spawning population. Ichthyoplankton samples were collected once or twice per week from April to June 1997 and 1998 from sample sites in the tributaries of the Coosa River above Weiss Reservoir. Fertilized striped bass eggs were collected from the Oostanaula River during both years and the Conasauga River in 1998 (the only year sampled). The peak spawning activity for both years occurred in mid-May, when the water temperature reached approximately 18 C. The peak density was lower in 1997 than in 1998, at 4.7 and 77.1 eggs/m3, respectively. The majority of the eggs collected were spawned less than 20 hours prior to harvest, indicating that much of the spawning was occurring within 76 km of Weiss Reservoir.

Buffalo Springs Reservoir is highly eutrophic with a dense population of large gizzard shad (Dorosoma cepedianum) and limited production of young gizzard shad. Fingerling (40 mm total length) striped bass (Morone saxatilis) were stocked into Buffalo Springs Reservoir in 1992 and 1993 at a rate of 550/ha to restructure the gizzard shad population. Three to 4 years after these stockings, density of large gizzard shad declined and a strong gizzard shad year class was produced. White crappie (Pomoxis annularis) recruitment paralleled gizzard shad recruitment. White crappie growth declined during 1991-1995 when age-0 gizzard shad were scarce or absent but increased in 1997, presumably a result of the abundant forage provided by the 1996 gizzard shad year class. No clear effects on largemouth bass (Micropterus salmoides) or bluegill (Lepomis macrochirus) growth were apparent, but relative abundance of both species increased.

Striped bass (Morone saxatilis) stocking rates were increased in 1988 and more restrictive harvest regulations were initiated in 1992 to increase striped bass abundance in Lake Normal, North Carolina. The objective of this study was to make inferences about current forage sufficiency for striped bass from length at age of capture and body condition of striped bass collected since 1994. Fish were obtained from November-January fishing tournaments and gillnet sets in 1994-1998. Late season samples were designed to collect fish after they had several months to recover from any reduced condition caused by summer temperature and dissolved oxygen related habitat stress. Striped bass were also collected with gill nets in late June or early July 1996-1998 to gather pre-summer habitat stress body condition information. Growth for fish older than age 2 had declined from pre-study data collected by Duke Power Company.

The use of infrared-triggered activity monitors in wildlife science has increased during recent years. In most cases, trail monitors are mounted to stationary objects. However, trail monitors in aquatic environments where change in elevation is necessary, because of varying water levels, curently restricts their use. In our study of Florida Key deer, we needed information on deer activities in areas influenced by tides. We developed an inexpensive device that allowed trail monitors to operate as tides changed. We propose that this device can expand the use of trail monitors to include aquatic environments.

Bobcats (Lynx rufus) and coyotes (Canis latrans) are sympatric in many areas; however, this sympatry has evolved relatively recently in the southeastern United States with coyote range expansion. Where the 2 species are sympatric, habitat selection and diets of bobcats and coyotes may overlap. Knowledge of seasonal variation in prey selection is required to assess interspecific competition and understand factors facilitating coexistence between sympatric species, yet long-term (>5 years) information on sympatric diets is unavailable. We collected and analyzed 1,183 scats (591 bobcat, 592 coyote) from 1991-1997 in central Mississippi. Diet was assessed using frequency information and frequency-based correction factors to determine seasonal prey consumption. Coyote diets were dominated by white-tailed deer (Odocoileus virginianus), rabbits (Sylvilagus spp.), and fruits; whereas, bobcats consumed primarily rabbits and rodents.

To facilitate recovery of the endangered red-cockaded woodpecker (RCW), management priorities and practices on >800,000 ha of Forest Service land in the Southeastern United States have been realigned to restore the pine-grassland system to which the RCW is adapted. Management regimes for RCW involve fire and mechanical suppression of hardwood understory and midstory. As such, RCW management practices might be expected to alter plant communities and associated wildlife populations. We examined differences in vegetation composition and structure between mature pine stands managed for red-cockaded woodpeckers and similarly-aged stands not managed specifically for RCWs on 2 national forests in southern Mississippi. During the growing seasons of 1997 and 1998, 123 vegetation plots were sampled on both study areas using nested circular plots. Pine sawtimber stands under RCW management differed in structure and composition from those under traditional U.S.

Schlumburger Geco-Prakla (SGP) conducted seismic operations in St. Mary and Terrebonne parishes of Louisiana, between October 1997 and March 1998. This area is located on the marsh-swamp interface and contains approximately 40% of the state's nesting bald eagles (Haliaeetus leucocephalus). The Louisiana Department of Wildlife and Fisheries and the U.S. Fish and Wildlife Service used spatial buffer zones with a 460-m radius to protect all eagle nests located on properties of the SGP seismic operations. Bald eagle production during seismic operations (1998) was compared to the previous season (1997) and the following season (1999). Production did not differ (P=0.85) for the interaction between year and area (seismic affected nests/non-seismic affected nests). Experimental airboat disturbances were used in the 1998 and 1999 nesting seasons to evaluate flush response at distances closer than 460 m. Flush response increased as the distance from an airboat to a nest tree decreased.

A detailed understanding of the spatial arrangement of northern bobwhite (Colinus virginianus) habitats would allow more focused efforts by wildlife managers. We used a 4-year average of northern bobwhite call-count data in conjunction with remotely sensed habitat maps to study landscape-level habitat associations. Landscape metrics were calculated for the landscape surrounding each stop and were used in 2 modeling exercises to differentiate between high and low northern bobwhite populations. Both pattern recognition (PATREC) and logistic regression models predicted levels of northern bobwhite abundance well for the modeled (73.5% and 73.9%, respectively) and independent (74.6% and 76.6%, respectively) data sets. The revised models were applied to the remotely sensed habitat maps of the eastern 2/3 of Virginia to develop maps expressing the quality of a landscape for supporting a high population of bobwhite based on existing land cover.

Multi-year studies in wildlife science and management can provide novel insights not detected in short-term investigations. Therefore, we continued a 2-year study by Stephens et al. (1998) to evaluate wood duck (Aix sponsa) reproduction in conventional and small nest boxes (i.e., approximately one-half conventional size) at Noxubee and Yazoo National Wildlife Refuges (NWRs) in Mississippi. Small nest boxes were designed to deter excessive dump nesting by wood ducks at these refuges. During 1994-1997, use of large boxes by wood ducks remained high (≥70%) at both study areas, but use of small boxes declined from 61% in 1994 to 34% in 1997 at Noxubee NWR. Concomitantly, use of small boxes by passerine birds increased from 14% to 65% at Noxubee NWR, but use of large boxes by passerines never exceeded 15%. Large boxes never were used by passerines at Yazoo NWR.

Tall fescue (Festuca arundinacea) grasslands do not provide quality habitat for northern bobwhite quail (Colinus virginianus). Converting tall fescue fields to other cool-season grasses and legumes (CSG) may provide more favorable wildlife habitat. We evaluated the effectiveness of spring and fall pre-emergence Round-Up PRO” applications with and without prescribed burns for eradicating tall fescue and establishing CSGs. Two tall fescue fields were located in western Kentucky. Prescribed burns, herbicide applications, and CSG seedings were applied in 2 0.05-ha treatment plots during spring and fall 1996 and spring 1997 at each site. Mean planted CSG cover at both sites was higher (P<0.05) in treatments with fall plantings (range=12.0%-124.5%) compared to treatments with spring plantings (range=0.0%-18.1%) during the first growing season.

Effect of exploitation on sympatric southern squirrel populations has not been documented. Additionally, opportunities to study effects of harvest on unexploited populations are rare. Consequently, we investigated the effect of exploitation on a previously unharvested population of fox (Sciurus niger) and gray squirrels (S. carolinensis) on Twin Oaks Wildlife Management Area, Mississippi, from 1993-1998. We annually collected age, body morphometrics, color, sex, and species composition of harvested squirrels at voluntary check stations. A population decrease was documented for both species with stabilization occurring at different times (P≤0.001). Fox squirrel age ratio changed following exploitation (P=0.031), but color and sex ratios did not (P>0.05). Gray squirrels exhibited no change in age, color, or sex ratios following exploitation (P>0.05). Body morpho-metrics of both species were affected by exploitation (P≤0.003) and demonstrated a density dependent response.

Scent-station surveys have been used to examine trends in felid and canid abundance throughout the Southeast. Scent station methods have been developed on the presumption that canids rely on olfactory stimuli and that bobcats (Lynx rufus) rely on sight and auditory stimuli. No studies have quantified the effects of various olfactory, auditory, and visual stimuli on scent-station visitation by bobcats and canids; however, such information could improve the effectiveness and ability of scent-station surveys. We established scent stations and track transects in intensively managed pine forests in east-central Mississippi from August 1989-May 1991 to evaluate the effectiveness of 4 attractants for eliciting response from 3 carnivores. We randomly allocated synthetic fatty acid scent, bobcat urine, an auditory stimulus, a visual stimulus, and a control to stations at monthly intervals.