Comparison of Abundance Indexes and Population Estimates for Wild Turkey Gobblers

Comparison of Abundance Indexes and Population Estimates for Wild Turkey Gobblers Author(s): John R. Lint, Bruce D. Leopold and George A. Hurst Source: Wildlife Society Bulletin (1973-2006) , Summer, 1995 , Vol. 23, No. 2 (Summer, 1995), pp. 164-168 Published by: Wiley on behalf of the Wildlife Society Stable URL: https://www.jstor.org/stable/3782783 REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/3782783?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms Wiley and Wildlife Society are collaborating with JSTOR to digitize, preserve and extend access to Wildlife Society Bulletin (1973-2006) This content downloaded from � � � � � � � � � � � � 128.227.206.28 on Fri, 13 Dec 2024 19:43:52 UTC�� All use subject to https://about.jstor.org/terms Wild turkey indexes Comparison of abundance indexes and population estimates for wild turkey gobblers John R. Lint, Bruce D. Leopold, and George A. Hurst Abstract Numerous indexes exist for wild turkeys (Meleagris gallopavo); however, few have been evaluated. We compared 3 indexes of gobbler abundance (harvested gobblers, har- vest/effort, and gobblers heard/day) and measured their relation to population estimates derived using a modified Jolly-Seber capture-recapture model. Harvested gobblers and harvest/effort were useful indexes of gobbler population estimates in our study (P < 0.1 0). We recommend that harvested gobblers and harvest/effort be used when examining gob- bler population trends. Gobblers heard/day (call counts) was related to harvested gob- blers and harvest/effort (P< 0.01). However, call counts possibly followed trends in num- bers and gobbler physical condition. We suggest these indexes be tested for applicability in other areas to verify or contrast our results. Key words abundance, call counts, capture-recapture, harvest, hunter effort, indexes, Meleagris gal- lopavo, trends, wild turkey gobblers A problem facing wildlife managers is measuring population sizes and trends. Often an exact popula- tion value may be unnecessary (Begon 1979:4), as in- dexes of trends are more feasible and usually provide adequate information to manage wildlife populations (Caughley 1977:14). Although gobbling call counts (Scott and Boeker 1972, Bevill 1975, Porter and Ludwig 1980) and harvest trends (DeGraff and Austin 1975, Lewis 1975, Menzell 1975, Glidden 1980) pro- vide indexes, relationships of indexes to gobbler population size or trends have not been determined. Few studies have validated the relationship between animal population size and commonly used indexes (Rotella and Ratti 1986), primarily because few stud- ies monitored population size and indexes over enough time (e.g., number of years). Wild turkey gobblers (Meleagris gallopavo) are in high demand by sportsmen in the spring gobbler hunting season. In Mississippi, hunting increased from 8,694 hunter-days in 1951 (Redmond 1952) to 381,077 in 1987 (Steffen 1987). Williams and Austin (1988) stated that an index that incorporated hunting effort was needed to compare gobbler harvest on dif- ferent areas. An accurate, standardized index of rela- tive abundance of wild turkey gobblers would facili- tate monitoring gobbler numbers and trends. Our 9 year study of wild turkey population dynamics pro- vided the framework to validate usefulness of gob- bler population indexes to monitor trends in wild turkey gobbler population. Ideally, indexes of population size should be com- pared to a known population (Davis and Winstead 1980:244). Seber (1982:560) stated that a common practice in capture-recapture evaluation was to obtain independent and unbiased estimates from harvest data for comparison. We sought to mitigate the problem of unknown population size by estimating population size with an open capture-recapture model. We com- pared our estimates to and examined similarities among 3 indexes: (1) total spring gobbler harvest, (2) spring gobbler harvest/hunter effort (harvested gob- blers/100 hunter-days), and (3) gobblers heard/day. All authors were with the Department of Wildlife and Fisheries, Box 9690, Mississippi State, MS 39762, USA. Present address for John R. Lint: U.S. Department of Agriculture Forest Service, Rt. 5 Box 1 57, Andalusia, AL 36420, USA. Wildlife Society Bulletin 1995, 23(2):164-168 Peer refereed This content downloaded from � � � � � � � � � � � � 128.227.206.28 on Fri, 13 Dec 2024 19:43:52 UTC�� All use subject to https://about.jstor.org/terms Wild turkey indexes * Lint et a!. 165 Study area The study area was a 14,140-ha tract in Tallahala Wildlife Management Area (WMA), Bienville National Forest, 16 km southeast of Newton, Mississippi. Topography was gently to moderately rolling with slopes from 0-15%. Mature pine (Pinus spp.) stands, pine-hardwood stands, and pine regeneration areas comprised 67% of the area. Loblolly pine (P. taeda) was the dominant species. Bottomland hardwood stands and hardwood regeneration areas comprised the remaining 33% of the area. Methods Capture We captured 271 gobblers using cannon nets or bait treated with alpha-chloralose from 7 January- 4 March (winter capture period) and I July-25 August (summer capture period) 1983-1992. Thirty-eight bait sites were maintained on gated Forest Service roads and in wildlife openings. Gobblers were double marked with numbered leg bands ad numbered pat al wng tags (cattle ear tags; Knowlton et al. 1964) and 135 re- ceived radio-transmitters. Gobblers were released at their capture site. Animal capture and handling was per- formed under Mississippi State University Institutional Animal Care and Use Committee No. 93-030. The Buckland (1980) modified Jolly-Seber capture-recapture model (jolly 1965, Seber 1965) was used to estimate gobbler abundance in an open popu- lation. Buckland (1980) offered an extension of the general jolly-Seber model where more data concerning marked individuals were used by including known deaths as part of the recapture sample in analyses. Many gobblers were known to die from telemetry or harvest, and these deaths were important for our model. Sample sizes increased, standard errors were reduced, and better estimates of population size were possible with the Buckland model (Lint 1990). Buckland (1980) found that his model had lower stan- dard errors that were not proportional to estimates of population size, and, when capture probability was low, estimates were less biased than Jolly-Seber. Additionally, Lint (1990) found that the 6 assumptions of the Buckland (1980) estimator were valid for wild turkey gobblers on Tallahala WMA. Although estimates of gobbler population size were derived from winter and summer capture periods, only estimates from win- ter capture periods were used in comparisons with the 3 indexes. Harvest and harvest/effort Hunters were required to report all harvested gob- blers at Tallahala WMA headquarters and to complete and return permit cards to 1 of 6 self-service stations located at entrances to Tallahala WMA. Hunter effort (hunter-days) was calculated from permit card counts. Hunter compliance was estimated at 95% for reporting harvested gobblers (Gribben 1986) and 85% for returning cards (Palmer et al. 1990) on Tallahala WMA. Harvested gobblers and harvest/ef- fort (harvested gobblers/hunter effort x 100) were used as indexes to gobbler population size. Gobbling call counts Gobbling call counts were conducted Monday, Wednesday, and Friday from 10 March-7 May, 1984-1992 (from 1 week before to I week after the spring gobbler hunting season). Two routes com- posed of 8 and 10 call-count stations, respectively, were sampled simultaneously from 30 minutes be- fore to 30 minutes after sunrise (Bevill 1975). Call- count stations were located about 1.6 km apart on Forest Service roads throughout the study area. Observers listened 4 minutes for gobbling at each sta- tion. Call counts were postponed during adverse weather (e.g., wind, rain) until the next possible day. This content downloaded from � � � � � � � � � � � � 128.227.206.28 on Fri, 13 Dec 2024 19:43:52 UTC�� All use subject to https://about.jstor.org/terms 166 Wildlife Society Bulletin 1995, 23(2):164-168 Average number of gobblers heard/day for each year was calculated as a relative abundance index. Data analysis Each index was compared to population size esti- mates using linear regression (Steel and Torrie 1980:239-243). Correlation matrix analysis (Kleinbaum and Kupper 1978:158-159) was used to examine relationships among the 3 indexes. An index was evaluated according to: (1) relation to gobbler population size estimates, (2) reflection of observed population levels and changes, (3) correla- tion to other indexes, (4) sources of data different from their respective correlates, and (5) practicality of data collection and analysis. Tests for significance were made at a = 0.10. Results Of 271 gobblers captured, 104 were recaptured, 9 were killed while trapping on initial capture and 1 on recapture, and 135 were returned through the har- vest. Thirty gobblers were found dead through telemetry or by hunters. Hunters reported 317 har- vested gobblers (tagged and untagged). Call counts were conducted on 189 days, and 527 gobblers were heard. Buckland estimates exhibited decreasing trend (r = -0.76, P = 0.018), averaged 82 gobblers over the 9- year study, and were estimated precisely (Table 1). Harvest/effort (r = 0.66, P = 0.053) and harvested gobblers (r = 0.63, P = 0.072) were related to esti- mates from the Buckland model, but gobblers heard/day was not (r = 0.53, P = 0.145). Harvest/ effort (r = -0.84, P = 0.005), harvested gobblers (r = -0.89, P = 0.001), and gobblers heard/day (r = -0.76, P = 0.017) all declined during the study (Table 1). Hunter effort did not decline over time (r = -0.50, P = 0.174). Harvested gobblers and harvest/effort were highly correlated (r = 0.94, P < 0.001). However, harvested gobblers and harvest/effort shared the same data, so a close relation was expected. Gobblers heard/day was correlated with harvested gobblers (r = 0.88, P = 0.002) and harvest/effort (r = 0.83, P = 0.006). Because harvested gobblers and harvest/effort were counts/year, standard errors were not available for within-year estimates. Discussion The turkey population on Tallahala WMA was be- lieved to have decreased during the study due to ad- verse environmental conditions (Lint 1990). Telemetry and population research conducted on the area simultaneously with this study (Seiss 1989, Lint 1990, Palmer 1990) indicated that bait-site use de- creased, nesting and reproduction decreased, preda- tion and disease increased, weights of harvested gob- blers declined, and average date for initiating incubation by hens was delayed because of drought and poor mast production in 1987-1988. Gobbler population trends matched total wild turkey popula- tion changes observed. We studied gobblers because they were readily measured using call counts and spring gobblers-only harvest, and gobbler captures were more consistent among seasons than hen cap- tures (Lint 1990). Although we did not know the exact number of gobblers on Tallahala WMA, we used the Buckland Table 1. Estimates and indexes of population size for wild turkey gobblers on Tallahala Wildlife Management Ar 1984-1992. Buckland model Gobblers heard/day Harvested Harvest/ Year N (SE)a gobblers 100 hunter-days x (SE)b n' Days 1984 123 (22) 59 12.4 5.7(1.6) 68 12 1 985 78 (1 2) 49 11.1 4.0 (0.8) 83 21 1986 98 (1 2) 50 10.1 6.9 (1.2) 1 25 1 8 1987 78 (20) 53 8.9 4.3 (0.7) 90 21 1988 102 (24) 21 5.2 0.9 (0.3) 20 23 1989 62 (1 2) 20 3.4 1.4 (0.4) 33 23 1990 78 (56) 28 6.7 1.2 (0.3) 28 24 1991 72 (29) 21 6.5 1.3 (0.4) 30 23 1992 49 (26) 16 4.6 2.1 (0.4) 50 24 a Estimated population size (standar - mean number of gobblers heard c n number of gobblers heard. This content downloaded from � � � � � � � � � � � � 128.227.206.28 on Fri, 13 Dec 2024 19:43:52 UTC�� All use subject to https://about.jstor.org/terms Wild turkey indexes * Lint et al. 167 capture-recapture model to estimate gobbler popula- tion size. The Buckland population estimates were used as the basis for index comparisons because they best reflected declining population trends, used the most biological information available on the gobblers which increased sample sizes by including known deaths, and had small standard errors. Environmental factors possibly affected physiolog- ical condition and, subsequently, population size, and may have contributed to reduced gobbling and lower gobbler density. Call counts possibly were af- fected by changes in physical condition as well as changes in gobbler numbers. Gobblers heard/day was not related (P = 0.14) to estimates of gobbler population size but was strongly correlated to harvest and harvest/effort. Harvested gobblers and harvest/effort were related to Buckland estimates. Harvest was an important source of information for this study; it provided a use- ful index of gobbler population trends and con- tributed much of the data on known deaths needed for the Buckland model. Godwin et al. (1991) stud- ied gobbler survival rates on Tallahala WMA and re- ported that 93% of gobbler mortality occurred during the spring harvest. Bailey (1980) stated that harvest data probably indicated trends better than popula- tion estimates. Incorporating hunter effort to mea- sure harvest/effort helped confirm total harvest. Catch/unit effort has been a useful index when capture (harvest) does not greatly influence popula- tion size and when it cross-checks other estimates of relative abundance (Caughley 1977:17). Because hunter effort did not decline during the study and because number of harvested gobblers was related to population size, we believe number of harvested gob- blers properly indexed gobbler population size. Management implications In Mississippi, harvest data from mail surveys is used (Steffen 1987) to make assumptions on gobbler populations. Many wildlife management areas may already be collecting data needed for the harvested- gobblers index. Managers can compile existing hunter effort information to examine possible trends and relations between harvested gobblers and hunter effort for their gobbler population. Indexes that are most useful are directly related to population size, use maximum amounts of available information, are easy for managers to obtain, and are easy to calculate. Based on cost of gathering neces- sary data, number of harvested gobblers was the least expensive index, followed by harvest/effort, gobblers heard/day, and capture-recapture. Although opera- tion of a check station was time consuming, harvest data were relatively easy and inexpensive to collect. Permit stations were maintained to gather data on hunter effort. A mandatory reporting policy was initi- ated for this study, but a permit card system or mail survey may provide necessary data on number har- vested where personnel are limited. However, re- quiring harvest reporting is recommended to better enumerate gobblers harvested, gather data on hunters and harvests, increase communication between wildlife agency personnel and the public, and allow hunters to become involved in management activities. Call counts required relatively intense effort for data collection, an adequate road system, and 2 peo- ple for approximately 24 mornings each spring har- vest season to collect data. Call counts might be use- ful in areas where there is limited hunting (low annual harvest, low hunting pressure). Acknowledgments. This is a contribution from the Mississippi Cooperative Wild Turkey Research Project that is supported by the Mississippi Department of Wildlife, Fisheries and Parks (Federal Aid in Restoration, Project W-48), National Wild Turkey Federation, U.S. Department of Agriculture Forest Service, National Forests in Mississippi, and Mississippi Agricultural and Forestry Experiment Station. We thank W. Hamrick, P. Phalen, K. Gribben, R. Kelley, R. Seiss, B. Palmer, L. Stacey, K. Godwin, K. Sullivan, S. Priest, and R. Flynt for help in data collection. Helpful editorial comments were provided by D. Steffen. Literature cited BAILEY, R. W. 1980. The wild turkey status and outlook in 1979. Proc. Natl. Wild Turkey Symp. 4:1-9. BEGON, M. 1979. Investigating animal abundance: capture-recap- ture for biologists. University Park Press, Baltimore, Md. 97pp. BEVILL, W. V. 1975. Setting spring gobbler hunting seasons by tim- ing the peak gobbling. Proc. Natl. Wild Turkey Symp. 3:198-204. BUCKLAND, S. T. 1980. A modified analysis of the Jolly-Seber cap- ture-recapture model. Biometrics 36:419-435. CAUGHLEY, G. 1977. Analysis of vertebrate populations. John Wiley and Sons, Inc., New York, N.Y. 232pp. DAVIS, D. E., AND R. L. WINSTEAD. 1980. Pages in S. D. Schemnitz, ed. Wildlife Management Techniques Manual. Fourth ed. The Wildl. Soc., Washington, D.C. 686pp. DEGRAFF, L. W., AND D. E. AUSTIN. 1975. Turkey harvest manage- ment in New York. Proc. Natl. Wild Turkey Symp. 3:191-197. GLIDDEN, J. W. 1980. An examination of fall wild turkey hunting statistics from different ecological areas of southwestern New York. Proc. Natl. Wild Turkey Symp. 4:76-85. GODWIN, K. D., G. A. HURST, AND R. L. KELLEY. 1991. Survival rates of radio-equipped wild turkey gobblers in east-central Mississippi. Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies 45:218-226. This content downloaded from � � � � � � � � � � � � 128.227.206.28 on Fri, 13 Dec 2024 19:43:52 UTC�� All use subject to https://about.jstor.org/terms 168 Wildlife Society Bulletin 1995, 23(2):164-168 GRIBBEN, K. J. 1986. Population estimates for the wild turkey in east-central Mississippi. M.S. Thesis, Mississippi State Univ., Mississippi State. 95pp. JOLLY, G. M. 1965. Explicit estimates from capture-recapture data with both death and immigration-stochastic model. Biometrika 52:225-247. KLEINBAUM, D. G., AND L. L. KUPPER. 1978. Applied regression analy- sis and other multivariable methods. Duxbury Press, Boston, Mass. 556pp. KNOWLTON, F. F., E. D. MICHAEL, AND W. C. GLAZENER. 1964. A mark- ing technique for field recognition of individual turkeys and deer. J. Wildl. Manage. 28:167-170. LEwIs, J. B. 1975. Evaluation of spring turkey seasons in Missouri. Proc. Nat. Wild Turkey Symp. 3:176-183. LINT, J. R. 1990. Assessment of mark-recapture models and in- dexes to estimate population size of wild turkeys on Tallahala Wildlife Management Area. M.S. Thesis, Mississippi State Univ., Mississippi State. 255pp. MENZELL, K. E. 1975. Population and harvest data for Merriams turkeys in Nebraska. Proc. Natd. Wild Turkey Symp. 3:184-189. PALMER, W. E. 1990. The relation of wild turkey hens and their habitat on Tallahala Wildlife Management Area. M.S. Thesis, Missississippi State Univ., Mississippi State. 120pp. PALMER, W. E., G. A. HURST, ANDJ. R. LINT. 1990. Hunter effort, suc- cess, and characteristics on Tallahala Wildlife Management Area. Proc. Nati. Wild Turkey Symp. 6:208-213. PORTER, W. F., AND J. R. LUDWIG. 1980. Use of gobbler counts to monitor the distribution and abundance of wild turkeys. Proc. Natd. Wild Turkey Symp. 4:61-68. REDMOND, H. R. 1952. Squirrel survey and investigation. Fed. Aid Proj. Rep. W-30.1R-3. Jackson, Miss. 148pp. ROTELLA, J. J., ANDJ. T. RBeTI. 1986. Test of a critical density index assumption: a case study with gray partridge. J. Wildl. Manage. 50:532-539. Scort, V. E., AND E. L. BOEKER. 1972. An evaluation of wild turkey call counts in Arizona. J. Wildl. Manage. 36:628-630. SEBER, G. A. F. 1965. A note on the multiple recapture census. Biometrika 52:249-259. SEBER, G. A. F. 1982. The estimation of animal abundance and re- lated parameters. Macmillan Publ. Co., New York, N.Y. 654pp. Soss, R. S. 1989. Reproductive parameters and survival rates of wild turkey hens in east-central Mississippi. M.S. Thesis, Mississippi State Univ., Mississippi State. 99pp. STEEL, R. G. D., ANDJ. H. TORRIE. 1980. Principles and procedures of statistics: a biometrical approach. Second ed. McGraw-Hill Book Co., New York, N.Y. 633pp. STEFFEN, D. E. 1987. Mississippi mail survey of game harvest and hunter effort for 1985-1986. Annu. Rep. Fed. Aid Proj. No. W- 48-33 (ob VI-1). Jackson, Miss. 49pp. WinuAms, L. E., JR., AND D. H. AuSTIN. 1988. Studies of the wild turkey in Florida. Florida Game and Fresh Water Fish Comm. Tech. Bull. No. 10. Univ. Press of Fla., Gainesville. 232pp. 1. Richard Lint (left) is a Wildlife Biologist and Timber Management Assistant for the U.S. Department of Agriculture Forest Service on the Conecuh National Forest in Andalusia, Alabama. He received his M.S. in Wildlife Ecology from Mississippi State University and a B.S. in Wildlife Resources from West Virginia University. Bruce D. Leopold (center) is an Associate Professor of Wildlife Ecology in the Department of Wildlife and Fisheries at Mississippi State University. He received his B.S. in Forest Science from Pennsylvania State University, his M.S. in Forestry from Mississippi State University, and his Ph.D. in Wildlife Ecology from the University of Arizona. His research in- terests include predator-prey ecology and interactions and habitat assessment and management. He currently serves as Associate Editor for the Wildlife Society Bulletin. George A. Hurst (right) is a Professor of Wildlife Ecology in the Department of Wildlife and Fisheries at Mississippi State University. He received his B.S. in Zoology/Wildlife Management and his M.S. in Animal Ecology from North Carolina State University, and his Ph.D. in Zoology and Wildlife Management from Mississippi State University. His research interests include upland gamebird ecology and forest-wildlife management relationships. Received 15July 1993. Accepted 29 August 1994. Associate Editor: Rotella This content downloaded from � � � � � � � � � � � � 128.227.206.28 on Fri, 13 Dec 2024 19:43:52 UTC�� All use subject to https://about.jstor.org/terms