USE OF LOBLOLLY PINE PLANTATIONS BY WILD TURKEY HENS IN EAST-CENTRAL MISSISSIPPI DAVID R. SMITH, Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, MS 39762 GEORGE A. HURST, Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, MS 39762 JOHN D. BURK, Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, MS 39762 BRUCE D. LEOPOLD, Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, MS 39762 M. ANTHONY MELCHIORS, Weyerhaeuser Company, P. 0. Box 1060, Hot Springs, AR 71902 Abstract: Conversion of mature pine-hardwood forests to intensively managed loblolly pine (Pinus taeda) plantations has caused concern that wild turkeys (Meleagris gallopavo) would not use the plantations. We determined habitat use by 55 wild turkey hens by telemetry from March 1986 through February 1988 in Kemper County, Miss. The study area (20,200 ha) consisted of pine plantations (45% ), mixed forest (27% ), hardwood forest (16% ), and nonforest (11%) habitats. Most (69%) plantations were midrotation-aged (13-19 years old). Average seasonal home ranges (minimum convex) varied from 394 to 872 ha. About 50% of telemetry locations were in plantations, and hens used 85% of the plantations in the study area equal to or greater than expected (P < 0.05). Hens (n = 16) that reached the incubation stage nested in plantations (.X = 17 years old) that had been commercially thinned (x = 4 years ago) and control burned (x = 4 years ago). Most (81%) telemetry locations of hens with broods were also in plantations (x = 16 years old) that had been commercially thinned (i = 4 years ago) and burned (.X = 3 years ago). Of 98 field observations of hens with broods, 95% were associated with plantations. Discriminant analysis identified spur roads in or adjacent to plantations as an important factor related to turkey use of plantations. Effects of large scale and intensive pine plantation management on wildlife in the southeastern U.S. have been discussed (McDowell 1954, Stoddard 1963). Several authors expressed concerns about responses of wild turkey populations to large-block, short- rotation, even-aged pine plantation management (Markley 1967, Schaffer and Gwynn 1967, Davis 1976). Only recently has research focused on the importance of plantation management to turkeys (Kennamer et al. 1980, Holbrook et al. 1985, Wigley et al. 1985, Exum et al. 1987). As southern pine-hardwood forests are converted to plantations, it is critical to understand the response of wild turkeys to these habitat changes. To make effective decisions, forest and wildlife managers need information on turkey use of plantations and effects of silvicultural treatments in plantations on turkeys. Objectives of this study were to (1) determine habitat use by turkey hens in intensively managed plantations, (2) determine 61 home ranges of turkey hens in plantations, and (3) describe hen use of plantations that received different silvicultural practices. This paper is a contribution of the Mississippi Cooperative Wild Turkey Research Project, and was supported by Weyerhaeuser Company, National Wild Turkey Federation, Gulf States Paper Company, Georgia-Pacific Corporation, East Mississippi Sportsmen Association, and Mississippi Agricultural and Forestry Experiment Station. STUDY AREA The study area was in east-central (Kemper County) Mississippi in the Interior Flatwoods land resource area (Pettry 1977). Cutting sawtimber from the mature pine-hardwood forests began in 1912 and continued until the late 1930s, when the timber supply was exhausted. Another timber company acquired 33,995 ha of this land in 1941 and used a selective cutting method to regenerate a Proceedings of the Sixth National Wild Turkey Symposium second-growth forest. In 1967, Weyerhaeuser Company purchased these lands and began converting mature pine-hardwood forests to intensively managed loblolly pine plantations (Smith 1988). The original study area (core area) consisted of 9,700 ha of which 66% were in plantations, 21% in mixed forest, 9% in hardwoods, and 4% in nonforest habitats. Average plantation age was 13 years (0-19) and plantation size averaged about 100 ha (65-202 ha). Weyerhaeuser owned 77% of this area. The study area was later expanded to include lands surrounding the core area that radio- equipped turkeys used. The expanded study area (20,200 ha) included Weyerhaeuser lands, privately owned pine plantations, pastures, fields, and a large area of mature hardwood- pine and bottomland hardwood forests, and fields (soybean) along the Sucarnoochee Creek flood plain. Percentages of habitat types of the expanded area were pine plantations (45.2% ), mixed forest (27.4% ), hardwood forest (16.3% ), and nonforest (11.1%). Weyerhaeuser ownership comprised 57% of the expanded area. Strips of mature hardwood forest, streamside management zones (SMZ) of various widths (20-100+ m) occurred within and adjacent to plantations and were scattered throughout the area. Following clearcutting and mechanical site preparation, plantations were established by hand-planting genetically superior loblolly pine seedlings. Hardwood brush control was accomplished by applying herbicides at plantation age 3-4 years, and many plantations were fertilized with urea (181 kg!ha) at age 8-10 years. Most plantations were pre-commercially thinned from an average of 1,483 to 741 trees/ha at age 7-9 years. Commercial thinning from 741 to 445 trees/ha occurred at age 15-16 years. Controlled burning of plantations began at age 9-10 years and should occur at a 3- to 5- year interval. A second commercial thinning is planned and then a final harvest cut will occur at age 25-30 years. Average site index for loblolly pine is 19.8 m (65ft) at age 25 years. Several all-weather roads traversed the area, and Weyerhaeuser gated spur roads extended into most plantations. Spur roads were unimproved and were covered by herbaceous vegetation. Some roadsides were disked in early fall and planted to wheat and rye grass by hunting clubs. 62 METHODS Wild turkeys were captured in the core area by cannon-netting on spur roads in plantations during January-March and July- August 1986 and 1987. Hens were fitted with transmitters ("backpack style") and released at capture sites. We used a hand-held 3-element Yagi antenna and a 1RX-1000S receiver (Wildlife Materials, Inc.) to locate turkeys by triangulation from 116 permanent telemetry stations established on roads throughout the area (Cochran and Lord 1963). To minimize error, we attempted to take bearings from 2 telemetry stations nearest the turkey. The dense road system permitted us to get close; most bearings were taken <0.6 km from the turkeys. Time intervals between consecutive locations were generally <5 minutes and many were between 2 and 3 minutes. Angles <25 degrees and > 155 degrees were generally not accepted. Some angles > 155 were accepted because the turkey was on the edge of a road between 2 stations on the road. Each turkey was located 3 times/day, 3 days/week throughout winter, spring, and summer; and 2 times/day, 3 days/week in the fall. The order in which hens were located was frequently changed and monitoring generally occurred in morning (0700-1000), mid-day (1100-1400), and afternoon (1500-1800). Accuracy tests were performed. Nesting hens were monitored several times each day. Hens with poults were located 3 times/day, morning, mid-day, and afternoon, for 14 consecutive days post-hatch. Ancillary observations of turkeys and associated habitat type( s) were recorded. Weyerhaeuser stand histories and inventory data files were used to characterize its lands into 4 major habitat types: pine plantation, mixed forest, mature hardwood forest, and nonforest. Other lands were classified into the same types based on aerial photographs and ground surveys. A base map containing all stands in the core area and expanded area was digitized from Weyerhaeuser stand maps and aerial photographs. All stands were assigned unique identifiers that corresponded to Weyerhaeuser stand numbers, histories, and conditions. Turkey habitat use and home ranges were analyzed by season: spring (Mar-May), summer (Jun-Aug), fall (Sep-Nov), and winter (Dec- Feb). A 2-sample test for equality of percentages (Zar 1984)) was used to compare (P = 0.05) turkey use of habitats to random expected use of habitats in the expanded study area. A line was drawn around the entire expanded study area to arrive at available habitats. This area included telemetry fiXes and home ranges for all turkeys. Minimum convex polygon method (Mohr 1947) and 80% harmonic mean transformations (Dixon and Chapman 1980) were used to calculate home ranges for turkeys with > 25 locations per season. A discriminant function analysis was used to determine if pine plantations that were used or not used by turkeys had distinctive silvicultural characteristics (Smith 1988). Prior to using discriminant analysis a correlation analysis was performed on all variables to eliminate intercorrelation between variables (Afifi and Clark 1984:246-284). RESULTS We took 7,353 locations on 55 turkey hens during the 24-month period (Mar 1986-Feb 1988). Number of locations and length of time individual turkeys were monitored varied. Telemetry accuracy tests, performed by the 2 observers who collected most (79%) of the data, produced mean estimated error polygons of 0.26 ha. Seasonal home ranges.-- We calculated 106 seasonal home ranges (Table 1 ). Average home ranges calculated by minimum convex method in 1986 ranged from 394 ha in summer to 611 ha Use of Pine Plantations by Wild Turkey Hens • Smith et al. in the fall (Table 2). In 1987 home range size varied from 419 ha in fall to 718 ha in spring. Home range sizes were decreased by over 50% using the harmonic mean method versus the minimum convex polygon method. Nesting effort.-- Four (25%) of the radio- equipped hens reached incubation stage in 1986, and 17 (71%) in 1987. Hatching success was 0% for 1986 and 46% for 1987. All nests were located in pine plantations. All hens except 1 selected 17- to 19-year-old pine plantations that had been commercially thinned and control burned in the past 6 years. One hen nested in a 9-year-old plantation. Brood habitat use and home range.-- Brood habitat information was obtained for 5 hens with broods < 14 days old. Broods used pine plantations (81.2% ), mixed forest (17.5% ), and hardwood forest (1.3% ). Habitat-type use was as expected (P > 0.05). Pine plantations used by hens with broods were 5-19 years old (:X = 16), had been commercially thinned within 0-6 years (x = 4); and had been control burned within 0-6 years (x = 3). We made 98 ancillary observations of broods, hens with poults, during the 2 years. Broods were associated with pine plantations (94.9% ), hardwood forest (4.1% ), and mixed forest (1% ). No broods were seen in nonforested habitats. Brood home ranges (n = 5) during 2 weeks post-hatch ranged from 50 to 169 ha (.X = 102 ha ). Movements from nest sites to brood range averaged <0.5 km. Six plantations used by brooding hens were also used for nesting, although 3 of these were used less than expected for brooding. Table 1. Numbers of radio-equipped wild turkey hens used for habitat and home range analysis in Kemper County, Miss., 1986-88. Turkeys used for Turkeys used for Yeara Season habitat analysis Locations (n) home range analysis 1986 Spring 6 352 5 Summer 8 218 2 Fall 12 420 9 Winter 11 537 8 1987 Spring 21 1,030 18 Summer 23 865 11 Fall 29 1,603 29 Winter 29 1659 24 a1986: March 1986-February 1987; 1987: March 1987-February 1988. 63 Proceedings of the Sixth National Wild Turkey Symposium Table 2. Average seasonal home ranges (ha) for wild turkey hens in Kemper County, Miss., 1986-1988. Minimum convex 80% harmonic Season X SD X SD 1986 Spring 413 316 186 134 Summer 394 221 50 13 Fall 611 282 243 96 Winter 515 287 202 107 1987 Spring 718 1,069 262 307 Summer 872 819 405 425 Fall 419 298 188 109 Winter 470 405 211 143 a1986: March 1986-February 1987; 1987: March 1987-February 1988. Habitat use.--In 1986 hens used pine plantations more than expected during fall and winter, and less than expected in summer (P < 0.05) (Table 3). In 1987, plantations were used more than expected for all seasons except winter, when plantations were used less than expected. Hardwood forests, including SMZs, were used more than expected in all seasons except summer. Mixed forests were used less than expected for all seasons. Nonforest habitats were used less than expected in all seasons except winter when they were used more than expected. In 1986 and 1987 some hens moved from pine plantations in late fall to the Sucarnoochee Creek bottom; they moved back into plantations in early spring. During the 2 winters, 36% (1986) and 56% (1987) of radio- equipped hens were found in the Sucarnoochee Creek area. Pine plantation use and treatments.-- Binomial comparisions (Zar 1984:395-397) of percentage use of pine plantations, by seasons, identified most pine plantations (85%, n = 464) as being used equal to or greater than expected (P = 0.05). All pine plantations in the core study area that were used less than expected or were not used at all (group 1, 119 plantations) were pooled and compared with pooled pine plantations used equal to or greater than expected (group 2, 142 plantations) for discriminant analysis. We chose variables representing stand characteristics for spring 1987 because they represented conditions of the middle of the study period. We used Mahalanobis (Klecka 1988:55) distance as the test criterion for group discrimination. The discriminant function correctly classi- fied 79.0% for group 1 and 50.7% for group 2 (P < 0.05). Variables that entered the stepwise selection included roads (does a spur road border or transect the pine plantation), fertilization (number of years since the plantation had been fertilized), and hardwood control (number of years since hardwood Table 3. Use of habitat types by radio-equipped wild turkey hens in Kemper County, Mississippi, 1986-1988. n Habitat wesa (%locations) Year Seasonb locations NF PP MF HF 1986 Spring 352 0.0 60.8 23.9 15.3 Summer 218 0.0 24.3-C 67.9+ 7.8+ Fall 420 5.5- 57.1+ 18.8 18.6 Winter 537 6.5- 46.7+ 21.0 25.7 1987 Spring 1,030 5.8- 56.2+ 12.2- 25.7+ Summer 865 3.5- 73.6+ 16.5- 6.4- Fall 1,603 10.4- 49.8+ 9.5- 30.3+ Winter 1659 18.1+ 35.7- 14.7- 31.5+ ~NF = nonforest, PP = pine plantation, MF = mixed forest, HF = hardwood forest. Spring (Mar-Ma~), Summer (Jun-Aug), Fall ~ep-Nov), Winter pee-Feb). c+ used greater t an availability, -used less t an availability (P < 0.05). 64 control with herbicides). The remammg variables did not significantly improve (P > 0.05) the model. Hardwood control was strongly correlated (r = 0.850) with fertilization. The pooled within-groups correlation matrix revealed that commercial thinning had the second highest correlation (r = 0.812) with fertilization. Commercially thinned also had an F value of 10.05 (P < 0.0017) second only to roads (F = 10.06, P < 0.0017). Ancillary observations.-- We recorded 570 observations of individual or flocks of wild turkeys. Most observations were made during the summer (43%) and the least in fall (13%) and winter (13% ). Only 9 observations were recorded in December 1986 and 1987. Habitat types associated with turkey observations, pooled throughout the study period, were pine plantations (88.5% ), mixed forest (2.9% ), hardwood forest (3.4% ), and nonforest (5.2% ). Large flocks of turkeys ( > 30), some of which had transmitters, were often observed in soybean fields in the Sucarnoochee Creek bottom during both winters. DISCUSSION Wildlife biologists have been concerned that wild turkeys would not use large-block, even-aged, short-rotation pine plantations. Our telemetry study documented wild turkey hen use of intensively managed plantations. In addition, over 88% of ancillary observations were of turkeys associated with plantations. All turkeys were captured on spur roads in plantations. Hens used plantations more than expected in most seasons. Hens used plantations more than expected in the winter of 1986 when there was a poor acorn crop, but used plantations less than expected in winter of 1987 when there was a large acorn crop. Some hens left the plantations in the winter and used a large creek bottom hardwood tract and associated soybean fields, but returned to the plantations in late winter. Movement of turkeys to river bottom forests has been reported (Dalke et al. 1946). All nest attempts and most brood ranges were in pine plantations. Some hens attempted to rear their broods in the same plantations used for nesting. Most (95%) observations of hens with broods were associated with plantations. Use of Pine Plantations by Wild Turkey Hens • Smith et aL 65 Plantations used by turkey hens for nesting, brood rearing, and foraging were mostly midrotation-aged, 13-18 years old. These plantations had been commercially thinned once and control burned at least once. Vegetative conditions in the plantations were not measured; however, we believe the burning and thinning improved understory plant conditions for turkeys (Hurst and Warren 1982). The pine plantation canopy was opened by reducing pine stocking, which allowed increased growth of herbaceous plants and soft mast-producing vines, both important turkey food sources. In addition, commercial thinning operations, which removed all trees in every fourth row, created travel lanes and seemed to increase abundance of forage and seed and soft mast-producing plants. Why turkeys used certain plantations was difficult to determine because most of the plantations were used equal to or more than expected, and they had received similar silvicultural treatments, thus yielding uniform conditions. These factors may explain why only 50.7% of plantations used equal to or more than expected were correctly classified by the discriminant function equation. Discriminant analysis identified spur roads in plantations as being an important factor. These roads and shoulders were about 20 m wide and were covered with native vegetation; some roadsides were planted to food (wheat, rye grass) plots by deer hunters. Spur roads were gated, locked, and posted. The roads provided travel corridors and food, such as green forage, grass seeds, blackberries (Rubus spp.), and insects (Kennamer et al. 1980, Exum et al. 1987). Turkeys seemed to frequent roads after a rain. Deer hunting from spur roads was intense from October through mid-January and probably decreased turkey use of roads during this period. Discriminant analysis identified control of hardwood brush and fertilization as important factors. We believe these practices, performed over 9 years before the study began, do not have any biological relation to turkey use of plantations. Most plantations in the core area received the treatments. Vegetative responses to these treatments have been obscured by pre- commercial and commercial thinning, and control burning once or twice. Hens used over 84% of the plantations equal to or more than expected. Most (69.3%) Proceedin~ of the Sixth National Wild Turkey Symposium plantations in the core area were > 12 years old, and only 18.3% of the plantations were in the 3- to 10-year-old age class. Most older plantations had been thinned and burned at least once. Exum et al. (1987) and Smith and Teitelbaum (1986) reported greater than expected use of plantations > 10 years old. MANAGEMENT IMPLICATIONS Turkey hens used intensively managed loblolly pine plantations. In fact, the most important phase of the wild turkey life cycle, the reproductive phase, occurred in pine plantations. Most plantations were 13-19 years old, and the important silvicultural practices seemed to be contol burning and commercial thinning. A burning and thinning rotation most favorable to turkey habitat has not been established but we can assume the earlier the better. Supplemental food in deer food plots may have improved the plantations for turkeys. The fact that all the study area is leased to hunting clubs and posted probably aids the turkey by decreasing poaching. High road spacing, about 1 road/0.5 km, is part of intensive pine plantation management, and roads were identified as important to turkey use of plantations. All roads should be gated, locked, and posted to control human access. Roads can be an asset (travel, food) to turkey habitat. Roadsides must be managed to prevent dominance by hardwood brush and volunteer pine trees. LITERATURE CITED Afiffi, AA, and V. Clark. 1984. Computer-aided multivariate analysis. Lifetime Learning Publications, Los Angeles. 458pp. Cochran, W.W., and R.D. Lord. 1%3. A radio- tracking system for wild animals. J. Wildl. Manage. 27:9-24. Dalke, P.D., AS. Leopold, and D.L. Spenser. 1946. The ecology and management of wild turkey in Missouri. Mo. Conserv. Comm. Tech. Bull. 1. 86pp. Davis, J. R. 1976. Management for Alabama wild turkeys. Special Rep. 5, Ala. Dep. Conserv. and Nat. Resour., Montgomery. 53pp. Dixon, KR., and J.A Chapman. 1980. Harmonic mean measure of animal activity areas. Ecology 61:1040-1044. 66 Exum, J.H., Jr., J.A McGlincey, D.W. Speake, J.L. Buckner, and F.M. Stankey. 1987. Ecology of the eastern wild turkey in an intensively managed pine forest in southern Alabama. Bull. 23, Tall Timbers Res. Sta., Tallahassee, Fla. 70pp. Holbrook, H.T., M.R. Vaughan, and P.T. Bromley. 1985. Wild turkey management on domesticated pine forests. Proc. Natl. Wild Turkey Symp. 5:253-258. Hurst, G.A, and R.C. Warren. 1982. Deer forage in 13-year-old commercially thinned and burned loblolly pine plantations. Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies 36:420-426. Kennamer, J.E., J.R. Gwaltney, and KR. Sims. 1980. Habitat preferences of the eastern wild turkey on an area intensively managed for pine in Alabama. Proc. Natl. Wild Turkey Symp. 4:240-245. Klecka, W.R. 1988. Discriminant analysis. Sage Publications, Beverly Hills, Calif. 71pp. Markley, M.H. 1%7. Limiting factors. Pages 199- 243 in O.H. Hewitt, ed. The wild turkey and its management. The Wildlife Society, Washington, D.C. McDowell, R.D. 1954. Productivity of the wild turkey in Virginia. Ph.D. Thesis. Virginia Polytechnic Inst., Blacksburg. 66pp. Mohr, C.O. 1947. Table of equivalents populations of North American small mammals. Am. Wildl. Nat. 37:233-249. Pettry, D.E. 1977. Soil resource areas of Mississippi. Miss. Agric. and For. Exp. Sta. Inf. Sheet 1278. 4pp. Schaffer, C.H., and J.V. Gwynn. 1%7. Management of the eastern wild turkey in oak-pine and pine forests of Virginia and the Southeast. Pages 303-342 in O.H. Hewitt, ed. The wild turkey and its management. The Wildlife Society, Washington, D.C. Smith, D.R. 1988. Use of midrotation-aged loblolly pine plantations by wild turkeys. M.S. Thesis. Mississippi State Univ., Mississippi State. 52pp. Smith, W.P., and R.D. Teitelbaum. 1986. Habitat use by eastern wild turkey hens in southeastern Louisiana. Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies 40:405-415. Stoddard, H.L. 1963. Maintenance and increase of the eastern wild turkey on private lands of the coastal plain of the deep Southeast. Tall Timbers Res. Sta. Bull. 3, Tallahassee, Fla. 49pp. Wigley, T.B., J.M. Sweeney, M.E. Gamer, and M.A Melchiors. 1985. Forest habitat use by wild turkeys in the Ouachita Mountains. Proc. Natl. Wild Turkey Symp. 5:183-197. Zar, J.H. 1984. Biostatistical analysis. Prentiss- Hall, Inc., Englewood Cliffs, N.J. 718pp.