Coursehero >> Texas >> Texas Tech >> ETD 06272008

Course Hero has millions of student submitted documents similar to the one below including study guides, homework solutions, papers, exam answer keys and textbook solutions.

USE HABITAT BY THE RIO GRANDE TURKEY ON THE SOUTHERN COASTAL PLAIN OF TEXAS by ELIZABETH J. WILBANKS, B.S.A., B.S.F.R. A THESIS IN WILDLIFE SCIENCE Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE Approved Chairperson of the Committee Accepted Dean of the Graduate School August, 2003 ACKNOWLEDGEMENTS First of all, I would like to dedicate both the work and the finished produd of my Master's degree to my family. Those family members that are here and those that are gone, those who are actually related by blood, and those who are related because we choose to care that much. I love you all, and there is no wa that I could have gotten through this with out your love, your support, and your ear when I needed to scream. This is as much your accomplishment as mine, I love you all and thanks for everything. Secondly, I would like to dedicate this work to Dr. Phillip J. Zwank, my original major professor. Dr. Zwank took a chance on me in giving me this position and for that I will be forever grateful to him. Unfortunately, he passed away in August of 2002, and never got to see the finished project. I hope that I have done him proud. Dr. Z, although we frequently butted heads, you were loved, and you are missed. First, to thank would definitely be the Welder Wildlife Foundation for their financial and technical support, access to records and property, a place to live, moral support and just generally being there for me when I needed all of you!! Dr. Drawe, thanks for stepping in when I needed you and offering to take on yet another student!! Terry, thanks for the assistance in my telemetry and all of the GIS questions. Selma, thanks for your unwavering support and all of your advice. Liz and Michelle, thanks for keeping everything running. Mr. Cox, thank for everything, I don't even know where to start. This project would never have gotten off of the ground without your unflagging support and technical assistance Second of course would be the Department of Range, Wildlife, and Fisheries Management at Texas Tech University for technical and logistical support. Dr. Mark Wallace, you will never know how much I appreciate your help, your support, and your willingness to also take on yet another student. Your ability to listen and support me in everything that I did was not unnoticed. You worked as hard as I did to get me out of here and I appreciate it!! To Kay, Louise, and Jeannine, you guys always take our crises and make them go away. A huge thank you goes out to all of my landowners. Jeff Rooke and Leonard Piland of the Rooke Ranch, thanks for the help, dragging me out ofthe mud, and answering all of my questions. Kenneth Hodges and Hugh Thomas, of the Aldrete Ranch and Alfred Niemann, thanks for the access to your property and for answering all of my questions. To the Devarry-Durrril Foundation, thanks for the access and the opportunity to affect other people's lives in a positive way. John and Tina Brunjes, my Georgia compatriots. John, the trapping and GIS genius, thanks for answering my endless questions and always being there when I had a crisis. Tina, you have been my sounding board and there is no way that I could have made it without you. Thanks to Chuck Canfield, Jim Wilbanks, Gail Wilbanks, Lisa Hunt, and lliana Pena for all ofthe trapping help and all ofthe hours spent in blinds. Special thanks to Chuck for the wonderful pictures that will forever remind me of "The Terrific Texas Turkey Tagging." Thanks to David Baine for making the packs to carry all of my stuff. Thanks to my Welder family, lliana Pena, Richard Heilburn, Tricia Fry, Emily Dacy, Amy Kresta, Laura Bonneau, and Ken Mix. Thanks for making me laugh, invading my room when I needed a break and just generally always being there for me, Ames, you are my sister in every way that counts. Thank you for always listening to me and for lending me your parents and your home. Thanks to Teresa, Maury, Caitlin, and Cora Benamy for your love and support. To my family...what can I say. You all are my rocks, my support lines, my lifelines. I love you all and I could never have accomplished this without you. This is as much your achievement as mine. Finally, thanks to Bailey, Sunny, Dancer, Banner, Jubilee, and Spot for carting me around for a year and being patient with me!! IV TABLE OF CONTENTS ACKNOWLEDGEMENTS LIST OF TABLES LIST OF FIGURES CHAPTER I. II. INTRODUCTION LITERATURE REVIEW, STUDY AREA, AND STUDY OBJECTIVES Introduction Study Area Study Objectives Literature Cited HABITAT USE BY RIO GRANDE TURKEYS ON THE SOUTHERN COASTAL PLAINS OF TEXAS Abstract Introduction Methods Results Discussion Management Implications Acknowledgements Literature Cited ij vii x 1 2 2 7 9 11 III. 16 16 17 19 23 28 31 32 33 IV. LAND USE CHANGES ON THE SOUTHERN COASTAL PLAINS OF TEXAS Abstract Introduction Methods Results Discussion Management Implications Acknowledgements Literature Cited 59 59 60 63 66 68 69 69 70 APPENDIX A Trapping record for Rio Grande turkeys from January 2002 to March of 2002 on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas Kaplan Meier survival rates for Rio Grande turkey hens on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas 79 B 83 VI LIST OF TABLES 2.1 3.1 Relationships between soils and range sites on the Welder Wildlife Refuge, San Patricio County, Texas Cause of mortalities for radio-transmittered Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Proportion of each vegetation class and the expected number of Rio Grande female turkey locations in each class and confidence intervals (CI) for the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Proportion of use, availability of vegetation types, and confidence intervals (CI) for annual use by Rio Grande females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002.... Winter vegetation type use in proportion to availability and confidence intervals for Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Spring vegetation type use in proportion to availability and confidence intervals (CI) for Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 200 Summer vegetation type use in proportion to availability and confidence intervals (CI) for Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Autumn vegetation type use in proportion to availability and confidence intervals (CI) on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Spring vegetation type use in proportion to availability and confidence intervals (CI) for surviving female Rio Grande turkeys on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 15 36 3.2 37 3.3 38 3.4 39 3.5 40 3.6 41 3.7 42 3.8 43 VII 3.9 Spring vegetation type use in proportion to availability and confidence intervals (CI) for female Rio Grande turkey mortalities on the Welder Wildlife Refuge and the Rooke Ranch in San Patricio and Refugio Counties, Texas in 2002... Summer vegetation type use in proportion to availability and confidence intervals (CI) for surviving female Rio Grande turkeys on the Welder Wildlife Refuge and the Rooke Ranch in San Patricio and Refugio Counties, Texas in 2002 Summer vegetation type use in proportion to availability and confidence intervals (CI) for Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Successful and unsuccessful nests based on nesting attempt for Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Successful and unsuccessful nests based on nest site location for Rio Grande turkeys on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Rio Grande turkey poult:female ratios for 9 years of turkey research on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas 44 3.10 45 3.11 46 3.12 47 3.13 48 3.14 49 3.15 Kaplan-Meier calculated weekly, overall nest survival rates, and confidence intervals (CI) for Rio Grande females turkeys on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 50 3.16 Vegetation type and cause of Rio Grande turkey nest Depredation on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 4.1 ^3 Change in proportional availability of vegetation types from 1969 to 1995 on the Welder Wildlife Refuge, San Patricio County, Texas ^2 Change in proportional availability of habitat types from 1969 to 1995 on the Rooke Ranch, Refugio County, Texas.... viii 4.2 73 A. 1 Trapping record for Rio Grande turkeys from January 2002 to March of 2002 on Rob and Bessie Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas Kaplan-Meier survival rates for Rio Grande turkey hens on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 80 B.I 84 IX LIST OF FIGURES 3.1 Comparison of monthly average rainfall from 1969 to 2001 and 2002 on the Welder Wildlife Refuge, San Patricio County, Texas Kaplan-Meier estimate of Rio Grande turkey female survival rates by week (26 January 2002 to 13 December 2002) on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Correlation of monthly rainfall averages (mm) and number of Rio Grande turkey female mortalities on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Comparison of number of locations and home range sizes for radio-telemetered Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Number of nesting attempts by month for transmittered Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002 Average yearly rainfall (1969 to 1995) for the Welder Wildlife Refuge, San Patricio County, Texas Comparison of regression analyses for brush, riparian, and grassland vegetation type changes from 1969 to 1995 on the Welder Wildlife Refuge, San Patricio County, Texas Comparison of regression analyses for brush, riparian, grassland, and agricultural field vegetation type changes from 1969 to 1995 on the Rooke Ranch, Refugio County, Texas.... Comparison of changes of brush vegetation types (1969 to 1995) on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas Comparison of the proportion of San Patricio and Refugio Counties, Texas being utilized as agricultural lands 54 3.2 55 3.3 56 3.4 57 3.5 58 4.1 74 4.2 75 4.3 76 4.4 77 4.5 78 CHAPTER I INTRODUCTION This thesis represents partial fulfillment ofthe requirements for the degree of Master of Science in Wildlife Sciences in the Graduate College at Texas Tech University. Chapter II is the literature review, description ofthe study area and the statement of study objectives. Chapters III and IV are manuscripts intended for publication in peer-reviewed journals. Chapter III discusses current habitat use by Rio Grande turkeys {Meleagris gallopavo intermedia) on the southern coastal plains of Texas. Chapter IV documents the changes in land use and land management practices that have occurred on the southern coastal plains of Texas since 1969. Chapters III and IV are intended for submission in peerreviewed journals following the completion of this thesis. This work represents my writing, analyzing and research abilities. With the guidance of my committee and particulariy my major advisors, I designed this study, collected the data, and synthesized the research into several manuscripts. Authorships for the following manuscripts were determined based on contributions. Authorship will be as follows: Chapter III. Elizabeth J. Wilbanks, Mark C. Wallace, Phillip J. Zwank, and D. Lynn Drawe; Chapter IV: Elizabeth J. Wilbanks, Mark C. Wallace, Phillip J. Zwank, and D. Lynn Drawe. CHAPTER II LITERATURE REVIEW, STUDY AREA, AND STUDY OBJECTIVES Introduction Rio Grande turkeys {Meleagris gallopavo intermedia) are distributed from Kansas through Oklahoma, Texas, and into northern and central Mexico (Beasom et al. 1992, Capel 1973, Glazener 1967, Schorger 1966). Kennamer and Kennamer (1990) estimated that there were more than 579,000 Rio Grande turkeys in Texas, Despite the relative abundance of this sub-species, knowledge concerning the Rio Grande turkey has not kept pace with that of other wild turkey sub-species. Peterson (1998) lamented the fact that, even though the Rio Grande sub-species is the second most numerous there were no papers presented on this taxon at the 7'^ National Wild Turkey Symposium. Most papers presented concerned either the eastern {M. g. silvestris) or Merriam's (M. g. merriami) turkeys. Hunters spend millions of dollars annually to hunt wild turkeys, making these birds an important game species (Baumann et al. 1990). In the southern coastal plain of Texas, over 91,000 turkeys are harvested annually (Kennamer and Kennamer 1995). Within this area, population numbers have been declining since the 1970s. For example, on the Rob and Bessie Welder Wildlife Refuge (WWR), there was an estimated population high of 700 birds in 1966-67, which declined to less than 100 birds by 1981 (Watts 1969, D. L. Drawe, WWR, personal communication). Recent years have seen slight increases in population numbers, but current estimates place the population at less than 140 birds (D. L. Drawe, WWR personal communication). During the 1980s WWR personnel observed few to no broods. Recently, a few small broods have been documented (T. L. Blankenship, WWR, personal communication). Surrounding ranches have not reported these same population trends. Ranch personnel on the Rooke Ranch, located directly across the Aransas River and to the north of WWR, have not noted these trends, but rather note a currently stable population (J. Rooke, Rooke Ranch, unpublished data). Personnel from WWR have often noted turkeys flying across the river to roost on the Rooke Ranch (D. L. Drawe, WWR, personal communication). Since 1970, ranching practices have changed in southern Texas. One such change has been a decline in the use of fire to control undesirable shrubs such as honey mesquite {Prosopis glandulosa), huisache {Acacia famesiana), and lime pricklyash {Zanthoxylum fagara). These species were previously present in the area, but suppressed by fire (Tharp 1926, Bogusch 1952, Johnston 1963), Drawe et al. (1978) noted a comparison of aerial photographs taken in 1939, 1950, 1960, and 1969 showed that brush has increased on the WWR. Scrifes (1979) reported in the mid-1970s that eariier brush eradication policies of ranchers had given way to an attitude of brush management. The philosophy was to manage woody vegetation on rangelands so as to optimize benefits for both livestock and wildlife. Woody vegetation has increased on the southern coastal plain of Texas. On WWR aerial photographs taken in 1970, 1983, 1989, and 1995 show that brush cover has increased throughout the refuge. Areas of dense brush may restrict turkey movements (Miller and Leopold 1992). The Rooke Ranch is more open as a result of more active brush control practices and more acreage being converted to croplands (Ransom et al. 1987). The optimal amount of brush cover per section (259 ha) for turkey habitat has received little study. Walker and Springs (1952) and Litton (1977) recommend that 50 percent coverage should be left to maintain turkey habitat (Litton 1977, Walker and Springs 1952). Turkeys are strongly affected by human disturbances (Gore 1973). Baker (1979) noted that hens raising broods primarily useriparianwoodlands and savannas, but a moderate amount of brush coverage is required, especially during brood rearing (DeArment 1959). Turkeys have not used area pastures that were dense with brush until the area was partially cleared leaving only live oaks {Quercus virginiana) and tall, scattered mesquite trees. After clearing, these areas were used by hens and broods, as well as, groups of hens and gobblers on a daily basis for feeding (Baker 1979, Baker et al. 1980). These dense brush areas also provide predators with excellent hiding cover (Miller and Leopold 1992). Areas that have received indiscriminate or complete brush removal are perceived as poor turkey habitat (Walker 1951). These practices not only remove the intended brush, and often remove roost sites as well as vital winter food sources. Glazener (1967) cites an example of a farmer carefully protecting roost trees, but removing all surrounding trees and shrubs, thereby destroying the habitat and resulting in the disappearance of his resident Rio Grande turkey population. Turkeys require roost trees, cover (often provided by woody vegetation), water sources, and a diverse grass and forb cover which provide a seed source and insect foods (Porter 1992). The trees provide daytime resting areas, food, and most importantly roost sites (Everett et al, 1979); whereas, the grasses offer food for the adults, as well as providing the poults with an area in which they can forage for insects (Litton 1977). Insect abundance is greater in open areas than in densely wooded areas (Porter 1992). Turkeys will leave the roost shortly after daybreak, feed throughout the morning and afternoon, and spend time wandering, loafing, preening, and dusting. The birds will then return to roost just prior to dusk (Melchiors 1991). During winter flocks will typically return to the same roost each night, barring any major disturbance (Logan 1970). During nesting the gobblers may rotate roost sites, while the hens remain with their young (Ransom et al. 1987). Close proximity to water is also vitally important as a habitat requirement, and therefore may be a factor in roost site selection (Crockett 1965). Other data suggest that tree height is of greater importance. Because taller trees have greater water requirements, they are thereby associated with permanent water sources (Crockett 1973, Haucke 1975). Rio Grande turkey hens often choose nesting sites that have good concealing cover and are near water (Glazener 1959). Cook (1972) found that 87% of nests in the Edwards Plateau were in cover taller than 45.7 cm, and Ransom et al. (1987) found that all nests located in shrubs and 50% of those located in herbaceous cover had nesting cover that was >45 cm in height. Ransom also reported that nests averaged 0.35 km from available water and 0.84 km from known roosts. These nest sites are likely chosen because of the improved survival rates and more favorable climatic conditions (Beasom 1970). These vegetation parameters have been shown to (1) reduce temperature extremes, (2) have higher relative humidity, and (3) reduce wind movement at the ground surface (Geiger 1950:230-240). Wild turkeys are edge-specialists, preferring areas that are the junctions of one or more types of habitat (Melchiors 1991). This offers them the greatest diversity of plants and insects as food sources. It also provides them with the required roost trees, as well as the woody vegetation and a combination of grasses and forbs (Beasom 1970). During brood rearing, turkeys most commonly use riparian woodlands and the open savannas and the edge between the two. During fall and winter the birds will decrease their use ofthe open areas and increase their utilization of the forested areas. Since mast is the primary food source during the fall and winter, the value ofthe habitat increases with the number of mast producing species (Porter 1992). However, during the fall and winter, green vegetation is also important; therefore, a forested area interspersed with herbaceous species would also be a potentially valuable habitat (Kennamer etal. 1980). We suspect that the increase in brush and croplands in the coastal plain of south Texas have negatively impacted Rio Grande turkey population in south Texas. While it has been documented that turkeys require a moderate amount of brush coverage, we suspect that the woody shrub density in the current landscape exceeds turkey requirements. We will assess the current habitat use by Rio Grande turkeys on WWR and the Rooke Ranch to determine the relationship between the shrub cover changes and the population trends on WWR. Studv Area The area is part of what historically was a grassland complex characteristic of the southern coastal plain bordering the Gulf of Mexico. Recently it has developed into a brush-grass complex, partially as a result of having been grazed historically by domestic livestock (Drawe et al. 1978). The southern coastal plain of Texas has an average annual rainfall of approximately 89 cm, with the majority falling in the late summer and fall hurricane season. The general climate of the area is considered to be humid and sub-tropical (Drawe et al. 1978). Relative humidity is high throughout the year because ofthe close proximity of the Gulf of Mexico. This proximity also contributes to the prevailing southeasteriy winds that are prevalent in all months of the year. Heavy dews occur almost nightiy and locally concentrated fog is common from late fall to eariy spring (Drawe et al. 1978). Soils in the area developed from the Beaumont and Lissie Pleistocene formations (Drawe et al. 1978). Local deposits from the Aransas River have further modified them (Kovar 1963). The majority ofthe soils on the refuge are of the Aransas, Nueces, Orelia, Sarita, Sinton, Willacy, and Victoria series (Table 2.1). They range in texture from clays to fine sands. Similarities in productivity and plant species composition have allowed grouping of these series under 10 range sites (Drawe et al. 1978). Victoria clay {Udic pellustert) is the most prevalent soil type ofthe area, covering almost 53 percent ofthe refuge (Kovar 1963). This area is part of a neariy level coastal plain in which the Aransas River and Moody Creek have carved sizable valleys. The major portion ofthe refuge IS about 12 m above sea level. Slope gradients on the upland portions are usually less that 1 percent, while slopes on the escarpment made by the stream and river terraces are somewhat greater (Drawe et al. 1978). WWR is a 3,158-ha private refuge located along the northern border of San Patricio County, approximately 4.3 km north of Sinton, Texas (N28 6'50" W97 26'15''). WWR was operated as a working cattle ranch, which has utilized a cow-calf operation in a variety of grazing systems and stocking rates. Prior to 1974, steers were used in a continuous grazing system (Glazener, WWR, unpublished data). It was operated as a cow-calf operation, with a stocking rate of 0.0791 AU per ha. At the time of this study the cows were managed in 3 herds, each with 4 available pastures. However, as a result of deferment for the burning program, only 3 pastures were actually in use for each herd (J. Cox, WWR, personal communication). The Rooke Ranch, a 4,237-ha ranch located in Refugio County is located directiy to the north, and across the Aransas River from the WWR (N28 7'50" W97 21'30"). Part ofthe Rooke Ranch has been converted to agricultural crop production, while the rest has remained in a cow-calf operation under a rotational grazing system. At the time of this study the cows were managed in 4 herds. Two herds were rotated through 3 pastures, and 2 herds were rotated through 2 pastures, although each herd occupied similar amounts of land. The stocking rate was 0.101 AU per ha. Studv Objectives Although there have been several studies ofthe Rio Grande turkey on the south Texas coastal plain (Glazener 1967, Watts 1968, Smith 1977, Baker 1979, Ransom 1984, Ransom et al. 1987, Glazener et al. 1990, and Beasom and Wilson 1992,), few have been conducted in recent years. There is a lack of data on how changes in land management practices may affect wild turkey populations. No reliable data exist on changes in turkey populations overtimein the region. Therefore we cannot directly determine if a correlation exists between the changes in habitat and the decline in population numbers. The primary goal of this research was to investigate the relationship between current habitat use by Rio Grande turkeys on the southern coastal plain of Texas and to relate these changes in the availability of vegetation types selected by Rio Grande turkeys over the 26 year period of 1969 to 1995. Study objectives were: 1. Document current coverage of vegetation types on the 2 study areas using aerial photographs and a Geographic Information System (GIS); 2. Use radio telennetry techniques to document current habitat use by Rio Grande turkeys on areas where vegetative types have been mapped; 3. Identify vegetation types and parameters of habitat selected for by Rio Grande turkeys on the southern coastal plain of Texas; and 4. Document changes in vegetation types over time from a series of aerial photographs (1969 to 1995). Specific hypotheses to be tested were as follows: 1. Rio Grande turkeys did not use the vegetation types in proportion to their availability. They used some types more than were proportionally available and others less. 2. Brush vegetation types will have increased to a greater extent on the WWR than on the Rooke Ranch. 10 Literature Cited Baker, B. W. 1979. Habitat use, productivity and nest predation of Rio Grande turkeys. Ph.D. Dissertation. Texas A&M University, College Station Texas. 46 pp. ., S. L. Beasom, N. J. Silvy. 1980. Turkey productivity and habitat use on south Texas rangelands. Proceedings National Wild Turkey Symposium 4:145-158. Beasom, S. L. 1970. Turkey productivity in two vegetative communities in south Texas. Journal of Wildlife Management 34:166-175. , and D. E. Wilson. 1992. Rio Grande Turkey. Pages 306-330/n J. G. Dickson, ed.. The wild turkey: biology and management. Stackpole Books, Harrisburg, Pennsylvania. Bogusch, E. R. 1952. Brush invasion in the Rio Grande plain of Texas. Texas Journal of Science 4(1):85-91. Capel, S. W. 1973. Introduction of Rio Grande turkeys into Kansas, Pages 11-17 in G. C. Sanderson and H. C. Schuiz, editors. Wild turkey management; current problems and programs. University of Missouri Press, Columbia, Missouri. Cook, R. L. 1972. A study of nesting turkeys in the Edwards Plateau of Texas. Proceedings of Annual Conference of Southeastern Association of Game and Fish Commission 26:236-244. Crockett, B. C. 1965. Quantitative evaluation of winter roost sites ofthe Rio Grande turkey in north central Oklahoma. M. S. Thesis. Oklahoma State University. Norman, Oklahoma. 45 pp. . 1973. Quantitative evaluation of winter roost sites ofthe Rio Grande turkey in north central Oklahoma. Pages 211-218 in G. C. Sanderson and H. C. Schuiz, editors. Wild turkey management; current problems and programs. University of Missouri Press, Columbia, Missouri. DeArment, R. D. 1959. Turkey hen-poult ratios as an index to reproductive trends. Proceedings ofthe National Wild Turkey Symposium 1:27-31. Drawe, D. L., A. D. Chamrad, and T. W. Box. 1978. Plant communities ofthe Welder Wildlife Refuge. Contr. No.5, Ser. B, Rev., Welder Wildlife Foundation, Sinton, Texas. 38 pp. 11 Everett p^D., D. W. Speake, and W. K. Maddox. 1979. Wild turkey ranges in Alabama mountain habitat. Proceedings ofthe Annual Conference ofthe Southeastern Fish and Wildlife Agencies 33:233-238. Geiger, R. 1950. The climate near the ground. Cambridge, Massachusettes' Harvard University Press. 482 pp. Glazener, W.C. 1959. Wild turkey research needs. Proceedings National Wild Turkey Symposium 1:177-182. , 1967. Management of the Rio Grande turkey. Pages 453-492 in O.H. Hewitt, editor. The wild turkey and its management. The Wildlife Society., Washington D.C. , D. Ransom Jr., J. R. Cary, and O. J. Rongstad. 1990. Demographic analysis of a Rio Grande turkey population. Southwestern Naturalist 35:23-27. Gore, H. G. 1973. Land-use practices and Rio Grande turkeys in Texas. Pages 253-262 in G. C. Sanderson and H. C. Schuiz, editors. Wild turkey management; current problems and programs. University of Missouri Press, Columbia, Missouri. Haucke, H. H. 1975. Winter roost characteristics ofthe Rio Grande turkey in south Texas. Proceedings ofthe National Wild Turkey Symposium 3:164169. Johnston, M. C. 1963. Past and present grasslands of south Texas. New Era Printing House, Lancaster, Pennsylvania. Kennamer, J. E., J. R. Gwaltney, and K. R. Sims. 1980. Food habits ofthe eastern wild turkey on an area intensively managed for pine in Alabama. Proceedings of the National Wild Turkey Symposium 4:246-250. , and M. C. Kennamer. 1990. Current status and distribution ofthe wild turkey 1989. Proceedings ofthe National Wild Turkey Symposium 6:1-12. 1995. Status and distribution ofthe wild turkey in 1994. Proceedings ofthe National Wild Turkey Symposium 7:203-211. Kovar, J. A. 1963. Physical, chemical, mineralogical, properties of some soils of the Welder Wildlife Foundation. M.S. Thesis. Texas A&M Univ., College Station, Texas. 88 pp. 12 Litton, G. W. 1977 Food habits ofthe Rio Grande turkey in the Permian Basin of Texas. Austin: Texas Parks and Wildlife Technical Series No. 18. 22 pp. Logan, T. H. 1970, Oklahoma Department of Wildlife Conservation. Oklahoma Federal Aid to Wildlife Restoration (P-R) Report, Project W-129-M, 47 pp. Melchiors, M. A. 1991. Wildlife management in southem pine regeneration systems. Pages 391-420 in M, L. Duryea and P. M. Dougherty, editors. Forest Regeneration Manual. Kluwer Academic Publishers, Netheriands. Miller, J. E. and B, D. Leopold. 1992. Population influences: predators. Pages 119-128 in J. G. Dickson, editor. The wild turkey: biology and management. Stackpole Books, Harrisburg, Pennsylvania. Peterson, M. J. 1998. Book review of Proceedings ofthe Seventh National Wild Turkey Symposium. Journal Wildlife Management 62:816-818. Porter, W. F. 1992. Habitat Requirements. Pages 202-213 in J. G. Dickson editor. The wild turkey: biology and management. Stackpole Books, Harrisburg, Pennsylvania. Ransom Jr., D. 1984. Nesting ecology of Rio Grande turkeys on the Welder Wildlife Foundation. M.S. Thesis. Texas A&M Univ., College Station, Texas. 35 pp. , O. J. Rongstad, and D. H. Rusch. 1987. Nesting ecology of Rio Grande turkeys. Journal of Wildlife Management 51:435-439. Schorger, A. W. 1966. The wild turkey: its history and domestication. University of Oklahoma Press, Norman, Oklahoma. 625 pp. Scrifes, C. J. 1979. Brush management: principles and practices for Texas and the southwest. Ph.D. Dissertation. Texas A&M University, College Station, Texas. 360 pp Smith, D. M. 1977. The social organization of Rio Grande turkeys in a declining population. Ph.D. Dissertation. Utah State Univ., Logan, Utah. 98 pp. Tharp, B. C. 1926. Structure of Texas vegetation east ofthe 98"" meridian. Austin: University of Texas Bulletin 2606. 100 pp. Walker, E. A. and A. J. Springs. 1952. Factors concerned with the success and failure of turkey transplants in Texas. Austin: Texas Game, Fish and Oyster Commission, Completion Report (P-R) Project W-44-R-2. 21 pp. 13 Watts, C. R. 1968. Rio Grande turkeys in the matting season. Transcripts of the North American Wildlife Conference 33:205-210. 1969. The social organization of wild turkeys on the Welder Wildlife Refuge. Ph.D. Dissertation, Utah State University, Logan, Utah 60 pp. 14 Table 2_1: Relationships between soils and range sites on the Welder Wildlife Kefuge, San Patricio County, Texas.^ Range Site Loamy Bottomland Soil Type Odem fine sandy loam {Cumulic haplustoll) Sinton loam {Cumulic tiaplustoll) Sinton complex {Cumulic haplustoll) Clayey bottomland Claypan prairie Aransas clay {Vertic haplaquoll) Orelia fine sandy loam {Typic octiraqualf) Edroy clay {Vertic haplaquoll) Orelia sandy clay loam {Typic ochraqualf) Blackland Victoria clay {Udic pellustert) Monteola clay {Typicpellustert) Monteola complex {Typic pellustert) Sandy loam Sandy prairie Loamy sand Tight loamy sand Deep sand Salty prairie Willacy fine sandy loam {Udic argiustoll) Sarita-Nueces complex {Grossarenic paleustalf) Delfina loamy fine sand {Aquic haplustalf) Papalote fine sandy loam {Aquic paleustalf) Nueces fine sand {Aquic haplustalf) Aransas clay, saline {Vertic haplaquoll) ^ USDA, Soil Conservation Service, Sinton, Texas, 1977 15 CHAPTER III HABITAT USE BY RIO GRANDE TURKEYS ON THE SOUTHERN COASTAL PLAIN OF TEXAS^ Abstract: We used radio-telemetry to assess habitat use of female Rio Grande turkeys {Meleagris gallopavo intermedia) on the Rob and Bessie Welder Wildlife Refuge (WWR), Sinton, Texas and the adjacent Rooke Ranch, Woodsboro, Texas during January to December 2002. Annual survival (23.5%) was determined from 29 radio-tagged turkeys. We assessed habitat availability from aerial photography. Females selected for riparian and against the brush habitats. Twenty-tiwo (85%) of 26 instmmented females nested at least once, 15tiwice,4 three times, and 1 fourth nesting attempt was made. Eleven hens successfully hatched broods. Successful nests were in riparian habitats. J. Wildl. Manage. 0(0):000-000 Key Words: habitat selection, habitat use, Meleagris gallopavo intermedia, mortality, nesting, telemetry ^ This article was submitted to the Journal of Wildlife Management. The authors are as follows: Elizabeth J. Wilbanks, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, Texas 79415-2125, Email: bwilbank@earthlinknet; Mark C. Wallace, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, Texas 79415-2125; Phillip J. Zwank (deceased), Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, Texas 79415-2125; and D, Lynn Drawe, Rob and Bessie Welder Wildlife Foundation, PO Box 1400, Sinton, Texas 78387, 16 Introduction The Rio Grande turkey population on the WWR has declined approximately 700 birds (Watts 1968) to less than 100 birds in 1981 (D. L. Drawe, WWR, unpublished data). It rebounded to a current estimate of 120 birds (T. L. Blankenship, WWR, unpublished data). During this same time, land use on the southern coastal plain of Texas shifted from traditional cattle ranching towards a mixture of cattle ranching and crop farming. There was also a reduction in the use of fire, mechanical, and chemical means to control the woody brush in the area (Scrifes 1979). Rio Grande turkeys in the southern coastal plains of Texas have been fairiy well studied (Glazener 1967, Watts 1969, Smith 1977, Baker 1979, Ransom 1984, Ransom etal, 1987, Glazener et al. 1990), but have received little attention in recent years. The apparent downward population trend has been attributed to decreases in nesting attempts, nest success, poult survival, and loss of habitat (Ransom et al. 1987, Glazener et al. 1990). This study was designed to assess current turkey habitat use on the southern coastal plains of Texas. We concentrated on nesting and brood rearing habitat as Ransom et al. (1987) and Glazener et al. (1990) both noted the lack of nesting success in this area. We hypothesize that if the apparent population decline was related to brush it would be apparent in the use and productivity of Rio Grande turkeys in nesting and brood rearing habitat. We hypothesize that (1) female Rio Grande turkeys select for specific habitats throughout the year and (2) different 17 vegetation types will provide Rio Grande turkey females with different survival and nesting success. The study was conducted on the WWR, a 3,158-ha private wildlife refuge located in San Patricio County Texas and the neighboring Rooke Ranch, a 4,237-ha private ranch located in Refugio County Texas and located north and across the Aransas River from WWR. This area was historically part of a grassland complex found along the Gulf of Mexico, however in the past 35 years it has developed into more of a brush-grass complex (Drawe et al. 1979). These vegetation changes were thought to be largely the result of the history of heavy, continuous grazing by domestic livestock and the elimination of fire (Drawe et al. 1978). Average rainfall was 89 cm, largely occurring in the late summer and eariy fall. Sixteen plant communities have been described on the area and there were 22 soil classes, ranging from fine sands to heavy clays (Drawe et al. 1978). Both WWR and the Rooke Ranch were working cattle ranches. WWR had been operated as a cow-calf operation since 1974 and used a 3 herd, 4 pasture deferred grazing system (J. Cox, WWR, personal communication). A lightgrazing regime was practiced on WWR (0.0791 AU per ha). In recent years the deferment has been dictated by other research projects being conducted on the property examining the effects of summer burning. The Rooke Ranch was also a cow-calf operation under a moderate, continuous grazing regime, (0.101 AU per ha). Additionally 1,776 ha ofthe Rooke Ranch had been converted to crop farming, primarily cotton {Gossypium spp.) and grain sorghum {Sorghum vulgare). Hunting was prohibited on the WWR, and only the family was allowed 18 to hunt the spring turkey season on the Rooke Ranch, but minimal poaching was known to occur on both areas. Methods Data were collected from December 2001 thru December 2002. Weather information for the field season was obtained fi-om the Crop Weather Program website (www.cwp.tamu.edu) provided by the Texas A&M University Cooperative Extension Service from data collected at WWR headquarters. Data we compiled included average daily temperatures, rainfall, and soil temperature at 25.4 mm. We compiled monthly averages to be compared against historic averages obtained for WWR (US Department of Commerce, 1969-1989, dimatological data, Texas). We defined seasons based on Rio Grande turkey behaviors as follows. Winter, December to January, when turkeys are predominantly associated with winter roosts (Thomas et al. 1966, Cook 1973). Spring was defined as February to April, and is characterized by dispersal from the winter roosts and the onset of mating season (Glazener 1967, Baker 1978). Nesting and brood rearing occurs in the summer. May to August (Watts 1968, 1969). Autumn, September to November, was distinguished by a concentration of birds of both sexes and all age classes in large flocks and the gradual return to winter roost areas (Baker 1979). Female turkeys were trapped 3 times betiween 21 January to 18 March 2002, using cannon nets and walk-in traps (Glazener 1967. Davis 1994) 19 (Appendix A). Birds were classified by sex and age (Williams 1961) and backpack-style radio transmitters (Advanced Telemetry Systems model #23412, 110 g) were attached using nylon over-braid harness material. Transmitters were equipped with an 8-hour mortality switch and ranged in frequency from 151.000 to 152.000 MHz. Locations of telemetered females were obtained by triangulation, with a minimum of 3 bearings for each location. Locations of females were recorded at intervals ^ 3 days during nesting (18 March to 20 August), and < every 5 days at all other times. At each location we recorded habitat type, activity, and association with other birds. Mortality checks were conducted every third night when birds were at roost and more easily detected. When mortality signals were detected the bird was located the next day and cause of death was determined by searching for evidence of predation and field necropsy (when feasible) (Vangilder and Kuzejeski 1995). Hen survival rates were calculated using Pollock (1989). Individuals not surviving < 2 weeks posttrapping were censored as capture related trapping mortalities. We correlated hen mortality and average monthly rainfall. We used LOAS (Imagine Software, Inc.) to calculate the triangulated Rio Grande turkey locations. We used a Global Position System (GPS; Eagle Map Guide Pro, Catoosa, OK) to document locations of known transmitter locations which we compared to estimated locations generated by telemetry and LOAS to generate mean error location error. Turkey capture and handling methods were approved by the Texas Tech University Animal Care and Use Committee permit #'s 01181 and 01181-11. 20 We mapped vegetation types on the study areas using a Geographic Information System (GIS). Coverages were developed from 1:24,000 U.S Geological Survey Digital Ortho-rectified Quarter Quadrangle Sheets (DOQQs) obtained from the USDA Aerial Photography Field Office. Of the 16 vegetation types described by Drawe et al. (1979), 8 were identifiable from the aerial photographs. Each vegetation type was determined by ocular estimation ofthe dominant species present. We delineated vegetation types in the GIS database using ERDAS Imagine 8.3 (Imagine Software, Inc.). To determine the accuracy ofthe vegetation types, we ground-truthed our mapping comparing visual verifications at random points to the mapped classifications. At points where the visual verification did not match the map we adjusted the map to reflect the visually verified vegetation type. We plotted turkey locations in our GIS and used the software BIOTAS to generate home ranges, 95% Minimum Convex Polygon (MCP) and 95% Full Kernel (FK). We used the GIS and the home ranges to determine use versus availability on 2 levels. The first level (Johnson [1980] second-order of selection) compares the calculated home ranges (FK) to total availability across the study area. We used Neu et al. (1974) to test if turkeys selected for home ranges with proportional vegetation types that differed from what was available across the entire study area. We also compared all individual locations to availability within home ranges (FK) (Johnson [1980] third-order of selection). This comparison was used to examine selection for certain habitat requirements related to 21 seasons (i.e., nesting habitat, brood rearing habitat, and roosting habitat). Finally, we compared variables related to nesting cover at the micro-site level. Females were considered to be nesting when ^ 2 subsequent re-locations of the same female were in the same place. Nest initiation was calculated as 14 days prior to the first documented nesting location (Williams et al. 1971). Nests were not approached until after nest abandonment or successful hatching. Nest depredation dates were identified as the last telemetry location tiiat indicated the female was still incubating. Cause of nest depredation was determined by examination ofthe nest sites and remaining debris (Hernandez et al. 1997). Clutch size was determined as the number of disfinct eggs (based on whole eggs or remaining pieces) remaining in the nest. Hatching success was calculated as the number of eggs hatched as a percentage of clutch size. We used repeated telemetry locations to determine that females were still on the nest, and then used Pollock (1989) to determine nest survival for the entire nesting period. Nests were considered successful if > 1 egg hatched. Poults for each female were counted by visual verification as soon after hatching as possible. Poult counts were conducted < every 2 days until fledging. Two-tailed, paired t-tests were used to evaluate differences in nest site selection between successful and unsuccessful nests. Height and density of vegetation were measured using a cover board observed from 5 m away in each cardinal direction from each nest locafion. The cover board was 2 m in height, with alternating black and white blocks, each measuring 305 mm). Height and density of cover was also measured at 4 points, 22 1 in each cardinal direction at 10 m from the nest. We compared height of nest cover, average visual obstruction, and the average visual obstruction 10 m from the nest for successful and unsuccessful nests with Students' Mests. We also used a 2-tailed, paired Mest to compare soil temperatures at 25.4 mm for successful and unsuccessful nests. We used the GIS to determine average distance from nests to water and known roost locations for successful and unsuccessful nests and tested for differences with Students' f-tests. Results Unusual weather patterns occurred for the southern coastal plain of Texas in 2002. Although rainfall normally occurs during all seasons, the largest amounts usually fall in late summer and early fall hurricane season. During 2002 we recorded only 108 mm of rainfl-omJanuary 26*^ to June 30'^ and 1,171 mm of rain from July 1 ' to December 13"^. Our monthly averages differ from the averages of previous 32 years. Rainfall was lower than average January to June, and greater than average July through December (R^ = 0.2112, y = 2.385 - 0.0054X) (Figure 3.1). We trapped 42 Rio Grande turkeys (38 females, 4 males). Of these we outfitted 29 female turkeys with backpack style transmitters (25 adults, 4 juveniles). We recorded 1,192 locations of transmittered females between 21 January 2002 and 13 December 2002. We triangulated the position of 176 known transmitter locations to compute the average telemetry error, 7.9 m ( 3.9 m). 23 We recorded 1,192 locations on 29 females to assess survival rates. One individual was censored because of trapping mortality; however, 8 females survived to the end of the study (Mean annual survival rate = 0.253) (Figure 3.2) and (Appendix B). Nineteen ofthe 21 (90.4%) mortalities occurred during the spring and summer seasons and the mortalities were inversely correlated to average monthly rainfall (r = -0.460) (Figure 3.3). Predators were responsible for 17 mortalities, 1 individual was poached, 1 died of apparent disease, and 2 died of unknown causes. Primary predators were coyotes and bobcats {Fells rufus), although feral hogs {Sus scrofa) and alligators {Alligator mississippiensis) did contribute to the mortalities (Table 3.1). A total of 8 vegetation types were distinguishable on the DOQQs and verifiable for ground truthing. We inspected 2,785 ground truthing points. Of these 93.9% were visually verified as mapped. Misclassified points (4.23%) were largely at the edges of mapped types. The remaining misclassified points (1.84%) were corrected on our maps. The proportion of each vegetation type occurring within the study area and the proportion of locations expected in each type were calculated (Table 3.2). We were able to calculate home ranges (FK) for 23 ofthe 29 telemetered birds. Of the other 6, 1 was censored because of capture related mortality, and for the others we were not able to collect enough locations to calculate the home ranges. The remaining birds averaged 49.7 locations for calculating home ranges. Calculating the optimal number of locations for calculating home range size, we determined that with s 45 locations we accounted for > 87% of the total 24 home range size (Figure 3.4). Therefore, we only used home ranges for those individuals for which we had > 45 locations. Home range size averaged 1,968,6 ha ( 814.3 ha), with a range of 564.3 to 3653.0 hectares (A/ = 23). We compared (Johnson [1980] second-order of selecfion) proportion of vegetation types within turkey home ranges to proportion of vegetation types available over the entire study area. Females selected (p < 0.001) for riparian types and against brushy habitats, like chaparral, huisache, and mixed brush. Other types were used in proportion to their availability (Table 3.3). We did not record the high roost fidelity recorded for other populations. We examined seasonal vegetation type selection (Johnson [1980] thirdorder of selection) by comparing female locations during a season to the proportion of vegetation type available within the home ranges. This analysis indicated that patterns were the same except that during winter (December and January) and spring (February to April) huisache was used in proportion to its availability (Table 3.4, Table 3.5). In summer (May to August), the vegetation types were used as they were for the whole year (Table 3.6). Finally, in autumn (September to November) chaparral was used proportional to availability (Table 3.7). As the majority of the mortalities occurred during the spring and summer, we compared habitat use versus availability for surviving females and those that were killed by predators. In spring the surviving birds selected riparian woodlands, over the mixed brush, and used everything else in proportion to availability (Table 3.8). The females that were killed in the spring used all ofthe 25 vegetation types in proportion to their availability, except for the mowed areas, which they selected over other types (Table 3,9). During the summer, surviving females used all types in proportion to availability, except the riparian woodlands that were selected for and the mixed brush, which was selected against (Table 3.10). The birds that did not survive the summer selected for the riparian woodlands and the mowed areas, against the mixed brush, and used all other types in proportion to availability (Table 3.11) Females began nesting 18 March 2002 and the last nest was initiated 19 July 2002. Nesting rate of monitored females was 84.6% (22 of 26 alive at the start ofthe nesting period). Many females (73%) nested > 2 times (Figure 3.5). First nest attempts were primarily made in the brush vegetation types and were mostly unsuccessful {N = 20). The successful first attempts {N = 2) were in the riparian vegetation types. Subsequent attempts were made primarily in the other habitats. First nesting attempts were successful 36.4% of the time {N = 4) (Table 3.12). Successful nests were located in the non-brush habitats 81.8% ofthe time (A/=9) (Table 3.13). Nest sites were categorized into the 3 broad vegetation types: brush, riparian, and other. Brush was dominated by honey mesquite {Prosopis glandulosa), huisache {Acacia famesiana), granjeno {Celtis pallida), or agarito {Berberis trifoliolata). The riparian areas included Texas persimmon {Diospyros texana), blackbrush acacia {Acacia rigidula), and brasil {Condalia obovata) with lateral screening cover provided by vines such as old man's beard {Clematis drummondii) and mustang grape {Vitis mustangensis), and residual grasses flrom 26 previous growing seasons. The vegetative type other, was dominated by herbaceous cover like mustang grape and old man's beard, as well as Canada wildrye {Elymus canadensis), silverieaf sunflower {Helianthus argophyllus), switchgrass panicum {Panicum virgatum), and spiny aster {Aster spinosus). Nesting cover was > 40 cm in height at all ofthe riparian locations {x = 41.5 cm, 3.2 cm), 60% ofthe other locations {x = 42.8 cm, 8.3 cm), but exceeded 40 cm at only 30% ofthe brush locations {x = 36.9 cm, 3.2 cm). Average visual obstruction at all nests was 59.56% ( 17.9%). Percent of visual obstruction at nests differed (f= -2.966, d.f = 40, p = 0.0051) between successful nests (A/ = 11) (x = 47.95%, 13.4%) and at unsuccessful nests (A/ = 31) (x = 64.38% ( 16.5%). Visual obstruction provided by vegetation in the vicinity (at 10 m away in each of the 4 cardinal directions) ofthe actual nest was less for (t = 2.274, d.f = 40, p = 0.0284) successful nests (x = 49.32%, 16.5%) and (x = 64.38% 19.6%) for unsuccessful nests. Nests were distributed across the study area, but concentrated near permanent water sources. Successful nests (x = 0.34 km, 0.13 km) differed from unsuccessful nests in that they were, on average, located closer to water and to known roosts (x = 0.41 km, = 0.20 km) in the average distance to water (t = 1.095, d.f = 40, p = 0.2802). Difference was also detected for the average difference from known roosts for successful nests (x = 0.70 km, 0.18 km) and unsuccessful nests (x = 0.83 0.241). Nests were primarily located in pastures that were either deferred from grazing or in light, continuously grazed pastures. 27 We documented females (A/ = 11) that hatched poults (A/ = 54), but only 8 successfully raised poults to fledge (A/ = 30). The number of poults associated with each female was recorded as the last number visually verified prior to the female and poults joining other individuals. This results in a poultfemale ratio of 1.36:1. After the females and poults joined groups of other birds, counts were recorded as the total number of hens and total number of poults, resulfing in a poult:female ratio of 1.06:1. In the 9 years of turkey research conducted on the WWR, this marks only the third recorded instance of a poult:female ratio >1 (Table 3.15). Additionally, 1 female raised 2 successful broods, hatching 3 and 8 poults and fledging 0 and 6 fledglings, respectively. Average clutch size (A/ = 11 nests) was 10 eggs, (range from 5 to 14). Weekly and overall nest survival rates were calculated (Table 3.16). Average weekly survival rate was 20.7% and overall was 8% (Table 3.17). Ofthe 42 nesting attempts 11 nests successfully hatched, Raccoons {Procyon lotor) were primarily responsible for depredating nests, but snakes, skunks {Mephitis mephitis), and coyotes {Canis latrans) were other nest predators (Table 3.18). Eight females raised poults (A/ = 30) to fledge. We were unable to ascertain the cause of these poult mortalities. Discussion Female Rio Grande turkeys in this study had lower survival rate on the WWR and tiie Rooke Ranch than reported in Watts (1968) and Ransom (1987). Mosby (1967) reported a 40% survival rate in a hunted population, but little 28 information exists on survival rates on non-hunted populations. Survival was correlated to the average monthly as rainfall, the majority of female mortalities occurred before the rains started in early July, Predation by coyote and bobcat was the major cause of adult female mortalities. Watts (1968), Baker (1979), and Ransom (1984) also found predation rates to be high. The Rio Grande turi<eys on the WWR and the Rooke Ranch selected for riparian woodlands and grasslands and against mixed brush, chaparral, and huisache. However, brushy vegetation types accounted for over 70% of the total available vegetation types calculated within the birds home ranges. During winter, the roosts are of high importance, although with the brevity of winter in south Texas, we did not note the high roost fidelity noted with other populations (Haucke 1975), although the birds did select for theriparianareas. The spring season vegetation type use was dominated by the breeding season, and the summer by nesting and brood rearing. The fall use was the transition between the summer brood rearing and the tradifional winter roosts. Females nested in brush,riparian,and other vegetation types in proportion to their availability. However, first nesting attempts were mostiy in brush and were largely unsuccessful. This may be the result of the lack of rainfall eariy in the year, resulting in brushy sites providing the only visual obstruction for eariy nesting females. Subsequent nesting attempts were primarily in the riparian and other vegetation types. There was an unusual number of second attempts in the other, primarily herbaceous cover, vegetation 29 type. These were mainly nests located in dry wetlands. Most of these attempts, however, failed as they werefloodedwhen the rains began in eariy July. Aberrant weather patterns observed during 2002 may be responsible for multiple nesting attempts. This propensity towards multiple nesting attempts is something that has not been previously reported for Rio Grande turkeys in this area (Keegan and Crawford 1993). The third and fourth attempts were primarily made after the initiation ofthe rain. Also, previously unreported is 1 hen successfully hatching 2 broods. The average clutch size (10) was consistent with previously reported on the WWR (Ransom 1987) and well within the range reported for wild turkeys (Mosby and Handley 1943, Mosby 1967). The height of the vegetation at the nest site, the amount of visual obstruction at the nest, and in the surrounding area was related to nest success. Successful nests were located in areas with less obstruction (mostly brush) and had less visual obstruction at the nest itself than the unsuccessful nests. Nests located in the brush habitats had a higher depredation rate, and although 3 females did hatch poults none of these females were able to raise fledglings. The only poults that survived to thefledglingstage were those that were raised in the riparian and other habitats. We could not adequately quantify poult mortality, but we do feel that predation on poults is limiting recruitment. Based on poult:female ratios, this year's recruitment rate would have resulted in a population increase. However, the recruitment ratesfluctuatewidely (Watts 1968, Smith 1977, Baker 1979, Ransom 1984). This may be largely caused by the highly variable weather conditions. 30 Female Rio Grande turkeys in this study avoided brushy vegetative types, apparentiy caused by higher risks of predation. However, because of the usually dry conditions during the eariy nesting period (March to June), nesting hens did use brushy types as cover for nesting. These eariy nests were not successful (only 4.9% hatched) and no poultsfledgedfrom any of these nests. Furthermore, 10% (19 of 21) ofthe adult female mortalities occurred during this period. Nests were more successful (48.7%) in non-riparian brush vegetative types and 61 % of poults successfully fledged. Riparian vegetation types provided the best nesting success. Management Implications Brush vegetation types dominated by mesquite, huisache, and chaparral species with > 64% visual obstruction is not good vegetafion types for female Rio Grande turkeys on the southern coastal plain of Texas. The high proportion of brush may also function as an ecological trap (Gore 1973) increasing adult female mortality and decreasing poult survival during dry nesting seasons when brushy vegetation may provide the only semblance of nesting cover available. Management for female Rio Grande turkeys in the southern coastal plain of Texas should focus on decreasing the amount of brush encroachment in remaining habitat and enhanceriparianhabitats. 31 Acknowledgements The Rob and Bessie Welder Wildlife Refuge (WWR), Sinton, Texas and the Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX providedfinancialsupport. We gratefully acknowledge the assistance of the faculty and other students of the department at Texas Tech, as well as the students and staff at WWR. We would like to thank Jeff Rooke and Leonard Piland ofthe Rooke Ranch, Kenneth Hodges and Hugh Thomas of the Aldrete Ranch, Alfred Neiman, and the Devarry-Durril Foundation for access to their property. 32 Literature Cited Baker, B. W. 1978. Ecological factors affecting wild turkey nest predation on south Texas rangeland. Proceedings of the Annual Conference of Southeastern Fish and Wildlife Agencies 32:126-136. 1979. Habitat use, productivity and nest predation of Rio Grande turkeys. Ph.D. Dissertation. Texas A&M University, College Station, Texas. 46 pp. Cook, R. L. 1973. A census technique for the Rio Grande turkey. Pages 279-283 in G. C. Anderson and H. C. Schultz, editors, Wild Turkey management: current problems and programs. Columbia: The Missouri Chapter of the Wildlife Society and University of Missouri Press. 355 pp. Davis, B. D. 1994. Breeding chronology of Rio Grande turkey females. W-126-R, Job Number 7.07, Texas Parks and Wildlife Department. Texas Federal Aid to Wildlife Restoration Report. Austin, Texas. Drawe, D. L., A. D. Chamrad, and T. W. Box. 1978. Plant communities ofthe Welder Wildlife Refuge. Contribution Number 5, Ser. B, Rev., Welder Wildlife Foundation, Sinton, Texas. 38 pp. Ecological Software Solutions. 1999. LOAS location of a signal. Ecological Software Solutions, Sacramento, California. Glazener, W.C. 1967. Management of the Rio Grande turkey. Pages 453-492 /n O.H. Hewitt, editor. The wild turkey and its management. The Wildlife Society., Washington D.C. , D. Ransom Jr., J. R. Cary, and O. J. Rongstad. 1990. Demographic analysis of a Rio Grande turkey populafion. Southwestern Naturalist 35:23-27. Haucke, H. H. 1975. Winter roost characteristics ofthe Rio Grande turkey in south Texas. Proceedings ofthe National Wild Turkey Symposium 3:164169. Hernandez, F., D. Rollins, and R. Cantu. 1997. Evaluating evidence to identify ground-nest predators in west Texas. Wildlife Society Bulletin.25:826-831. Johnson, D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61(1):65-71. Keegan, T. W. and J. A. Crawford. 1993. Renesting by Rio Grande turkeys after brood loss. Journal of Wildlife Management 57:220-230. 33 Mosby, H. S., and C. 0 . Handley. 1943. The wild turkey in Virginia: it's status, life history, and management. Virginia Commission on Game and Inland Fisheries, Richmond, Virginia. 281 pp. Mosby, H. S. 1967. Population dynamics. Pages 113-136 in O.H. Hewitt, editor. The wild turkey and its management. The Wildlife Society., Washington D.C, Neu, C. W., C. R. Byers, J. M. Peek, and V. Boy. 1974. A technique for analysis of utilization-availability data. Journal of Wildlife Management 38:541-545 Pollock, K. H., S. R. Winterstein, C. M. Bunck, and P. D. Curtis. 1989. Survival analysis in telemetry studies: the staggered entry design. Journal of Wildlife Management 53:7-15, Ransom Jr., D. 1984. Nesting ecology of Rio Grande turkeys on the Welder Wildlife Foundation. M.S. Thesis. Texas A&M Univ., College Station, Texas. 35 pp. , O. J. Rongstad, and D. H. Rusch. 1987. Nesting ecology of Rio Grande turkeys. Journal of Wildlife Management 51:435-439. Scrifes, C. J. 1979. Brush management: principles and practices for Texas and the southwest. Ph.D. Dissertation. Texas A&M University, College Station, Texas. 360 pp Smith. D. M. 1977. The social organization of Rio Grande turkeys in a declining population. Ph.D. Dissertation. Utah State Univ., Logan, Utah, 98 pp. Thomas, J. W., C. V. Hoozer, and R. G., Marburger. 1966. Wintering concentrations and seasonal shifts in range of the Rio Grande turkey. Journal of Wildlife Management 30:34-49. U.S. DepL of Commerce, 1956-2000. Climatological Data, Texas. National Climate Data Center, Asheville, NC, Volumes 61(1) -105(5). Vangilder, L. D. and E. W. Kurzejeski. 1995. Population ecology ofthe eastern wild turkey in northern Missouri. Wildlife Monographs 130:1-50. Watts, C. R. 1968. Rio Grande turkeys in the matting season. Transcripts of the North American Wildlife Conference 33:205-210. . 1969. The social organization of wild turkeys on the Welder Wildlife Refuge. Ph.D. Dissertation. Utah State University, Logan, Utah. 60 pp. 34 Williams, L. E., Jr. 1961, Notes on wing molt in the yeariing turkey. Journal of Wildlife Management 25:439-440. , D. H. Austin, T. E, Peoples, and R. W. Phillips. 1971. Laying data and nesting behavior of wild turkeys. Proceedings of the Conference of Southeastern Association of Game and Fish Commissions 25:90-106. White, G. C. and R. A. Garrott. 1990. Analysis of wildlife radio-tracking data. Academic Press, San Diego, California, 383 pp. 35 Table 3.1 Cause of mortalities for radio-transmittered Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. Season^ Predator Winter Spring 4 3 0 0 1 0 0 8 Summer Autumn Total Coyote Bobcat Feral Hogs Alligator Disease Poaching Unknown Totals 0 0 0 0 0 0 0 0 3 3 2 1 0 1 1 11 1 0 0 0 0 0 1 2 8 6 2 1 1 1 2 21 ^ Seasons were defined as: Winter - December to January, Spring - February to April, Summer - May to August, and Autumn - September to November 36 Table 3.2 Proportion of each vegetation class and the expected number of Rio Grande female turkey locations in each class and confidence intervals (CI) for the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. Vegetation Class Percentage of Study Area Expected # of Locations Lower CI Upper CI chaparral grasslands huisache mixed brush mowed areas riparian grasslands riparian woodlands wetiands 13 2 16 47 1 3 14 156 24 192 558 12 37 0.081 -0.057 0.078 0.383 -0.086 -0.105 0.180 0.097 0.245 0.553 0.106 0.167 166 -0.037 0.316 4 47 -0.011 0.090 37 <U C N c o c o C O C O C O CD CM c 01 X CO TO C3) (0 C D 13 D" C D c C D D) C D C C D U) C D C D 13 CT C D CD CT C D ^ CO (U c :3 o 0) Q. h~ CN i D O lO CO CN CN , CN CO CD CN o a> o CD CD CO CD O <!r CD d (D Q O D . 0) oQ^ to C CD "> o 1^ CN O O O O o ^.2 o CO CD I o cb o CO o "^ o CD o CD CO CD o d 1 0) o C i: 8 8 ' CD Q. c o o c < C o D o CO sc: C ^ O B .o o CO ^ o CD CO CD O CM CO CO C3 3 ll il iS c Si x> - -^ (0 o o c "S G o X "S CJ CD c o CD T UJ o CO C M Oi ^ "t r^ co in CO co CD 0) (0 > CO O 0) o fo E > >.a: (0 OJ CO o O C O o CN O O 1^ O Qi.E . C D c O) O ID O o o CD o CO o CM O o o CD CD 13 m ] 0) 5 c g c o o O. CO o ^ O C M - O o CO o o CO o cx> CN CD CO o CO o CO T3 CL ^ CO c g CO ' ^ CD 1_ o _g) (0 "S CD CO o. C D CL C D _CD CO CO c CD a CD "o x : T3 ^ CO o T3 CO CD i_ (0 o 'D $ O CD 0) C _CD .2 To C g D E ro "5 38 CN O c g o CD CO C O T C O c . 2 CO CD ^ TO C3) C D (D D cr C D "D C D DC D c C D C3 3 C D CD D CT CD CD D CT CD c CD C D CO CD (D g C 0 QL D O 05 O CD CD o d 05 CM CN CN O) CN CO CX) O 0 cz^ 2 d ,029 O O) ^ ^ _C0 Q) .016 ,006 CD 0.011 CN 5o CD Q : CO o 1 CD CD o dI o d lO O) C3) CN O c o 8I 0 ) CD C o D 3 O < o CN CO in CM 00 CD CN 1^ CD S CO JD = > CD <D (D O 01 O CD Q: O CD > o c "G g <i> -Si P 0 CD CO S : CD -' CL JO LU ^ CD O CN in Ci N Q. C D O O) > CD > CD c: *- .^ ^ = i_ o to E >> ^ ^ (D..2 O CO o CO o 12 '.i= CD 01.2 .2 C D C O) o CN o O CD o d d o d o CO o o o -^ o CD CD - * CJ) <D > -^ u. C C O D J3 CD o o rr CO o o o o D ID CO o CO o in o CJ> co o hd o -^ o >i? > C C O CD d D C C CD Q_ C ^ CO _g> JQ g ro 2 CD CD CD nds _(D JtD (D i_ o c CO CO CO T3 CO C O D C D J= " 2 C O D E o CO H CD 05 CD C3) C L a o CD CO "o sz o 'D CF) Eh 39 T3 0) $ O E CO CD 0) CO .2 To Iff c Q. O ^ . - (D 0 *= l_ ^ CN c g C O T C O c CD T> C CD CO CD X CD CD CO C O O) C O C D D ID cr C D D CT C D c C D C3) C O (0 D CT 0 CD D .0 cr 0 ICD Q. Q . II 0 i_ O 3 O o o CD o d CN CN CM CN o CD o CO d d o d CO in in CN C % O 037 CD O .2 c c <o g 8o (D CD o CO o o d o d o o 209 0 TO a : o CO o CD o CO o 1^ CN O 0 d 1^ S T3 C CO CD CO o o c -CD 8 < o CO c x: o m o a> in CD 05 o CO CN C M > < CD = o: Jo ^^ |i Q. C D O C3) o <i> "S .2 .2 "C c Q . to LU ^ o c in h- TT- 1CD in CM CD in CD CM .2 CD CO Q: CD * r > C D o ro E >* CO CD : = ? c ^ J3 CD CL.2 O t5:.2 .2 C D C O) 0 "^ T- 0 0 CM O r^ r- o o d o CO o CN o d o o ^ CD - * O ) 0) < -2 D > c CJ) E 5 QL X : o CO O o o T- o CO o CO o o CO CO c a CO o CO d c g "S C D > C3) Q . CD > , d CO o x: o CO CO 0 o d CO o in o d CO D ID 0 CO D C C CO CO (0 in E 2 H D T3 T3 J= CO E o 0 x: o CO a CO c JO CO CO o x: TD 0 T3 CO 5 E CD CO CO (0 E 'D x: Eh 40 O 2^ as L: C3) C O o c JO "0 2 o ^ 0 c o C O C O C O CN B. CO J? (D X 0 CO 0 0 CO c D C CO C O CJ) C O (0 D c cr 0 C O D) C O C D CJ) C D c To D CJ" 0 (0 D CJ0 o g ,109 221 160 0 Q. Q . _ 249 CM CO O CO CO o g 01 O 0 ID O o o o d in CJ) CD O d iJ "D "c CD c 9 ? -o J 5 CM 8 c -* o o d o o dI CN O CN O d o d ,037 236 o in o O) CM O o d i;= 0 "O *i_ .4-J C Q. C D C (D d c x:" >, o c z^z (0 X3 C O T3 C O C C D 8 < o C D C O O D O^ O Vfr- tion O C O CO o in CM CO O) in 00 1^ oo eg O) a: ^ o C O C O o c "S C T3 0 c t3 g CL (D CN CO CD O) CM in in 00 O. CO Q Q . CJ) 2 0 UJ ^ 0 0 Q^ > C O o o E > O D CL5 g 0) '^^ CJ)> 0 0 0 ^ O CO o sr o CM O O O o O o CO o o CM Q:.2 d d o "^ o d .2 0) o O > 0 C O CO 0 E E D o CO o o d TJ o CO o in o d o o T ^ CM O o d d co C O o C D c "O C O D c CO _0 X3 CO .0 C O (0 Q. CD jD CO CO CO I c x: o CD CO 0 5^ "0 0 C3) Ci. TD .c 0 O C O CD 0 i_ CO .12 w C x: o D CJ) Eh 41 E CJ) Q. 9 ^ ^ 9 ? ^ nds C O CO To CD E g m CM g C O C O D,2 0 CO CD i _ C O D CT 0 C O D CT 0 C D CJ) C D c C C O CJ) C D (0 D CT 0 D CT 0 CD g CO 0 0 CL 01 c D O) o CD in CO CN o CD CD CO 3 0 CI d in O) O d d o CM o dI CJ) i2v2 C 0 D 0 5o ^ C O C o D o d CM o o I o d CM in o o ""it o CJ) CJ) I CJ) CM d d o XI IS 8 T3 (0 si CO < o in CN o CM CM o CO CO 'go. 0 c O C O TJ CO >> c JO (1) 1 o - -' 0 TJ 0 o c CO c o .g CL ro CO CM CO 00 CM CO C M in o CN UJ o I? O C O o 0 CL D .2 0 CDQ^ > C O o C O E >> O CO CD 1^ C 0 D 0 < 0 0 i > D |t 0 TJ Q..2 0 o CO o CM O o r^ T- o Q:.2 o d o CO o o CM o ^r o d 0 D ID D C C C O C O c CJ) o c CO o 0 C3 S d o CD d o o d o CM Q. x: o d o O) o d C O o c CO CO r^ o o d o oo o co C O o sr o d 3 D < CO J0 XI C O i par iras 5^ CJ) CD. .c o CD L- CO etI ^ C O _0 C D "^ g 0 0 c 2 CO Q. CO (0 'D 0 ^ CO CO CJ) x: E.^ 42 o a> E CD C O g C CJ) Q ,- iff JO JZ 0 TJ CO CJ o "O x : TJ cD C c C O C O tn TJ o L_ O C l_ E 0 cD C CO QL 0 0hCD co" c g C O 0 CO (D D E.g> ^ cr 0 C O D CT 0 C O D CT 0 C C O D) (0 C O D 0 CD cr D cr 0 C O D CT 0 ^o 235 415 109 160 329 144 D O) C ^ O 0 CL Oct :Z"T) JO g C o O 3 O o o o o d d d d -0.024 -0.029 -0.020 Lower CI 0.005 0.105 ts 0 T C D in 0 CJ) 00 0 CJ) CM O .2 8^ ^ CO 0 r-" o C O d C O 8-2 0 < c CO CM CD CM CO 16 S 3 0 8DC TJ 0 ^ 0 T T ' m CO r^ ' 00 in o CJ) C 2 O JD i CD CO 0 TJ CO 0 c c O 0 CJ) C 01 O =: o t) UJ tj g 0 '> Q. CD X 0 ^ -siCD CJ) h- in o CM CO CM in 2 ^ > C D O C O 0:5 *. CD DI CL.d O C O o o CM O o O c C o O 0 0 | i.2 0 c d d o o d CO o CM o o o d ^ g c CO CJ) I- o 0 o CM E <u cvi Q.TJ O Jr C D Q_ x : CD O CO o h~ O o o CD CM o o o a> o ^ CM d C O c g jD "0 0 CJ) C2. 2 x: o CD CL CD TJ _0 X3 C O H- CO CO CO CJ) par iras c JO x: o CD CO "D 0 TJ TJ C O CO O 0 0 TJ c=o CO CO c C D C c C O C O g x: Eh 43 E i_ CD CJ) 069 o CM o TJ CO 391 c JD 0 5 0 CO CO X 0 g co CO 0 is _0 0 CO C O D CT 0 C O D CT 0 C D D CT 0 C O D CJ" 0 C O D CT 0 C D D C7 0 C D D C7 0 Q: O .5 o I CJ) IQ: 0 in CO CM ^ ^ ^ ~ a. Q. 13 O o d in T CD T O) d d ^r d o d T C> J CM CO ,,_ O) CO C3) CD O d d d as J2 2 (0 ' i : 0 in 1*= TJ CO c x: o 0 c C> c 0 TJ C CO D ^O O C C 3 ^ c tion 8 C CO CO o _l $ o __ C O o o d O CD d I CJ) CM O T in o O) 00 o CJ) CM O 8o o 0 .^ o o < o c CD in CN CO CO CM CM 01 x: X3 JD TJ CD C CD > CD 0 CJ) "S " S LU g CO J2 C 0 .OQ: Q. ^ X X CM CD 1- in ^r CO c ,_ 0 TJ > CD 0 c 2? O C O E > ' O C O S3 0 > o la-I- CN o o r^ <- o o d CO CO 0.9J $ 01.2 .2 C D c O) o E 2 o o d o CO o o CM o o d C D CJ) c: c CL o E '^ CM > o 8.^ Q. x: c g ' 4 > o CM T- o CD O o 1^ O o CD CM o o O CJ) o CM o CM o CO TJ d TJ CO CO * - X3 CD ^ ^ c TO 2 I TJ CO c JD i_ "0 0 CJ) O. CD c^ C D JZ CO CO CO CJ) x: o 'D 0 CO CO TJ SZ I- o x: Eh 44 TJ 0 C O CO O 0 ^ 10 .5 ^ C D T3 o CO E C D C CJ) c JD "0 a: C O n \ CO i> X CO OJ C g o 0 CO E"^ C O CO- C D D ^.^ .2 c cr 0 C D D CJ0 C D D CT 0 c 'co CJ) C D cr 0 CD D CO D cr 0 (0 D 0 cr I8 C O O t- :3 p ^ .P**V CJ) 0 0 CL CD. o CC ""^TJ C C O 3 O CM CM C M C C O O s CM CO o d ^ o CJ) o CO CD CM C M t 0 ^ o o CO d 0 d d m o d 8 c o J ^ ^ O C O c 0 C D TJ CO C TJ CD C 01 C 0) O o 5 o c ID ^D C o D JO c c d o o o dI CM 1^ o d o d CJ) o CO i^ CO o d1 o o d ^ in CM d o d = o X3 O JD Q: CD 0 o o < o c 0 .4-J > CD TJ C CD C c m CO CD in -"^ tOO in 00 CO g o 0 CJ) D " 2 C C L CO 0 X g UJ .2 m > C O 2^ 1 0 CO in in in CO CJ) CO CO > c 2 O CO > 0 C ^ E > O D O C O o <r 2 0 tD 0 c D) O 5 C O <0 o CM o o o d a: .2 .2 0 c O) C 3 o o d o CO o o CM o o ^ 0 ^ E 2 8.^ o t g CO CO Q. CD o in o CO o -sr o in CM o O o CM d O o d CO d CO o o d CO TJ CO C C D M ;;5o.2 X) CD 4 ' C D 0 0 CJ) Q. c JO CO CO TJ x: o 0 TJ x: o CD ICJ) CD CO 3 CJ .c ^ ^ TJ 0 CO $ CD TJ c JD C g O C CJ) Eh 45 O 2 E CD "0 5 0 ^ TJ X C 0 cotCD co- c g o JD 0 CO C O D C O D ^ JD o (0 O ^ O ) cr 0 0 cr C O D CT 0 C O C (0 O) (0 C O D 0 CD cr cr 0 D E o ai ^01 0 CL Q._ Z30 263 ^Y JO C D C o O 0 J3 0 o CJ) o .127 .221 403 CO CD CM CN in 0 0 o 0 0 gx:c o 8 C D TJQ^ 0 CO CO c o 0 0 1^ CM h-- hCJ) 0 d1 0 CO o 0 .2 .2 < o m CD C O O d 1 d d o d I CO in o d d o d I CM < .^8 1^ (0 . i ; JD 0 C -^ D S " o c TJ 0 CO m 00 CO CO 00 CD CO to C O C O ^ 0 CD 5 g cu -i^i X g LU C D si CO I- 00 CO > C O O CD CM CM CN CD CD in 02 D 0 E > .2 0 ^ &3 o C 0 o o CM O O ro c "0 o CJ) > 0 0 Q. o d o d o CO o o CM o o E o E D CO 0 CO c CJ) g c CD o 0 o t c g 0 0 CJ) QL o CO o CM o CO O o in CM o CO o O 0 hCM 0 T 0 ^CNi >>o 0 o d CO d TJ d CO * - J0 X} CD 2 x: o C D CL C D L_ c JD CO CO CO L. x: o to CO 0 TJ ^ TJ 0 C O I- C3) 3 x: Eh 46 5 o E CD 0 CD ai C CJ) JZ (0 Q. c i TJ 0 0 ^ ^ 0 ^ nds C O ^ ^CN TJ C O TJ Table 3.12 Successful and unsuccessful nests based on nesting attempt for Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. First Attempt Successful nests Unsuccessful Nests Totals 4 18 22 Other Attempts 7 11 18 47 Table 3.13 Successful and unsuccessful nests based on nest site location for Rio Grande turkeys on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. Brush Successful Nests Unsuccessful Nests Totals 2 20 22 Non-brush 9 11 20 48 Table 3.14 Rio Grande turkey poult:female ratios for 9 years of turkey research on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas. Year Poultfemale ratio Reference 1966 1970 1971 1972 1976 1977 1983 1984 2002 1.4 0.3 0.1 0.1 0.5 1.17 0.21 0.93 1.36 Watts, 1969 Smith, 1977 Smith, 1977 Smith, 1977 Baker, 1979 Baker, 1979 Ransom, 1987 Ransom, 1987 This study 49 CO 0 C D II -^ g O D) rn C O '-' 0 2^. .0 00 000 027 919 540 JO "5 C cu D C C D 8g c "is g -J o L CJ) CO CJ) CD CM o CO o CM - ^ T - o o d d d .0 00 .6 66 465 152 084 068 O C O O CO Exf co o co" CD DO 0 CL C2._ 039 033 o o "^ T o o o o o o o 0 "O T Q: CD 692 846 346 084 216 168 18 SI C O "co "O D CO - T o o o o o o o TJ 0 CD 0) C C D = 0 C CJ) D > ^ o a: 0 B li i_ TJ 0 0 0 C D 0 p X3 . 2 o in CD CM CO C M CD ^ 0 TD al S 0 CD "0 o x: 8c 0 ^ 0 o CO > o E CM CO CO CM CO c 0 JO _ Q. D CD 0 E ^ D. CJ) CO -I- 00 O O CJ) CM O) m E O CO XD CD CD CM 0 0 CJ) CM CO in CD 00 50 068 076 > .0 00 .0 00 0 - .^ 025 2 c c !> CO o o o 0 .6 07 o o d CD in o d CD in o d CD o o d CM CO o d in CN ^ r o d ^ in 't^ o d o d CD ^ O O CN O 3 0 0 Q. CL_ CO o d T- OO o d O o o d o o d o o J^ o o d CM CO CO CN ,^ o o d o o d o o d o o d o o d o 1 CD ,028 .2 08 .2 08 .017 014 008 008 008 d O in D CO o d o d o o o o o o d d d 0 ' *-> X) 1_ TJ ^ "O T E D L_ $ 0 Z c CM CM CM CO T3 0 0 ^ > E 7 D 0 CM CO O T- CM CD D _C TJ 0 0 X3 L_ E -^ D . CD CO o o in C CO in in 00 oo CD CO _0 X2 CD 0 5 CD ^ o CM CM CM CM CO CM "^ CM in CM CD CM hCM CO CM O) CM O CO 51 008 r^ m CO CO 008 > .0 03 CO \U\ OUJ 0 O CM O d 0 CD. CO o o IDO Q L _ CD > D CO o o d OO l_ TJ 0 ^-rf .CO 0 X3 E ^ D 0 ini ^^-f Z c N k - TJ 0 0 XJ > 0 jr TJ 4 o C O E ^ O x: o ^ O CM E D 0 CD 3 ^1o |8 < 5 D C O TJ 0 D C 0 X3 in CM E -^ D. c o o in CO z cc CM 1_ 1 CD I - oB ID o TJ o 0 CM CM TJ o TJ 0 a 0 j0 X) h- CD I 0 00 ^ LU CO 52 Table 3.16 Vegetation type and cause of Rio Grande turkey nest depredation on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. Predator Raccoon Coyote Snake Skunk Other Brush 8 3 3 4 4 Non-brush 3 2 1 2 1 Totals 11 5 4 6 5 53 Figure 3.1 Comparison of monthly average rainfall from 1969 to 2001 and 2002 on the Welder Wildlife Refuge, San Patricio County, Texas. 54 Figure 3.2 Kaplan-Meier estimate of Rio Grande turkey female survival rates by week (26 January 2002 to 13 December 2002) on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. 55 l 5 o 0 50 100 150 200 250 300 350 400 450 500 Average Monthly Rainfall (mm) Figure 3.3 Correlation of monthly rainfall averages (mm) and number of Rio Grande turkey female mortalities on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. 56 25000 20000 m 15000 01 a c ra a. E o X 10000 D) 01 s > < 5000 0) 10 20 30 40 50 60 70 80 90 Number of Locations Figure 3.4 Comparison of number of locations and home range sizes for radio -telemetered Rio Grande turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. 57 14 12 12 I. 1 0 E l# of 1st attempts (n = 22) I # of 2nd attempts (n = 15) I # of 3rd attempts {n = 4) < a z 01 8 01 ot .a I 4 ^ l# of 4th attempts (n = 1) h March April May Month June July Figure 3.5 Number of nesting attempts by month for transmittered Rio Grande Turkey females on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. 58 CHAPTER IV LAND USE CHANGE AND RIO GRANDE TURKEYS ON THE SOUTHERN COASTAL PLAIN OF TEXAS^ Abstract: Aerial photographs were obtained for 1969, 1975, 1982, 1989, and 1995 for the Welder Wildlife Refuge (WWR) and the Rooke Ranch in San Patricio and Refugio counties, Texas. Using ERDAS Imagine Software 8.3 and ARC/INFO these photographs were analyzed for proportional changes in 6 vegetation types. Brush vegetation increased by 52%, riparian vegetation declined by 13%, and grasslands declined by 32% on the WWR. On the Rooke Ranch, brush vegetation decreased by 2%, riparian vegetation decreased by 16%, and grasslands decreased by 14%. However, the Rooke Ranch converted 28% of available land to agricultural cropland, but no such conversion took place on WWR. Analysis of historical management techniques (1974-1995) showed that the WWR conducted 46 treatments, whereas the Rooke Ranch conducted 47 treatments. The average size of the treatments was larger on the Rooke Ranch. Rio Grande turkeys {Meleagris gallopavo intermedia) on the southern coastal plain of Texas select for riparian and against the brush vegetative types. The ^ This article was submitted to the Journal of Wildlife Management. The authors are as follows: Elizabeth J. Wilbanks, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, Texas 79415-2125, Email: bwilbank@earthlink.net; Mark C. Wallace, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, Texas 79415-2125; Phillip J. Zwank (deceased), Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, Texas 79415-2125; and D, Lynn Drawe, Rob and Bessie Welder Wildlife Foundation, PO Box 1400, Sinton, Texas 78387, 59 long-term loss of turkey habitat is a large factor in Rio Grande turkey population decline on the southern coastal plain of Texas. J. Wildl. Manage. 0(0):000-000 Kev Words: aerial photographs, GIS, habitat selection, habitat use, Meleagris gallopavo intermedia. Introduction Cattle ranching in south Texas has been the primary land use since at least 1768 (Drawe et al. 1978). Little is known about the vegetative communities of that period, except for an account by Heller (1895) in which he noted the general lack of vegetation. In his opinion this was a result of the heavy overgrazing, a claim supported by photographs (Coffey 1909) of San Patricio County, Texas. This area was historically a grassland complex that occurred along the Gulf of Mexico. Box (1961) described the area as a coastal grassland climax, while Thomas (1975) described it as a tall grass prairie. Today, it has developed into a brush-grass complex with heavy, woody brush invading all but some deep sandy sites. Bray (1901, 1906), Young et al. (1948), and Allred and Mitchell (1955) suggested that there was considerable brush encroachment during the latter part of the 19*^ century. However, others are of the opinion that the brush has not moved into the area, but rather it has increased in density (Box 1959, Johnston 1963). 60 Woody vegetation increase has been noted at the WWR and by surrounding landowners (Drawe et al. 1978). Since 1970, ranchers in southem Texas have reduced the use of fire to control undesirable brush species such as honey mesquite {Prosopis glandulosa), huisache {Acacia famesiana) and lime prickly ash {Zanthoxylum fagara) (J. Cox, WWR, personal communication). Scrifes (1979) reported that in the mid-1970's brush eradication policies of ranchers had given way to an attitude of brush management. This new philosophy was to manage the woody vegetation on rangelands so as to optimize benefits for wildlife and livestock. During this time, the Rio Grande turkey population on the WWR declined from a high of approximately 700 birds (Watts 1968) to less than 100 birds in 1981 (L. Drawe, WWR, unpublished data). It has rebounded to a current estimate of 120 birds (T. Blankenship, WWR, unpublished data). The apparent population trend on the WWR has been attributed to decreased nesting attempts, nest failure, poult survival, and loss of habitat (Ransom et al. 1987, Glazener et al. 1990). This population has been fairiy well studied (Glazener 1967, Watts 1969, Smith 1977, Baker 1979, Ransom 1984, Ransom et al. 1987, Glazener et al. 1990), but has received little attention in recent years. Walker and Springs (1952) and Litton (1977) suggested that 50% brush should be left as cover for Rio Grande turkeys to reduce effects of human disturbances (Gore 1973). A moderate amount of brush cover is required, especially during brood rearing (DeArment 1959). Porter (1992) stated that good brood habitat for eastern wild turkeys (M. g. sylestris) is an area that provides 61 cover for the poults to hide in, but allows the adult female an unobstructed line of sight Areas that have high amounts of brush coverage were less desirable because of the inability of the birds to move through the area (Miller et al. 1999). Brush provides predators with excellent hiding cover and can increaserisksof predation for eastern turkeys (Miller and Leopold 1992). Areas that have received indiscriminant brush removal are perceived as poor turkey habitat (Walker 1951). These practices remove the intended brush, but often include roost sites as well as vital winter food sources. Glazener (1967) discusses one farmer who carefully protected roost trees, but removed all surrounding trees and shrubs, thereby destroying the habitat and resulting in the disappearance of his resident Rio Grande turkey population. We attempt to relate the Rio Grande turkey population decline on the southern coastal plain of Texas to changes in land use practices in the area. Our objective was to identify cover from available aerial photographs (1969,1972, 1982, 1989, and 1995) and to determine the changes over time in percent composition of vegetative types used by Rio Grande turkeys. The study was conducted on the WWR, a 3,158-ha private wildlife refuge in San Patricio County Texas and the adjacent Rooke Ranch, a 4,237-ha private ranch in Refugio County Texas north across the Aransas River from WWR. This area was historically part of grassland complex found along the Gulf of Mexico; however, in the past 35 years it has developed into a brush-grass complex (Drawe et al. 1979). These vegetation changes are thought to be largely caused by the history of grazing by domestic livestock and a reduction in brush control 62 (Drawe et al. 1978). Climate in this region was sub-tropical. Average rainfall was 89 cm, largely occurring in the late summer and eariy fall. Drawe et al. (1978) described a total of 16 plant communities and 22 soil classes, ranging from fine sands to heavy clays. Both WWR and the Rooke Ranch were working cattle ranches. WWR has been operated as a cow-calf operation since 1974 and used a 3 herd, 4 pasture deferred grazing system (J. Cox, WWR, personal communication). A lightgrazing regime (0.0791 AU per ha) was practiced on WWR. In recent years, the deferment has been dictated by research of summer burning being conducted on the property. The Rooke Ranch was also a cow-calf operation under a moderate (0.101 AU per ha), continuous grazing regime. Additionally 1,776 ha ofthe Rooke Ranch had been converted to farming, primarily cotton {Gossypium spp.) and grain sorghum {Sorghum vulgare). Hunting was prohibited on the WWR, and only the family was allowed to hunt, during the spring turkey season, on the Rooke Ranch, but minimal poaching was known to occur on both areas. Methods We obtained historical weather data for the area from WWR personnel (T. Blankenship, WWR, unpublished data) for comparisons overtime. Only rainfall data was available for the entire time period (1969-1995). We compared yearly rainfall totals. We obtained 1:24,000 U.S. Geological Survey (USGS) Digital Orthorectified Quarter Quadrangle Sheets (DOQQs) (Sinton East, Sinton West, 63 Cranell, Rincon, and Woodsboro quadrangles) from the Texas Natural Resource Information Service (TNRIS). The DOQQs were used for the 1995 photographs. Digital scans of older aerial photographs were obtained from the U.S. Department of Agriculture (USDA) Aerial Photography Field Office for the years 1969, 1975,1982, and 1989. The USDA photographs were ortho-rectified in a Geographic Information System (GIS) using ERDAS Imagine Software 8.3 (Imagine Software, Inc.). Easily identifiable, permanent reference points (A/ ^ 7 points per picture) were located in the field and geo-referenced on site using a Global Position System (GPS) (Eagle Map Guide Pro, Catoosa, OK). We used a total of 587 points to ortho-recitfy the images in the GIS. We delineated 6 vegetation types on each set of photographs: brush, riparian, grasslands, mowed areas, wetiands, and agricultural fields. These types were defined by historical transect data available from WWR (T. Blankenship, WWR, unpublished data) and which of these were identifiable on the photographs. Coverages were then transferred to ARC/INFO. The total amount and proportion of each vegetation type was calculated for each time period represented by a set of aerial photographs. We mapped vegetation from the 1995 photographs first because we could ground truth them. We then attempted to use the ERDAS algorithm to define vegetation types on the other sets of photographs. However, projected vegetation types were not accurate when compared to the historical transect data. Therefore the vegetation types were delineated by hand. We used regression analysis to quantify the changes in the vegetation type changes over time. 64 Rio Grande turkey habitat use was determined by telemetry during 2002 (January to December) (Wilbanks 2003). We compared use to availability (Neu et al. 1974) to determine important vegetation classes for Rio Grande turkeys (Wilbanks 2003). Rio Grande turkeys selected for riparian and against brushy vegetation types (Wilbanks 2003). Histories of land management practices were obtained from WWR (S. Glasscock, WWR, unpublished data). These included burning, oil, and herbicide treatments. The majority of these treatments were in conjunction with other research projects that had been conducted on WWR. Often these do not involve entire pastures, but rather delineated plots. Land management histories were also obtained from personnel on the Rooke Ranch (J. Rooke, Rooke Ranch, unpublished data). These treatments included burning, mechanical disturbance, and herbicide treatments. The Rooke Ranch maintains a burning program that involves deferred grazing for 1 year prior to burning, with a mechanical disturbance occurring during that year. Pasture sizes were not consistent throughout. Therefore, disturbance was recorded as the number events occurring within time periods coinciding with the photographs (1974-1975, 19761982, 1983-1989, 1990-1995). The average size of treatment area (ha) for the WWR and the Rooke Ranch could only be calculated for the entire time period (1974-1995). We also obtained the amount of each county (San Patricio and Refijgio) in agricultural land for 1969, 1975, 1982, 1989, and 1995 (D. Tinkler, Texas Cooperative Fish and Wildlife Agency, unpublished data). This was used to 65 generate the proportion of each county in agricultural land for each year. We used regression analysis to quantify the changes. Results We detected no trend or change in annual rainfall betiween 1969 and 1995 (Figure 4.1). Proportion of vegetation types on the WWR, determined from aerial photographs, changed for each ofthe 6 vegetation types (Table 4.1) resulting in 5 1 % more brush in 1995 than in 1969. Brush increased at an average rate of 2.0% per year. There was a concurrent 13% (averaging 0.5% per year) decline in the riparian vegetation. Grasslands also declined by a yeariy average of 1.4% or a total of 37%. Brush vegetation increased (slope = 13.1, R^ = 0.9802), whereas riparian and grassland vegetation types decreased (slope = -3.3, R^ = 0.9764 and slope = -9.9, R^ = 0.9641 respectively) (Figure 4.2). Changes also occurred on the Rooke Ranch (Table 4.3). Brush vegetation decreased by 2%, (0.07% decrease per year). Riparian vegetation decreased by 16%, (2.7% per year). Grasslands also declined by a yeariy average of 0.53% per year, or a total of 14%. Agricultural fields increased by 28% (0.8% per year). Brush, riparian, and grasslands vegetation types decreased (slope = -1.1, R^ = 0.3689; slope = -4.1, R^ = 0.9617; and slope = -2.6, R^ = 0.3428 respectively), whereas the agricultural fields increased (slope = 6.8, R^ = 0.7605) (Figure 4.3). 66 Based on analysis of 1995 DOQQs and telemetered Rio Grande turkey use of vegetation types during 2002 Rio Grande turkeys select for the riparian and against brush vegetation types (Wilbanks 2003). In 2002 female Rio Grande turkeys nested eariy in the brush habitats (90.9%, 20 of 22), but were mostly unsuccessful (90%, 18 of 20). Nesting in brush despite the poor success was attributed to lack of hertDaceous vegetative cover elsewhere, due to lack of precipitation in 2002. Subsequent nesting attempts were primarily made in the non-bmsh vegetation types (90%, 18 of 20) and were more successful (45%, 9 of 20). The majority (66.7%, 14 of 21) of Rio Grande female mortalities recorded in 2002 were located in the brush vegetation types. Brushy vegetation increased more {t - 3.746, d.f. = 8, p = 0.0057) on the WWR than on the Rooke Ranch (Figure 4.4). Significant differences were not detectable for either the grassland orriparianvegetation types. Additionally, 28% ofthe Rooke Ranch was converted to agricultural croplands betiween 1973 and 1981, while there has been no such change on WWR. Historical management data were not available prior to 1974 for the WWR; therefore all comparisons are made betiween 1975 and 1995. The Rooke Ranch has maintained a brush control regime throughout the past 70 years. They utilize burning and mechanical means to control the woody brush density on the property, but discontinued use of chemicals in 1976. Conversely, the WWR had largely abandoned brush control practices. Some fire, mechanical, and chemical treatments were conducted in small areas on the property in conjunction with other research occurring on the property. Since 1996 they have adopted a more 67 aggressive summer burning program, in conjunction with long-term prescribed fire research being conducted on the property. The numbers of treatments made on the WWR (N = 46) and on the Rooke Ranch (N = 47) were similar. However, the average size of treatment was larger {t = -2.531, d.f. = 91, p = 0.0131) on the Rooke (x = 1019.1 ha 716.7 ha) than on the WWR (x = 342.5 ha 1683,5 ha). In San Patricio County in 1969, 96% ofthe county was in agricultural land, while in 1995 it had decreased to 85% (Figure 4.3). Refugio County, in 1969, had 58% of the county in agricultural land, and now has 89%. In 1969, 73.6% of the total area was in agricultural land, and had increased to 87.4% by 1995. Discussion The changes in land use practices, particularly in brush management noted (Scrifes 1979) have impacted the landscape on the southern coastal plain of Texas. Rio Grande turkey populations on the southern coastal plain have declined as these changes have taken place. The loss of riparian and grasslands vegetation types that turkeys are selecfing for, and where the majority of the successful nesting occurs is negatively impacting the population. Furthermore, brushy vegetation types that turkeys select against have increased more than any other vegetation type. Brushy vegetation is also where the majority of predation occurred. Vegetative changes have been greater on the WWR than on the Rooke Ranch, seemingly due to more active brush control measures on the Rooke Ranch. 68 Management Implications The increase in the brush vegetation types on the southern coastal plains of Texas has altered the landscape. This change is a probable cause for the Rio Grande turkey population decline. The decrease ofthe riparian vegetation types that the birds select for and the increase in the brush vegetation types that they select against is likely a factor in the declining population. Furthermore, the lack of suitable habitat for Rio Grande turkeys is a large factor in the inability of the population to return to it's former size. Acknowledgements The Rob and Bessie Welder Wildlife Foundation (WWR), Sinton, TX provided financial support and the Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX provided logistical and financial support. We gratefully acknowledge the assistance of the faculty and other students of the department at Texas Tech, as well as the students and staff at WWR. We would like to thank Jeff Rooke and Leonard Piland of the Rooke Ranch, Kenneth Hodges and Hugh Thomas ofthe Aldrete Ranch, Alfred Neiman, and the Devarry-Durril Foundation for access to their property. 69 Literature Cited Allred, B. W. and H. C. Mitchell. 1955. Major plants types in Arkansas, Louisiana, Oklahoma, and Texas their relations to climate and soils. Texas Journal of Science 7:7-19. Box, T. W. 1959. Relationships between soils and vegetation of four plant communities on the Welder Wildlife Refuge in south Texas. Ph.D. Dissertation, Texas A&M University, College Station, Texas 100 pp. 1961. Relationships between plants and soils of four range communities in south Texas. Ecology 42:794-810. Bray, W. L. 1901. The ecological relationship ofthe vegetation of western Texas, Botanical Gazette 32:195-217 .1906. Distribution and adaptation ofthe vegetation of western Texas. University of Texas Bulletin. 82. 108pp. Coffey, G. L. 1909. Reconnaissance soil survey of south Texas. Pages 10291130, Map 33 in Field operations ofthe Bureau of Soils. U.S. Department of Agriculture. DeArment, R. D. 1959. Turkey hen-poult ratios as an index to reproductive trends. Proceedings ofthe National Wild Turkey Symposium 1:27-31. Drawe, D. L., A. D. Chamrad, and T. W. Box. 1978. Plant communities ofthe Welder Wildlife Refuge. Conti. No.5, Ser. B, Rev., Welder Wildlife Foundation, Sinton, Texas. 38 pp. Glazener, W.C. 1967. Management of the Rio Grande turkey. Pages 453-492 in O.H. Hewitt, editor. The wild turkey and its management. The Wildlife Society., Washington D.C. Heller, A. A. 1895. Botanical explorations in southern Texas. New Era Printing House, Lancaster, Pennsylvania. Johnson, D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61(1):65-71. Johnston, M. C. 1963. Past and present grasslands of south Texas and northeastern Mexico. Ecology 44:456-465. Litton, G. W. 1977. Food habits ofthe Rio Grande turkey in the Permian Basin of ' Texas. Austin: Texas Parks and Wildlife Technical Series No. 18. 22 pp. 70 Miller, D. A., G. A. Hurst, and B. D. Leopold. 1999. Habitat use of eastern wild turkeys in central Mississippi. Journal of Wildlife Management 63:210-222. Miller, J. E. and B, D. Leopold. 1992. Population influences: predators. Pages 119-128 /n J. G, Dickson, editor. The wild turkey: biology and management. Stackpole Books, Harrisburg, Pennsylvania. Neu, C. W., C. R, Byers, J. M. Peek, and V. Boy. 1974, A technique for analysis of utilization-availability data. Journal of Wildlife Management 38:541-545 Porter, W. F. 1992, Habitat Requirements. Pages 202-213 in J. G. Dickson, editor. The wild turkey: biology and management. Stackpole Books, Harrisburg, Pennsylvania. Scrifes, C. J. 1979. Brush management: principles and practices for Texas and the southwest Ph.D. Dissertation. Texas A&M University, College Station, Texas. 360 pp. Thomas, G. W. 1975. Texas plants - an ecological summary. Pages 7-14 in Texas Agricultural Experiment Station Miscellaneous Publication 585. Revised. Young, V. A., F. R. Andenwald, and W. G. McCully. 1948. Brush problems on Texas ranges. Texas Agricultural Experiment Station Miscellaneous Publication 21. 19 pp. Walker, E. A. 1951. Land use and wild turkeys. Texas Game and Fish 9(11): 1216. and A. J. Springs. 1952. Factors concerned with the success and failure of turkey transplants in Texas. Austin: Texas Game, Fish and Oyster Commission, Completion Report (P-R) Project W-44-R-2. 21 pp. Wilbanks, E. J. 2003. In Press. Habitat use by the Rio Grande turkey on the southern coastal plain of Texas. Master's Thesis. Texas Tech University, Lubbock, Texas. 71 Table 4.1 Change in proportional availability of vegetation types from 1969 to 1995 on the Welder Wildlife Refuge, San Patricio County, Texas. Year Mowed Brush Grasslands Areas Riparian Wetiands Ag Fields 1969 1975 1982 1989 1995 27 41 56 70 78 39 30 19 1 1 1 1 1 27 24 19 17 14 6 4 5 7 5 0 0 0 0 0 5 2 72 Table 4.2 Change in proportional availability of vegetation types from 1969 to 1995 on the Rooke Ranch, Refugio County, Texas. Mowed Brush Grasslands areas Ag Fields Year 1969 1975 1982 1989 1995 Riparian 37 32 26 23 21 Wetiands 21 23 17 16 19 36 19 20 21 22 1 1 1 1 1 5 9 8 11 0 16 28 28 28 9 73 a> a> at CO CA CD X Ol o> CO Ol C o O o o "TO OS CO 00 Ol ^ c C D C O (1) O] u> oo o> 3 0) CO Ol ra - >- ^ 00 Ol a> o < U TJ (U ^ C D JZ .t-^ > - 197 a> CO O at TIs. Ol lO CD ^ CO h. Ol a> o T" ^" r U Ol CJ) CO CD ^ Ol ^^ '*-^' ^ C D C C D L_ U. ^ n o> y^ l Ol >, C D C D ^ Ol > % C D Ui CD CD 1970 T" > < T ^f (uiui) ijeiuje^ |e)Oi C D L 3 D) 74 O) CD C3) , o i 1 / / L>* E o CO s s ^ CD CJ) c i3 CD 1 :/ 1 / o CI CD ! /'' fn^l \ >1 c If 1 > . C O >^ c C D (D D) CD 1976 < N 0.9 Ix. > n c ; CM , l\ 1 1 1 970 II c J O CO CO ; to CD C3) . ddAj. uonEi969A u e p e d i u | o u o | ( j o d o j d CD CD s CM ID c %o o n *-< C D 1 o Q_ <* C O O -J ses c CD C < T >; ! 3t 9 1 <r > T- Ti <3 f\1 1990 r _ C O CO >> D _ CD C C D in OB c > (M c c ! 't II ear o a: C 0) O C O M *^ reg k. 1980 > - o O D t C D TJ C D 5 1 at o C O LUOO ^ C U) C C D D Q.JZ o CM o SLI , o ^ o 5 o 80 o o n a M c ^ "^ u. C35 CD T s s s s a s s ? " " " " ' d A i uoi)ei86eA pueissaio fo uojModojd C D adAj. uoite^BaA Msnjg jo uoiuodojd ID C3) U_ 75 X <D , cn (O c* I w O A > / / ^ 4Aj| M i 6 A u i e d i | j | o o | ) * M i o j ^ C 2000 , o C4 CD .5- ( D - ce 1986 > Year 1990 I 80 in S 3 O C M I n 1996 1995 ^ CD <*! CD 1990 (D CO O O 1 ' - d ro a) CD c c g o CO I T ) / 11 1985 Year 5(? CD gj CT) 1980 1975 1980 } / *- s * d d to n D C CO 1975 X c 5 lA S o 3 m < N o w w . odAi. U0|iE)e6eA spuB(ssejo *o uo((iod(Ud CO C C35 O -c- ^ E CO o Q - ^ t C O O C D O CD CO TO 1970 1970 65 1965 ^ C D D CD O C N O . C* O * " sdAi uo(;E|aB3A gsiue V uoDJodojd 76 Figure 4.4 Comparison of changes of brush vegetation types (1969 to 1995) on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas. 77 ._ Proportion of County In Agricultural Land -A e O o O o ^ ^ San Patricio -O Refugio p o p o p o p o 1969 1975 1982 Year 1989 1995 Figure 4. 5 Comparison of the proportion of San Patricio and Refugio Counties, Texas being utilized as agricultural lands. 78 APPENDIX A TRAPPING RECORD FOR RIO GRANDE TURKEYS FROM JANUARY 2002 TO MARCH OF 2002 ON THE WELDER WILDLIFE REFUGE AND THE ROOKE RANCH, SAN PATRICIO AND REFUGIO COUNTIES, TEXAS. 79 Table A.I Trapping record for Rio Grande turkeys from January 2002 to March of 2002 on Rob and Bessie Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas. Transmitter 151.003 151.023 151.034 151.052 151.075 151.075'^ 151.094 151.133 151.145 151.191 151.223 151.243 151.252 151.252' 151.261 151.282 151.282' 151.302 Lea Band # Sex^ Aae^ Trap Type 2699 3103 2696 2697 2688 2680 2693 2689 2695 2690 3109 3106 2694 2677 2691 3108 2686 2692 F F F F F F F F F F F F F F F F F F A A A A A A A A J A J A A A A A A A Walk-In Location Rooke WWR WWR WWR WWR Rooke WWR WWR WWR Rooke WWR WWR WWR Rooke WWR Rooke Rooke Rooke Date 2-6-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Walk-In 2-6-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Walk-In 2-6-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Walk-In 3-18-02 Cannon-Net 1-26-02 Walk-In Walk-In Walk-In 2-6-02 3-18-02 2-6-02 80 Table A. 1 Continued: Transmitter 151.323 151.343 151.362 151.362' 151.402 151.463 151.483 151.523 151.542 151.562 151.582 Lea Band # Sex^ Aae^ Trap Type 3111 3112 3101 3132 3105 3110 3107 3104 3102 2700 2698 2678 2679 2681 2682 2683 2684 2685 3119 ...- , Date Location WWR WWR WWR Rooke WWR WWR Rooke Rooke WWR WWR WWR WWR Rooke Rooke Rooke Rooke Rooke Rooke Rooke Rooke Rooke F F F F F F F F F F F F F F F F F F F M M J A A A A A A A A J A A A J J J J J J A A Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Walk-In 3-18-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Walk-In Walk-In 2-6-02 2-6-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Cannon-Net 1-26-02 Walk-In Walk-In Walk-In Walk-In Walk-In Walk-In Walk-In Walk-In Walk-In 2-6-02 2-6-02 2-6-02 2-6-02 2-6-02 2-6-02 3-18-02 3-08-02 2-6-02 3120 3129 81 Table A.I Continued: Transmitter Lea Band # Sex^ Aae^ Trap Type 3130 3131 3135 M M F J A A Walk-In Walk-In Walk-In Date 2-6-02 2-6-02 3-18-02 Location Rooke Rooke Rooke ^ F - female, M - Male ^ A - Adult, J - Juvenile ^ Denotes a re-set transmitter 82 APPENDIX B KAPLAN-MEIER SURVIVAL RATES FOR RIO GRANDE TURKEY HENS ON THE WELDER WILDLIFE REFUGE AND THE ROOKE RANCH, SAN PATRICIO AND REFUGIO COUNTIES, TEXAS. 83 Table B.I Kaplan-Meier survival rates for Rio Grande turkey hens on the Welder Wildlife Refuge and the Rooke Ranch, San Patricio and Refugio Counties, Texas in 2002. No. iat No. Week 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Dates Jan 21 - Jan27 Jan 28 - Feb 3 Feb 4 - F e b 10 Feb 11 -Feb 17 Feb 18-Feb 24 Feb 25 - Mar 3 Mar 4 - M a r 10 Mar 11 -Mar 17 Mar 18-Mar 24 Mar 25 - Mar 31 Apr 1 - Apr 7 Apr 8 - A p r 14 Apr 15-Apr 21 Apr 22 - Apr 28 Apr 29 - May 5 May 6 - M a y 12 May 13-May 19 May 20 - May 26 May 27 - Jun 2 Jun 3 - J u n 9 Jun 10-Jun 16 Jun 17-Jun 23 Jun 24 - Jun 30 Jul 1 - Jul 7 Jul 8-Jul 14 Jul 15-Jul 21 Jul 22 - Jul 28 Jul 29 - Aug 4 Aug 5 - Aug 11 Aug 12-Aug 18 Aug 19-Aug 25 Aug 26 - Sep 1 Sep 2 - Sep 8 risk 19 18 25 25 25 24 23 22 24 24 24 22 22 21 20 19 17 17 16 15 13 12 12 12 12 11 11 11 10 10 10 10 10 No. new deaths added Survival 1.000 1 0.944 0.944 7 0.944 0.944 0.905 1 0.866 1 0.826 1 0.792 1 3 0.792 0.792 0.720 2 0.720 0.686 1 0.651 1 0.617 1 0.545 2 0.545 0.510 1 0.476 1 0.403 2 0.370 1 0.370 0.370 0.370 0.336 1 0.336 0.336 0.302 1 0.302 0.302 0.302 0.302 Upper CI Lower CI 1.000 1.000 1.047 0.842 0.857 1.032 0.857 1.032 0.857 1.032 0.794 1.017 0.736 0.995 0.683 0.970 0.647 0.936 0.647 0.936 0.647 0.936 0.879 0.561 0.879 0.561 0.850 0.521 0.820 0.483 0.789 0.445 0.719 0.370 0.719 0.370 0.685 0.335 0.651 0.302 0.572 0.234 0.536 0.204 0.536 0.204 0.536 0.204 0.536 0.204 0.498 0.174 0.498 0.174 0.498 0.174 0.459 0.146 0.459 0.146 0.459 0.146 0.459 0.146 0.459 0.146 84 Table B l Continued No. at No. No. new Week Dates risk deaths added Survival Sept 9-Sept 15 10 0.302 34 35 Sept 16-Sept 22 9 1 0.269 Sept 23 - Sept 29 9 0.269 36 37 Sept 30 - Oct 6 9 0.269 9 38 Oct 7-Oct 13 0.269 39 Oct 14-Oct 20 9 0.269 40 Oct 21 - Oct 27 9 0.269 0.269 41 Oct 28 - Nov 3 9 0.269 Nov 4 - Nov 10 9 42 0.235 Nov 11 -Nov 17 8 43 0.235 Nov 18-Nov 24 8 44 0.235 Nov 25 - Dec 1 8 45 0.235 8 Dec 2 - Dec 8 46 0.235 8 Dec 9-Dec 13 47 Upper CI Lower CI 0.459 0.146 0.419 0.119 0.419 0.119 0.119 0.419 0.419 0.119 0.119 0.419 0.419 0.119 0.119 0.419 0.119 0.419 0.378 0.093 0.093 0.378 0.093 0.378 0.093 0.378 0.093 0.378 85 PERMISSION TO COPY In presenting this tliesis in partial fulfillment ofthe requirements for a master's ciegree at Texas Tech University or Texas Tech University Health Sciences Center, I agree that the Library and my major department shall make it freely available for research purposes. Permission to copy this thesis for scholarly purposes may be granted by the Director ofthe Library or my major professor. It is understood that any copying or publication of this thesis for financial gain shall not be allowed without my further written permission and that any user may be liable for copyright infringement. Agree (Permission is granted.) 1 ^ ^ - , '^ ^ " = "-^ Student. oi ():ure Date Disagree (Permission is not granted.) Student Signature Date

Find millions of documents here - Study Guides, Homework Solutions, Papers, Exam Answer Keys and more. Course Hero has millions of course related materials that will enable you to learn better, faster and get an A in all your courses.
Below is a small sample set of documents: