Home

Courses

Turkeys

Prairie Chickens

RCWs

Publications

EVALUATION OF SURVEY TECHNIQUES FOR RIO GRANDE WILD TURKEY POPULATIONS IN THE SOUTHERN GREAT PLAINS

Abstract:

Few studies have assessed methods of estimating abundance, density, and trends for wild turkey (Meleagris gallopavo) populations and most wild turkey survey efforts have been limited to small scales, unstandardized, and unsuccessful.  However, successful large-scale monitoring programs are important to management decisions and evaluating management activities.  Recent evidence has suggested some Rio Grande wild turkey (M. g. intermedia) populations in the Southern Great Plains have began to decline, emphasizing the importance of population monitoring.  Thus, my objectives were to develop and evaluate abundance estimation techniques for Rio Grande wild turkeys and determine if useful trends in population change are detectable on ecoregion scales.

Counting Rio Grande wild turkeys at winter roosts is a technique commonly used to index abundance because they congregate in specific roosts throughout winter.  I compared 5 techniques for counting wild turkeys on winter roosts and found advanced technologies such as night vision devices, thermal infrared cameras, and automated video monitoring systems were ineffective.  Though the more traditional morning counts were best, additional research concerning sampling design and identification of roosting habitat are needed.

Aerial surveys have been used to estimate abundance for several wild bird species including wild turkeys.  I used inflatable turkey decoys and wild turkey flocks that contained radio-tagged individuals to evaluate detectability of flocks and individuals within flocks during fixed-wing and helicopter surveys.  I conducted computer simulations to evaluate the performance of aerial surveys and examined power to detect trends in population change.  My simulations suggested fixed-wing surveys would underestimate abundance by about 10% to 15% with a relative variability of 2.0% to 4.8%, but helicopter surveys would underestimate population abundance by about 5.6% with a relative variability of only 4.6%.  Also, power analyses suggested aerial surveys can provide sufficient power to detect a population change of 10% to 25% over a 4- to 5-year period.  Though, estimates from fixed-wing and helicopter surveys had similar biases and relative variability, helicopter surveys can cost as much as 6 times that of fixed-wing surveys.

Many states use opportunistic poult-hen counts from roads to index wild turkey population parameters such as reproduction, recruitment, and density.  I conducted an intensive poult-hen counting effort in the Southern Great Plains.  I found reproductive and recruitment parameters were correlated (r² > 0.349, 9 > n < 10, P < 0.05) to my local scale poults/hen estimates.  However, ecoregion scale poults/hen estimates obtained from Texas Parks and Wildlife Department (TPWD) were not correlated to reproductive or recruitment parameters (r² < 0.143, 13 > n < 16, P > 0.10) and poult-hen counts cannot be used to index density.  The inadequacy of TPWD poults/hen estimates probably resulted from small sample sizes and poor sampling of the ecoregion.  Ground-based surveys can be improved with line transect based distance sampling from roads.  However, because wild turkeys may avoid or be attracted to roads, I examined their distributional patterns around roads and found autumn midday and winter AM were the best times to conduct road surveys in the Southern Great Plains.

I used inflatable turkey decoys to evaluate detectability of flocks and individuals within flocks during road surveys.  I conducted computer simulations to evaluate the performance of line transect based distance sampling from roads and examined power to detect trends in population change.  My simulations suggested density estimates may be biased by about –18% during winter (13.4% CV) and by about –34% during autumn (16.1% CV).  Winter surveys tended to have less bias, lower relative variability, and greater power than autumn surveys.  During winter surveys, power was sufficient to detect a 10% to 25% change in population density in 5 to 8 years.

Aerial surveys are more expensive than line transect based distance sampling from roads.  However, estimates from road surveys had greater bias and relative variability when compared to aerial techniques.  Also, responsive movements and avoidance-attraction behaviors by wild turkey flocks and difficulty obtaining representative samples may present problems in road based surveys.  Thus, if fiscal restrains permit, I suggest managers use fixed-wing aircraft to monitor wild turkey populations in the Southern Great Plains.

Project Cooperators:

Texas Parks and Wildlife Department
Kansas Department of Wildlife and Parks
National Wild Turkey Federation
Texas Tech University
Houston Safari Club

Visit the
Texas Tech Rio Grande Wild Turkey Project Homepage
for more information.

See Publications for more information.

Butler, M. J.  2006.  Evaluation of survey techniques for Rio Grande wild turkey populations in the Southern Great Plains.  Dissertation, Texas Tech University, Lubbock, USA. [pdf]