Estimation of Northern Bobwhite Densities in South Texas
Disclaimer: This work has been submitted by a student. This is not an example of the work written by our professional academic writers. You can view samples of our professional work here.
Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UK Essays.
Published: Fri, 08 Jun 2018
Bart M. Ballard, Caesar Kleberg Wildlife Research Institute, Texas A&M University-Kingsville, Kingsville, Texas 78363.
Fidel Hernández, Caesar Kleberg Wildlife Research Institute, Texas A&M University-Kingsville, Kingsville, Texas 78363
Leonard A. Brennan, Caesar Kleberg Wildlife Research Institute, Texas A&M University-Kingsville, Kingsville, Texas 78363
Northern Bobwhites (Colinus virginianus) are a wide-ranging species and are commonly hunted throughout Texas. Their population has been declining since about 1880 and are decreasing in abundance in over 75% of their range in the United States (Leopold 1931, Errington and Hammerstrom 1936, Lehmann 1937, Guthery 2002). Texas has been seen as one of the remaining strong holds in North America (Rollins 2002) but recent evidence shows that populations are also declining within Texas (DeMaso et al. 2002). These declines have spurred an increase in research and management of the species and an improved understanding of the species (Guthery 2002). Declining numbers however can be difficult to understand because of their natural boom-bust population cycles and influence of weather on the population (Lehmann 1953, Keil 1976, Guthery et al. 1988, Bridges et al. 2001, Lusk et al. 2002). In order to properly manage quail within south Texas where there is large annual variation in precipitation and temperatures local population trends are needed.
Current hunting regulations are set at the state level where a liberal hunting framework (15 birds/day over 120 season; Brennan 2014) is assumed to have is little to no impact on the population (Guthery et al. 2004b). However, these state-wide regulations are not appropriate when managing at the fine scale where such a liberal harvest quota could negatively affect local populations (Roseberry and Klimstra 1984, Peterson 1999, Brennan et al. 2014). Bobwhites variable population cycles make it necessary for local land managers to set harvest limits based on local population trends (Brennan 2014).
By setting sustainable harvest limits based on local population densities the likelihood of population crashes goes down and there is quicker recovery following drought conditions and natural population declines (Brennan 2014). Recent recommendations for harvest values were made by Brennan in 2014 for south Texas where there is extreme variability in precipitation and temperatures compared to other regions of their range. These recommendations are based on the assumption of good environmental conditions, a 20% harvest rate, and are density depended. It is also recommended to conduct fall surveys in late November-mid December when detection is highest and basing the harvest on pre-hunt population numbers to minimize the probability of local extinction (Guthery et al. 2000, Sands 2010).
Estimating yearly and seasonal population densities can be difficult for many reasons including observer variability, local habitat variability and change between years, environmental factors such as weather, and species characteristics (Rusk et al 2007). Common methods include estimating abundance using indices or using distance sampling (Rusk et al. 2007). However, the accuracy of indices is sensitive to changes in detection (Anderson 2001, 2003, Rosenstock et al. 2002; Thompson 2002). During bobwhite population lows it becomes even more difficult to estimate population density due extremely low encounter rates (1 covey/7km; Kuvlesky et al. 1989). Distance sampling allows for varying detection probabilities while estimating densities and is a popular method that has been used successfully for bobwhites in many studies in south Texas (Brennan and Block 1986, Shupe et al 1987, Guthery 1988, Guthery and Shupe 1989, DeMaso et. al. 1992, Rusk et al 2007).
Unlike census techniques that are based on the assumption that all individuals within the survey area are counted, distance sampling works under the assumption that more animals are missed the farther you get from a transect (Brennan and Block 1986). To calculate density within a survey area, the perpendicular distance from a transect to an animal is recorded and then used to calculate a probability density function (Burnham et al. 1980, Buckland 2004) from which the density throughout the study area can be estimated using Program Distance (Thomas et al. 2010). Assumptions of distance sampling which must be met include: 1) all animals on the transect are detected, 2) animals are detected at their original locations prior to any movement in response to the observer, and 3) distances are measured accurately (Buckland 1992).
These assumptions can be difficult to meet in field condition but most issues with these assumptions can be addressed using proper survey design, post processing of the data, and statistical analysis. Assumption one can be relaxed if needed by incorporating a double observer design in which two counts are occurring simultaneously (Laake and Borchers 2004) or by applying an adjustment term. Assumption two can be violated if animals have the chance to respond to the surveyors by running, coming closer, or learning to hide (Buckland et al. 2001). Careful analysis of data can help determine if and how this assumption is violated and certain techniques can be implemented to account for animal responses such as truncation of data close to the line in cases were animals run (Fewster et al. 2008). Assumption three can be violated by untrained observers, lack of proper technology, or inaccurate estimates of cluster sizes if animals are clustered (Buckland et al 2001). Another assumption which can be violated includes independence between animal observations which can be an issue if surveys are done on roads or too close together (Thomas et al 2009).
Careful survey design is crucial to accurately estimating population densities and local knowledge of habitat, densities, and environmental gradients help when designing surveys. Once densities are estimated for a region careful consideration and local knowledge is needed to make the proper recommendations for hunting regulations and habitat management. Given accurate densities, harvest can be optimized at the ranch or pasture levels while also decreasing the likelihood of local population extinction.
The purpose of our study is to design a repeatable helicopter line transect survey for the King ranch study location which will be implemented over a three-year period from Sept 2018-December 2021. From this data fall bobwhite quail densities will be estimated using Program Distance from which management recommendations can be made. Specifically, our goals are;
- Develop a repeatable helicopter line-transect survey protocol for bobwhite quail
- Implement survey over three fall survey seasons
- Use Distance software to develop detection probability functions and estimate fall densities which can be used to aid in management and conservation decisions.
The study area includes a 25,000 acre section of the King Ranch (King Ranch, Kingsville TX) located south west of Kingsville (Figure1). The study area is located in the South Texas Plains ecoregion and may include parts of the Gulf Prairie and Marshes ecoregion (Gould 1975) Within this region there is high variability in rainfall (Correl and Johnsonston 1979; Omernik 1987) causing local populations to exhibit strong boom-bust population cycles. Major plant communities present on the King Ranch include blue stem prairie (Schizachyrium scoparium), mesquite-granjeno thornbrush (Prosopis glandulosa- Celtis pallida), mesquite-bluestem savannah, oak-bluestem (Quercus virginiana, Quercus stellata) (McLendon 1991, Fulbright and Bryant 2002). Major land uses on the King Ranch include commercial hunting and cattle production (Schnupp et al 2013). Annual rainfall is on average 65.4 cm with monthly values ranging from 1.4-13cm (Williamson 1983).
Figure 1. Divisions of the King Ranch (green), located in south Texas. Habitat includes but are not limited to shrub land, grasslands, mesquite-woodlands, oaklands, freshwater wetlands, and saltmarsh. Habitat is managed for cattle, white tail deer, and quail.
To estimate fall densities within the survey area we will first develop a three-year helicopter based, line-transect count survey. We will develop the transects in such a way that if desired, the surveys can continue past three years. Spatial layers will be made for the study area boundary and line transects in ArcGIS 10.3 (Environmental Systems Research Institute, Redlands, CA). Sample transects were placed parallel to each other leading north to south (Figure 2) and further stratification will be done using post processing techniques if desired. Transects were places at a distance of 400 meters from each other with a random starting location and giving a survey coverage of 50 percent. Given this design we have 30 transects of 8 km in length and a total survey length of 240 km (Figure 2).
Figure 2. Sample study area with transects (n=30) with survey zones of 100m to each side of the transect. Landcover includes woodland/shrublands (dark green), grasslands (light green), agriculture (light brown), wetlands (blue) and urban (red).
Given previous encounter rates of 1 covey/0.95km observed during a comparable study on another section of the King Ranch (Rusk et al. 2007), this design would yield an estimated 250 observations. However, encounter rates have been reported to be much lower during population lows; Rusk 2007 documented as low as 1 covey/7.38 km while walking transects in a low year verses 1 covey/ 1.96 km in an abundant year. Given flying transects gives roughly twice the number of detections per km, we will assume a helicopter flight on a low population year would give an encounter rate of Â¼ km and 60 observations.
To make sure that this transect design will provide a 25% or less Coefficient of Variation for population density estimates we can plug use the equation (Buckland 1993):
Where L= total line length needed, = the coefficient of variation for population density estimate, and L0/n0 = encounter rate, or the number of quail detected per km of transect. The value b is typically between 1.5-3 (Burnham 1980) and it is most frequently assumed that b=3 (Buckland 1993).
Given this equation, under an assumed encounter rate of 1 covey/.95 km and a 25% CV the minimum total transect length is:
However, when the encounter rate is dropped to 1 covey/4 km during a population low, the needed length becomes:
By conducting more surveys than is needed to achieve a 25% CV there is less of a chance that during a dry year we will not be able to estimate density because of lack of encounters. After year one we will re-evaluate transect design by incorporating the first year’s encounter rate to help determine transect lengths for years two and three (Buckland 1993).
Surveys will take place in the first week of October to give enough time to provide updated recommendation for harvest quotas before the onset of quail regular season on October 29th (TPWD, Outdoor Annual). October 1st of each year a mock survey will be done in which tools are calibrated and extra surveyor training done if need following protocols similar to Schnupp et al. (2013). This test flight will occur along a 3 km transect with 16 targets (Otto and Pollock 1990, Shnupp et al. 2013). Each side of the transect will have 8 targets (dove decoys suspended at 1.2m) distributed randomly between 10-70m at 10m intervals from the line and spaced 300m apart along the transect (Schnupp et al.). This will help reduce potential errors counts due to equipment malfunction and surveyor error.
The full survey will begin the day after the mock survey and all transects will be surveyed once per year. If detections for an entire survey are below 80 then a second survey will be done. Surveys will take place in the first 3 and last three hours of day light when possible and the start location will vary each survey. From the start location, every other transect will be sampled to reduce the probability of over counting and then returning to skipped transmitters as soon as possible. We will use a four person helicopter such as the Robinson R-44 (Robinson Helicopter Company, Torrance, California) or similar models equipped with a parallel swathing lightbar for navigation (2005; Raven RGL 600, Raven Industries, Sioux Falls, South Dakota). Surveys will be conducted at approximately 48 km/hour and at a height of 18 m (Shupe et al. 1987, Rusk 2007) One observer will be facing forward counting coveys directly on the line and two rear-facing observers counting quail which flush on the sides or behind the helicopter. When a covey is spotted, the helicopter will hover briefly to allow observer to use the range finder and count the number of quail in the covey.
The forward facing technician in addition to counting coveys will help navigate to the transects, and will start and stop the survey recordings (Schnupp 2013). The two rear observers will collect data as well as enter data for all surveyors. Covey counts and covey size will be recorded for 100 meters to each side of the helicopter using laser electronic range finders, differential global positioning systems, personal tablet computers, and keypads (Schnupp et al. 2013). Tablets will be installed with ArcPad (Environmental Systems Research Institute, Redlands, CA), and connected to the laser range finders with sub meter accuracy. The differential global positioning system will collect 5 points/second to track the flight path. Electronic range finders will be synced to tablets using blue tooth and will measure distance to covey, compass bearing, angle of inclination and horizontal offset of covey from the helicopter for each covey. Key pads were also used to record sizes of coveys. Raw survey information is then imported into ArcMap 10.3 for data processing and then imported to Program Distance.
Using distance survey data collected over three years we will calculate densities and variance estimates in Program Distance 7.0 similar to Rusk et al 2007. Program Distance calculates estimated densities and variances as
Where is density, n is the number of coveys detected, is the effective half-band width, cv is the coefficient of variation, L is the length of transects, and E(s) is average covey size. Effective half widths with be calculated in distance by fitting detection functions to histograms of distances and covey counts. To improve model fit, 5% of the right hand data will be truncated (Buckland et al. 2001; Shnupp 2013) and data will be evaluated visually for any signs of violation of the basic assumptions. We will consider a variety of detection functions (uniform, half-normal, and hazard-rate with several series adjustments) and choose the best fitting model using Akaike’s Information Criterion values (AICc) and chi-square analysis (Buckland 2001; Shnupp 2013).
We will then develop a global detection function for each year to estimate fall densities and use confidence intervals and coefficient of variation reported from distance. If stratification by pasture is desired and there are enough observations to do so, then detection functions will be built at the pasture level otherwise a global detection function will be applied to each pasture. A coefficient of variation of less than 20% is recommended for bobwhite density estimates (Guthery 1988) but we will consider a coefficient of variation of 25% acceptable.
EXPECTED RESULTS AND BENEFITS
From these three fall bobwhite quail surveys, we will be able to report yearly bobwhite density estimates and begin to understand local population trends. Once funding is approved, exact methods will be refined using actual ranch and pasture boundaries and habitat gradients. Survey design will be reviewed by quail researchers at the Caesar Kleberg Wildlife Research Institute to ensure proper design.
Yearly encounter rates, detection functions, estimated population density, and recommendations for harvest rates will be provided in annual reports. A final report will be submitted in the form of a dissertation chapter within one year after the completion of the last fall survey. This chapter will summarize yearly results as well as trends observed throughout the study region and will include recommendations for sustainable harvest limits. Research results may be presented at professional meetings and will include one or more King ranch employees as authors and King Ranch will be acknowledged as the primary funding contributor.
Project deliverables include:
- P.h.D dissertation chapter and corresponding scientific publication
- Scientific presentations
- Spreadsheets of density estimates and recommended harvest rates
ENDANGERED SPECIES CONSIDERATIONS
Not applicable to the proposed project.
ETHICAL USE OF ANIMALS
Animal and Care Use form is not required
This study will be a cooperative project between the Caesar Kleberg Wildlife Research Institute (CKWRI) and the King Ranch. Drs. Bart M. Ballard, Fidel Hernandez, and Leonard A. Brennan will be primary investigators. This project will include one P.h.D. student who will act as project coordinator and field supervisor. The graduate student hired will also be responsible for hiring part-time student technicians to aid in surveys. The student hired will conduct fall densities surveys on the King Ranch as a partial fulfillment of P.h.D contract and will also be conducting other quail research in assistance of other projects.
Jan-April: Await funding
April-May: Search for P.h.D candidate
June-Sept:Hire student, coordinate field surveys and hire part-time surveyors for Survey week
October: Fly surveys and estimate fall densities
November: Further data analysis and reporting
Aug-Sept: Refine transects/protocol if needed, hire technicians for Survey week
October:Fly surveys and estimate fall densities
November: Further data analysis and reporting
Aug-Nov:Same schedule as above
December: Provide final analysis and Report
2 Electronic distance estimators ($18,000 each) =36,000
2 Tablets w/accessories: = $1,600
2 Keypads: $100
1Raven Cruiser: $2,000
Rounded Estimate: $40,000
-P.h.D student stipend: $1500 with fringe benefits at .7% of salary and medical (up to 250$/month)
-2 Short term technicians: 100$/day during fall surveys. Total=2 technicians*$100*7 days a year= 1400/year
-Helicopter time: 500$/hr *estimated 10 hrs per year= $5,000/year
-Driving costs: $0.50/mi+ gas. Exact distance to site in unknown, preliminary estimate= $10,000/year
Summary of Annual Cost:
Anderson, D. R. 2001. The need to get the basics right in wildlife ï¬eld studies. Wildlife Society Bulletin 29:1294-1297.
Anderson, D. R. 2003. Response to Engeman: index values rarely constitute reliable information. Wildlife Society Bulletin 31:288-291.
Brennan, L. A., and W. M. Block. 1986. Line transect estimates of mountain quail density. Journal of Wildlife Management 50:373
Brennan, L.A., F. Hernandez, E.D. Grahmann, F. C. Bryant, M.J. Schnupp, D.S. Delaney, and R. Howard. 2014. Quail Harvest Guidelines for South Texas: Concepts, Philosophy, and Applications; Wildife Technical Publication No. 3 of the Caesar Kleberg Wildlife Research Institute Texas A&M University-Kingsville.
Bridges, A. S., M. J. Peterson, N. J. Silvy, F. E. Smeins, and X. B. Wu. 2001. Differential influence of weather on regional quail abundance in Texas. Journal of Wildlife Management 65:10-18.
Buckland, S. T. 1992. Fitting density functions using polynomials. Applied Statistics. 41:63.
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers, and L. Thomas. 2001. Introduction to distance sampling estimating abundance of biological populations. Oxford University Press, New York, USA.
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers, and L. Thomas. 2004. Advanced distance sampling: estimating abundance of biological populations. Oxford University Press, New York, New York, USA.
Burnham, K. P., D. R. Anderson, and J. L. Laake. 1980. Estimation of density from line transect sampling of biological populations. Wildlife Monographs 72.
Correll, D. S., and M. C. Johnston. 1979. Manual of vascular plants of Texas. The University of Texas Printing Division, Austin, Texas, USA.
DeMaso, S. J., F. S. Guthery, G. S. Spears, and S. M. Rice. 1992. Morning covey calls as an index of northern bobwhite density. Wildlife Society Bulletin 20:94-101.
DeMaso, S. J., M. J. Peterson, J. R. Purvis, N. J. Silvy, and J. L. Cooke. 2002. A comparison of two quail abundance indices and their relationship to quail harvest in Texas. Proceedings of the National Quail Symposium 5:206-212.
Errington, P. L., and F. N. Hammerstrom, Jr. 1936. The northern bob-white’s winter territory. Iowa State College of Agriculture and Mechanical Arts Research Bulletin 201:305-443.
Fewster, R.M., Southwell, C., Borchers, D.L., Buckland, S.T. & Pople, A.R. 2008. The influence of animal mobility on the assumption of uniform distance in aerial line transect surveys. Wildlife Research 35:275-288.
Fulbright, T. E., and F. C. Bryant. 2002. The last great habitat. Caesar Kleberg Wildlife Research Institute, Special Publication No. 1, Kingsville, Texas, USA
Guthery, F. S. 1988. Line transect sampling of bobwhite density of rangeland: evaluation and recommendations. Wildlife Society Bulletin 16:193-203.
Guthery, F. S., N. E. Koerth, and D. S. Smith. 1988. Reproduction of northern bobwhites in semiarid environments. Journal of Wildlife Management 52:144-149.
Guthery, F. S., and T. E. Shupe. 1989. Line transect vs. capture-removal estimates of bobwhite density. Proceedings of the Annual Conference Southeast Association of Fish and Wildlife Agencies
Guthery, F. S., M. J. Peterson, and R. R. George. 2000. Viability of northern bobwhite populations. Journal of Wildlife Management 64:646âˆ’662. Sands 2010
Guthery, F. S. 2002. The technology of bobwhite management: the theory behind the practice. Iowa State University Press, Ames, Iowa, USA Guthery, F. S. 2002. The technology of bobwhite management: the theory behind the practice. Iowa State University Press, Ames, Iowa, USA.
Guthery et al. 2004b Guthery, F. S., M. J. Peterson, J. J. Lusk, M. J. Rabe, S. J. DeMaso, M. Sams, R. D. Applegate, and T. V. Dailey. 2004. Multi-state analysis of fixed, liberal regulations in quail harvest management. Journal of Wildlife Management 68:1104-1113
Hernández, F., F. S. Guthery, and W. P. Kuvlesky. 2002a. The legacy of bobwhite research in south Texas. Journal of Wildlife Management 66:1-18.
Kiel, W. H. 1976. Bobwhite quail population characteristics and management implications in south Texas. Transactions of the North American Wildlife and Natural Resources Conference 41:407-20.
Kuvlesky, W.P., B.H. Koerth,andN.J.Silvy.1989.Problemsofestimating northern bobwhite populations at low density. Proceedings of the Annual Conference Southeast Association of FishandWildlifeAgencies43:260-267.
Laake, J.L. and Borchers, D.L. 2004. Methods for incomplete detection at distance zero. Advanced Distance Sampling (eds S.T. Buckland, D.R. Anderson, K.P. Burnham, J.L. Laake, D.L. Borchers and L.Thomas). pp. 108-189. Oxford University Press Oxford.
Lehmann, V. W. 1937. Increase quail by improving their habitat. Texas Game, Fish, and Oyster Commission, Austin, Texas, USA.
Lehmann, V. W. 1953. Bobwhite population fluctuations with vitamin a. Transactions of the North American Wildlife Conference. 18:199-246
Leopold, A. 1931. Report on a game survey of the north central states. Democrat Printing Company, Madison, Wisconsin, USA.
Lusk, J. J., F. S. Guthery, R. R. George, M. J. Peterson, and S. J. DeMaso. 2002. Relative abundance of bobwhites in relation to weather and landuse. Journal of Wildlife Management 66:1040-1051.
McLendon, T. 1991. Preliminary description of the vegetation of south Texas exclusive of coastal saline zones. Texas Journal of Science 43: 13-32.
Otto, M. C., and K. H. Pollock. 1990. Size bias in line transect sampling: a ï¬eld test. Biometrics 46:239-245.
Peterson, M. J. 1999. Quail harvest management in Texas: a rational approach. Pages 124-133 in K. A. Cearly, editor. Preserving Texas’ Quail Hunting Heritage into the 21st Century. Texas Agricultural Extension Service, Texas A&M University, College Station, USA.
Rollins, D. 2002. Sustaining the ‘quail wave’ in the southern great plains. Proceedings of the National Quail Symposium 5:48-56.
Roseberry, J. L., and W. D. Klimstra. 1984. Population Ecology of the Bobwhite. Southern Illinois University Press.
Rosenstock, S. S., D. R. Anderson, K. M. Giesen, T. Leukering, and M. F. Carter. 2002. Landbird counting techniques: current practices and an alternative. Auk 119:46-53.
Rusk, J.P., F. Hernandez, J.A. Arredondo, F. Hernandez, F.C. Bryant, D.G.Hewitt, E.J. Redeker, L.A Brennan, R.L. Bingham. 2007. The Journal of Wildlife Management 71:4(1336-1343).
Shupe, T. E., F. S. Guthery, and S. L. Beasom. 1987. Use of helicopters to survey northern bobwhite populations on rangeland. Wildlife Society Bulletin 15:458-462.
Thomas, L., S. T. Buckland, E. A. Rexstad, J. L. Laake, S. Strindberg, S. L. Hedley, J. R. B. Bishop, T. A. Marques, and K. P. Burnham. 2010. Distance software: design and analysis of distance sampling surveys for estimating population size. Journal of Applied Ecology 47:5-14.
Thompson, W. L. 2002. Towards reliable bird surveys: accounting for individuals present but not detected. The Auk 119:18-25.
Williamson, D. L. 1983. Soil survey of Brooks County, Texas. United States Department of Agriculture, Soil Conservation Service, Washington, D.C., USA.
Cite This Work
To export a reference to this article please select a referencing stye below: