Translator Disclaimer
1 December 2006 EFFECTS OF RANGELAND MANAGEMENT ON COMMUNITY DYNAMICS OF THE HERPETOFAUNA OF THE TALLGRASS PRAIRIE
Dustin J. Wilgers, Eva A. Horne, Brett K. Sandercock, Allan W. Volkmann
Author Affiliations +
Abstract

The Flint Hills of Kansas and Oklahoma are the largest remaining tracts of tallgrass prairie in North America. This area has undergone major changes in land management practices in the past 30 years. Traditional season-long cattle stocking with variable burn schedules has diversified to include intensive-early cattle stocking accompanied by annual burning. To understand how different land management practices affect the herpetofauna of a tallgrass prairie, we used mark-recapture statistics to analyze herpetofaunal community dynamics. We analyzed survey data collected over a 15-year time span (1989–2003) from a rangeland site in Cowley County, KS, USA. A modified Jolly-Seber open population model, POPAN-5, was used to estimate four community parameters: probability of species loss (φ;′), probability of detection (P), probability of entry (Pent), and species richness (N). The top models included burn status as a covariate for species loss rate, while cattle stocking received moderate support as a covariate. Rates of species loss were higher during burn years (φ;′  =  0.04, 95% CI: 0.02 to 0.08) than nonburn years (φ;′  =  0.00, 95% CI: 0.00 to 0.01). Analysis of the impacts of different management practices was difficult due to confounding effects of changes in both burning and grazing. Declines in species richness tended to be steepest during a period of season-long stocking, but results were not statistically significant. Though our limited data set does not allow us to draw strong conclusions on the effects of land management on herpetofaunal populations, the mark-recapture models illustrated in our study should prove to be a valuable tool in future analyses of similar data.

The tallgrass prairie has been reduced to less than five percent of its pre-European settlement extent, making it the most heavily impacted ecosystem in North America (Samson and Knopf, 1994). The largest contiguous section of remaining tallgrass prairie occurs in the Flint Hills of northern Oklahoma and Kansas (Knapp and Seastedt, 1998). Much of this region is unsuitable for row-crop agriculture because shallow soils are associated with limestone bedrock. However, the Flint Hills provide extensive rangeland for cattle. Much of the rangeland experienced a marked change in land management in the early 1980s. Traditionally, ranchers practiced season-long cattle stocking, which allowed a limited number of cattle to graze year-round, accompanied by rotation of burned and unburned areas every 2–3 years. Under intensive-early stocking, cattle stocking rates were roughly double that of the previous practice for only 90–120 days in the growing season with annual spring burns, which increased daily steer gains and gains per acre (Robbins et al., 2002; Smith and Owensby, 1978). Intense cattle grazing has been found to have a significant effect on the diversity of native plant species, reducing overall cover of indiangrass (Sorghastrum nutans), and increasing Kentucky bluegrass (Poa pratensis) and big bluestem (Andropogon gerardii) basal area (Owensby et al., 1988).

Fire is an essential component in the development and persistence of the tallgrass prairie ecosystem (Axelrod, 1985). Regular burning of the tallgrass prairie increases plant productivity, decreases aboveground biomass by removing dead litter, and limits woody vegetation encroachment (Briggs and Knapp, 1995; Heisler et al., 2003; Hulbert, 1988). However, annual burning eliminates aboveground cover, which may increase short-term predation risk for larger-bodied species of snakes (Wilgers, 2005).

Species richness estimates can be used as an indicator of disturbance effects on communities by comparing the numbers of species pre- and post-disturbance within in a single area. Past studies have determined that reptile and amphibian abundance and community composition are influenced by fire in grassland and shrub ecosystems, including desert (Fyfe, 1980), sandhill (Mushinsky, 1985), sand-pine scrub (Greenberg et al., 1994), spinifex grassland (Masters, 1996; Pianka, 1996), and tropical savanna (Braithwaite, 1987; Griffiths and Christian, 1996; Hannah and Smith, 1995; Trainor and Woinarski, 1994), as well as other ecosystems in North America (Pilliod et al., 2003; Russell et al., 1999). In the tallgrass prairie, past projects have studied the effects of fire on arthropods (Evans, 1984, 1988; Joern, 2004, 2005; Seastedt, 1984a,b), birds (Beutler, 1993; Powell, 2004; Zimmerman, 1992, 1997), and mammals (Clark and Kaufman, 1990; Clark et al., 1989; Kaufman et al., 1990; McMillan et al., 1995). A limited number of studies have examined the distribution of herpetofauna in tallgrass prairie (Busby and Parmelee, 1996; Busby et al., 1994; Heinrich and Kaufman, 1985), but the effects of fire and grazing on herpetofauna inhabiting the tallgrass prairie ecosystem are poorly understood (Cavitt, 2000; Setser and Cavitt, 2003).

To estimate species richness of an area (N ˆ), most studies have used unadjusted count data (Braithwaite, 1987; Fredericksen and Fredericksen, 2002; Hannah and Smith, 1995), Shannon-Wiener indices (Chao and Shen, 2003; Greenberg et al., 1994), or species accumulation curves (Thompson et al., 2003). All of these techniques assume that capture probabilities (P) equal one. Previous mark-recapture studies of amphibians and reptiles show that P < 1 for both individuals and cryptic species (Bailey et al., 2004a,b,c; Schmidt, 2003; Schmidt et al., 2002), which can result in a biased estimator of N ˆ (Pollock et al., 1990). Mark-recapture statistics can be used to analyze a variety of data types, which include but are not restricted to the typical mark-recapture methods, which utilize the marking of individuals. Mark-recapture models have been used to estimate species richness in spiders (Follner and Henle, 2001) and birds (Boulinier et al., 1998; Nichols et al., 1998), but not reptile or amphibian species. This is the first study to examine community dynamics of reptiles and amphibians using mark-recapture models. Our goals in this study were two-fold: (1) to estimate species richness and probabilities of detection for reptiles and amphibians, and (2) to determine the effects of land management techniques on community dynamics.

Materials and Methods

Study Site

Fourteen herpetofaunal surveys were conducted during a 15-year period (1989–1990 and 1992–2003) at Bud Jan Nitschke (BJN) cattle ranch in Cowley County, Kansas (37°15′ N, 96°43′ W). The 650 ha ranch is located in the southern portion of the Flint Hills, and the vegetation is dominated by a matrix of perennial warm-season C4 grasses, such as big bluestem (A. gerardii), little bluestem (A. scoparius), indiangrass (S. nutans), and switchgrass (Panicum virgatum). A diverse mixture of other less abundant species includes warm-season and cool-season grasses, composites, legumes, and other forbs. Woody species such as leadplant (Amorpha canescens), buckbrush (Symphoricarpos orbiculatus), New Jersey tea (Caenothus herbaceous), and smooth sumac (Rhus glabra) are locally common (Freeman and Hulbert, 1985).

Average monthly temperature ranges from a low in January (−2.7 C) to a high in July (26.6 C). Average annual total precipitation ranges from ~750 to 850 mm with 75% falling during a summer growing season (March–July; Bark, 1987). One common characteristic of the region is protruding limestone outcrops on hilltops, which provide suitable habitat for several species of amphibians and reptiles.

During the 15-year study, two management practices were implemented at the BJN ranch. Traditional season-long stocking (200 hundred cows with their calves, ~0.6 animals/ha) was implemented the first 10 years (1989–1998) and pastures were burned in alternate years. Starting in 1999, the land management practice switched to intensive-early cattle stocking (approximately 650 yearlings, ~1.0 animals/ha, maintained for 3 months starting in late spring) combined with annual burning.

Field Protocol

Herpetofaunal surveys were conducted on one day between 19–30 April each year. The survey route consisted of the same 4-km walk which took approximately four hours (1000–1400 h). Walking surveys is an effective and widespread practice for monitoring amphibian and reptile communities (Doan, 2003; Woodford and Meyer, 2003). The route included normal tallgrass prairie habitats, the shorelines of a pond, and intermittent streams. Amphibians and reptiles were detected by turning over rocks and other debris, and by sighting animals in the open. Each year, we recorded the number of individuals of each species encountered, and environmental conditions including: midday air temperature (C), water temperature (C), cloud cover, burn conditions (burned versus unburned), and the presence of flowing water in the intermittent streams (Table 1).

Table 1

Burn status and climatic conditions of surveys at Cowley County site.

i0018-0831-62-4-378-t01.gif

Statistical Analysis

Mark-recapture models can be used to analyze a wide variety of data, which include but are not restricted to the more traditional mark-recapture methods requiring the marking of single individuals. More recent mark-recapture statistics enable us to estimate community parameters such as species richness, emigration, and immigration via species presence/absence data across repeated sampling efforts with no individual marking. The reference of mark-recapture statistics and models throughout our paper does not imply the use of traditional mark-recapture methods.

Species richness is used to compare the complexity of communities between two or more habitats or to compare before and after a perturbation. When measuring community dynamics, two additional parameters are of interest: (1) the probability of species loss (local extinction or permanent emigration), and (2) the probability of a new species entering the community (immigration). Mark-recapture statistics were used to estimate these initial parameters along with the species richness of the herpetofaunal community of the study site. Initial detection data were transformed into presence/absence data per species by year, creating a distinct encounter history for each of the species recorded across all 14 years of data (Table 2).

Table 2

Species detected during herpetofaunal surveys at Cowley County site (1989–2003).

i0018-0831-62-4-378-t02.gif

Program POPAN-5, a Cormack-Jolly Seber (CJS) model design was fit to the data structure (Schwarz and Arnason, 1996). This model assumes an open population (allows immigration, emigration, births, and deaths in the community) throughout the entire study period with near instantaneous sampling periods. Four parameters are estimated by POPAN-5: φ;, species retention; P, probability of any species detection given the species is in the area; Pent, probability of a new species entering the community; and N, species richness. Rates of species loss (φ;′) were calculated as the complement of retention (1 − φ;). Two other parameters can be derived from the parameter estimates above: b, number of species entries into the community (Equation 1); and N ˆ, species richness estimates for each individual time step (i) within the study starting at time zero with N  =  0 (Equation 2).

1

i0018-0831-62-4-378-e01.gif

2

i0018-0831-62-4-378-e02.gif

Mark-recapture analyses were conducted using Program MARK (Version 4.1; White and Burnham, 1999) in three-steps. First, we selected a global model. Species retention (φ;), probability of detection (P), and probability of entry (Pent) were all modeled with time dependency (φ;t, Pt, Pent t, N) because we expected variation in annual conditions might impact these parameters. This was the most parameterized model (no. of parameters  =  43).

Second, we tested the global model for possible overdispersion. Program RELEASE was used to estimate a variance inflation factor (ĉ), which was calculated by dividing the sum of all the chi-square tests by the sum of the degrees of freedom over all the component tests of Test 2 and 3 (Burnham et al., 1987).

Last, we fit reduced models where annual variation in the parameter estimates was modeled as a function of environmental covariates or as a constant. We modeled φ; as a function of burn status because spring fires are known to cause direct mortality of reptiles (Erwin and Stasiak, 1979; McLean, 1990). Probability of detection (P) was modeled as a function of temperature and cloud cover because amphibians and reptiles are ectothermic and may have reduced activity or seek refugia under cool conditions. Burn status and cloud cover were treated as Bernoulli conditions (1  =  yes, 0  =  no). In POPAN-5, species richness (N) is estimated as a single parameter, and the derived N per occasion cannot be modeled as a function of a time dependent environmental covariate. In surveys of vertebrate communities, most species are detected in the first few years, and new detections become progressively more difficult as time goes by. Thus, the probability of entry of new species into the community is likely to decline, and we modeled Pent as a linear function.

Models were constructed using the following link functions: logit (φ; and P), Mlogit1 (Pent), and log (N). The logit link constrains species retention and probability of detection from zero to one, which is necessary because both parameters are probabilities. The Mlogit1 link constrains the sum of all time steps (i.e., 1–14) from zero to one, which is necessary for a probability of entry (Pent) estimate due to its similarities with detection functions. The log link function does not constrain the estimate within a bounded, which is desired because estimate of species richness (N) are >1.

All possible combinations between our global model and the environmental covariates yielded a candidate set of 25 models. Model fit was assessed using quasi-Akaike's Information Criteria (QAICc), which corrects for overdispersion and small sample size (Burnham and Anderson, 1998). Model selection was based on the difference between QAICc values (ΔQAICc). The best-fit model had a ΔQAICc of zero, but models with a ΔQAICc ≤ 2 were considered equally parsimonious.

Parameter estimates, their corresponding standard errors, and 95% confidence intervals for the top models were calculated in Program MARK using all models. Derived estimates of N for each year were averaged across all models using the model averaging procedure of Program MARK. In a post-hoc analysis, we looked at the effect of land management practices on species richness by pooling years to give species richness estimates for three blocks of time: (I) initial years (1989–1992), (II) season-long stocking (1993–1998), and (III) intensive-early stocking (1999–2003). We did not include the initial years (1989–1992) in the land management estimates because high probability of species entry during the first three surveys indicated that initial surveys missed >5% of the species in the area. This model was not used for model selection or parameter estimates. Significant differences in species loss rates were tested using program CONTRAST (Hines and Sauer, 1989). Significance testing was performed using t-tests assuming unequal variances, and α was set at 0.05.

Results

Over the 14 years of this study, 1563 individuals representing 35 species of amphibians and reptiles were detected at the study site. The sample included species representing 4 orders of heterothermic tetrapods (Caudata, Anura, Testudinata, and Squamata) with two suborders of squamates (Sauria and Serpentes; Table 2). Components of Program RELEASE (Test 2 + Test 3) yielded a moderate variance inflation factor (χ2  =  64.7, df  =  23, ĉ  =  2.81) for the global model (φ;t, Pt, Pent t, N), indicating minor overdispersion of the count data.

Model Selection

The best fit models included constant parameters and the environmental covariate, burn status. The two top models had good support (wi > 0.20) and were considered to be equally parsimonious (ΔQAICc ≤ 2), while eight other models containing the additional environmental covariates (cattle stocking, air temperature, and cloud cover; Table 1) showed a moderate fit to the data (ΔQAICc ≤ 7, wi  =  0.01 to 0.10; Table 3). The global model was a poor fit to the data (φ;t, Pt, Pent t, N; ΔQAICc  =  79.9). The 3 time block model (φ;3tb, Pc, Pent, N), model 3, was only moderately supported by the data (ΔQAICc  =  4.16, wi  =  0.033).

Table 3

Summary of model selection for species richness estimates of reptiles and amphibians in tallgrass prairie, Cowley County, Kansas.

i0018-0831-62-4-378-t03.gif

Of the candidate models, φ; was best modeled as a constant, or with the burn status covariate. Land management showed moderate support with ΔQAICc  =  2.1 and garnished 10% of the data's support (Table 3). Probability of detection (P) was best modeled as a constant or with the two environmental covariates: mid-day air temperature and cloud cover (Table 3). Modeling Pent as a linear function was a better fit to the data than time dependence (ΔQAICc ≥ 20; Table 3).

Community Dynamics

Community demographics φ; and P were taken from the top two equally parsimonious models. The φ; parameter for model 1 yielded 1 rate across all 14 years (φ;′  =  0.025, 95% CI  =  0.011 to 0.048), while model 2 yielded 2 rates according to burn status. Species loss rates were significantly higher during burn years (φ;′  =  0.04, 95% CI: 0.02 to 0.08) when compared to years without burning (φ;′  =  0.0, 95% CI  =  0.00 to 0.01; Χ;2  =  8.16, df  =  1, P  =  0.004). Probability of detection was moderate in the top two models: model 1, P  =  0.581 (95% CI  =  0.524 to 0.636), and model 2, P  =  0.586 (95% CI  =  0.529 to 0.640). Probability of species entry into the community was best modeled as a linear function over time and declined from Pent  =  0.19 in 1989 to Pent ≤ 0.03 by 1993 (Fig. 1a).

Figure 1

(A) Probability of new herpetofaunal species entering Cowley County tallgrass prairie site from 1989–2002.

Parameter estimates ( ± 1 SE) were calculated by program MARK through model averaging of all candidate models (Table 1). (B) Estimates of all herpetofaunal species richness from 1989–2003 at Cowley County tallgrass prairie site. For post-hoc analysis, the surveys were divided into three time periods: (I) Initial years (1989–1992) where Pent ≥ 0.05, (II) Traditional season-long stocking (1993–1998), (III) Intensive-early stocking (1999–2003). Parameter estimates ( ± 1 SE) were calculated by three time blocks model. Overall estimate (N ˆ) was calculated across all 14 years of study.

i0018-0831-62-4-378-f01.gif

The overall estimate for species richness (N ˆ), across the entire time series did not significantly differ between the two top models, model 1 (N ˆ  =  35.1) and model 2 (N ˆ  =  35.0). These estimates matched the total number of species observed throughout the study. Annual estimates of species richness are a derived parameter and were averaged across all candidate models and years to estimate the average number of species in the community each year. Species richness estimates for each individual year did not reach the overall estimate across all years. Yearly species richness estimates were at a maximum in 1995 (N ˆ  =  32.0, 95% CI  =  29.4 to 34.5), with subsequent declines by the end of the study in 2003 (N ˆ  =  26.6, 95% CI  =  21.0 to 32.1; Fig 1b).

In the three block model, estimates of species loss rates were not different between the three time periods (φ;′I  =  0.02, 95% CI  =  0.0 to 0.08; φ;′II  =  0.03, 95% CI  =  0.01 to 0.08; φ;′III  =  0.01, 95% CI  =  0.01 to 0.39). Species richness estimates declined faster during the season-long stocking compared to the intensive-early stocking period (Fig. 1b), however, the variability in species richness estimates are high and greatly overlap.

Discussion

This is the first study to use mark-recapture statistics to analyze effects of land management practices on the community dynamics of herpetofauna. Overall, detection rates of herpetofaunal species were low throughout the study (P < 0.59). Mark-recapture statistics were a useful approach here because not all herpetofaunal species were detected every year and the probability of detection was affected by environmental covariates (climatic conditions), (Schmidt et al., 2002). Analyses assuming probability of detection close to unity (P  =  1) would have resulted in biased estimates of N ˆ (Schmidt, 2003). Some of the annual variation in detection rates was due to environmental variables such as air temperature and cloud cover.

In this study, total species richness of the herpetofaunal community was estimated at 35 species across a 14-year time series in southeast Kansas. By comparison, 57 herpetofaunal species have been reported from Cowley County, and 97 species are on the current state list for Kansas (Collins, 1993; Potts and Collins, 2005). According to the probability of species entry into the system (Fig. 1a), our sampling detected a majority of species occurring in the area within the first three years of the study.

Effects of Management Practices

This retrospective analysis limits our ability to draw strong conclusions about the effects of land management on prairie herpetofauna. One sampling effort per year may not sufficiently explain species presence or absence in the area due to variations in activity based on climate or other variables. The trends of herpetofaunal community dynamics found in this study may be widespread throughout the area, but the lack of a reference sample does not allow us to determine this.

However, based on our limited data, the mark-recapture models we used suggest that different land management practices may have an effect on herpetofaunal species richness in the Flint Hills ecosystem. Intensive-early stocking, which is a combination of repeated annual burning and intense grazing, has been shown to have a negative effect on other tallgrass prairie fauna, such as birds (Robbins et al., 2002; Rohrbaugh et al., 1999; Zimmerman, 1997) and mammals (Kaufman and Kaufman, 1997).

Species loss rates were significantly higher in years after a spring burn. During the first 10 years of season-long stocking, species richness peaked around 31. Species loss rates declined during both management practices. Declines were steepest during season-long stocking, however variability in species richness estimates were greater during intensive-early stocking. The management practice and three time block models were only given moderate support from the data, and were not among the top models. Failure of the three time block model to improve the model fit indicates minimal influence of the initial detection time period on the overall community dynamics estimates. Possible confounding effects between the observed negative impact of burning and the unknown impact of grazing may have decreased the overall impact of the two stocking practices, thus weakening the model fit to the data. Thus, management practice may influence herpetofaunal community dynamics, but, further research is necessary to determine any significant differences.

Herpetofaunal species loss rates were significantly higher following burn years across the entire 14-year data set. Fire and grazing disturb the vegetation structure similarly, leaving much of the disturbed area devoid of grass more than a few centimeters high for much of the spring (Robbins et al., 2002). Intensive-early stocking provides grasses a chance to recover during later months of the year. This results in a continuous fuel source and more intense fires (Smith and Owensby, 1978). Direct mortality of herpetofauna due to fire can have large effects on certain species within the herpetofaunal community (Erwin and Stasiak, 1979; McLean, 1990), and more intense fires can lead to an increase in direct mortality of individuals in the area. Other negative responses observed in the herpetofaunal community can be addressed by changes to the macrohabitat (vegetation structure; Mushinsky, 1986), microhabitat (ground temperature, soil moisture, etc.; Blair, 1997; Knapp and Seastedt, 1986), or declines in prey abundances (most often due to lower abundances of arthropods in burned sites when compared to unburned sites; Evans et al., 1983). Annual burning is stressful to certain species, leading to local permanent emigration or local extinction. Past studies on effects of burning on herpetofauna have found variable responses among species. In tropical humid forests, herpetofaunal species richness was significantly higher in fire disturbed areas compared to undisturbed areas (Fredericksen and Fredericksen, 2002). Disturbances on small spatial scales at low frequencies may increase species richness by providing new habitat locally (McLeod and Gates, 1998; Mushinsky, 1985; Pilliod et al., 2003; Russell et al., 1999). However, there are also many studies which have reported that species respond negatively to burning (Cavitt, 2000; Rudolph and Burgdorf, 1997; Setser and Cavitt, 2003), or have a neutral response to disturbance by fire (Greenberg et al., 1994; Hannah and Smith, 1995; Woinarski et al., 1999). If immigration of new species or new individuals of an extirpated species is not possible, species richness levels decline.

Three caveats apply to the results of this study. First, we used an open population model based on the Jolly-Seber model. This model does not allow for heterogeneous capture probabilities across all different species, which may have been violated (Pledger et al., 2003). A violation of this assumption can lead to biases in parameter estimates, including underestimates of species richness (Huggins et al., 2003). A single sampling day per year was used to determine species occurrence, and some species may have been present but overlooked. Missing a species in multiple consecutive years may be due to permanent emigration or local extinction. Multiple survey days within a year would increase the probability of detection and allow for closed population modeling and estimation of species-specific capture probabilities. Last, our analysis was retrospective and an unplanned experiment with no replication. Lack of reference areas does not allow us to evaluate the possibility that declines in species richness were due to regional factors other than land management. Nevertheless, different decline rates in herpetofaunal species coincided with a switch in management practices (burning and stocking practices) from 1997–1999.

Management decisions based upon these results should be applied with caution, and future research with improved sampling methods is needed. Despite the caveats of our retrospective analysis, this 14-year data set remains one of the few long-term records of the herpetofaunal communities in the Flint Hills region of eastern Kansas.

The mark-recapture statistics used in this study provide an ideal method of analyzing community demographics of a multitude of ecosystems disturbed by a variety of human land-use practices.

Acknowledgments

Thanks to B. and J. Nitschke who own BJN Ranch and gave A. Volkmann permission to survey their land. Thanks to all other survey participants including N. Cubbage, J. Greider, R. Greider, J. Lent, E. McCarrier, J. Clark, R. Previtera, G. Volkmann, T. Volkmann, and S. Wiechmann. Financial support was provided by the Divison of Biology of Kansas State University.

Literature Cited

1.

D. I. Axelrod 1985. Rise of the grassland biome, central North America. Botanical Review 51:163–201. Google Scholar

2.

L. L. Bailey, T. R. Simons, and K. H. Pollock . 2004a. Estimating site occupancy and species detection probability parameters for terrestrial salamanders. Ecological Applications 14:692–702. Google Scholar

3.

L. L. Bailey, T. R. Simons, and K. H. Pollock . 2004b. Estimating detection probability parameters for plethodon salamanders using the robust capture-recapture design. Journal of Wildlife Management 68:1–13. Google Scholar

4.

L. L. Bailey, T. R. Simons, and K. H. Pollock . 2004c. Spatial and temporal variation in detection probability of plethodon salamanders using the robust capture-recapture design. Journal of Wildlife Management 68:14–24. Google Scholar

5.

D. Bark 1987. Konza Prairie Research Natural Area, Kansas. 45–50. In D. Greenland , editor. ed. The Climates of the Long-term Ecological Research Sites. Occasional paper no. 44, Institute for Arctic and Alpine Research, University of Colorado. Boulder, Colorado, U.S.A. Google Scholar

6.

D. K. Beutler 1993. The effects of burning tallgrass prairie on the breeding bird community. M.S. Thesis,. Kansas State University. Manhattan, Kansas, U.S.A. Google Scholar

7.

J. M. Blair 1997. Fire, N availability and plant response in grasslands: a test of the transient maxima hypothesis. Ecology 78:2539–2368. Google Scholar

8.

T. Boulinier, J. D. Nichols, J. R. Sauer, J. E. Hines, and K. H. Pollock . 1998. Estimating species richness: the importance of heterogeneity in species detectability. Ecology 79:1018–1028. Google Scholar

9.

R. W. Braithwaite 1987. Effects of fire regimes on lizards in wet-dry tropics of Australia. Journal of Tropical Ecology 3:265–275. Google Scholar

10.

J. M. Briggs and A. K. Knapp . 1995. Interannual variability in primary production in tallgrass prairie: climate, soil moisture, topographic position and fire as determinants of aboveground biomass. American Journal of Botany 82:1024–1030. Google Scholar

11.

K. P. Burnham and D. R. Anderson . 1998. Model Selection and Inference: A Practical Information-theoretic Approach. Springer-Verlag. New York, New York, U.S.A. Google Scholar

12.

K. P. Burnham, D. R. Anderson, G. C. White, C. Brownie, and K. H. Pollock . 1987. Design and analysis methods for fish survival experiments based on release-recapture. American Fisheries Society Monograph 5:1–437. Google Scholar

13.

W. H. Busby and J. R. Parmelee . 1996. Historical changes in a herpetofaunal assemblage in the Flint Hills of Kansas. American Midland Naturalist 135:81–91. Google Scholar

14.

W. H. Busby, J. R. Parmelee, C. M. Dwyer, E. D. Hooper Jr, and K. J. Irwin . 1994. A Survey of the Herpetofauna of the Fort Riley Military Reservation, Kansas. Kansas Biological Survey of Reptiles. No. 58. Lawrence, Kansas, U.S.A. Google Scholar

15.

J. F. Cavitt 2000. Fire and a tallgrass prairie reptile community: effects on relative abundance and seasonal activity. Journal of Herpetology 34:12–20. Google Scholar

16.

A. Chao and T. J. Shen . 2003. Nonparametric estimation of Shannon's index of diversity when there are unseen species in sample. Environmental and Ecological Statistics 10:429–443. Google Scholar

17.

B. K. Clark and D. W. Kaufman . 1990. Short-term responses of small mammals to experimental fire in tallgrass prairie. Canadian Journal of Zoology 68:2450–2454. Google Scholar

18.

B. K. Clark, D. W. Kaufman, E. J. Finck, and G. A. Kaufman . 1989. Small mammals in tallgrass prairie: patterns associated with grazing and burning. Prairie Naturalist 21:177–184. Google Scholar

19.

J. T. Collins 1993. Amphibians and Reptiles in Kansas. 3rd Edition. The University of Kansas Natural History Museum. Lawrence, Kansas, U.S.A. Google Scholar

20.

T. Doan 2003. Which methods are most effective for surveying rain forest herpetofauna? Journal of Herpetology 37:72–81. Google Scholar

21.

W. J. Erwin and R. H. Stasiak . 1979. Vertebrate mortality during the burning of a reestablished prairie in Nebraska. The American Midland Naturalist 101:247–249. Google Scholar

22.

E. W. Evans 1984. Fire as a natural disturbance to grasshopper assemblages of tallgrass prairie. Oikos 43:9–16. Google Scholar

23.

E. W. Evans 1988. Grasshopper assemblages of the tallgrass prairie: influences of fire frequency, topography, and vegetation. Canadian Journal of Zoology 66:1495–1501. Google Scholar

24.

E. W. Evans, R. A. Rogers, and D. J. Opfermann . 1983. Sampling grasshoppers on burned and unburned tallgrass prairie: trapping vs. sweeping. Environmental Entomology 12:1449–1454. Google Scholar

25.

K. Follner and K. Henle . 2001. Closed mark-recapture models to estimate species richness: an example using data on epigeal spiders. Journal of Agricultural, Biological, and Environmental Statistics 6:176–182. Google Scholar

26.

N. J. Fredericksen and T. S. Fredericksen . 2002. Terrestrial wildfire responses to logging and fire in a Bolivian tropical humid forest. Biodiversity and Conservation 11:27–38. Google Scholar

27.

C. C. Freeman and L. C. Hulbert . 1985. An annotated list of the vascular flora of Konza Prairie Research Natural Area, Kansas. Transactions of the Kansas Academy of Sciences 88:84–115. Google Scholar

28.

G. Fyfe 1980. The effect of fire on lizard communities in central Australia. Herpetofauna 12:1–9. Google Scholar

29.

C. H. Greenberg, D. G. Neary, and L. D. Harris . 1994. Effect of high-intensity wildfire and silvicultural treatments on reptile communities in sand-pine scrub. Conservation Biology 8:1047–1057. Google Scholar

30.

A. D. Griffiths and K. A. Christian . 1996. The effects of fire on the frillneck lizard in Northern Australia. Australian Journal of Ecology 21:386–398. Google Scholar

31.

D. S. Hannah and G. C. Smith . 1995. Effects of prescribed burning on herptiles in Southeastern Queensland. Memoirs of the Queensland Museum 38:529–531. Google Scholar

32.

M. L. Heinrich and D. W. Kaufman . 1985. Herpetofauna of the Konza Prairie Research Natural Area, Kansas. Prairie Naturalist 17:101–112. Google Scholar

33.

J. L. Heisler, J. M. Briggs, and A. K. Knapp . 2003. Long-term patterns of shrub expansion in a C-4 dominated grassland: fire frequency and the dynamics of shrub cover and abundance. American Journal of Botany 90:423–428. Google Scholar

34.

J. E. Hines and J. R. Sauer . 1989. Program CONTRAST: a general program for the anlaysis of several survival or recovery rate estimates. United States Fish and Wildlife Service Fish and Wildlife Technical Report 24:1–7. Google Scholar

35.

R. Huggins, H. C. Yang, A. Chao, and P. S. F. Yip . 2003. Population size estimation using local sample coverage for open populations. Journal of Statistical Planning and Inference 113:699–714. Google Scholar

36.

L. C. Hulbert 1988. Causes of fire effects in tallgrass prairie. Ecology 69:46–58. Google Scholar

37.

A. Joern 2004. Variabilities in grasshopper densities in response to fire frequency and bison grazing from tallgrass prairie. Environmental Entomology 33:1617–1625. Google Scholar

38.

A. Joern 2005. Long-term disturbance from fire and bison grazing modulates grasshopper species assemblages (Orthoptera) in tallgrass prairie. Ecology 86:861–873. Google Scholar

39.

G. A. Kaufman and D. W. Kaufman . 1997. Ecology of small mammals in prairie landscapes. 207–243. In F. L. Knopf and F. B. Samson , editors. eds. Ecology and Conservation of Great Plains Vertebrates. Ecological Studies 125. Springer-Verlag. New York, New York, U.S.A. Google Scholar

40.

D. W. Kaufman, E. J. Finck, and G. A. Kaufman . 1990. Small mammals and grassland fires. 46–80. In S. Collins and L. L. Wallace , editors. eds. Fire in North American tallgrass prairies. University of Oklahoma Press. Norman, Oklahoma, U.S.A. Google Scholar

41.

A. K. Knapp and T. R. Seastedt . 1986. Detritus accumulation limits productivity of tallgrass prairie. Bioscience 36:662–668. Google Scholar

42.

A. K. Knapp and T. R. Seastedt . 1998. Grasslands, Konza Prairie, and long-term ecological research. 3–15. In A. K. Knapp, J. M. Briggs, D. C. Hartnett, and S. L. Collins , editors. eds. Grassland Dynamics, Long-Term Ecological Research in Tallgrass Prairie. Oxford University Press. Oxford, U.K. Google Scholar

43.

P. Masters 1996. The effects of fire driven succession on reptiles in spinifex grasslands at Uluru National Park, Northern Territory. Wildlife Research 23:39–48. Google Scholar

44.

S. McLean 1990. Reptile mortalities caused by a grass fire. Lammergeyer 41:46. Google Scholar

45.

R. F. McLeod and J. E. Gates . 1998. Response of herpetofaunal communities to forest cutting and burning at Chesapeake Farms, Maryland. American Midland Naturalist 139:164–177. Google Scholar

46.

B. R. McMillan, D. E. Brillhart, D. W. Kaufman, and G. A. Kaufman . 1995. Short-term responses of small mammals to autumn fire in tallgrass prairie. Prairie Naturalist 27:158–166. Google Scholar

47.

H. R. Mushinsky 1985. Fire and the Florida sandhill herpetofaunal community: with special attention to responses of Cnemidophorous sexlineatus. Herpetologica 41:333–342. Google Scholar

48.

H. R. Mushinsky 1986. Fire, vegetation structure and herpetofaunal communities. 383–388. In Z. Rocek , editor. ed. Studies in Herpetology, Proceedings of the 3rd European Herpetological Meeting. Charles University. Prague, Czechoslovakia. Google Scholar

49.

J. D. Nichols, T. Boulinier, J. E. Hines, K. H. Pollock, and J. R. Sauer . 1998. Inference methods for spatial variation in species richness and community composition when not all species are detected. Conservation Biology 12:1390–1398. Google Scholar

50.

C. E. Owensby, R. Cochran, and E. F. Smith . 1988. Stocking rate effects on intensive-early stocked Flint Hills bluestem range. Journal of Range Management 41:483–487. Google Scholar

51.

E. Pianka 1996. Long-term changes in lizard assemblages in the Great Victoria Desert: dynamic habitat mosaics in response to wildfires. 191–213. In M. L. Cody and J. A. Smallwood , editors. eds. Long-term Studies of Vertebrate Communities. Academic Press. New York, New York, U.S.A. Google Scholar

52.

D. S. Pilliod, R. B. Bury, E. J. Hyde, C. A. Pearl, and P. S. Corn . 2003. Fire and amphibians in North America. Forest Ecology and Management 178:163–181. Google Scholar

53.

S. Pledger, K. H. Pollock, and J. L. Norris . 2003. Open capture-recapture models with heterogeneity: I. Cormack-Jolly-Seber model. Biometrics 59:786–794. Google Scholar

54.

K. H. Pollock, J. D. Nichols, C. Brownie, and J. E. Hines . 1990. Statistical inference for capture-recapture experiments. Wildlife Monographs 107:1–97. Google Scholar

55.

G. D. Potts and J. T. Collins . 2005. A Checklist of the Vertebrate Animals of Kansas. 3rd Edition. Sternberg Museum of Natural History. Hays, Kansas, U.S.A. Google Scholar

56.

A. F. L. A. Powell 2004. Effects of prescribed burning and grazing on breeding bird abundances in tallgrass prairie. M.S. Thesis,. Kansas State University. Manhattan, Kansas, U.S.A. Google Scholar

57.

M. B. Robbins, A. T. Peterson, and M. A. Ortega-Huerta . 2002. Major negative impacts of early intensive cattle stocking on tallgrass prairies: the case of the greater prairie-chicken (Tympanuchus cupido). North American Birds 56:239–244. Google Scholar

58.

R. Rohrbaugh Jr, D. L. Reinking, D. H. Wolfe, S. K. Sherrod, and M. A. Jenkins . 1999. Effects of prescribed burning and grazing on nesting and reproductive success of three grassland passerine species in tallgrass prairie. 165–170. In P. D. Vickery and J. R. Herbert , editors. eds. Ecology and Conservation of Grassland Birds of the Western Hemisphere. Studies in Avian Biology No. 19, Cooper Ornithological Society. Camarillo, California, U.S.A. Google Scholar

59.

D. C. Rudolph and S. J. Burgdorf . 1997. Timber rattlesnakes and Louisiana pine snakes of the West Gulf Coastal plain: hypothesis of decline. Texas Journal of Science 49:11–122. Google Scholar

60.

K. R. Russell, D. H. Van Lear, and D. C. Guynn Jr . 1999. Prescribed fire effects on herpetofauna: review and management implications. Wildlife Society Bulletin 27:374–384. Google Scholar

61.

F. Samson and F. Knopf . 1994. Prairie conservation in North America. Bioscience 44:418–421. Google Scholar

62.

B. R. Schmidt 2003. Count data, detection probabilities, and the demography, dynamics, distribution, and decline of amphibians. Comptes Rendus Biologies 326:119–124. Google Scholar

63.

B. R. Schmidt, M. Schaub, and B. R. Anholt . 2002. Why you should use capture-recapture methods when estimating survival and breeding probabilities: on bias, temporary emigration, overdispersion, and common toads. Amphibia-Reptilia 23:375–388. Google Scholar

64.

C. J. Schwarz and A. N. Arnason . 1996. A general methodology for the analysis of capture-recapture experiments in open populations. Biometrics 52:860–873. Google Scholar

65.

T. R. Seastedt 1984a. Microarthropods of burned and unburned tallgrass prairie. Journal of the Kansas Entomological Society 57:468–476. Google Scholar

66.

T. R. Seastedt 1984b. Belowground microarthropods of annually burned and unburned tallgrass prairie. American Midland Naturalist 111:405–408. Google Scholar

67.

K. Setser and J. F. Cavitt . 2003. Effects of burning on snakes in Kansas, USA, tallgrass prairie. Natural Areas Journal 23:315–319. Google Scholar

68.

E. F. Smith and C. E. Owensby . 1978. Intensive-early stocking and season-long stocking of Kansas Flint Hills range. Journal of Range Management 31:14–17. Google Scholar

69.

G. G. Thompson, P. C. Withers, E. R. Pianka, and S. A. Thompson . 2003. Assessing biodiversity with species accumulation curves; inventories of small reptiles by pit-trapping in Western Australia. Austral Ecology 28:361–383. Google Scholar

70.

C. R. Trainor and J. C. Z. Woinarski . 1994. Responses of lizards to three experimental fires in the savanna forests of Kakadu National Park. Wildlife Research 21:131–148. Google Scholar

71.

G. C. White and K. P. Burnham . 1999. Program MARK: Survival estimation from populations of marked animals. Bird Study 46:120–139. Google Scholar

72.

D. J. Wilgers 2005. Impacts of land management practices on tallgrass prairie herpetofauna. M.S. Thesis,. Kansas State University. Manhattan, Kansas, U.S.A. Google Scholar

73.

J. C. Z. Woinarski, C. Brock, A. Fisher, D. Milne, and B. Oliver . 1999. Response of birds and reptiles to fire regimes on a pastoral land in the Victoria River District, Northern Territory. Rangeland Journal 21:24–38. Google Scholar

74.

J. E. Woodford and M. W. Meyer . 2003. Impact of lakeshore development on green frog abundance. Biological Conservation 1:277–284. Google Scholar

75.

J. L. Zimmerman 1992. Density-independent factors affecting the avian diversity of the tallgrass prairie community. Wilson Bulletin 104:85–94. Google Scholar

76.

J. L. Zimmerman 1997. Avian community responses to fire, grazing, and drought in the tallgrass prairie. 167–180. In F. L. Knopf and F. B. Samson , editors. eds. Ecology and conservation of the Great Plains vertebrates. Springer-Verlag. New York, New York, U.S.A. Google Scholar
Dustin J. Wilgers, Eva A. Horne, Brett K. Sandercock, and Allan W. Volkmann "EFFECTS OF RANGELAND MANAGEMENT ON COMMUNITY DYNAMICS OF THE HERPETOFAUNA OF THE TALLGRASS PRAIRIE," Herpetologica 62(4), 378-388, (1 December 2006). https://doi.org/10.1655/0018-0831(2006)62[378:EORMOC]2.0.CO;2
Accepted: 1 June 2006; Published: 1 December 2006
JOURNAL ARTICLE
11 PAGES


SHARE
ARTICLE IMPACT
Back to Top