Printer Friendly

Differential use of grazed and ungrazed plots by Dipodomys elator (Mammalia: heteromyidae) in north central texas.

Abstract. - Two populations of Texas kangaroo rat (Dipodomys elator) were compared at a grazed and ungrazed pasture in Wichita County, Texas in order to examine differences in vegetation and Texas kangaroo rat population size. Within vegetation quadrats, species richness, height, percent grass, forbs, woody, and bare ground were examined. Vegetation height and type were examined along transects centered in the quadrat and extending outward in the four cardinal directions. Dipodomys elator was found inhabiting areas of short, sparse grasses with little overhead woody cover. Trapping revealed a higher concentration of D. elator at the grazed site.

Dipodomys elator is unusual because the habitat in which it is found is not typical among kangaroo rats. Dipodomys elator seems to prefer soils with high clay content which support overgrazed or short grasses (Dalquest & Collier 1964; Roberts & Packard 1973; Dalquest & Horner 1984; Schmidly 2004; Goetze et al. 2007) and is rarely recorded in locations with dense vegetation.

The historic range of D. elator spanned across the convergence of two physiographic regions, the Rolling Plains to the west, and the West Cross Timbers to the east (Carter et al. 1985; Jones et al. 1988; Martin 2002). Martin (2002) surveyed the entire historic range of D. elator and found this species in only five counties in Texas: Archer, Childress, Hardeman, Motley, and Wichita, all within the Rolling Plains region of Texas. The perceived decline in D. elator has led to some limited protective status. D. elator is listed as a threatened species by the Texas Parks and Wildlife Department (Martin 2002; Schmidly 2004).

Researchers need to quantify habitat critical to D. elator survival (Martin & Matocha 1972; Jones et al. 1988). Therefore, vegetation and corresponding population size were compared for two D. elator populations found in two habitats whose structure was influenced by grazing, or lack thereof, within Wichita County.


Study sites.--The grazed site (approximately 15 ha) is located 10 km N of Lake Buffalo Creek Reservoir, Wichita Co., Texas, immediately adjacent to Burnett Ranch Rd. (property entrance Lat: 34.03028709 Long: 98.76656946) (Figure 1). This is a location of previous D. elator research involving trapping and burrow surveys, as well as descriptions of the associated habitat (Stangl et al. 1992; Goetze et al. 2007). The grazed site is continuously grazed by cattle at a stocking rate of about 0.83 animals per ha. As a result, vegetation is relatively short and unburned brush piles within this pasture have collapsed over time and accumulated soil, creating mounds which are favored as burrowing sites for D. elator (Stangl et al. 1992; Goetze et al. 2007).


The ungrazed site (approximately 16 ha) is located 3.7 km W of the junction of FM 2345 and FM 368, Wichita County, Texas (property entrance Lat: 34.06475002 Long: 98.71983118). This site has not been grazed by livestock since at least 1999 and has not recently been subjected to brush control. As a result, herbaceous vegetation is thick and rank, and residual cover remains from year to year. The dominant woody vegetation is dense, mature mesquite (Prosopis glandulosa).

Population surveys.-Burrow trapping was conducted by placing three 7.5 by 8.75 by 30 cm Sherman live traps immediately around each active burrow entrance (within 0.10 to 0.50 m of each burrow entrance), with the open end of the trap facing the entrance (Cross & Waser 2000). For the grazed site, trapping was conducted throughout the spring and summer months of 2005 during the following dates (parentheses indicate number of trap nights): 18-24 May (534), 21-25 June (210), 6-7 July (84), and 19-22 July (168). Trapping was conducted on the ungrazed site throughout the spring and summer months of 2005 during the following dates: 15-18 March (90), 10-12 May (141), 24 May (30), 5-7 June (249), 21-25 June (210), 6-7 July (84), and 19-22 July (168).

Captured D. elator were tagged with passive integrated transponder (PIT) tags in order to determine recapture rates. PIT tags were implanted subcutaneously just posterior to the cranium with a 12ga. syringe fitted with a plunger. In order to minimize handling time and other stresses, anesthesia was not used (Schooley et al. 1993). Syringes were disinfected between implantations with 91% alcohol.

Vegetation.--A survey of vegetation was conducted in May when most species were in bloom. A rectangular quadrat (1[m.sup.2]) was placed directly over each of 10 known D. elator burrows at the grazed site. Also, three known D. elator burrows and seven additional sites within the ungrazed area were surveyed. Within each quadrat, dominant species were noted, and species richness recorded. Percentage of grass, forb, woody vegetation, bare ground, and average herbaceous vegetation height (obtained by taking four samples 15 cm interior to each corner of the quadrat) were noted. If woody vegetation was present, its height was measured.

To quantify vegetation of the habitat surrounding sampled sites, two 20 m north-south and east-west transects, bisecting each quadrat were evaluated. Vegetation type (grass, forb, bare ground, woody), height, and vertical intercept of woody vegetation (distance from ground to woody species in cm directly over meter-point) was recorded at each meter-point along transects.

Data analysis.-Vegetation data from within quadrats and between quadrats and transects were compared between sites utilizing the Wilcoxon Mann-Whitney test of SAS 8.0 (SAS Institute 1999). Height values were compared with a paired T-test of SAS 8.0 (SAS Institute 1999). Parameters of herbaceous height, and percentages of grasses, forbs, bare ground, woody, and species richness were all evaluated for statistically significant (P<0.05) differences.

Vegetation data recorded along transects was compared between sites, percentage vegetation type, height, transect direction (North, East, South or West), and transect meter-point distance (1-10 m), using the generalized linear model procedure (Proc GLM) of SAS 8.0 (SAS Institute 1999). Parameters of herbaceous height and percentages of grasses, forbs, bare ground, woody and other were all evaluated for statistically significant (P<0.05) differences across direction, distance, and site.


Burrow trapping.--There was a total of 1968 trap nights (trap night = 1 trap/night) for the two sites combined. At the grazed site, 996 trap nights over 18 trapping periods resulted in a total of 67 captures, 45 (67%) of which were D. elator. Eighteen different individuals accounted for the 45 D. elator captures, caught at 22 different burrows. There were 22 non-target captures (species other than D. elator) at the grazed site representing four species of rodents. Of the non-target species, Chaetodipus hispidus was the most common (10 captures), followed by Spermophilus tridecemlineatus (six captures), Neotoma micropus (four captures), and Peromyscus leucopus (two captures).

At the ungrazed site, 972 trap nights resulted in a total of 220 captures, of which eight were D. elator (Table 1). Two different individuals accounted for the eight D. elator captures caught at three different burrows. There were 212 non-target captures at the ungrazed site with the most common species being Sigmodon hispidus (119 captures) followed by Peromyscus maniculatus (36 captures), P. leucopus (28 captures), N. micropus (16 captures), and C. hispidus (13 captures).
Table 1. Average vegetative height, mean richness and mean percent
cover of quadrats compared between sites. Statistics reported for
percentages and richness were evaluated by a Wilcoxon Mann-Whitney test
(SD). Statistics reported for height data were compared by a paired
T-test (SD).

   Parameter          Grazed Site          Ungrazed Site         P

Mean % Bare      49.9 ([+ or -] 24.0)  18.2 ([+ or -] 25.5)   0.027 *

Mean % Forbs     17.0 ([+ or -] 12.9)  13.8 ([+ or -] 17.7)   0.168

Mean % Grasses   25.1 ([+ or -] 18.9)  61.5 ([+ or -] 31.1)   0.010 *

Mean % Woody      6.0 ([+ or -] 15.8)   4.5 ([+ or -] 12.6)   0.479

Mean Richness     5.8 ([+ or -] 2.2)    6.5 ([+ or -] 2.8)    0.230

Avg. Herbaceous   7.1 ([+ or -] 6.8)   27.7 ([+ or -] 11.0)  <0.001 *
Height (cm)

Avg. Woody       15.9 ([+ or -] 38.8)  20.1 ([+ or -] 44.4)   0.842
Height (cm)

* Indicates statistical significance at (P<0.05).

Vegetation.-Little barley (Hordeum pusillum) was the dominant grass species recorded at all ten quadrats within the grazed site. Virginia pepperweed (Lepidium virginicum) was the dominant forb the majority of the time (six quadrats). Other dominant forbs within quadrats at the grazed site were western ragweed (Ambrosia psilostachya), common broomweed (Gutierrezia dracunculoides), and hog potato (Hoffmannseggia glauca). One mesquite (Prosopis glandulosa) and one lotebush (Ziziphus obtusifolia) occurred in two of the quadrats at the grazed site.

Within the ungrazed site, Japanese brome (Bromus japonicus) was the dominant grass species in nine out of ten quadrats. Little barley was the dominant grass in one quadrat. Western ragweed was the dominant forb in most quadrats (eight quadrats). Other dominant forbs were common broomweed and nightshade (Solarium spp.). Mesquite was the only woody species occurring in two quadrats.

Significant differences were observed in the quadrats sampled at each site (Table 1). Percentage bare ground was significantly higher within quadrats at the grazed site (mean 50%) than at the ungrazed site (mean 18%). The mean percentage of grass at the ungrazed site was significantly different than the grazed site (62% as compared to 25%). Herbaceous height within quadrats was significantly different between sites and was, on average, three times higher within the ungrazed site than quadrats within the grazed site (average herbaceous height 27.7 cm and 7.1 cm respectively). Percentage coverage of forbs and woody vegetation and of woody vegetation height did not differ significantly between sites. Species richness was not significantly different between sites.

Within the grazed site there was significantly more bare ground within quadrats (49.9%) than along transects (22.3%) (Table 2). Also, there was significantly less grass within quadrats (25.1%) than along transects (54.3%). At the grazed site, there were no significant differences between average herbaceous and woody height, percent forbs, and woody vegetation within quadrats and along transects. Within the ungrazed site, there were no significant differences between vegetation sampled within quadrats and along transects (Table 2).
Table 2. Mean percentages bare ground, forbs, grasses, and woody
vegetation as well as average herbaceous and woody vegetation height
compared between quadrats and transects. Statistics reported for
percentages were evaluated by a Wilcoxon Mann-Whitney test (SD).
Statistics reported for height data were compared by a paired T-test

                                            Grazed Site

    Parameter            Quadrats             Transects           P

Mean % Bare        49.9 ([+ or -] 24.0)  22.3 ([+ or -] 8.7)   0.010 *

Mean % Forbs       17.0 ([+ or -] 12.9)  22.9 ([+ or -] 9.8)   0.177

Mean % Grasses     25.1 ([+ or -] 18.9)  54.3 ([+ or -] 9.5)   0.002 *

Mean % Woody        6.0 ([+ or -] 15.8)   0.5 ([+ or -] 1.6)   0.256

Avg. Herbaceous     7.1 ([+ or -] 6.7)    7.2 ([+ or -] 3.7)   0.934
Height (cm)

Avg. Woody Height  15.9 ([+ or -] 38.8)  10.2 ([+ or -] 32.3)  0.700

                                           Ungrazed Site

    Parameter            Quadrats             Transects           P

Mean % Bare        18.2 ([+ or -] 25.5)  17.5 ([+ or -] 20.4)  0.197

Mean % Forbs       13.8 ([+ or -] 17.7)  13.0 ([+ or -] 5.9    0.119

Mean % Grasses     61.5 ([+ or -] 31.1)  67.9 ([+ or -] 22.2)  0.500

Mean % Woody        4.5 ([+ or -] 12.6)   0.7 ([+ or -] 1.2    0.442

Avg. Herbaceous    27.7 ([+ or -] 11.0)  27.6 ([+ or -] 11.5)  0.957
Height (cm)

Avg. Woody Height  20.1 ([+ or -] 44.4)  32.5 ([+ or -] 58.2)  0.112

* Indicates statistical significant diffenence (P<0.05)


Species composition of the rodent populations at the two sites is different. The grazed site is predominately inhabited by D. elator, whereas the ungrazed site is predominately inhabited by Sigmodon hispidus and Peromyscus spp. The high population of S. hispidus (119 captures) and extremely low occurrence of D. elator (2) at the ungrazed site concurs with Packard & Roberts (1973) observation that the Texas kangaroo rat and hispid cotton rat rarely co-occur. Other notable species compositional differences between sites were the absence of P. maniculatus and the presence of S. tridecemlineatus at the grazed site. These species reflect the habitat differences between the sites. Peromyscus maniculatus prefers areas where vegetation offers concealment, such as tall forbs and grasses (Schmidly 2004). Spermophilus tridecemlineatus is an inhabitant of short-grass prairies (Schmidly 2004) and prefers heavily grazed pastures (Jones 1964; Streubel & Fitzgerald 1978). Therefore, it is reasonable to conclude that the grazed site is more indicative of suitable D. elator habitat.

At the grazed site, D. elator utilized existing mounds (old brush piles, fence rows, and abandoned farm equipment that has accumulated soil) for burrow construction, but it was observed that D. elator will readily excavate burrows on natural, slightly raised areas. Thirty-three percent of the burrows at the grazed site were associated with these types of areas. Contrary to Dalquest & Collier (1964), who invariably associated D. elator burrows with mesquite at their study site, only 3.0% of burrows at the grazed site were associated with mesquite. When burrows associated with lotebush (Zizyphus obtusifolia) are added, the percentage of burrows associated with woody species rises to 7.5%. Goetze et al. (2007) found burrows associated with mesquite 6.0% of the time. When burrows associated with lotebush were added, the percentage of burrows associated with woody species rose to 15.0% (Goetze et al. 2007). Other researchers have noted an association of D. elator with mesquite (Dalquest & Collier 1964; Chapman 1972; Roberts & Packard 1973; Carter et al. 1985; Schmidly 2004). However, this investigation concurs with Stangl et al. (1992) and Goetze et al. (2007) in suggesting that woody vegetation is not essential for D. elator burrows. Burrow site selection by D. elator seems to be based primarily on a disturbance regime and the presence of bare ground (mean 49.9% at the grazed site), in agreement with Stangl et al. (1992), Martin (2002) and Goetze et al. (2007). Within the grazed site, 50% of burrows were associated with disturbance (fence rows, brush piles, and/or human structures). Goetze et al. (2007) reported 56.0% of burrows associated with disturbance.

Overall, D. elator habitat is dominated by short vegetation (between 3.5 cm and 10.9 cm in height at the grazed site) with very little overhead woody cover (grazed site < 1.0% coverage). The microhabitat immediately around burrows at the grazed site contained significantly more bare ground and less grass than transects sampled further from the burrows (Table 2). The overall habitat values from this investigation compare favorably with percent grass (avg. 75.5%), forbs (avg. 15.8%), and bare ground (avg. 18.9%) data collected in Hardeman County, Texas for D. elator habitat (Martin 2002). There is general agreement that D. elator requires a sparse, short-grassland habitat (Dalquest & Collier 1964; Roberts & Packard 1973; Carter et al. 1985; Stangl et al. 1992; Martin 2002; Goetze et al. 2007).

Much land is under cultivation within the range of D. elator. Routine tillage and the resulting agronomic monocultures render such areas uninhabitable to this species (Stangl et al. 1992). Many areas that are not in crop production have been developed for gas and oil exploration. Associated disturbances, such as road construction and discarded equipment that accumulates soil, are thought to be beneficial for kangaroo rats (Roberts & Packard 1973; Stangl & Schafer 1990; Stangl et al. 1992; Martin 2002; Goetze et al. 2007). The use of fire to control woody species is uncommon and mechanical means are often too costly. These circumstances allow areas to develop dense stands of mesquite and the herbaceous understory to become dense, as noted at the ungrazed site in this study. Based upon rodent species composition at the two sites, lack of grazing and these other extrinsic factors may have favored dense mid-grass to woodland vegetational communities that include 5. hispidus and P. maniculatus as noted by Schmidly (2004). In 1985, D. elator was recorded from two separate locations in Hardeman County, Texas. When these sites were visited again in 1990, the vegetation had become denser and D. elator had been extirpated from both locations (Stangl & Schafer 1990).

It is speculated that historically the short, sparse, grassland habitat that D. elator requires was maintained by buffalo (Bos bison), and/or prairie dogs (Cynomys ludovicianus) and naturally-occurring wildfires (Stangl et al. 1992). Currently, neither of these species or fires have a major impact on the environment. To develop and maintain favorable habitat for D. elator, systems of moderate to intense grazing pressure need to be implemented that mimic grazing and disturbance by bison and prairie dogs (Stangl et al. 1992; Nelson et al. 2009).

In addition to these management practices, continued research of D. elator is needed to assure the future stability of the species. Specific habitat requirements need to be defined from additional localities within the geographical range of D. elator. Genetic variation and minimum sustainable population size will need to be assessed.


Sincere thanks are owed to Ernest, Oscar and Edith Goetze for access to the study sites located in Wichita County, Texas. We also thank Mike Miller of the Texas Parks and Wildlife Department and Roger Wittie in the College of Agriculture and Human Sciences at Tarleton State University who provided field equipment and technical advice. We thank Robert E. Martin and an anonymous reviewer for their thoughtful comments that improved this manuscript.


Carter, D. C., W. D. Webster, J. K. Jones JR., C. Jones & R. D. Suttkus. 1985. Dipodomys elator. American Society of Mammalogists, Mammalian Species, 232:1-3.

Chapman, B. R. 1972. Food habits of Loring's kangaroo rat, Dipodomys elator. Journal of Mammalogy, 53:877-880.

Cross, C. L. & P. M. Waser. 2000. Estimating population size in the banner-tailed kangaroo rat. Southwestern Naturalist. 45:176-183.

Dalquest, W. W. & G. Collier. 1964. Notes on Dipodomys elator, a rare kangaroo rat. Southwestern Naturalist, 9:146-150.

Dalquest, W. W. & N. V. Horner. 1984. Mammals of north-central Texas. Midwestern State University Press. Wichita Falls, Texas, 261 pp.

Goetze, J. R., W. C. Stasey, A. D. Nelson & P. Sudman. 2007. Habitat attributes and population size of Texas kangaroo rats on an intensely grazed pasture in Wichita County. Texas. Texas Journal of Science, 59:11-22.

Jones, J. K., Jr. 1964. Distribution and taxonomy of mammals of Nebraska. University of Kansas Publications, Museum of Natural History, 16:1-356.

Jones, C., M. A. Bogan & L. M. Mount. 1988. Status of the Texas kangaroo rat (Dipodomys elator). The Texas Journal of Science, 40:249-258.

Martin, R. E. 2002. Status and long term survival of the Texas kangaroo rat, Dipodomys elator. Sect. 6, Project 70,. Unpublished report prepared for Texas Parks and Wildlife Department. Austin. Texas, 44 pp.

Martin, R. E. & K. G. Matocha. 1972. Distributional status of the kangaroo rat, Dipodomys elator. Journal of Mammalogy, 53:873-877.

Nelson, A. D., J. R. Goetze, E. Watson & M. Nelson. 2009. Changes in vegetation patterns and its effect on Texas kangaroo rats (Dipodomys elator). Texas Journal of Science, 61(2): In Press

Packard, R. L. & J. D. Roberts. 1973. Observations on the behavior of Texas kangaroo rat. Dipodomys elator Merriam. Mammalia. 37:680-682.

Roberts, J. D. & R. L. Packard. 1973. Comments on movements, home range and ecology of the Texas kangaroo rat, Dipodomys elator Merriam. Journal of Mammalogy. 54:957-962.

SAS Institute. 1999. SAS companion for the Microsoft windows environment, version 8. NC SAS Publishing. Cary, North Carolina.

Schooley, R. L., B. Van Home & K. P. Burnham. 1993. Passive integrated transponders for marking free-ranging Townsend's ground squirrels. Journal of Mammalogy, 74:480-484.

Schmidly, D. J. 2004. The mammals of Texas. Texas Parks and Wildlife Press, Austin, Texas, USA, 501 pp.

Stangl Jr., F. B. & T. S. Schafer. 1990. Report on the current status of the Texas kangaroo rat, Dipodomys elator, in north-central Texas. Unpublished report prepared for Texas Parks and Wildlife Department, Austin, Texas, 17 p.

Stangl, Jr., F. B., T. S. Schafer, J. R. Goetze & W. Pinchak. 1992. Opportunistic use of modified and disturbed habitat by the Texas kangaroo rat (Dipodomys elator). The Texas Journal of Science. 44:25-35.

Streubel, D. P. & J. P. Fitzgerald. 1978. Spermophilus tridecemlineatus. American Society of Mammalogists, Mammalian Species, 103:1-5.

JRG at:

William C. Stasey, Jim R. Goetze *, Philip D. Sudman and Allan D. Nelson

Department of Biological Sciences, Tarleton State University Stephenville, Texas 76402 and * Natural Sciences Department, Laredo Community College Laredo. Texas 78040
COPYRIGHT 2010 Texas Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Stasey, William C.; Goetze, Jim R.; Sudman, Philip D.; Nelson, Allan D.
Publication:The Texas Journal of Science
Article Type:Report
Geographic Code:1U7TX
Date:Feb 1, 2010
Previous Article:Population dynamics of an established reproducing population of the invasive apple snail (Pomacea insularum) in suburban southeast Houston, Texas.
Next Article:Distribution of the pallid shiner, Hybopsis amnis (Cypriniformes: cyprinidae), in Arkansas.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters