Printer Friendly

Preferential Burrow entrance placement in the Dulzura kangaroo rat, Dipodomys simulans.

Abstract.--This study examined burrow location preferences in the Dulzura kangaroo rat, Dipodomys simulans. Burrow entrances known to be occupied by individual kangaroo rats on two experimental sites were identified and the distances to the nearest shrub were determined. These data were then compared to a distribution of random points delineated on those same plots to test whether this species appears to have a spatial preference for burrow placement in relation to vegetative cover. The results suggest that this species prefers to locate its burrow entrances in the open, but within 2 m of vegetative cover.


Kangaroo rats generally coexist with several species of murids including deer mice, Peromyscus spp, and wood rats, Neotoma spp (Brown and Leiberman 1973; Rosenzweig and Winakur 1969). Other studies report coexistence with other heteromyids, the pocket mice, Perognathus spp. and Chaetodipus spp. (Jones 1993; Brown and Harney 1993; Thompson 1982). The structure of these rodent communities appears to be organized with microhabitats and food resources partitioned in such a manner that abundant and diverse species may be maintained. Brown and Munger (1985) showed that the population density of smaller seed-eating rodents increases after removal of the local kangaroo rat population, indicating that competition for resources may be responsible for the structure of these communities.

Cover utilization also seems to be an important factor in determining community structure, with some heteromyids preferring open areas, some dense cover, and others seeming to have no preference. Experiments have shown that the distribution of species in a community can be altered by artificially manipulating the density of cover (Rosenzweig 1973). Rosenzweig and Winakur (1969) found that the kangaroo rats Dipodomys merriami and D. spectabilis prefer open areas for foraging, whereas the pocket mice tend to utilize dense cover almost exclusively for this purpose. Rosenzweig (1973) further states that this difference in habitat use patterns is crucial to their coexistence. Other studies support these findings, reporting that another species of kangaroo rat (D. deserti) also prefers to forage in the open and that this choice is made in response to competitive pressure from coexisting pocket mice (Eisenberg 1963).

Kangaroo rats possess adaptations such as large auditory bullae tuned to low frequencies (Lay 1993), bipedality, and eyes located dorsolaterally, all of which enhance the animals' ability to escape predation (Thompson 1982). With these adaptations, kangaroo rats have an advantage over other coexisting rodent species, enabling them to better exploit open habitats.

Although there has been substantial work done on foraging habits of heteromyids in relation to vegetative cover, there appears to be very little concerning burrow site selection in relation to vegetative cover. Tremor and Haas (2000) report that D. merriami maintains multiple entrances to a complex burrow system that it preferentially places at the base of shrubs. The Texas kangaroo rat, D. elator, places its burrow at the base of mesquite plants, often with the root system forming one side of the entrance (Davis and Schmidly 1997).

In this report, Dulzura kangaroo rat (Dipodomys simulans) burrow entrance location data collected during a food augmentation study conducted by Yvonne Moore were utilized to examine spatial dispersion of burrow entrances in relation to vegetative cover.


I studied burrow locations at two sites in Riverside County, California where Moore has conducted long-term studies: one at the Motte Rimrock Reserve in Perris, the other at the Shipley Ecological Preserve in Hemet. Both sites consist of coastal sage scrub with the Motte Reserve dominated by California buckwheat (Eriogonum fasciculatum), brittle bush (Encelia farinosa), and black sage (Salvia mellifera). Buckwheat, white sage (S. apiana), and the wild flower (Keckiellia antirrhinoides) dominate the Shipley Preserve. At the Motte Reserve, eight plots (Plots 1-8), each 40 by 40 m with a 5 by 5 grid of live-trapping stations spaced 10 m apart, had been previously established by Moore. There were four plots at the Shipley Preserve, each 50 by 60 m with a 6 by 7 grid of live-trapping stations spaced 10 m apart (Plots 13, 14, 18, and 20).

Burrow entrance data were obtained from Moore, who used methods described by Boonstra and Craine (1986) to track animals back to their burrows upon release. The method employs silk thread bobbins attached to the hindquarters of the animals with glue. The loose end of the thread is tied to a stake or branch and the animal is left to make its way back to its burrow, sometimes after further foraging. The thread can later be traced to the burrow. The burrows were then numbered and flagged for future focal trapping.

The present study characterized burrow entrances as either under the cover of shrubs, or in the open (beyond the dripline of a shrub). The location of burrows in the open was measured as the distance from the center of the burrow entrance to the dripline of the nearest shrub. For this study I characterized cover as that which would appear to effectively conceal rodents from predators. Therefore, shrubs were only considered cover if their foliage and branches exceeded 1 cubic foot or [(30 cm).sup.3] in volume.

In order to obtain a random distribution of points to use for comparison to the actual burrow locations, 240 random numbers between 0 and 1 to six decimal places were generated using Microsoft Excel in which numbers are generated from a uniform distribution. These numbers were then divided evenly into two separate columns, and designated as x and y coordinates. Next, both columns were divided into groups of ten, giving twelve sets of ten rectangular coordinates, one set of 10 pairs for each plot. Each column was then multiplied by the dimensions of its corresponding plot producing a system in which every point in the plot was potentially represented. The perimeter boundaries of two sides of each plot were delineated by driving stakes at three corners of the plot and then pulling twine along two perpendicular sides and tying it to the stakes. These lines were used as x and y-axes from which the random points could be measured. Ten random points were determined on each plot, and then the distance from each point to the nearest shrub was measured using the same criteria as that used for the actual burrows.

The statistical computer software package used to analyze all the data of this project was JMP, by SAS Institute Inc.

All data were tested for normality using the Shapiro-Wilk W Test. The data for actual burrow location distances from the nearest shrub were non-normal (W = 0.7027, P < 0.0001). Random point data also were non-normal (W = 0.7398, P < 0.0001). Therefore, all data were log-transformed [ln(distance+ 1)] and tested for normality. The transformed data were still distributed non-normally: burrows W = 0.9228, P < 0.0001; random points W = 0.7610, P < 0.00001. Non-parametric statistical methods were then employed to analyze the data in three ways:

1. The Mann-Whitney U test was used for the two-sample test between the burrow and random data for the single combined data set.

2. Differences between burrow and random points on each individual plot were also tested using the Mann-Whitney U test.

3. In order to test the difference in burrow placement between plots, the Kruskal-Wallis test was used.


1. Combined data set: Overall, the pattern of burrow distribution differed significantly from the random distribution when all data were analyzed together ([U.sub.0.05(1),74,120] = 18.344, P < 0.0001, S.[D.sub.burrow] = 0.968, S.[D.sub.random] = 0.477). This indicates that this species of kangaroo rat generally prefers to locate its burrows in a non-random fashion, favoring open areas to those beneath the vegetative canopy.

2. Individual plots: Actual burrow entrance measurements from each plot were compared with those random point distributions from the same plot (note: there were no burrows recorded on plot 4). These results varied from plot to plot (Table 1). There were significant differences in Plots 1, 2, 5, 7, 14, and 18 between burrow and random point distances to nearest shrub. Plot 7 had the greatest difference ([U.sub.0.05(1),5,10] = 9.912, P = 0.0008) with a mean burrow distance from shrub cover of 0.832 m (n = 5) and all ten random points located under cover (mean random point distances to nearest shrub = 0.0, Table 1).

Plots 3, 6, 8, 13, and 20 did not differ significantly. However, Plots 3 and 6 had only two and three burrows respectively, so the Mann-Whitney U test had very little power to detect differences (Table 1).

3. Between plot analysis: Burrow-to-shrub distances for each plot were compared to test for between plot variation. The non-parametric Kruskal-Wallis test showed no significant difference ([X.sup.2.sub.0.05] = 14.801, P = 0.1395) (Note: Since plot 4 had no burrows, 11 plots are left in the analysis, and H is approximated by [X.sup.2] with 10 df.).

Plot 1 had an uncharacteristically high mean distance (mean = 2.286 m) due to three outliers of 3.05, 3.66, and 6.10 m. All remaining plots had mean distances between 0.3 m and 0.9 m. Sparse vegetation on plot 1 could be responsible for this distribution. If this were the case, it should also be reflected by the random distribution. However, of the 10 random points on Plot 1, none was greater than 1 m from the nearest shrub cover. Burrows located more than 1 m from cover were unusual in this study, occurring only on Plot 1 and representing 16% of burrow data. Only 4% of the total burrow data set was beyond 3 m. It should also be noted that 23% of the burrows measured were under cover, while 55% of the random points landed under shrubs (Figure 1). Forty-nine percent of the burrows were within 1 m of cover and 65% between 0 and 2 m. Of the random points, 33% were within 1 m and 44% were between 0 and 2 m. There was only one random point beyond 2 m from cover (<1%), whereas 8% of the actual burrow entrances were beyond 2 m from cover (Figure 1). These data indicate a preference for burrow placement at distances of less than 2 m, but not under the canopy of the nearest shrub.



Evidence provided by this study indicates that D. simulans places its burrows in a non-random fashion, favoring open areas to those under the canopy. This is indicated by the individual plot data as well as the combined data set. These findings suggest that this kangaroo rat has different burrow site selection requirements than those reported for D. merriami (Tremor and Haas 2000) and for D. elator (Davis and Schmidly 1997). These sources reported that the distribution was non-random, but the preference for burrow entrance locations was near the base of shrubs, rather than in the open.

On all plots, except Plot 3 and 8, the trend was that the mean burrow distances were greater than those of the random points. The data from all plots combined (Table 1) indicate a preference for burrow entrances located within 1 m of the nearest shrub, but not directly beneath shrub cover. Whether or not these trends are due to spatial partitioning as the result of interspecific competition for higher quality sites is not known. The data suggest that preference of this species for open areas leaves a majority of the areas under the canopy open for settlement by its competitors, Chaetodipus and Perognathus. Jones (1993) reports that most heteromyids defend their burrows and burrow defense is an indication of competitive pressure.

In order to address the question of what is influencing D. simulans to choose open areas for burrow entrance placement, a study similar to this one should be conducted on the species of pocket mice coexisting with them. If the Chaetodipus and Perognathus species are indeed making use of the areas under the canopy of the local vegetation for their burrows, then it will be possible to design a study to test whether this preference is due to interspecific competition. By altering the structure of the shrub cover in an area selected for its occupation by a community with a known distribution of these heteromyids, and observing any changes in the distribution and abundance of these species over time, it should be possible to determine what factors influence the decisions these animals make in regards to burrow placement.

Although this study did detect trends in burrow placement in relation to the nearest shrub cover, spatial distribution, density, and species composition of the local vegetation were not quantified. An examination of these features might demonstrate whether or not this kangaroo rat has a preferred shrub species or a preferred side of the shrub. This may be important if the burrow is north of the nearest shrub because on a moonlit night the shrub might shield the burrow entrance from the moonlight, thereby making it safer for the kangaroo rat to emerge.

It also would be interesting to study a location with a different composition of heteromyids to see how resources are partitioned. Does the larger species exploit the open areas in other regions with a different species composition? Is it typically the kangaroo rats that prefer the open areas for placement of their burrows?

Finally, do these same patterns apply to other taxonomic groups as well? If so, perhaps optimality and game theory models can be applied to this system so that predictions can be made about the behavior of these rodent communities just by having some information on the species composition.
Fig. 1 Percentage of total number of burrows or random points from
each data set at various distance to cover.

Percent Burrows/Random Points

Distance from Shrub (m)

 Burrow Random

0 23 55
0 to 1 49 33
1 to 2 16 11
2 to 3 4 0.83
>3 4 0

Note: Table made from bar graph.


I wish to thank Yvonne Moore for allowing me the use of her study sites, and Wendy Binder and Stephen Adolph for their support during the project. Ronald Quinn, Emil Morhardt, Glenn Stewart, David Moriarty, Gary Carlton, and Jim des Lauriers commented on earlier drafts of the paper. I also thank the reviewers for their comments on the manuscript.

Literature Cited

Boonstra, R., and I. T. M. Craine. 1986. Natal nest location and small mammal tracking with a spool and line technique. Can. Jour. Zoology 64:1034-1036.

Brown, J. H., and B. A. Harney. 1993. Population and community ecology of heteromyid rodents in temperate habitats. In Biology of the Heteromyidae. The Amer. Assoc. of Mammalogists, H. H. Genoways and J. H. Brown (ed.) Albuquerque, New Mexico, pp. 618-651.

--, and G. Leiberman. 1973. Resource utilization and coexistance of seed-eating rodents in sand dune habitats. Ecology 54(4):788-797.

--, and J. C. Munger. 1985. Experimental manipulation of a desert rodent community: food addition and species removal. Ecology 66(5): 1545-1563.

Davis, W. B. and D. J. Schmidley. 1997. The Mammals of Texas: Online Edition. Texas Tech University. Available at: Accessed 26 June 2003.

Eisenberg, J. F. 1963. The behavior of heteromyid rodents. Univ. of Calif. Pub. Zoology 69:1-100.

Jones, W. T. 1993. The social system of heteromyid rodents. In Biology of the Heteromyidae. The Amer. Assoc. Mammalogists, H. H. Genoways and J. H. Brown (ed.) Albuquerque, New Mexico, pp. 575-595.

Lay, D. M. 1993. Anatomy of the heteromyid ear. In Biology of the Heteromyidae. The Amer. Assoc. Mammalogists, H. H. Genoways and J. H. Brown (ed.) Albuquerque, New Mexico, pp. 270289.

Rosenzweig, M. L. 1973. Habitat selection experiments with a pair of coexisting heteromyid rodent species. Ecology 54(1):111-117.

--, and J. Winakur. 1969. Population ecology of desert rodent communities: habitats and environmental complexity. Ecology 50(4):558-572.

Thompson, S. D. 1982. Microhabitat utilization and foraging behavior of bipedal and quadrupedal heteromyid rodents. Ecology 63(5): 1303-1312.

Tremor, S. and B. Haas. 2000. Dipodomys merriami: Merriam's Kangaroo Rat. San Diego Natural History Museum. Available at: Accessed 26 June 2003.

Accepted for publication 30 September 2003.

Randy Striplin

Department of Biological Sciences, California State Polytechnic University, Pomona, 3801 West Temple Blvd., Pomona, California 91768
COPYRIGHT 2004 Southern California Academy of Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Striplin, Randy
Publication:Bulletin (Southern California Academy of Sciences)
Geographic Code:1U9CA
Date:Dec 1, 2004
Previous Article:Ecological impacts of seed-harvester ants on the phenological events of seven Mojave Desert shrub species in Southern Nevada.
Next Article:Age and growth of young-of-the-year (YOY) California halibut (Paralichthys californicus), from Northwestern Baja California.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters