While many animals respond to seasonal variation in their environment, animals such as pocket gophers Thomomys bottae that live below ground might seem to be buffered against such variation. In some areas, however, the patterns of burrowing activity by pocket gophers are tied to the seasons, one factor of which is rainfall. Variation in activity patterns may result from the ease of digging in moist soil or increased food availability during the wet season. Previous simulation modelling work suggests that food availability influences burrowing patterns, while soil conditions do not. Thus, field experiments were used to investigate how soil conditions and food availability influence seasonal burrowing activity. Results indicate that an increase in soil moisture initiates activity. After this initial increase in activity, mound production declines and reaches a steady rate, which can be supported by vegetation availability, in agreement with previous model results. Our findings support the idea that moist soil promotes a burst of digging activity, potentially for burrow maintenance when soil becomes easily workable, and the eventual growth of vegetation provides the food necessary to support continued activity.

For many burrowing species of mammals, the shape of the burrow may affect the success with which resources are acquired. Parameters used to characterize burrows have included: number of branches, mean segment length, turn angle, and linearity. However, the interactions among these parameters are unclear, making it desirable to find a single measure to characterize burrow geometry. For this reason, fractal dimension has been proposed as a measure of burrow geometry. However, the ways in which fractal dimension relates to other burrow metrics are not obvious. In this study, we examine correlations between fractal dimension and other measures of burrow geometry, using burrows of the pocket gopher Thomomys bottae. We also investigate whether burrows excavated by pocket gophers of different age, sex, and at two locations differing in food abundance have different fractal dimensions. We show that there are strong correlations between fractal dimension and other measures of burrow geometry, but that replacing fractal dimension with any single measure of burrow geometry discards important information about burrow structure. We conclude that fractal dimension offers a useful way of describing burrow geometry. No differences were found in burrow fractal dimension between pocket gophers of different ages or sex or in sites of different productivities, and we conclude that the factors that determine the pattern of burrowing in this species are invariant with respect to these parameters.