27th Annual Meeting and Symposium of the
Desert Tortoise Council, March 22-24, 2002
Abstracts

Revised Techniques for Estimating Desert Tortoise Abundance in the Fort Irwin National Training Center Expansion Area in 2001 and the Results of those Surveys

Alice E. Karl
P.O. Box 74006, Davis, CA 95617

In an ongoing effort to examine impacts to desert tortoises that could occur from the expansion of Fort Irwin's National Training Center (NTC), the U.S. Department of the Army (Army) directed studies in 2001 to examine tortoise density. Studies were concentrated in the western expansion area (Superior Valley area) where the most intensive military training is anticipated, with more limited sampling in the southern portion of the expansion area, in the area (9-0 Area). Six 1.0 km² mark-recapture plots and one 1.44 km² mark -recapture plot were completed in Spring, 2001, and 568 belt transects were completed the following summer. Of the 258 km² in the western proposed expansion area, we sampled 115 km² by either transects or mark-recapture plots. While this was 44 % of the kilometers in the western area, the concentration of transects resulted in a sampling of 61 % of the square kilometers in the main-use area.

Techniques of both data collection and analysis were modified from earlier studies to improve the accuracy of the results in estimating tortoise density. For mark-recapture plots, these included: (1) choosing a 1.0 km² plot area, based on a comparison of different plot sizes attempted during the previous studies on the NTC and their attendant difficulties with edge effects and effective coverage; and (2) using a sampling crew of eight people to complete the mark and recapture sampling in approximately four days, thereby ensuring the closed population assumption for the statistical analysis, as well as minimizing sampling difficulties associated with tortoise movement. For belt transects, the 2.4 km transect length and triangular shape of the 2001 transects were identical to earlier transects, but improvements for statistical accuracy included: (1) increasing the sampling rate from the typical rate of 0.39 transects per square kilometer (1 transect per square mile), to 4.0 transects per kilometer (10.4 transects per mile; (2) analyzing only the size group that was actually sampled - the adult cohort (i.e., tortoises >179 mm in carapace length) - rather than all sizes of tortoise; (3) truncating transect width by sign type and individually for each observer; (4) using sign types that provided the highest correlation coefficients to known tortoise density; and (5) conducting calibration transects on mark-recapture plots conducted the same year as the transects.

Tortoise densities were low throughout the western expansion area and those portions of the 9-0 area sampled. Nowhere did adult tortoise densities exceed 18 tortoises/km² (47 tortoises/mi²) and broad areas were calculated to have one or fewer tortoises per square kilometer. For the entire western expansion area, the number of adult tortoises was estimated to be < 700 animals.

It is likely that the drought cycle of the past 13 years has resulted in substantial mortality and concomitantly depressed reproduction and recruitment. Adult annualized mortality rates (AM) on mark-recapture plots with enough live tortoises to be meaningful were 11.8-13.8 % for the previous 2-year period and 9.6-10.8 % for the previous 4-year period. One plot with low adult densities had a disproportionately high absolute number of carcasses and a 42.0 % AM for previous two years. However, it is likely that tortoise densities have never been high in the western expansion area, due to factors associated with the relatively high elevation there. By contrast, the 9-0 area is higher quality habitat and probably hosted substantially more tortoises prior to the severe recent drought.