Date of Award


Degree Name

MS in Biological Sciences


Biological Sciences


John D Perrine


Island foxes (Urocyon littoralis) are currently listed as federally endangered on four of the six Channel Islands to which they are endemic. The Santa Rosa Island (SRI) population declined by 99% during the 1990’s due to non-native golden eagle (Aguila chrysaetos) predation and is currently the lowest fox population (~280) and density (0.86 foxes/km2) of any of the Channel Islands. The goals of this study were to assess new miniaturized GPS technology and to quantify home range and habitat use of the SRI population. This is only the second use of Global Positioning System (GPS) collars on Channel Island foxes and provides essential baseline data for the recovering population. These results can be used to guide management decisions and future habitat restoration efforts after the recent removal of non-native ungulates.

In fall 2009, 14 GPS collars were deployed on male foxes on the east side of SRI. Nine collars and three remote download datasets were recovered in 2010. The collars’ battery life was 40% lower than expected at an average (±SE) of 16.5 ± 1.7 weeks but had high performance in precision and fix rate. Collars yielded an average of 347 ± 33 locations with a fix rate of 82.3% ± 2.1% and 88% of locations categorized as high precision.

From these data, 95% minimum convex polygon (MCP) home ranges and 95% kernel density isopleth (KDI) home ranges were created. The average 95% MCP home range size was 3.39 ± 0.59km2 and the area of overlap with adjacent home ranges had a median of 5.3%. The average 95% KDI home range size was 3.82 ± 0.68km2 with a median overlap of 6.0%. These home range sizes are almost triple the size reported in other island fox studies, likely due to the low fox densities in the recovering SRI population.

Habitat analysis was performed using KDI home ranges and a Euclidian distance analysis (EDA) method to assess habitat selection within the study area, the home range and the core area. Results showed selection for lupine within the study area, which no previous studies have documented. There was no significant habitat selection within the home ranges or core areas. Foxes selected for valley bottom topography and for bare and grassland habitat at night. One shortcoming of EDA is that its reliance on random points for determining second order selection can lead to unused areas being identified as selected habitat. The lack of significant selection within home ranges and core areas may be attributed to small sample sizes, use of male foxes only and the timing of the study in relation to fox reproductive biology.

I recommend further investigation in the use of lupine habitat and associated resources through prey inventory studies to further assess these findings. When densities reach historic levels of 4 foxes/km2, follow up studies should be conducted to reassess home range size, overlap and habitat use to determine if home range sizes have decreased and overlap has increased. Future studies should incorporate spring and summer seasons and females to determine if foxes select a particular habitat within the core area during denning and pupping periods.

Included in

Biology Commons