Friday, August 7, 2009

PS 77-16: Islands of fear: Effects of wooded patches on habitat suitability of the striped mouse (Rhabdomys pumilio) in a South African grassland

Mohammad A. Abu Baker and Joel S. Brown. University of Illinois at Chicago

Background/Question/Methods

We used live trapping,  giving-up densities (amount of food left behind in artificial food patches -GUDs), and GIS analyses to test for spatial and temporal habitat selection by the four striped grass mouse (Rhabdomys pumilio)  inhabiting a grassland containing distinct “islands” of woody vegetation (c. 30 m in diameter).  We investigate how these woody islands influence the quality of the grassland in terms of activity patterns, feeding opportunities, predation risk and habitat suitability.
Results/Conclusions

Rhabdomys pumilio avoided the wooded patches, just one of 267 captures occurred within a forest patch, and only 8 on the wood margin.  Similarly no food patches were ever foraged within the wooded islands.  Suggesting fear of wooded patches, GUDs declined with distance from edge.   While more striped mice were captured during the day (153) than at night (114), the number of patches foraged was significantly higher during the night (533) than day (429), suggesting a shift in activity towards night when foraging opportunities are available.  GUDs did not vary with night or day suggesting similar predation risks.  Within the grassland, the striped mouse had significantly lower GUDs in portions dominated by ferns than areas dominated by grasses.  Furthermore, the presence of fern resulted in mice foraging closer to the wooded patches.  By measuring GUDs during the day at 2-hour intervals, we found the highest GUDs between 1-3pm and the lowest between 7-9am.  GIS analyses of the 23.5 ha site revealed 59%, 14% and 27% of grass, fern and woody islands habitats, respectively.  But, because of the negative edge effects of the wood islands on the striped mice (unsuitable habitat up to 1m into the fern and 3 m in to the grass; and lower habitat quality up to 9 m), the suitable habitat available to the mice declines to 45 % of high quality habitat, 19% of lower quality, and 36% of unsuitable habitat.  These techniques and concepts should be broadly applicable by conservationists and ecologists interested in understanding and managing edge habitats.