Record Details
Field | Value |
---|---|
Title | Predicting the distribution of the fisher (Martes pennanti) in northwestern California, U.S.A. : using survey data and GIS modeling |
Names |
Carroll, Carlos
(creator) Noss, Reed F. (advisor) |
Date Issued | 1997-07-01 (iso8601) |
Note | Graduation date: 1998 |
Abstract | Forest carnivores such as the fisher have frequently been the target of conservation concern due to their association with older forests and assumed sensitivity to landscape-level habitat alteration. Although the fisher has been extirpated from most of its former range in the western U.S., it is still found throughout much of northwestern California. However, fisher distribution is still poorly known in the majority of this region where surveys have not been conducted. In order to predict fisher distribution across the region, a multiple logistic regression model was created using data from 682 previously surveyed locations and a GIS vegetation coverage created from satellite imagery. A moving-average function was used to derive landscape level indices of vegetation variables from the GIS layer. Moving averages of canopy closure, tree size class, and percent conifer were found to have strong correlations with fisher presence. Regional gradients as represented by either precipitation or a trend surface derived from spatial coordinates were also significant predictors in the model. The model was validated with new data collected from 240 survey locations and proved to be accurate in predicting fisher presence in unsurveyed areas. The model was used to generate hypotheses as to the mechanisms controlling habitat selection and the scales at which these operate and to evaluate the representation of fisher habitat in existing protected areas. These insights may be valuable in designing conservation reserve networks that insure the long-term viability of forest carnivore populations. |
Genre | Thesis/Dissertation |
Topic | Fisher (Mammal) -- California -- Geographical distribution -- Computer simulation |
Identifier | http://hdl.handle.net/1957/33705 |