Date of Award


Document Type


Degree Name

Master of Science (MS)



First Advisor

Richard Sweitzer


A small group of mule deer were introduced to Santa Catalina Island in the early
1930s. The mule deer population increased through the 1940s and 1950s and remains widespread and abundant today. Large browsing ungulates are not native to Santa Catalina Island and there is concern that nonnative mule deer are damaging endemic trees and shrubs on the island when foraging. This study was designed to develop quantitative information on diets, foraging behavior, movements, and population biology of mule deer on Catalina Island.

Diet analysis indicated mule deer are consuming a wide variety of native and
endemic plants found on Santa Catalina Island in all seasons of the year. Data from browse transects indicated that mule deer consumed all species of trees/shrubs that were available to them, but preferentially foraged on Island mountain mahogany, Island bigpod ceanothus (Ceanothus megacarpus var. insularis), Catalina holly, and chamise (Adenostomafasciculatum). Several of these preferred shrubs lost over 20% of their current annual growth stems during the July to November period when browse represented 70-82% of the diets of mule deer on the island. A particularly compelling manifestation of the negative impacts of nonnative mule deer on island endemic trees/shn1bs was the heavy browse pressure exerted on two species of Catalina Island endemic shrubs (Catalina island bush poppy, Dendromecon harfordii, and felt-leaf ceanothus, Ceanothus arboreus) after several exclosures in the Goat Harbor area were altered to expose several individuals of each shrub to browsing by mule deer. Intensive browsing by mule deer killed one of three felt-leaf ceanothus trees within 15 months of being exposed. Data from 16 seedling plots revealed that physical disturbances by bison and mule deer were the two most important causes of mortality for young oak trees on Catalina Island but browsing by mule deer on oak seedlings did not kill many seedlings outright.

I used GPS radio-telemetry collars and resource selection probability functions to
identify areas of the island where high concentrations of deer use overlap known rare plant locations. Mule deer were highly variable in their habitat selection and I found no significant correlation between predicted deer use and locations of rare plants.

After estimating the mule deer herd on the island at approximately 2400 deer I
used a stage-based stochastic population model to determine the effects of different harvest strategies on the population dynamics of mule deer on the island. Modeling results indicated harvesting the deer herd at the maximum allowed under the current hunting program will not lead to a sustainable herd. In addition harvesting proportionately more females from the population will be the most efficient method to reduce the population. My models also predict the number of specific age/sex classes of deer ,required to be harvested in order to reduce and maintain the population at 50%, 75%, and 90% of its estimated size in 2006.

Included in

Biology Commons