The western purple martin (Progne subis arboricola) is a species of conservation concern throughout the Pacific Northwest. In western Oregon, the purple martin nests in three major ecosystem types: inland open-water, coastal, and upland forest. The availability of suitable breeding habitat is a major limiting factor for western purple martin populations and has likely decreased as a result of timber harvest reductions under the Northwest Forest Plan. As a cavity-nesting species that forages on airborne insects, purple martins require nesting structures in open habitat with high insect productivity. From anecdotal observations, it is assumed that dragonflies are a major prey item of purple martins. Prey limitations are unknown, as studies on the diet of western purple martin have not yet been done. An understanding of how availability of both suitable nesting cavities and prey resources limit populations is critical for developing a conservation strategy for this species.
The goal of my first study was to provide an assessment of the current status and distribution of the purple martin population nesting in upland forest in western Oregon. The objective of my habitat study was to determine the probability of occupation of purple martins at potentially suitable nesting sites in western Oregon, and to develop a prediction model of habitat suitability based on a comparison of habitat attributes at multiple spatial scales measured at used and unused sites. Each of
the variables that we used to characterize purple martin habitat were significantly different between used and unused nest sites except for snag DBH. We found strong statistical evidence that the odds of purple martins occupying a snag are affected by the area of early seral habitat within a stand. Despite the major limitation of lacking contiguous snag-level data for the study area, our prediction model of habitat suitability scores a “B” on the traditional academic point system with an AUC fit index of 0.8134. We intend this model to be useful in estimating the amount of suitable habitat available to support purple martin colonies both currently and under future timber harvest regimes.
The goal of my second study was to provide the foundational information on diet that will be necessary for further studies on nest site and prey limitations. The objective of my diet study was to illuminate diet composition and differences in diet composition between ecosystem types. I sampled prey composition at each ecosystem type to provide an index for prey availability. I utilized the noninvasive and innovative method of metabarcoding to produce a list of insect prey for purple martins in western Oregon. The abundance of dragonfly prey was higher at inland open-water sites compared with coastal and upland forest sites. Small insect biomass did not significantly differ between ecosystem types, though a trend of large insect hatches appeared only in open-water ecosystems.
My study fills important information gaps about suitable nesting habitat for the western purple martin and how nest site influences diet. I intend the predictive model of habitat suitability to be useful in estimating the amount of suitable habitat available to support purple martin colonies both currently and under future timber harvest regimes. I identified major limitations in statewide data availability for purple martin nesting habitat. It is recommended that further research aims to improve the continuity of snag data in western Oregon. Now that we understand the differences in diet among ecosystem types, we can target important prey taxa in sampling efforts to determine if and where prey limitations may exist. This study will guide wildlife
managers as to where their efforts to protect or bolster existing populations will be most effective.