Logging and topography effects on artificial nest predation by rats in laurel forests

Abstract

We performed a general assessment of predation by rats (chiefly Rattus rattus) on artificial nests in context of vegetation structure at two subtropical laurel forests on Tenerife (Canary Islands). We selected two different types of forest stands (logged vs unlogged) differing in vegetation pattern and conservation degree to test for the effect of logging on rat predation. We placed ground and tree nests on four plots per forest stand, each plot containing 40 nests (80 nests per treatment, 160 nests in all). We distributed nests among available sites with contrasting topography (plain areas, slopes, ridges and ravines). We tested for logging effects, topography and their interaction with survival analysis, comparing survival and hazard functions. We assessed the relationships between nest survival and vegetation structure with Cox survival regression. Vegetation structure was consistently different between slopes of logged (early successional) and unlogged (mature-like) forest. We found higher nest predation rates at the unlogged forest, in sloping areas, and in patches with dense tree canopy cover and an intermediate successional development. Rat nest predation on forest slopes (the dominant topographic landscape feature) was higher in mature (unlogged) than in early successional (logged) ones, suggesting that rats have adapted well to mature habitat and would prefer mature forest for foraging.