In this research, I address lemur conservation by extending our ecological knowledge of Lemur catta and Propithecus verreauxi population density and spatial distribution outside of protected lands. The goal was to better understand how the two species utilize complex forest landscapes characterized by multiple forest classes with protected and unprotected areas around Beza Mahafaly Special Reserve. The objectives of this study were to create a forest cover map of the study area that characterizes the diversity of forest classes on the landscape; assess L. catta and P. verreauxi population densities across a 3 km dry forest gradient; and to characterize forest structural and lemur population parameters in adjacent grazed and ungrazed forest sites. Multi-seasonal satellite data were acquired, from which vegetation indices and textural filters were derived, and then the "randomForest" package in the R statistical program was used to classify the imagery. Using built-in cross-validation, overall classification accuracy was 85% with class accuracies ranging from 63 - 91%. Accuracy estimates were highest for dry deciduous and gallery forest classes and lowest for spiny forest. This new map may help expand our definition of lemur habitat in this study area. The study of southern dry forest lemurs has been largely restricted to small reserves, yet, the majority of the region's lemur populations reside outside of protected areas. Despite their sympatry (i.e., occupation of the same habitat), no previous studies have compared the concurrent densities in any single forest class--much less across forest classes. This study assessed L. catta and P. verreauxi population densities across a 3 km dry forest gradient in southern Madagascar. The highest densities recorded for each species were in the protected area. L. catta density was highest in protected gallery forest and P. verreauxi density was highest in protected dry deciduous forest. Outside of protected areas, this partioning of habitat was not observed. Both species' density was relatively high in unprotected gallery forest. Results of this study indicate that patterns of lemur density in protected areas are not representative of patterns in disturbed forests; this also suggests that we cannot fully understand the ecological constraints facing primate species by studying them only in protected areas. This research highlights the value of pairing the study of regional patterns of species distribution with local ecological interpretations; information from only one level may give an incomplete view of the community. The primate-livestock issue in the dry forests of southern Madagascar is a fairly unique case of wildlife-livestock interaction. Grazing impacts on forest structural parameters appear to be more pronounced in the dry forest than in gallery. There was no significant difference in either lemur species' densities between the grazed and the ungrazed sites; however, comparisons by forest class captured differences that otherwise would have gone undetected. Results suggest that grazing activities that do not significantly alter forest structure may support lemur densities similar to those in ungrazed areas. However, P. verreauxi, in particular, seems sensitive to changes in forest structure in their preferred habitat.
Read
