Contrasting performance of Lidar and optical texture models in predicting avian diversity in a tropical mountain forest

Abstract

Abstract Ecosystems worldwide are threatened by the increasing impact of land use and climate change. To protect their diversity and functionality, spatially explicit monitoring systems are needed. In remote areas, monitoring is difficult and recurrent field surveys are costly. By using Lidar or the more cost-effective and repetitive optical satellite data, remote sensing could provide proxies for habitat structure supporting measures for the conservation of biodiversity. Here we compared the explanatory power of both, airborne Lidar and optical satellite data in modeling the spatial distribution of biodiversity of birds across a complex tropical mountain forest ecosystem in southeastern Ecuador. We used data from field surveys of birds and chose three measures as proxies for different aspects of diversity: (i) Shannon diversity as a measure of α-diversity that also includes the relative abundance of species, (ii) phylodiversity as a first proxy for functional diversity, and (iii) community composition as a proxy for combined α- and β-diversity. We modeled these diversity estimates using partial least-square regression of Lidar and optical texture metrics separately and compared the models using a leave-one-out validated R2 and root mean square error. Bird community information was best predicted by both remote sensing datasets, followed by Shannon diversity and phylodiversity. Our findings reveal a high potential of optical texture metrics for predicting Shannon diversity and a measure of community composition, but not for modeling phylodiversity. Generalizing from the investigated tropical mountain ecosystem, we conclude that texture information retrieved from multispectral data of operational satellite systems could replace costly airborne laser-scanning for modeling certain aspects of biodiversity.