Affiliation:
1. Research School of Biology Australian National University Canberra Australia
2. Land and Water, Commonwealth Scientific and Industrial Research Organization Winnellie Australia
3. Department of Geodesy and Geoinformatics, Tashkent Institute of Irrigation and Agricultural Mechanization Engineers NRU Tashkent Uzbekistan
4. School of Biological Sciences University of Bristol Bristol UK
5. Fenner School of Environment and Society, Australian National University Canberra Australia
Abstract
AbstractRemotely sensed measures of vegetation structure have been shown to explain patterns in the occurrence and diversity of several animal taxa, including birds, mammals, and invertebrates. However, very little research in this area has focused on reptiles and amphibians (herpetofauna). Moreover, most remote sensing studies on animal–habitat associations have relied on airborne or satellite data that provide coverage over relatively large areas but may not have the resolution or viewing angle necessary to measure vegetation features at scales that are meaningful to herpetofauna. Here, we combined terrestrial laser scanning (TLS), unmanned aerial vehicle laser scanning (ULS), and fused (FLS) data to provide the first test of whether vegetation structural attributes can help explain variation in herpetofauna abundance, species richness, and diversity across a woodland landscape. We identified relationships between the abundance and diversity of herpetofauna and several vegetation metrics, including canopy height, skewedness, vertical complexity, volume of vegetation, and coarse woody debris. These relationships varied across species, groups, and sensors. ULS models tended to perform as well or better than TLS or FLS models based on the methods we used in this study. In open woodland landscapes, ULS data may have some benefits over TLS data for modeling relationships between herpetofauna and vegetation structure, which we discuss. However, for some species, only TLS data identified significant predictor variables among the LiDAR‐derived structural metrics. While the overall predictive power of models was relatively low (i.e., at most R2 = 0.32 for ULS overall abundance and R2 = 0.32 for abundance at the individual species level [three‐toed skink (Chalcides striatus)]), the ability to identify relationships between specific LiDAR structural metrics and the abundance and diversity of herpetofauna could be useful for understanding their habitat associations and managing reptile and amphibian populations.
Funder
Australian Research Council
Subject
Nature and Landscape Conservation,Computers in Earth Sciences,Ecology,Ecology, Evolution, Behavior and Systematics
Reference63 articles.
1. Australian Government Department of Environment. (2013)Vegetation assessment guide. Australian Government.
2. Use and categorization of Light Detection and Ranging vegetation metrics in avian diversity and species distribution research
3. Fitting Linear Mixed-Effects Models Usinglme4
4. Bin Abdul Rahman R.(2018)Nest site selection by flatback sea turtles: characterization of nesting beach topography with airborne LiDAR. Honours Murdoch University.