Detecting Sheer Rock Outcrops Using Digital Elevation Models for Ecological Conservation

Rock outcrops serve as unique habitats for several specialist species, yet their identification and conservation are challenged due to logistical and computational constraints. This study aimed to improve the accuracy and efficiency of locating sheer rock outcrops (>60 cm in height) using high- and medium-resolution digital elevation models (DEMs) and conditional inference trees, a recursive binary partitioning approach to predict an outcome. In our application, the outcome is the presence of sheer rock outcrops. We used slope and topographic position index (TPI) as predictors. The conditional inference tree models estimated slope thresholds of 38.4° for high-resolution DEM and 18.14° for medium-resolution DEM as indicative of sheer rock outcrop presence. These findings align with a previous study, adding a layer of validity. In addition, a TPI >0.137 in the high-resolution DEM provided additional accuracy of sheer rock outcrop presence and the medium-resolution (∼10 m) DEM can be used when high-resolution (1 m) DEM is not available. This is particularly relevant for large-scale conservation efforts where computational resources may be limited. However, high-resolution DEM is recommended for more precise quantification of the total area of sheer rock outcrops. Our study has significant implications for efficiently locating sheer rock outcrops and identifying putative areas for protection to help conserve species that rely on these unique habitats while minimizing the logistical constraints of previous methods. The modular methodology developed here could serve as an invaluable approach for broader rock outcrop conservation. Furthermore, our findings add momentum to further innovate remote sensing of unique habitats.