Remote sensing of three‐dimensional coral reef structure enhances predictive modeling of fish assemblages

Abstract

Listen

Translate article

AbstractLiDAR (light detection and ranging) allows for the quantification of three‐dimensional seascape structure, which is an important driver of coral reef communities. We hypothesized that three‐dimensional LiDAR‐derived covariables support more robust models of coral reef fish assemblages, compared to models using 2D environmental co variables. Predictive models of coral reef fish density, diversity, and biomass were developed using linear mixed effect models. We found that models containing combined 2D and 3D covariables outperformed models with only 3D covariables, followed by models containing only 2D covariables. Areas with greater three‐dimensional structure provide fish more refuge from predation and are crucial to identifying priority management locations that can potentially enhance reef resilience and recovery. Two‐dimensional seascape metrics alone do not adequately capture the elements of the seascape that drive reef fish assemblage characteristics, and the application of LiDAR data in this work serves to advance seascape ecology theory and practice in the third dimension.