Multisensor UAS mapping of Plant Species and Plant Functional Types in Midwestern Grasslands

Uncrewed aerial systems (UASs) have emerged as powerful ecological observation platforms capable of filling critical spatial and spectral gaps in plant physiological phenological traits that been difficult to measure from space-borne sensors. Despite recent technological advances, the high cost drone-borne sensors limits widespread application UAS technology across scientific disciplines. Here, we evaluate tradeoffs between off-the-shelf sophisticated for mapping species functional types (PFTs) within a diverse grassland. Specifically, compared PFT accuracies derived hyperspectral, multispectral, RGB imagery fused with light detection ranging (LiDAR) or structure-for-motion (SfM)-derived canopy height models (CHM). Sensor–data fusion were used consider either single period near-monthly frequencies integration information (i.e., phenometrics). Results indicate overall classification PFTs highest hyperspectral LiDAR-CHM fusions (78 89%, respectively), followed by multispectral phenometric–SfM–CHM (52 60%, respectively) SfM–CHM (45 47%, respectively). Our findings demonstrate clear economical versus exorbitant sensor networks but highlight may achieve comparable those integrating phenometrics into machine learning image classifiers.