Merging land-marine realms: Spatial patterns of seamless coastal habitats using a multispectral LiDAR

Abstract

A lack of spatially, structurally and thematically accurate habitat data complicates conservation and management planning as well as ecological research within structurally complex littoral enviromnents. The Scanning Hydrographic Operational Airborne Light Detection And Ranging (LiDAR) Survey (SHOALS) has considerable potential to provide such data, by means of its proficiency in generating high-resolution measurements of emerged and immersed elevations and to create thematic maps. Return signals, i.e., waveforms, contain signatures and structural information of salt-marsh and benthic cover. This paper focuses on the capability of the SHOALS to assess the structural complexity of emerged and immersed coastal habitats, and to define the contribution of SHOALS data, both elevation and intensity, in order to accurately and seamlessly map these habitats from supratidal to nearshore levels. The study area was selected based upon the variety of littoral cover types, encompassing kelp habitat, eelgrass meadow, beach, salt-marsh, farm and urban coastal environments. Firstly, the LiDAR-derived green waveform, through an ad hoc decision-tree, satisfactorily assessed the structural complexity of littoral habitats (r=0.75, p<0.01). Secondly, synergistic combination between NDLVI (LiDAR-derived red and near-infrared data) and topographic/bathymetric surfaces (LiDAR-derived near-infrared and green, respectively) enabled the accurate mapping of 19 littoral habitats (overall accuracy and kappa coefficient equaled to 90.09% and 0.8883, respectively). Accordingly, a single multispectral (three bands) LiDAR survey holds significant potential for monitoring small-scale structural complexity, as well as constructing spatially and temporally seamless patterns of coastal habitats.