Using object-based image analysis with multi-temporal aerial imagery and LiDAR to detect change in temperate intertidal habitats

Intertidal habitat maps are needed at both fine and coarse scales to monitor change and inform conservation and management, but current methods of field survey and expert interpretation of aerial imagery can be time-consuming and subjective. Object-based image analysis (OBIA) of remote sensing data is increasingly employed for producing habitat or land cover maps. Users create automated workflows to segment imagery, creating ecologically meaningful objects, which are then classified based on their spectral or geometric properties, relationships to other objects and contextual data. This study evaluates the change-detection capability of OBIA in the intertidal environment by developing and comparing two OBIA methods for quantifying change in extent and distribution of habitats from freely available multi-temporal aerial imagery and LiDAR data. Despite considerable variability in the data, pre- and post-classification change detection methods had sufficient accuracy (mean overall accuracy from 70.5 to 82.6%) to monitor deviation from a background level of natural environmental fluctuation. This insight into spatial and temporal patterns of natural cyclical change and their detectability by OBIA could inform use of remote sensing for regular, rapid coastal assessment, providing an alert system to direct survey resources to areas of ecologically relevant change.