Investigation of Remote Sensing for Potential Use in Dike Inspection

Dikes are critical barriers against storm surges, sea level rise, and flooding by rivers. Monitoring of these structures is most often done by visual inspection. This paper investigates remote sensing as a tool to facilitate the dike inspection process for grass-covered dikes. Two inspection criteria were considered, namely soil moisture and the quality of the dike cover. Four types of remote sensing data were used, obtained in a ground-based campaign with hand-held sensors. These were thermal, visible, multispectral, and hyperspectral remote sensing data. Relationships were found between the remote sensing data and the inspection criteria. In particular, afternoon thermal remote sensing data showed a negative correlation to soil moisture and broadband multispectral remote sensing data, especially using the near-infrared band, showed a negative correlation to dike cover quality. Limitations for the use of thermal remote sensing are also discussed.