Utility of an inexpensive near-infrared camera to quantify beach surface moisture

Abstract

Surface moisture content is an essential factor that must be considered when studying aeolian sediment transport on a sandy beach. In recent years, near-infrared (NIR) remote sensing sensors have shown promise for obtaining accurate surface moisture data; however, prior studies utilized instruments with extreme costs. This study assesses the capability of an inexpensive NIR digital camera to measure surface moisture at two sandy beach environments – Tybee Island, Georgia and Pensacola Beach, Florida – that exhibit varying sediment hue characteristics. To account for temporal variations in solar atmospheric conditions, we normalized the raw sediment surface reflectance data against a white reflectance card and a sample of oven dry sand representative of each study site. This is a necessary step to account for solar atmospheric conditions. Calibration results illustrate that the NIR camera is capable of producing accurate representations of beach surface moisture; analyses from both study sites produced R2 values greater than 0.76 with error estimates at ±1–2% moisture. No statistical difference in calibration relationships were observed for data collected over multiple days and times of day. Calibration data for the reflectance card produced more robust relationships with smaller prediction errors than the oven dry sand analyses; however. Overall, this study illustrates that an inexpensive digital camera modified to record NIR radiation is capable of producing robust and accurate measurements of beach surface moisture.