Irrigated agriculture constitutes one of the largest uses of freshwater resources. However, monitoring extraction, ensuring compliance and legislating for the sustainable management of water resources remains an ongoing challenge. Of particular interest is the legality and efficient use of water that is intercepted and stored in On-Farm Storages (OFSs), which are large, shallow constructed dams used for agriculture in the semi-arid zones of Australia. 

In this study, a rapid monitoring and reporting methodology using remote sensing methods was applied via a case study of OFS surface water status, spanning a historic flooding event in the northern inland catchments of New South Wales, Australia. The Modified Normalised Difference Water Index (mNDWI) was applied to publicly available Sentinel-2 imagery (~10m pixel resolution) to extract surface water extent; surface water extent to volume estimates were then calculated using LiDAR-derived storage capacity curves. To validate these estimates, a comparative assessment was also performed, using high-resolution PlanetScope imagery (~3m pixel resolution). Our findings showed a less than 5% difference in computed volumes between both imagery types, demonstrating a high degree of relative precision between both datasets.

The application of our methods demonstrated that change detection in surface water extent and volume can be reliably monitored from a scale of individual storages to entire catchments. Further, as Sentinel-2 imagery is freely available and subject to rapid turnaround times (48-72 hours), the methodology has global applicability for use in near-real time water resource monitoring and compliance. This is especially relevant where preliminary evidence is required to conduct detailed and targeted investigations in resource-constrained environments.