Remote sensing of river discharge: where are we now, and where are we going?

Abstract:

Remote sensing of river discharge (RSQ) is rapidly growing field gaining in recognition and importance in hydrology. However, this field is intellectually and methodologically diverse, and as a result there is confusion among the broader hydrology community, and even among the broader remote sensing hydrology community, about the exact methods, assumptions, and ramifications of RSQ, particularly as regards ungauged basins. We argue that not all RSQ methods are appropriate for all applications, and that careful consideration is needed in choosing an RSQ approach for a particular use case. In this presentation, we summarize current trends in RSQ research and present a synopsis of the field as it stands by parsing existing literature into methods appropriate for gauged, semi-gauged, regionally gauged, politically ungauged, and totally ungauged basins: defining what it means to be ungauged in an RSQ context. This framework is designed to present RSQ to a wider scholarly community and unify seemingly disparate methodological schools of thought within RSQ. Unlike previous summaries of remote sensing in hydrology, we find summarizing by sensor capabilities (i.e., active, passive, multi- or hyper-spectral) to be counterproductive in understanding what is and what is not possible in RSQ. Finally, we also contextualize RSQ within a broader effort of remote sensing of hydrology, and highlight connections between models, observations, and improved hydrologic understanding. We conclude with a view to a future in which machine learning and big-data literacy skills are set to become important if not necessary for RSQ practitioners. Remotely sensed data are increasing in volume and complexity, and we thus argue for our framework as a meaningful way to organize the discipline of RSQ moving forward.