Retrieving hidden water contribution to river from space: focus on the Seine basin

Image of the Month - December 2018

Discharge (in grey), baseflow reference estimate (black), and two Swot estimates, depending on the number of revisits (red and blue). Even in the case of only one measurement on 21 days, the Swot estimate is close to the reference one (Credits Mines ParisTech)
Seine basin (France)
Number of Swot revist planned during the Science phase over the Seine Basin.
Water inputs for river discharge (Credits Mines ParisTech)
Discharge components - baseflow and quickflow (Credits Mines ParisTech)

Rain is not the only provider of water for rivers either by direct precipitation or runoff. The baseflow, i.e. the river discharge with no rain, represent the minimum flow a river can have. It is coming from groundwater and also delayed flow (e.g. snowmelt). Its quantification is of primary importance to evaluate the impact of climatic and anthropogenic stresses on the availability of water resources; it is also of importance for the formation of cold thermal refuges for some fish species. This is difficult to measure directly, and only possible in situ, but baseflow can be estimated using the overall river discharge, or even estimated at basin scale. Since discharge is one of the expected outputs of Swot for rivers over 50 to 100-m wide - not direct measurement either, but using modelling and the surface water heights which will be the actual Swot measurements -, we can hope to be able to monitor continuously baseflow at global and basin-scale using Swot-derived data.
But will Swot measurements be frequent enough for such an estimation to be interesting? Will they sufficiently sample the river basins? A test was made over the Seine basin in France, using discharge time series simulated at a daily time-step by a coupled hydrological-hydrogeological model to obtain the reference base flow estimations. The Swot sampling, even when considering the worse case of only one measurement every 21 days, seems to retreive well enough the baseflow; it can miss high amplitude and high frequency peaks and sharp variations, but those are mostly part of the quickflow.

A whole new field of applications is being developed even now, three years from Swot launch, to be ready when the measurements will be coming in. Teams all around the world are working on those, and meet several times a year to check on their progress.

See also: