|DFG Projekt Direct WaterBalance|
|Project manager:||Prof. Dr. rer.nat. Dr.-Ing. András Bárdossy, Dr. rer.nat. Johannes Riegger|
|Research assistants:||Dr.-Ing. Mohammadjavad Tourian|
|Duration:||14.1.2007 - 28.2.2013|
|Funding:||DFG SPP 1257 “Mass transport and Mass distribution in the System Earth”|
- Institute of Geodesy (University of Stuttgart)
- Institute for Meteorology and Climate Research Garmisch (University of Karlsruhe)
Abstract:Measuring River Discharge from Space
River discharge on a global scale is one of the essential inputs (“musts”) for any global scale description of the water balance on the land masses and in the atmosphere and their interaction and also serves as one of the driving forces for ocean circulation. Despite their immeasurable importance the number of river gauging stations has been permanently decreasing worldwide since the 1970s and has nearly decreased by 50 percent since then. Modelling of river discharge does not solve the problem as it is flawed with very big uncertainties. Thus there is an exceeding necessity for spaceborne methods, which allow to measure river discharge and thus overcome possible technical or political constraints in the future.
A recent study on the relationship between measured river discharge on the one hand and liquid water storage measured by the gravity satellite GRACE and remote sensing on the other hand now opens up new possibilities to determine river discharge from space directly.
“GRACE gravity measurements of water storage variations mean a quantum leap in global scale hydrology both for the understanding of large scale systems behaviour and the global water cycle as well as for applications” emphasize Johannes Riegger and Mohammad Tourian from Stuttgart University. They have utilized GRACE observations of the time dependent mass distributions over the Earth, which allow to directly measure the monthly state of water storage and thus to directly determine the water balance for large scale catchments.
This for the first time allows a direct comparison of measured river discharge and GRACE derived water storage on global scale catchments. The investigations reveal complex, yet periodic characteristics with hysteresis clearly distinct for different climatic zones. For fully humid tropical conditions like for the Amazon catchments simple linear relationships exist between runoff and storage, phase shifted by a time lag due to runoff routing. “ The discharge behaviour of the largest river basin of the world is comparable to the drainage of a bath tub, just with a delayed response”, Riegger explains.
For catchments from other climatic zones such as boreal or seasonally dry tropical regions the relationship is much more complex. “Of course, only liquid and hydraulically connected or coupled water storage can contribute to river discharge and not unconnected storages components like snow/ice, soils, uncoupled flooded areas, surface ponds, etc., from where the water is not released to the drainage system.” says Riegger.
The effect of an unconnected storage can be clearly seen in the Runoff-Storage diagram (Fig.1) of total storage for the boreal catchments of Siberia and North America, which show an increase in total mass by snow accumulation during winter times yet no increase in runoff. “So, if we want to determine river discharge from space with this method we have to somehow separate the liquid / coupled and uncoupled storages with the help of remote sensing approaches”, Riegger underlines.
In order to estimate the uncoupled solid water storage for the boreal catchments they used MODIS snow coverage, assuming that all water underneath the snow covered areas is totally frozen. Indeed, the separation of coupled / liquid and uncoupled / solid storage contributions by MODIS satellite imagery reveals a linear relationship between runoff and coupled/liquid storage (Fig.1). Thus the researchers conclude that the relationship of runoff versus liquid/coupled storage can be assumed to be linear in general, while nonlinear contributions between runoff and total mass emerge from uncoupled storage components.
On this basis, river discharge can be directly determined from liquid storage measured by the GRACE gravity signal and separated by remote sensing. For the fully humid Amazon catchments this allows to directly determine river discharge by GRACE with an unprecedented accuracy of 5.5 mm/mo RMSE or ~6% of the mean signal, while for the boreal catchments ~3mm/mo RMSE or 18% of the mean is achieved. Vice versa liquid or total mass can be determined from runoff measurements and remote sensing, which allows to bridge possible gaps between the gravity missions.
“We are not at the end of a development but rather at the beginning. The behaviour of the seasonally dry catchments for example is not understood so far and we urgently need new methods to utilize the potential of remote sensing. However, now we knowwhat we have to search for”! Riegger emphasizes. “Quantifying the uncoupled storage components and thus determining the liquid / coupled storage by use of GRACE means a totally new approach to measure river discharge by remote sensing. This opens up a completely new field of research in hydrology and spaceborne methods and hopefully triggers improvements and new developments in satellite gravimetry and remote sensing.”
Detailed information on data, methodology and results can be found in :
Riegger, J., and M. J. Tourian (2014), Characterization of runoff-storage relationships by satellite gravimetry and remote sensing. Water Resour. Res.,50,doi:10.1002/2013WR013847.