WaterGAP (Water Global Assessment and Prognosis) is a hydrological model used to model water shortage, groundwater depletion, and floods and droughts (e.g. as impacted by climate change) over the land area of the globe. The Frankfurt hydrologist Prof. Petra Döll has examined how good a fit this model provides, using GPS observations and data from the GRACE satellite, which measures the gravitational field of the Earth. The study, published in the current issue of the scientific journal Surveys in Geophysics indicates that WaterGAP needs to be modified.
"In most regions of the globe, WaterGAP underestimates seasonal continental water storage fluctuations and does not retain rainwater for a sufficiently long time on the continents", is how Petra Döll from the Institute of Physical Geography at Goethe University sums up the findings. "So more water is being stored than the model predicts."
On a daily basis and with a spatial resolution of approx. 50 km, WaterGAP measures various forms of water flux such as evaporation, groundwater recharge and water flow in rivers, as well as the amount of water stored in the ground, in groundwater, in surface water bodies and as snow. Here, water abstraction for drinking water, industry and agriculture is also taken into account. A variety of data is used in calculating the models: climatic data, data concerning vegetation and soil, as well as socioeconomic and many other data besides.
Due to the uncertain data that is fed in and the simplifications required in a global-scale model, the findings are unreliable. Until now, river flow data has been employed to calibrate and check the quality of the model, but unfortunately this data does not exist for all major rivers. Besides, a model also needs to accurately map the dynamics of the stored water in order, for example, to be able to detect water abstraction by humans.
For this reason, to check the model, Petra Döll decided to use the influence of water masses on the deformation of the Earth's crust and the gravitational field of the Earth. Periodic changes in water masses deform the Earth's crust, which causes the position of permanently installed GPS antennae to vary by millimetres. Simultaneously, varying water masses also lead to significant variations in the Earth's gravitational field. These can be estimated using the GRACE satellite.
Working together with Dr. Mathias Fritsche, a geodesist in Dresden specialising in the data analysis of GPS observations, and Dr. Annette Eicker, a geodesist in Bonn dealing with gravity field computation, Petra Döll checked the WaterGAP calculations for the dynamic of continental water storage and could thus identify the shortcomings of the model.
In the study, the changes in position of about 200 GPS worldwide-distributed antennae were measured and compared to the changes in position that should – according to the WaterGAP calculations – have occurred due to the variations in water masses. In addition the researchers compared the seasonal variations of the continental component of GRACE gravitational fields to the results from WaterGAP. The result showed that WaterGAP underestimates seasonal variations in continental water storage and so in the future it will need to be modified.
A further result of the study is that seasonal variations in the Earth's gravitational field cannot be used to measure human water abstraction. The reason for this is that there are too few permanently installed GPS antennae, and the accuracy and spatial resolution of GRACE's field of gravity is too poor. "Only if water abstraction leads to groundwater depletion, i.e. where the amount of water abstracted is greater than the inflow, can GRACE satellite measurements be used to support the quantification of water abstraction", explains Prof. Döll. This possibility was used in a follow-up study, which has not yet been published.
Explore further: Predicting fire activity using terrestrial water storage data
Döll, P., Fritsche, M., Eicker, A., Müller Schmied, H. (2014): Seasonal water storage variations as impacted by water abstractions: Comparing the output of a global hydrological model with GRACE and GPS observations. Surv Geophys. DOI: 10.1007/s10712-014-9282-2