Real-time forecasting of water table depth and soil moisture profiles
article
We present a method for real-time forecasting of water table depth and soil moisture profiles. The method combines a simple form of data-assimilation with a moving window calibration of a deterministic model describing flow in the unsaturated zone and local as well as regional drainage. The local drainage level is calibrated on-line using a moving window calibration. Assigning more weight to the last available measurements then yields a form of model adaptation that is in between on-line calibration and data-assimilation (i.e. a simplified form of Newtonian nudging). Five-day hydrological forecasts are performed based on 5-day weather forecasts, while on-line observations of phreatic level and soil moisture content are assimilated on a daily basis. Advantages of the proposed method are that it improves the real-time forecasts compared to off-line calibration and ordinary moving window calibration and that it yields physically consistent soil moisture profiles. © 2005 Elsevier Ltd. All rights reserved.
Topics
TNO Identifier
239242
ISSN
03091708
Source
Advances in Water Resources, 29(5), pp. 692-706.
Pages
692-706
Files
To receive the publication files, please send an e-mail request to TNO Repository.