Comparative discharge prediction from a small artificial catchment without model calibration: Representation of initial hydrological catchment development
2011 (English)In: Die Bodenkultur, ISSN 0006-5471, Vol. 62, no 1-4, 23-29 p.Article in journal (Refereed) Published
Ten conceptually different models were applied to predict the discharge from the 6 ha artificial Chicken Creek catchment in Lausatia, North-East Germany, which has been created in an open cast mining area. The study consisted of three steps to make a model intercomparison with the objective of a priori prediction of the water balance and the discharge dynamics. In order to test the ability of each model and modeller to predict water flows in an ungauged catchment, only soil texture, topography, vegetation coverage and climate data were provided to the modellers in the first step. Hydrological data on discharge, soil moisture and groundwater levels were withheld. This enabled us to assess the predictive capabilities of the models under sparse data conditions. The predicted components of the water balance varied in a wide range. None of the model simulations came close to the observed water balance for the entire 3-year study period. Discharge was mainly predicted as subsurface flow with little surface runoff. In reality, surface runoff was a major flow component despite the fairly coarse soil texture. In the second step, additional process knowledge was gained during a joint field visit. The occurence of gully erosion and surface crusting was detected and implemented into the models. Consequently, model predictions changed considerably. The previous simulations dominated by subsurface flow changed to surface flow-dominated simulations. Additional data, provided in the third step, mainly confirmed the parameterisations and assisted in a better definition of initial conditions and subsurface storage. The comparison indicates that, in addition to model philosophy, the personal judgement of the modellers was a major source of the differences in the model results. The model parameterisation and choice of initial conditions depended on the modeller's judgement and were therefore a result of the modellers' experience in terms of model types and case studies.
Place, publisher, year, edition, pages
2011. Vol. 62, no 1-4, 23-29 p.
A priori prediction, Additional quantitative data, Artificial catchment, Initial ecosystem development, Model intercomparison, Process understanding, Ungauged catchments
Earth and Related Environmental Sciences
IdentifiersURN: urn:nbn:se:kth:diva-150598ScopusID: 2-s2.0-84858691489OAI: oai:DiVA.org:kth-150598DiVA: diva2:744399
QC 201409082014-09-082014-09-082014-09-08Bibliographically approved