Incremental distributed modelling investigation in a small agricultural catchment: 1. Overland flow with comparison with the unit hydrograph model

Abstract

Abstract:A distributed overland flow model is presented and the test results compared with those of the unit hydrograph (UH) model. Infiltration excess in the overland model was calculated using both a modified Green and Ampt (G-A) method and a more complicated method that keeps track of the soil moisture content. The two-dimensional partial differential flow equations with kinematic flow approximation were solved using both backward-central explicit and implicit finite-difference schemes. Moreover, a baseflow component was added to the flow model. Each part of the model was built by replacing a process by a parallel process or by adding a completely new component. When the model was developed and validated in small agricultural fields in southern Finland, comparisons were made between corresponding processes and the significance of added components were estimated in order to find out whether increased model complexity improves the model performance. Apart from the basic model, all the other distributed models outperformed the UH approach. The implicit scheme was clearly more accurate than the explicit scheme, and the baseflow component improved the results significantly. There was no major difference between the performances of the infiltration models, but the G-A method was more stable and much faster. The approaches that were best in the calibration were also the best in the verification. The full distributed model provides a potential method for running water-quality models and testing their different processes.

Incremental distributed modelling investigation in a small agricultural catchment: 2. Erosion and phosphorus transport

“…As mentioned by Taskinen and Bruen (2006), baseflow in the study field is assumed to consist of drainage from groundwater bodies, whose water tables are above the levels of the streambeds, and interflow as rapid subsurface flow through pipes, macropores and seepage zones in the soil. Therefore, sediment in baseflow comes mainly from bank erosion and maybe from subsurface drainage pipes and during overland flow also from the field surface.…”

“…Details of the overland flow models driving the erosion model are presented in Taskinen and Bruen (2006). In five events, the model with the implicit scheme and the modified Green and Ampt infiltration equation (Chu, 1978) was used because it is so much faster than the one with the soil moisture and infiltration equation of Corradini et al (1997).…”

“…SS measurements of six storms from the Hovi catchment run by the Finnish Environment Institute (see Taskinen and Bruen (2006) for a detailed description of the catchments) were used to calibrate the erosion model and to test the effect of different processes. The results are presented in full by Taskinen (2002) and the main results are summarized here.…”

“…The land use of the Hovi catchment is presented in Taskinen and Bruen (2006: figure 4, table II). In the case of the erosion modelling, grouping of the events was done on the basis of the land use of area 1 (near side to the measurement weir) only.…”

“…These models and many others have been used in many applications with promising results. In their discussion of overland flow models, Taskinen and Bruen (2006) noted that it is often difficult to see through the fitted results to identify the principle factors causing a certain model performance. This is particularly difficult when the models have been applied in large catchments with various different land uses, soil types, rainfall conditions, etc.…”