Rainfall-runoff modelling with high temporal resolution on the arable catchment Skuterud : assessing the performance of the Distance Distribution Dynamics model

Runoff from arable land is recognised as the main reason for water degradation, eutrophication and reduced water quality. The aim of the study is to model runoff on a 1h temporal resolution from an arable catchment using the Distance Distribution Dynamics model. Research objectives are to compare the model with and without artificial drainage system as part of the model parameters and introduce two alterations to the recession analysis of the pre-processing routine. Model inputs are precipitation and temperature. Pre-processing includes recession analysis, to determine subsurface storage and subsurface flow velocities i.e. celerities, and distance distribution analysis. The study area is the small agriculturally dominated catchment Skuterud, which is a part of The Norwegian Agricultural Environmental Monitoring Programme (JOVA). The calibration period is from 2000-2004, and the validation period 2005-2009. Time series from 2000-2004 were aggregated to 3h, 6h, 12h and 24h temporal resolution to check the assumption of time invariance for the recession characteristics. The results show an improvement of Nash-Sutcliffe efficiency (NSE) and Kling-Gupta Efficiency (KGE) criterion when artificial drainage is included in the model, and further improvement using the altered recession analysis methods. Limitation and uncertainties with the model include, but is not limited to, instrumental limitation for runoff measurements at low flow, hydrological impact on soil characteristics or seasonal influences. The overall best result is an NSE of 0.56 and KGE of 0.74