A study of alternative Hydrological Response Units (HRU) configurations in the context of Geographical Information Systems (GIS) - based distributed hydrological modeling

Abstract

A new methodology for delineating Hydrological Response Units (HRUs), based on the widely used runoff Curve Number (CN) concept, is proposed, aiming to provide a systematic and physically-consistent procedure for the delineation of HRUs in the context of hybrid semi-distributed hydrological models. Given that the hydrological and engineering community has great experience in estimating the CN parameter on the basis of easily-retrieved geographical information, a methodology for delineating HRUs is proposed and tested, based on distributed CN maps, along with guidelines for its optimal use. Their formulation is an extension of the standard SCS approach, with the use of an empirical expression accounting for three major physiographic characteristics, by means of indices of: (a) soil permeability, evaluated according to the hydraulic properties of the soil and the unsaturated zone, and the dominant geological formations; (b) land use/land cover characteristics, typically expressed in terms of vegetation density; and (c) drainage capacity, evaluated according to the geomorphological characteristics of the basin (mainly the terrain slope) and the existence of runoff retention structures. The map of CN classes is eventually used within model parameterization, to identify the essential number and spatial extent of HRUs and, consequently, the number of control variables of the calibration problem. The proposed approach aims, on the one hand, at reducing subjectivity introduced by the definition of HRUs and providing parsimonious modelling schemes, on the other. In particular, the modified CN-based parameterization: (1) allows the user to assign as many parameters as can be supported by the available hydrological information, (2) associates the model parameters with anticipated basin responses, as quantified in terms of CN-classes across HRUs, and (3) reduces the effort for model calibration, simultaneously ensuring good predictive capacity. The proposed approach is demonstrated and tested in semi-arid river basins with intermitted low. In the hydrological simulation of the Nedontas River Basin, Greece, parameterizations of different complexities are employed in the HYDROGEIOS modelling framework. A modelling experiment with a varying number of HRUs, where the parameter estimation problem was handled through automatic optimization, showed that the parameterization with three HRUs, i.e., equal to the number of flow records, ensured the optimal performance. Similarly, tests with alternative HRU configurations confirmed that the optimal scores, both in calibration and validation, were achieved by the CN-based approach, also resulting in parameters values across the HRUs that were in agreement with their physical interpretation. The approach is further tested in two other river basins, Yialias and Kouris river basins, Cyprus, two watersheds of different sizes that vary in terms of physiographic characteristics and meteorological stresses, ideal to evaluate the performance of the method in diverse environments. Different classification schemes were implemented in creating the CN sub-sets to delineate the final HRUs, in an attempt to emphasize the advantage of the association of each HRU response to the corresponding parameter values in terms of CN, thus, allowing for a more efficient and objective model set up, assuring the parameters’ physical meaning and realistic representation of the hydrological behaviour of the basin. Hence, through a proper classification of CNs, the user can determine a priori a reasonable and relatively narrow range of feasible parameter bounds, which is of key importance towards ensuring effective and efficient calibrations. Finally, the sensitivity of the approach on each of the three major physiographic characteristics was investigated to provide further insight as to what impact each characteristic has on the HRU delineation processes, and thus, model performance.