The effect of spatial discretization in Semi-distributed Hydrological modelling, the case of JGrass-NewAGE Model

Friday, 26 September 2014: 3:40 PM
Wuletawu Abera Worku1, Giuseppe Formetta2 and Riccardo Rigon1, (1)University of Trento, Civil, Environmental and Mechanical Engineering, Trento, Italy, (2)University of Calabria, Environmental Engineering, Calabria, Italy
Abstract:
Due to limiting factors of fully distributed hydrological modelling such as high computational demands, unrealistic process representation domain(i.e. pixel), the lack of data, etc (Garbrecht and Martz 2000, Bathurst 2002), semi-distributed hydrological modelling are preferred for water resource assessment and water simulation studies in medium and large basin. Bogaart and Troch (2006) indicates that the use of grid cell ignores the large scale spatial correlations structures of geomorphic features, and the use of sub-basin discretizationetization approach respect large and meaningful geomorphic structures.

Hence, discretizationetization of basin into dierent homogeneous units is the base for most distributed and semi-distributed hydrological models. Different models use different approaches of the subbasin partitions. For instance, KINEROS (Woolhiser et al., 1990) is based on partitions of the basin into topographical variables. In KINEROS model, the topological parameters of the segments of the channel link and delineated hillslopes is characterized by slope, length, geometry. In the case of TOPMODEL, the use of topographic variables for hydrological modelling is represented by the the socalled topographic index, . The use of topographic index, as topographic information for hydrological modelling is studied in many researches (Ambroise et al., 1996; Quinn et al.1995;Wolock and McCabe, 1995; Hjerdt et al., 2004). The WEPP(Flanagan and Nearing,1995), AGNPS, PRMS(Leavesley et al., 1983) assumes detail delineation of sub-basins into hillslopes using contributing area. Whereas SWAT (Neitsch et al. 2002) uses sub-basin discretization and further divide the sub-basins to the homogeneous units called HRUs, based on the combination of topographic elements(slope), landuse, and soil maps.

JGrass-NewAGE is a semi-distributed hydrological model which uses channel-hillslopes discretizationetization for representation of hydrological processes. Hillslopes are where all the meteorological forcing data and hydrological processes such as rainfall-runoff ptocesses is averaged (Formetta et al., 2011). The hydrological simulation is based on the appropriate hillslope and channel retization, and topological integration of the two entity. In NewAGE-JGrass, hillslopes are the part of the basin that drains to a particular channel link. However, the hillslopes can be seen as too coarse and diversified in process and responses to be used as modelling units. To investigate the effects of hillslope geometries, in JGrass-NewAGE model system, we used the three alternatives methods of channel extraction and channel segmentations from DEM that is provided by uDig Spatial Toolbox GIS. Threshold value on the contributing areas, threshold value on both slope and total contributing area, and threshold value for the concave sites are available in uDig GIS spatial toolbox (Formetta et al., 2014). Since the latter two methods are based on geomorphological based relations on the formation of channel, the use of the two methods could improve head upslope extractions with similar slopes and curvatures due to the similar channel initiation slopes and curvatures. In addition to the effects of those three methods of channel-hillslope extraction in spatial discretization, the use of different level of hillslope sizes in a particular methods, such as the choice of critical area on threshold area method, could also affect process representation, hence we have considered the effects of hillslope size on the hydrological processes and responses of a basin.

Different researchers investigated the effect of spatial discretization on basin rainfall-runoff modelling, and the results in general are inconclusive. It is also know that the calibration processes of most model, could hinder the exact effect of the spatial units for hydrological modeling. On the contrary, the use of Width function to characterize the geomorphological response of the basin, could be used as a direct way of measuring how those spatial discretization of hydrological modelling units affects the hydrological modelling. The idea is that if we understand the difference between the fully (pixel by pixel) and semidistributed (hillslope, or subbasin, or HRU) geomorphological hydrograph, it can be argued that this difference is the effect of subbasin discretization on hydrological modeling. Hence, we used the rescaled width function of grid by grid and subbasin topographic partition to anlayze the effect of the topologca. geomorphological and geometrical effect of different partitioning. Figure 1a shows that the difference in rescaled width function of a basin between grid by grid (as in the fully distributed model) and subbasin (as in the case of semi-distributed model). The moments of errors due to the use of subbasin discretization for process and input representation in hydrological model can be shown in figure 1b.

Reference

Ambroise, B., Beven, K., Freer, J., 1996. Toward a generalization of the topmodel concepts: Topographic indices of hydrological similarity. Water Resources Research 32 (7), 2135-2145.

Bogaart, P., Troch, P., 2006. Curvature distribution within hillslopes and catchments and its effect on the hydrological response. Hydrology and Earth System Sciences Discussions 3 (3), 1071-1104.

Dymond, J., Derose, R., Harmsworth, G., 1995. Automated mapping of land components from digital elevation data. Earth Surface Processes and Landforms 20 (2), 131-137.

Formetta, G., Antonello, A., Franceschi, S., David, O., Rigon, R., 2014. Hydrological modelling with components: A gis-based open-source framework. Environmental Modelling & Software 55, 190-200.

Formetta, G., Mantilla, R., Franceschi, S., Antonello, A., Rigon, R., 2011. The jgrass-newage system for forecasting and managing the hydrological budgets at the basin scale: models of flow generation and propagation/routing. Geoscientific Model Development 4 (4), 943-955.

Gao, H., Hrachowitz, M., Fenicia, F., Gharari, S., Savenije, H., 2013. Testing the realism of a topography driven model (flex-topo) in the nested catchments of the upper heihe, china. Hydrology & Earth System Sciences Discussions 10 (10).

Hack, J. T., Goodlett, J. C., 1960. Geomorphology and forest ecology of a mountain region in the central Appalachians. US Government Printing Office Washington, DC.

Hjerdt, K., McDonnell, J., Seibert, J., Rodhe, A., 2004. A new topographic index to quantify downslope controls on local drainage. Water Resources Research 40 (5).

Mantilla, R., Gupta, V. K., 2005. A gis numerical framework to study the process basis of scaling statistics in river networks. Geoscience and Remote Sensing Letters, IEEE 2 (4), 404-408.

O'loughlin, E., 1986. Prediction of surface saturation zones in natural catchments by topographic analysis. Water Resources Research 22 (5), 794-804.

Quinn, P., Beven, K., Lamb, R., 1995. The in (a/tan/) index: How to calculate it and how to use it within the topmodel framework. Hydrological processes 9 (2), 161-182.

Wolock, D. M., McCabe, G. J., 1995. Comparison of single and multiple flow direction algorithms for computing topographic parameters in topmodel.Water Resources Research 31 (5), 1315-132

Wood, E. F., Sivapalan, M., Beven, K., Band, L., 1988. Eects of spatial variability and scale with implications to hydrologic modeling. Journal of Hydrology 102 (1), 29-47. 14

<< Previous Abstract | Next Abstract