HydPy-H (HBV)

HydPy implements a very close emulation of the central routines of the famous HBV96 model, introduced by Lindstroem et al [1]. As discussed in [2], the results of version 1 application models of HydPy-H and the IHMS-HBV96 implementation of the Swedish Meteorological and Hydrological Institute agree very well for the majority of the tested river basins 1. This documentation focusses on technical aspects of HydPy-H and tries to be as precise as possible regarding the implementation of the process equations. The background of the scientifical development of the HBV96 model is more thoroughly covered in Lindstroem et al [1].

HydPy-H is divided into three base models, which can be used to compile different application models:

So far the following application models are compiled:

All these application models are stand-alone models, which can be combined freely with all other models implemented in HydPy.

Footnotes

1

Unfortunately, the report [2] is only available in German so far. But inspecting pictures 2.2 to 2.11 should be instructive nonetheless. The simulated runoff values are virtuelly identical for catchments in high and low mountain ranges as well as in lowland areas (pictures 2.2 to 2.5). The same is true for all internal states, beginning with the interception storage (picture 2.6) and ending with the lower groundwater storage (picture 2.11). Table 2.5 evaluates the aggreement between the results of HydPy-H and the IHMS-HBV96 implementation of the SMHI for the whole Rhine river basin, using the Nash-Sutcliffe efficiency. Relevant differences occur only within the Main river basin. These differences result from implausible evaporation values calculated by the IHMS software. If these is due to a bug of the IHMS software or due to an incorrect configuration of the HBV forecasting model of the German Federal Institute for Hydrology could not be clarified so far.