A collection of python scripts to discretize a watershed shapefile into sub-areas to account for geospatial attributes.
Authors: 2020-2021 Andy Wood and Hongli Liu designed and wrote the codes.
HL did initial code prototyping for the specified design and AW upgraded codes through revission, debugging, streamlining, testing and documentation.
The experimental application goal was to provide an effective discretization of watersheds based on 3 primary watershed attributes influencing hydrologic runoff variability: elevation, vegetation, and solar radiation exposure.
In addition, the application targeted use in SUMMA modeling efforts, in which watersheds are viewed as 'grouped response units' (GRUs) with the sub-watersheds termed 'hydrologic response units' (HRUs) -- a naming convention used throughout the code.
Rather than use data-driven clustering approaches for arbitrary attributes to derive HRUs, known controlling factors are applied in a binary fashion to create 8 potential discretization levels (all permutations of the 3 factors) for the GRUs.
Small HRUs can be eliminated (based on area fraction or area thresholds).
The overarching objective is to support the implementation of watershed models that represent spatial and process heterogeneity with a computationally frugal approach -- i.e., 1-8 HRU elements per watershed in this case -- although this limit is not prescribed and the code is extensible to allow for the introduction of other factors.
The code is organized into subdirectories as follows:
The recommended workflow for appying this code is the following:
Andy Wood, andywood@ucar.edu Hongli Liu, hongli.liu@usask.edu
The development of this code base and the associated experimental project (led by A. Wood) was funded by the US Bureau of Reclamation under Cooperative Agreement #R16AC00039.
The initial application of the code is described in:
Liu, H, AW Wood, D Broman, G Brown, and J Lanini, 2021. Impact of SUMMA hydrologic model discretization on the representation of snowmelt and runoff variability. J. Hydromet. (in prep, target submission Nov 2021)..
We thank Genevieve Brown of the University of Waterloo for providing the code implementation of the radiation algorithm that was included in the radiation preparation step.
We also thank Wouter Knoben for providing a MERIT-based DEM file and Naoki Mizukami for providing the landcover dataset file and soiltype files that were used in the initial experiment & development.