Spatial modeling for resources framework (SMRF): A modular framework for developing spatial forcing data for snow modeling in mountain basins

Authors: Havens, Scott; Marks, Daniel; Kormos, Patrick; Hedrick, Andrew

Submitted to: Computers and Geosciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/29/2017
Publication Date: 9/1/2017
Citation: Havens, S.C., Marks, D.G., Kormos, P.R., Hedrick, A. 2017. Spatial modeling for resources framework (SMRF): A modular framework for developing spatial forcing data for snow modeling in mountain basins. Computers and Geosciences. 109:295-304. https://doi.org/10.1016/j.cageo.2017.08.016.
DOI: https://doi.org/10.1016/j.cageo.2017.08.016

Interpretive Summary: This paper presents a modular software toolkit for accessing, evaluating, correcting, downscaling and distributing weather data from stations or regional atmospheric models, so that these data can be used to force physics-based snow, hydrologic or ecosystem models. We present the software developed, methods used, the required computer resources, and information on acquisition and installation. SMRF provides direct access to a much wider range of station data and distribution methods than previously available, and allows application of research-quality models over large basins.

Technical Abstract: In the Western US and many mountainous regions of the world, critical water resources and climate conditions are difficult to monitor because the observation network can be very sparse. The most important of these resources are water from the mountain snowpack, streams, and reservoirs utilized for irrigation, flood control, power generation, and ecosystem services. Water supply forecasting in a rapidly changing climate has become increasingly difficult due to “extreme” conditions and in response, operational water supply managers have begun to move from statistical techniques towards using physically based models. As we begin to transition physically based models from research to operational use, we must address the most difficult and time-consuming aspect of model initiation: the need for robust methods for developing and distributing the input forcing data. In this paper we present a new open source framework, the Spatial Modeling for Resources Framework (SMRF), which automates and simplifies the most common forcing data distribution methods. It is computationally efficient and can be implemented for both research and operational applications. Currently, SMRF is able to generate all of the forcing data required to run physically based snow or hydrologic models at 50–100 m resolution over regions of 500–10,000 km2, and has been successfully applied in real time and historical applications for both the Boise River Basin in Idaho, USA and the Tuolumne River Basin in California, USA. These applications use meteorological station measurements and numerical weather prediction model outputs as input data. SMRF has significantly streamlined the modeling workflow, decreased model set up time from weeks to days, and made near real-time application of physics-based snow and hydrologic models possible.