Modular Assessment of Rainfall-Runoff Models Toolbox (MARRMoT) v1.0: an open- source, extendable framework providing implementations of 46 conceptual hydrologic models as continuous space-state formulations

Wouter J. M. Knoben, Jim E. Freer, Keirnan J. A. Fowler, Murray C. Peel, Ross A. Woods
2019 Geoscientific Model Development Discussions  
<p><strong>Abstract.</strong> This paper presents the Modular Assessment of Rainfall-Runoff Models Toolbox (MARRMoT): a modular open-source toolbox containing documentation and model code for 46 existing conceptual hydrologic models. The toolbox is developed in Matlab and works with Octave. Models are implemented following several best practices: definition of model equations (the mathematical model) is kept separate from the numerical methods used to solve these equations (the numerical model)
more » ... to generate clean code that is easy to adjust and debug; the Implicit Euler time-stepping scheme is provided as the default option to numerically approximate each model's Ordinary Differential Equations in a more robust way than (common) Explicit schemes would; threshold equations are smoothed to avoid discontinuities in the model's objective function space; and the model equations are solved simultaneously, avoiding physically unrealistic sequential solving of fluxes. Generalized parameter ranges are provided to assist with model inter-comparison studies. In addition to this paper and its Supporting Materials, a User Manual is provided together with several workflow scripts that show basic example applications of the toolbox. The toolbox and documentation are available from <a href=" https://github.com/wknoben/MARRMoT"target="_blank">https://github.com/wknoben/MARRMoT</a> (DOI: <a href="https://doi.org/10.5281/zenodo.2482542" target="_blank">https://doi.org/10.5281/zenodo.2482542</a>). Our main scientific objective in developing this toolbox is to facilitate the inter-comparison of conceptual hydrological model structures which are in widespread use, in order to ultimately reduce the uncertainty in model structure selection.</p>
doi:10.5194/gmd-2018-332 fatcat:avt64ufsdjcjdgaiy5ou2ndcia