Information theoretic fingerprinting of hydrologic Models

The multiplicity of hydrologic models with varying degrees of process fidelity are the product of a wide range in research interests, computing capabilities, data availability, and scientific philosophy within the field of hydrology. Ideally, this multiplicity would benefit the community, allowing for the proper selection of a model based on research and logistical constraints. In practice, this is not the case, as there are simply too many models, process parameterizations, and parameter sets to be able to reasonably consider all options available.

One avenue being developed to make this diversity more accessible to users is based on concepts of models with flexible model structures. The Structure for Unifying Multiple Modeling Alternatives (SUMMA) is one such modeling framework. SUMMA provides flexible representations for spatial arrangements, flux parameterizations, and parameter values. This makes it an ideal modeling framework for investigating the effects of different model structures.

We explore how information theory can be used to evaluate model behavior with the ultimate aim of model selection based on available information. In this work, we instantiate a large number of SUMMA runs with different modeling options across sites in diverse hydrometeorological regimes and analyze the full set of energy and mass balance terms to quantify the effects that model structure has on model output. To do so, we calculate information theoretic process networks from the timeseries outputs of the model instantiations. These high-dimensional networks are then analyzed via clustering and dimensionality reduction algorithms to provide high-level insights into the model structure space that is covered.