Applying metrics, measures or objective functions is at the heart of Hydrology and Hydrological Modeling in its widest sense. They are important to
• analyze and classify the systems we are dealing with
• identify the scales at which to separate implicit and explicit representations of structures and processes
• calibrate models by comparison of their simulations to observations
• evaluate the quality of forecasts and simulations by comparison to observations
• quantify information (and information transfer) about hydrological processes or models
In its origins, hydrological modeling was mainly focused on reproduction of observed discharge timeseries on catchment scale. Accordingly, the available repertoire of metrics was small and mainly based on space-and-time-aggregated average measures of performance error, e.g. the well-known Nash-Sutcliffe efficiency. In recent years, hydrology has moved far beyond its beginnings, both with respect to detail and thematic boundaries: Hydrology becomes more and more interlinked with atmospheric, vegetation, erosion, groundwater etc. modeling. With this, more than discharge hydrographs are asked from Hydrology, as spatio-temporal patterns of transport and storage form the interface between the model disciplines. Accordingly, we need new ways to both analyze and describe the systems we are dealing with and also to evaluate the models used to represent these systems.
In this session, we welcome contributions to the following topics:
• Metrics to evaluate hydrological timeseries (forecasts, simulations) by comparison with observed timeseries, especially metrics that move from standard value-based comparison (such as the Nash) towards multi-criteria, comprehensive and informative comparison.
• Metrics for (1-, 2-, 3-d) spatial and temporal analysis of hydrological systems, such as landscape structure and flow path connectivity
• Metrics that explicitly quantify the information, relevant in a hydrological context, embedded in the data (and in our hydrological models)
• Metrics for functional similarity
• Metrics for optimality, such as Maximum Entropy Production (MEP) and Maximum Energy Dissipation (MED) to explain the formation and maintenance of structure and gradients. In short, any metrics that might help unraveling rules that drive and govern evolution of landscape structure.
• Metrics suitable for probabilistic/ensemble information, expanding deterministic/single approaches.