On the need for catchment classification

The rapid growth in our ability to generate new descriptions of catchment heterogeneity, using both experimental and modeling approaches, is now outstripping our ability to use the knowledge effectively in theory and practice. Studies in experimental catchments have identified the idiosyncrasies of many river basins, and produced very complex, yet highly qualitative descriptions of catchment behavior. However, the sporadic attempts to regionalize those studies have been of limited value, because of the difficulty in producing concise, easily understood explanations of different basin behaviors. In a similar way, the variety of simulation models has continued to grow as new data sources and computing resources have become available, and new processes have been included in models. Some models have very detailed representations of processes which are difficult to validate except in the most intensively instrumented sites, while other models are simpler and easier to parameterize, but harder to validate against observations at the appropriate spatial scale. No clear guidance is available on which model or model structure is appropriate to any particular catchment type or management question. As a result, model structures often seem to be ad hoc. Similarly, no clear guidance is available on which dominant processes and mechanisms are operating in a given catchment type. As a result, the way that we instrument catchments and seek process understanding often also seems to be ad hoc. We suggest that the underlying cause of many of our problems in catchment hydrology is the tremendous variability in space, time and process, which is present in natural hydrological systems all around the globe. It is enormously ambitious for