No abstract

No abstract

Precipitation simulated by regional climate models (RCMs) is generally biased with respect to observations, especially at the local scale of a few tens of kilometers. This study investigates how well two different RCMs are able to reproduce the spatial correlation patterns of observed summer precipitation for the central United States. On local scales, gridded precipitation observations and simulated precipitation are compared for the period of the 1987 First International Satellite Land Surface Climatological Project Field Experiment (FIFE) campaign. The results show that spatial correlation length scales on the order of 130 km are found in both observed data and RCM simulations. When simulations and observations are aggregated to different grid sizes, the pattern correlation significantly decreases when the aggregation length is less than roughly 100 km. Furthermore, the intermodel standard deviation between simulations with different domains or resolutions increases for aggregation lengths below ;130 km. Below this length scale there is a high level of randomness in the precise location of precipitation events. Conversely, spatial correlation values increase above this length scale, reflecting larger predictive certainty of the RCMs at larger scales. The findings on aggregated grid scales are shown to be largely independent of the underlying RCMs grid resolutions but not of the overall size of RCM domain. With regard to hydrological modeling applications, these findings indicate that precipitation extracted from the present RCM simulations at a catchment scale below the intermodel standard deviation length cannot be expected to accurately match observations.

Distributed hydrological models can make predictions with much finer spatial resolution than the supporting field data. They will, however, usually not have a predictive capability at model grid scale due to limitations of data availability and uncertainty of model conceptualizations. In previous publications, we have introduced the Representative Elementary Scale (RES) concept as the theoretically minimum scale at which a model with a given conceptualization has a potential for obtaining a predictive accuracy corresponding to a given acceptable accuracy. The new RES concept has similarities to the 25‐year‐old Representative Elementary Area concept, but it differs in the sense that while Representative Elementary Area addresses similarity between subcatchments by sampling within the catchment, RES focuses on effects of data or conceptualization uncertainty by Monte Carlo simulations followed by a scale analysis. In the present paper, we extend and generalize the RES concept to a framework for assessing the minimum scale of potential predictability of a distributed model applicable also for analyses of different model structures and data availabilities. We present three examples with RES analyses and discuss our findings in relation to Beven's alternative blueprint and environmental modeling philosophy from 2002. While Beven here addresses model structural and parameter uncertainties, he does not provide a thorough methodology for assessing to which extent model predictions for variables that are not measured possess opportunities to have meaningful predictive accuracies, or whether this is impossible due to limitations in data and models. This shortcoming is addressed by the RES framework through its analysis of the relationship between aggregation scale of model results and prediction uncertainties and for considering how alternative model structures and alternative data availability affects the results. We suggest that RES analysis should be applied in all modeling studies that aim to use simulation results at spatial scales smaller than the support scale of the calibration data.

No abstract