s u m m a r yThe Office of Hydrologic Development (OHD) of the National Oceanic and Atmospheric Administration's (NOAA) National Weather Service (NWS) conducted the second phase of the Distributed Model Intercomparison Project (DMIP 2). After DMIP 1, the NWS recognized the need for additional science experiments to guide its research-to-operations path towards advanced hydrologic models for river and water resources forecasting. This was accentuated by the need to develop a broader spectrum of water resources forecasting products (such as soil moisture) in addition to the more traditional river, flash flood, and water supply forecasts. As it did for DMIP 1, the NWS sought the input and contributions from the hydrologic research community. DMIP 1 showed that using operational precipitation data, some distributed models could indeed perform as well as lumped models in several basins and better than lumped models for one basin. However, in general, the improvements were more limited than anticipated by the scientific community. Models combining so-called conceptual rainfall-runoff mechanisms with physically-based routing schemes achieved the best overall performance. Clear gains were achieved through calibration of model parameters, with the average performance of calibrated models being better than uncalibrated models. DMIP 1 experiments were hampered by temporally-inconsistent precipitation data and few runoff events in the verification period for some basins. Greater uncertainty in modeling small basins was noted, pointing to the need for additional tests of nested basins of various sizes.DMIP 2 experiments in the Oklahoma (OK) region were more comprehensive than in DMIP 1, and were designed to improve our understanding beyond what was learned in DMIP 1. Many more stream gauges were located, allowing for more rigorous testing of simulations at interior points. These included two new gauged interior basins that had drainage areas smaller than the smallest in DMIP 1. Soil moisture and routing experiments were added to further assess if distributed models could accurately model basininterior processes. A longer period of higher quality precipitation data was available, and facilitated a test to note the impacts of data quality on model calibration. Moreover, the DMIP 2 calibration and verification periods contained more runoff events for analysis. Two lumped models were used to define a robust benchmark for evaluating the improvement of distributed models compared to lumped models. Fourteen groups participated in DMIP 2 using a total of sixteen models. Ten of these models were not in DMIP 1. This paper presents the motivation for DMIP 2 Oklahoma experiments, discusses the major project elements, and describes the data and models used. In addition, the paper introduces the findings, which are covered in a companion results paper (Smith et al., this issue). Lastly, the paper summarizes the DMIP 1 and 2 experiments with commentary from the NWS perspective. Future papers will cover the DMIP 2 experiments in the western...
Results indicate that in the two study basins, no single model performed best in all cases. In addition, no distributed model was able to consistently outperform the lumped model benchmark. However, one or more distributed models were able to outperform the lumped model benchmark in many of the analyses. Several calibrated distributed models achieved higher correlation and lower bias than the calibrated lumped benchmark in the calibration, validation, and combined periods. Evaluating a number of specific precipitation-runoff events, one calibrated distributed model was able to perform at a level equal to or better than the calibrated lumped model benchmark in terms of event-averaged peak and runoff volume error. However, three distributed models were able to provide improved peak timing compared to the lumped benchmark. Taken together, calibrated distributed models provided specific improvements over the lumped benchmark in 24% of the model-basin pairs for peak flow, 12% of the model-basin pairs for event runoff volume, and 41% of the model-basin pairs for peak timing. Model calibration improved the performance statistics of nearly all models (lumped and distributed). Analysis of several precipitation/runoff events indicates that distributed models may more accurately model the dynamics of the rain/snow line (and resulting hydrologic conditions) compared to the lumped benchmark model. Analysis of SWE simulations shows that better results were achieved at higher elevation observation sites. Although the performance of distributed models was mixed compared to the lumped benchmark, all calibrated models performed well compared to results in the DMIP 2 Oklahoma basins in terms of run period correlation and %Bias, and event-averaged peak and runoff error. This finding is noteworthy considering that these Sierra Nevada basins have complications such as orographicallyenhanced precipitation, snow accumulation and melt, rain on snow events, and highly variable topography. Looking at these findings and those from the previous DMIP experiments, it is clear that at this point in their evolution, distributed models have the potential to provide valuable information on specific flood events that could complement lumped model simulations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.