You are here:
How does spatial variability of climate affect catchment streamflow predictions?
Citation:
Patil, S., P. Wigington Jr., S. Leibowitz, E. Sproles, AND R. Comeleo. How does spatial variability of climate affect catchment streamflow predictions? JOURNAL OF HYDROLOGY. Elsevier Science Ltd, New York, NY, 517:135-145, (2014).
Impact/Purpose:
The ability of a hydrologic model to predict the streamflow response of rivers serves several important needs of our society, such as flood protection, irrigation demand, domestic water supply, and preservation of fish habitat. Based on its spatial configuration, a hydrologic model can roughly be classified as either lumped or distributed. Spatially lumped models conceptualize catchments as a single control volume, whereas spatially distributed models conceptualize catchments as a collection of multiple control volumes. For the spatially lumped models, an important assumption is that the meteorological inputs to the catchment are uniformly distributed over its entire drainage area. However, this has the potential to introduce significant uncertainty in catchments that have high spatial variability of climate, and can negatively affect streamflow predictability. In this study, our goal was to better understand the climatic conditions of catchments for which a spatially distributed model does (or does not) provide better streamflow predictions than a spatially lumped model. We implemented both lumped and distributed versions of the EXP-HYDRO model at 41 meso-scale catchments located across the Pacific Northwest (Oregon, Washington, and Idaho) and analyzed the performance difference between them. Results showed that the distributed model performed better than the lumped model in majority of the catchments. We further found that the spatial variability of moisture distribution alone is insufficient to explain the observed patterns of model performance improvement. Based on the results presented this study, we conclude that the use of spatially distributed meteorological inputs in hydrologic models has the potential to substantially improve streamflow predictions. Our use of spatially uniform model parameter values within a catchment ensured that any improvement obtained with the distributed model was solely based on the spatially distributed representation of meteorological inputs. However, this assumption will have to be relaxed for future investigations of the effects of spatially variable land use, soil types, and/or geology on catchment streamflow predictions.
Description:
Spatial variability of climate can negatively affect catchment streamflow predictions if it is not explicitly accounted for in hydrologic models. In this paper, we examine the changes in streamflow predictability when a hydrologic model is run with spatially variable (distributed) meteorological inputs instead of spatially uniform (lumped) meteorological inputs. Both lumped and distributed versions of the EXP-HYDRO model are implemented at 41 meso-scale (500 – 5000 km2) catchments in the Pacific Northwest region of USA. We use two complementary metrics of long-term spatial climate variability, moisture homogeneity index (IM) and temperature variability index (ITV), to analyze the performance improvement with distributed model. Results show that the distributed model performs better than the lumped model in 38 catchments, and noticeably better (>10% improvement) in 13 catchments. Furthermore, spatial variability of moisture distribution alone is insufficient to explain the observed patterns of model performance improvement. For catchments with low moisture homogeneity (IM < 80%), IM is a better predictor of model performance improvement than ITV; whereas for catchments with high moisture homogeneity (IM > 80%), ITV is a better predictor of performance improvement than IM. Based on the results, we conclude that: (1) catchments that have low homogeneity of moisture distribution are the obvious candidates for using spatially distributed meteorological inputs, and (2) catchments with homogeneous moisture distribution benefit from spatially distributed meteorological inputs if those catchments have high spatial variability of precipitation phase (rain vs. snow).