Seasonal Hydroclimate Predictions for the Western United States

Predicted precipitation at a 36 x 36 km resolution for December 1997-February 1998. The large-scale data are from a 2.5° x 2° UCLA-GCM seasonal forecast.

Research Objectives

Seasonal-scale hydroclimate predictions are important for improving water resources management and assessment, and preparation for natural disasters (e.g., flooding and landslide potential). Seasonal climate prediction is a challenging task due to the chaotic behavior of global numerical models. The main objective of this collaborative research is to examine the potential for seasonal-scale predictions and application of seasonal forecast data toward impact assessments such as streamflow and water budget.

Computational Approach

In partnership with the National Center for Environmental Prediction (NCEP), University of California (Scripps Institution of Oceanography, UCLA), and the National Weather Service (NWS), we have nested the Regional Climate System Model (RCSM) within the UCLA General Circulation Model (GCM). The NCEP has provided seasonal-scale tropical sea surface temperature (SST) anomaly forecasts for the 1997-1998 winter season, when the intense El Niño event occurred. The UCLA-GCM has simulated multiple realizations at a 2.5° and 2° resolution in the zonal and meridional directions, respectively. The RCSM was nested using two different resolutions, 20 x 20 km and 36 x 36 km, to examine the effects of domain size and spatial resolution. In addition to the seasonal predictions, a diagnostic simulation using the NCEP global analysis was performed and evaluated for a reference simulation.

Accomplishments

We have produced two predictions and one diagnostic simulation over the western United States. The UCLA-GCM successfully predicted large-scale circulation patterns during December 1997 and February 1998. However, the GCM did not simulate synoptic-scale fields as accurately during the winter.

Due to the characteristics of the GCM fields, the RCSM yielded better predictions over the larger domain than the small domain, despite the lack of spatial resolution. The reference simulation successfully simulated streamflow at the Russian River.

Significance

Seasonal hydroclimate predictions using a physically based model are crucial for improved decision making in water resources management, preparation for natural disasters, agriculture, etc. Despite its significance, this work is still in its early stage and is an internationally recognized growth area.

Impact assessment requires very high-resolution information, and the computational requirements increase to the third power as the resolution increases (e.g., doubling the spatial resolution requires about 8 times the computational load). Hence, high performance computing is a crucial component to successful improvements in seasonal hydroclimate research.

Publications

J. Kim, N. Miller, J. Farrara, D. Cayan, and K. Mo, "Winter-season hydroclimate study for the western U.S. using the Regional Climate System Model (RCSM)," Proceedings of the 11th Conf. on Applied Climatology (in press, 1999).

N. Miller, J. Kim, J. Farrara, K. Mo, and D. Cayan, "Short-term and seasonal streamflow predictions for a California Coastal basin during the 1997-1998 winter," Proceedings of the 14th Conf. On Hydrology (in press, 1999).