|
Implications of Hydrologic Variability and Change for Western Water Management
Dennis Lettenmaier
Civil and Environmental Engineering
University of Washington
Implementation of traditional hydrologic forecast methods uses what might be termed a "bottom up" approach, in which hydrologic simulation models are applied to river basins, typically with drainage areas ranging from a few hundred to a few thousand km2, each of which are defined by a stream gauge. Parameters are estimated locally via a process of trial and error (or in some cases, using automated search algorithms). Because the parameters of the so-called conceptual simulation models that are used operationally are only loosely related to physically measurable quantities, attempts at transferring parameters among basins have had only minimal success, and the entire process is quite time consuming. For this reason, implementation of hydrologic forecast tools compatible with evolving climate prediction methods like ensemble forecasts for large river basins has been at best a painstaking process.
Macroscale hydrology models, like the Variable Infiltration Capacity (VIC) model, have evolved over the last 10 years, and take what might be termed a "top down" approach. They are implemented over large areas, usually defined by grid meshes typically having resolutions of fractions of degrees latitude by longitude. They have as their primary objective representation of the surface hydrology of large continental river basins, and secondarily the response of major tributaries, which are resolvable by the specified grid resolution.
Our experience in development and testing of the VIC model is described, followed by two applications. The first is seasonal streamflow forecasting over the eastern U.S. during spring and summer, 2000, and the Columbia River basin during the 2000-2001 drought. These exploratory seasonal streamflow forecasts utilized ensemble climate forecasts (precipitation and average temperature for six-month lead times, updated monthly), produced by the NCEP/CPC Global Spectral Model (GSM). Evolution of a region of anomalously low soil moisture over the southeastern U.S. in spring and summer 2000, and its implications for summer streamflow, are illustrated. The second case study utilizes downscaling methods developed as part of the seasonal streamflow forecasting work, and applied to simulate hydrologic changes associated with three NCAR/DOE Parallel Climate Model ensemble climate predictions, each of length 105 years, downscaled over the western U.S. The archived PCM ensemble outputs (monthly total precipitation and average temperature at T42 grid resolution) were first bias-corrected, disaggregated from a monthly to daily time step and then downscaled to 1/8 or 1/4-degree spatial resolution using the VIC model. Water resource simulation models (for the Columbia (CRB) and Sacramento-San Joaquin (SSJB) were then used to predict, on a monthly time-step, the effects of the climate change scenarios on streamflow timing and volume.
The main results are: a) CRB hydrology, and water management objectives, were more robust to the PCM scenarios than was the SSJB; b) decadal-scale variations in precipitation were as large a driver of hydrologic effects as were the relatively modest PCM temperature changes, especially over the next 50 years; c) the temperature-driven seasonality changes in streamflow found in prior climate change studies were apparent for the SSJB but were present to a lesser extent for the CRB and Colorado River basin; d) in the CRB, agricultural withdrawals, hydropower generation and minimum streamflow requirements associated with salmon recovery efforts would be threatened over the next 50 years; while in the SSJB, the predicted changes in future streamflows would decrease reliability for hydropower generation, water supply, releases for fisheries support and flood control.
About the Lecturer
Dennis Lettenmaier received his B.S. in Mechanical Engineering (summa cum laude) at the University of Washington in 1971, his M.S. in Civil, Mechanical, and Environmental Engineering at the George Washington University in 1973, and his Ph.D. at the University of Washington in 1975. He joined the University of Washington faculty in 1976.
In addition to his service at the University of Washington, he spent a year as visiting scientist at the U.S. Geological Survey in Reston, VA (1985-86) and was the Program Manager of NASA's Land Surface Hydrology Program at NASA Headquarters in 1997-98.
He is a member of the American Geophysical Union, the American Water Resources Association, the American Meteorological Society, and the American Society of Civil Engineers. He was a recipient of ASCE's Huber Research Prize in 1990, is a Fellow of the American Geophysical Union and American Meteorological Society, and is the author of over 100 journal articles. He is currently Chief Editor of the American Meteorological Society Journal of Hydrometeorology. Dr. Lettenmaier's research interests include Hydroclimatology, Surface Water Hydrology, and GIS & Remote Sensing.
More Information
http://www.hydro.washington.edu/
http://www.ce.washington.edu/faculty/bios/lettenmaier_d.html
|
Dennis Lettenmaier
|
