Hydrology and Water Resources: Current Research

Reconciling Projections of Future Colorado River Streamflow



Within the Upper Colorado River Basin, reductions in naturalized streamflow (water management effects removed) by the mid 21st century have been projected to range from 6 to 45% in published studies, and a recent analysis of future P-E (a proxy for runoff) suggests an “imminent transition to a more arid climate in southwestern North America”. While the range of projections may be of intellectual interest and stimulate scientific debate, to users and decision makers at the federal level, in the seven basin states, and internationally, providing what appears to be conflicting information on future conditions is a serious impediment to drought and climate change planning.

To better understand the reasons for the wide range of projections, we have undertaken a systematic intercomparison of methodologies and models to understand why different modeling approaches produce such different levels of flow reduction. To date, we have evaluated annual elasticities of annual streamflow (fractional change in runoff divided by fractional change in precipitation) as inferred from observations and from three spatially distributed land surface schemes: the Variable Infiltration Capacity (VIC) model, the NOAH land scheme, and a grid-based version of the NWS Sacramento soil moisture accounting model. From the models, we have also evaluated the sensitivity of runoff to temperature changes as fractional changes in annual runoff per °C of (uniform) temperature increase.

From observations, the inferred elasticity of the Colorado River discharge at Lees Ferry is about 1.5, whereas the range from the three models is about 1.75 to 1.90. Temperature sensitivities cannot easily be inferred from observations, however the model sensitivities at Lees Ferry range from about 2.2 to 2.85% per °C for equal changes in daily maxima and minima (which implies no change in downward solar radiation in the method used to prescribe model forcings), whereas the range is about 3.3 to 4.1% for a one degree uniform increase in daily temperature when the daily minima were unchanged (constant dew point). Ongoing work is diagnosing the spatial distribution of elasticities and sensitivities across the basin for the three models, and is extending the analysis to other models applied to smaller subcatchments of the Colorado.


Western Water Assessment (WWA), California Applications Project (CAP), Climate Assessment for the Southwest (CLIMAS)

Primary Funding


Related Publications

For publications on climate impacts on PNW water resources, please see CIG Publications.

Hoerling, M., D. Lettenmaier, D. Cayan, and B. Udall. 2009. Reconciling projections of Colorado River streamflow. Southwest Hydrology 8(3): 20-21, 31.