Use of the Palmer index and other water supply indexes for drought monitoring in Colorado

Abstract

The Colorado Drought Response Plan of 1981 assigned drought monitoring responsibilities to a special intergovernmental technical working group called the Colorado Water Availability Task Force (WATF). The intent of this group is to use existing data sources and information products to monitor Colorado's water supplies. The information assembled and interpreted by the WATF is then used by State decision makers to guide State government's response to drought. The Palmer Index, developed in the 1960's, has become a credible tool for monitoring drought and assessing drought severity on the national scale. It reasonably depicts soil moisture conditions using a simple hydrologic balance accounting for precipitation, evapotranspiration, runoff and soil moisture recharge. However, experiences of the WATF have revealed that Palmer Index values, currently generated weekly through the growing season by the National Climatic Data Center for 5 climatic divisions in Colorado, were only marginally useful for drought monitoring. The regions were too large and climatically diverse, and input temperature and precipitation data were not adequately controlled to produce consistent and meaningful results. With the encouragement arid cooperation of the WATF this project was undertaken to adapt the Palmer Index model to Colorado. The original program was brought to Colorado, the state was broken down into 25 climatically similar regions, and a simple routine for adjusting input data to correct for missing data and station moves was implemented. The existing model was then used to generate 30 years of monthly Palmer Index values for all 25 regions of the state. A thorough examination of these new Palmer Indexes has been performed. Comparisons with the original indexes show noticeable differences and considerable small scale detail which previously could not be resolved. With the new smaller regions it is now reasonable to use contour analysis of Palmer Index values to visually describe local variations in drought severity across Colorado. Two case studies were conducted to show how the new indexes compared to original index values during severe drought situations: 1) the end of the 1956 drought on the Eastern Plains, and 2) the 1976-1977 winter drought in the Colorado mountains. A particular application of the Palmer Index was given special attention. Palmer Index values were correlated with dryland winter wheat yields. The best correlations with annual yields were obtained using June 1 or July 1 Palmer Index values. Good correlations were obtained in most of the major wheat growing areas but especially in the northeastern counties of Colorado. Better correlations were obtained using indexes calculated for the new areas than were obtaining using the original index values. The WATF agreed that the capability to calculate Palmer Indexes here in Colorado, with our own choice of climatic divisions, greatly increases the utility of this drought monitoring tool. More refinements are possible, and further study conducted jointly with agricultural interests would be desirable. This index is already of sufficient value to the WATF to justify the low cost required to produce it on a routine monthly basis.