Abstract

The Palmer Drought Severity Index (PDSI) is used for quantifying present and historical drought conditions based on a simple water balance model. The PDSI is broadly accepted in the U.S and is routinely used in water management decision-making. However, in the Western U.S. a significant portion of precipitation falls as snow, which is not considered in common PDSI methods. The timing and rate of snowmelt is fundamentally different from rainfall, occurring later in the year and often at higher daily rates. In this study we ask the question: “How does snowmelt modify the severity and forecastability of hydrological drought?” To address this question, we developed a metric of effective precipitation that included both snowmelt and rainfall at 462 SNOTEL sites across the Western U.S. We then ran classic PDSI algorithms using both the standard precipitation metric and our effective precipitation metric. We found that incorporating snowmelt into the PDSI resulted in the moderation of drought, especially intense drought conditions. During extreme drought (PDSI between -4 and -5) we found a 15% reduction in PDSI by including snow and a 20% reduction during exceptional drought (PDSI < -5). One complication to including snowmelt is that snow-adjusted PDSI methods typically indicated drought later than normal PDSI methods due to the delay in water release from snowmelt. To validate our methods, we plan to compare the PDSI results to soil moisture measurements at the SNOTEL sites and to nearby streamflow. Our results suggest that extreme drought is likely to increase from earlier snowmelt and shifts from snow to rain. Prediction of drought timing and severity at regional scales could benefit dramatically from including snowmelt, which remains a challenge for future research.

Citation

Weiner, J.; Harpold, A.A.; Safeeq, M. (2016): Is Snow a Drought Buster? The Need to Incorporate Snow into Common Drought Indices. Fall Meeting, American Geophysical Union, December 2016. Abstract H21G-1508..