Since it first debuted online in 2014, the Drought Risk Atlas has provided decision-makers with historical drought information and web-based tools to help visualize and assess drought risk. Following a significant upgrade, the Atlas now also tracks long-term changes in drought and streamflow levels for specific locations across the U.S. The National Drought Mitigation Center, based at the University of Nebraska-Lincoln, added trend analysis to the Atlas in July.
For each station in the Atlas with a sufficiently long, uninterrupted record, the Atlas now provides an assessment as to whether it is getting drier or wetter, and if so, whether the change is statistically significant. Trends are visible either as lines for individual stations, or as color-coded dots on a map.
“The new trends analysis is a faster way to help people see if there are long-term changes in drought in their area,” NDMC climatologist Brian Fuchs said.
Curtis Riganti, NDMC climatologist and U.S. Drought Monitor author, led the research, working closely with Mark Svoboda, NDMC climatologist and director, and Fuchs, the USDM author who leads NDMC’s Monitoring program area. Riganti presented on the initial concept of trends analysis at the American Meteorological Society meeting in January 2019.
Riganti then worked extensively with Chris Poulsen and Jeff Nothwehr of NDMC’s Information Technology, GIS and web team on the analysis, database structure and web display needed to implement the concept and display the 7,992,610 new calculations and visualizations.
“This is a good example of the research-to-application approach that the NDMC brings to drought science,” Svoboda said. “We are building on the high-quality data in the Drought Risk Atlas and finding more ways to answer the fundamental question that the Atlas is designed to answer: How do present conditions compare with the past?”
Getting to a trend line or map within the Atlas requires several decisions beyond picking a climate station: which drought index to use; whether to look at calendar years, seasons, or the growing season; how far back to start; and what counts as significant change (the p-value on a Mann-Kendall test). Although looking at a longer record produces more meaningful results, it also reduces the number of stations with uninterrupted data.
New trend tabs appear for climate and hydrological data stations in the Atlas, online at droughtatlas.unl.edu. Trend analyses are for several measures of drought: precipitation, Standardized Precipitation Index, Standardized Precipitation and Evapotranspiration Index, the Palmer and self-calibrating Palmer Drought Indexes, and dry streaks, defined as the number of days in a row without rain. Trend analysis for streamflow is based on the Standardized Streamflow Index, and if in doubt as to which SSI distribution to choose, Riganti recommends the Tweedie over the gamma.
Funding for the Drought Risk Atlas has come from the National Integrated Drought Information System and from the U.S. Department of Agriculture.
Although some stations show trends and some maps appear to show patterns, Riganti cautioned that more research is needed before drawing definitive conclusions. “It’s a wealth of data that we have now,” he said.
- Kelly Helm Smith, NDMC Communications