In the late 1990s, National Drought Mitigation Center founding director Don Wilhite assigned Mark Svoboda to find every drought-related index, indicator and tool that existed, and request access to the data that was used to create them. Unfortunately, Google didn't debut until after he began his search.
“There wasn’t a whole lot out there, and I remember the response to my request for operational data was getting a hard copy map in the mail of the Palmer Drought Severity Index from the National Climatic Data Center,” Svoboda said. “That wasn’t even delivered digitally at the time.”
With scarcity of information in mind, Svoboda presented on drought mapping at the 1998 American Meteorological Society annual meeting. Another presenter at the session, Douglas Le Comte of the Climate Prediction Center, was interested in combining various drought indices into one map. The two talked after the meeting about joining forces.
“That’s where the idea was born to make a higher resolution map made from combining several indicators together that shows where drought is and how severe it is,” said Svoboda, who is now the NDMC director.
Their collaboration spearheaded the creation of the U.S. Drought Monitor, which celebrates its 20th anniversary this year. Every week since the Drought Monitor was unveiled at a White House press conference on Aug. 11, 1999, the NDMC, U.S. Department of Agriculture (USDA) and the National Oceanic and Atmospheric Administration (NOAA) have teamed up to release an update of the USDM.
An extensive network from an array of agencies has contributed data and on-the-ground observations to produce more than 1,000 maps, and the USDM has grown to include all U.S. states and territories, including the additions of the U.S.-affiliated Pacific Islands and the U.S. Virgin Islands in 2019. It has triggered billions of dollars in federal aid and low interest loans. Federal, state, tribal, local and basin-level decision makers use it to detect emerging droughts.
And it all started as a map made with CorelDRAW 8.
“I think I have a curled-up map that actually shows one of the original drafts of the Drought Monitor,” Le Comte, now retired from the CPC, recently said from his Arlington, Virginia, home. A few minutes later, he found the map.
Dated July 13, 1999, the prototype features some classifications familiar to those who have used the USDM over its 20-year existence. Yellow blobs indicating abnormally dry areas covered much of the Southwest and Northeast. Encircled in red were portions of the Pacific Northwest, Alaska, Hawaii, the Northeast and the Mid-Atlantic, including all of Maryland, Virginia and Washington D.C. These were the only two colors on the draft, though, with red being an all-encompassing indicator of drought. (Each level of drought now has its own designated color.) Arrows specified the class and types of drought in those locales, with one pointed directly at our nation’s capital. That drought, the USDM’s early authors believe, helped provide the project with a big green light.
“Serendipity is the word,” Le Comte said.
Not long after creating that mid-July map, a secretarial briefing regarding the USDM was held at the White House. The USDM’s proponents told officials that it could help heighten awareness of drought as an environmental hazard, provide the public and decision-makers vital information about the creeping disaster and decrease response lags to drought, like the rare one building in the Northeast in the summer of ‘99.
“The Palmer wasn’t showing that drought evolving nearly fast enough,” said Svoboda, who was a USDM author for 17 years. “Our new prototype showed potential to pick up the signal earlier given we weren’t solely relying on any one drought indicator in particular. So they informed us that this new prototype drought indicator was going to go operational this summer. After production of the first operational map in early August, the very next week, the experimental label was off the map. So I think that might be the shortest experimental product in government history. That drought is really what made it all happen, in a way. So we quickly ramped up from two authors to six authors in the span of just a few months.”
The first six USDM authors were Svoboda and Michael Hayes from the NDMC, Le Comte and Rich Tinker from NOAA’s Climate Prediction Center (CPC), and Brad Rippey and David Miskus, who was on assignment from the CPC at the USDA, where he joined Rippey. Nearly 30 authors have taken two-week shifts creating the map over its 20-year history. Since late 2000, once the map is released each Thursday at droughtmonitor.unl.edu, the author’s name has been included alongside it. Tinker (135 shifts), Miskus (122) and Rippey (96) have authored the most so far.
The map is now created with GIS software, and authors consider data from more than 50 sources, including precipitation, temperature, evapotranspiration, the Palmer Drought Severity Index, the Standardized Precipitation Index, soil moisture indicators, hydrologic data, snowpack data, satellite-based assessments of vegetation health, land-data assimilation models and many more. Some of those sources have been vital to the map’s creation since its early stages, when the final drought report was essentially hand-drawn onto the maps utilizing late-’90s graphic design software.
“Maps all over my desk,” Svoboda recalled. “Maps on the floor. And you’re trying to piece them together in your head. That’s hard to do for 50 states in just over two days. Once you get into GIS, everything’s digital. You can overlay those together and make a much quicker assessment of the situation. You really start to see the patterns and determine where they agree or disagree. And the subject matter expertise is vitally important when those areas diverge to determine which indicators are going to be the best ones telling the story.”
Le Comte said he realized early on that the map was going to be a vital tool when he saw versions of it broadcast on the Weather Channel and reprinted in The New York Times, USA Today, The Washington Post and elsewhere.
“It is really something I enjoyed doing,” Le Comte said. “I felt like a little bit of a pioneer doing this, because it was a feeling that this is something important, and that probably would be widely used if done correctly.”
Rippey saw the first sign that the weekly publication could be a vital aid trigger in late 2002, when then-USDA Chief Economist Keith Collins invited him to his office in the midst of a drought in the High Plains.
“They said we’ve got this drought going on, and we’ve got some nonfat dry milk to give away to these drought-ravaged producers,” Rippey recalled. Rather than base the program eligibility on state-level pasture condition reports, as had happened previously, Collins authorized the USDM to trigger aid for livestock producers with the 2003 Surplus Non-fat Dry Milk Sales for Feed Program.
“And that was the first time that anybody in a position to make high-level decisions had come to me as an author and asked if we should use the Drought Monitor (as a trigger), and I said yes,” Rippey said.
In the summer of 2006, with nearly half of the U.S. experiencing drought, attention once again turned to the USDM’s drought designations as a trigger for aid in the form of $50 million in state block grants for livestock producers. The USDM has been written into the Farm Bill since 2008 as a trigger for drought relief under the Livestock Forage Disaster Program, and after widespread drought in 2012, it became a trigger for fast-track Secretarial Disaster Designations. As of 2019, the USDM had been used to distribute approximately $7.2 billion in aid to livestock producers. The USDM helps producers receive aid faster, said Brian Fuchs, NDMC Monitoring Coordinator and USDM author since 2006.
“Back in the early days when USDA would try to have these different aid programs, a lot of times it was tied to the Palmer Drought Severity Index, and that’s a monthly tool at that,” Fuchs said. “With the Drought Monitor being this consolidation of evidence, you’re getting that signal and the information is coming through more rapidly because of all the different tools that we’re using, and you’re getting the best of all the indicators and not relying on a single indicator.”
Along with multiple datasets, USDM authors have come to rely on the team of local, state and regional experts on the Drought Monitor network listserv, where climatologists and evaluators provide updates from their locations and also respond to drafts of the map as publication dates near. They often also share news stories about experiences of drought, like a village in Alaska that recently ran out of stored water as the state grappled with persistent drought throughout 2019.
“I think if the Alaska drought that is going on now had happened 20 years ago, we might have missed it,” Rippey said. “There's no drought that's going to happen anymore without somebody knowing about it. And that’s a good thing.”
Svoboda said that as computing evolves and allows for further combination of drought indicators using deep learning, that will add to the Drought Monitor process, but not override it.
“I think we have a process called the Drought Monitor,” he said. “It also involves ownership of people on the ground, those 420 or so evaluators that are now part of the Drought Monitor network. Once they felt that they have a voice, and they have ownership, then we had the buy-in and credibility on the ground, and no single indicator or model integrated validation on the ground better than the USDM.”
Added Fuchs: “It’s this process of data and people coming together, and the end result is the map.”
-- By Cory Matteson, NDMC Communications Specialist