Record heat for the United States

Could the Imminent U.S. Heat Wave Trigger a Flash Drought?

By: Bob Henson

Wunderground,

19 July, 2016

A massive upper-level high will envelop most of the contiguous U.S. in the last half of July, setting up what could be a prolonged bout of extreme heat for millions of Americans. If the scorching weather persists into August, the odds of a “flash drought” in the nation’s heartland will rise sharply (along with the odds that the U.S. will notch its hottest summer on record, in line with what’s very likely to be Earth’s warmest year on record). Even though it appears that heat and humidity will combine to put residents, pets, and livestock through the wringer, it’s quite possible that the croplands of the Midwest and Plains will fare better than one might expect, thanks to a fortunate confluence of factors.

There’s certainly no mercy in the pattern projected for the next few days at upper levels. As shown in Figure 1, the upper high at 500 millibars (about four miles up) encompasses nearly all of the contiguous U.S. by Thursday night, July 21. Because air expands as it warms, a higher 500-millibar surface is associated with a warmer air mass at lower levels. Although the most extreme 500-mb heights projected in earlier model runs have dropped just a bit, both the GFS and ECMWF models have been consistent with their portrayal of a mammoth upper ridge centered near the nation’s midsection for at least the next week to ten days, perhaps longer. The atmospheric variables predicted to take shape later this week are similar to those observed during some of the nation’s most notorious heat waves of recent decades, according to the CIPS Analog website from the Cooperative Institute for Precipitation Systems/Saint Louis University. At the time period shown below in Figure 1 (8 PM Thursday, July 21),the closest analog for conditions predicted for the Midwest (taking into account temperatures, moisture, and other factors at various layers) is July 13, 1995--the second day of a catastrophic five-day heat wave that took more than 700 lives in the Chicago area. The top analogs also include July 4, 2012, and August 7, 1980, two peak days from the devastating central U.S. drought years of 2012 and 1980. Excessive heat watches for later this week have already been issued by the National Weather Service for parts of Illinois, Iowa, Kansas, Minnesota, Nebraska, North Dakota, and South Dakota. We can expect many other areas to follow suit as the week unfolds.

Figure 1. 500-millibar heights (in decameters, or tens of meters) predicted for 8:00 PM Thursday, July 21, 2016, by the 12Z Monday GEFS (the ensemble run of the GFS model). The colors show how much the predicted 500-mb height deviates from the average for this time of year. Image credit: tropicaltidbits.com.

Blazing temps and wilting humidity: a risky combo

With such strong model support for a high-end upper ridge, one might expect surface temperatures to be correspondingly extreme. Indeed, readings near or above the century mark (100°F) are likely to encompass large parts of the Great Plains by midweek--perhaps even topping 110°F in some spots--with 90°F to 100°F readings over a far larger part of the nation. As an upper-level impulse rides along the north side of the ridge later this week, a burst of heat and humidity will be shunted eastward, approaching the East Coast toward the weekend. In a post on Monday, Capital Weather Gang observed that both the GFS and ECMWF models are predicting highs in the vicinity of 105°F in the Washington, D.C., area for Saturday, perhaps approaching the city’s all-time high of 106°F (set on August 6, 1918 and July 20, 1930). It’s important to keep in mind that long-range models often struggle in nailing down the exact location and strength of high temperatures this extreme, especially beyond two or three days. The key message here isn’t the precise forecast for Saturday, but the overall signals pointing toward an intense and possibly historic round of heat affecting most of the central and eastern U.S.

Another important element of this heat wave: some of the energy that would otherwise go into heating up the lower atmosphere will be diverted into evaporating moisture. Plants and soils are quite moist in many areas thanks to recent rains, especially through a belt from the Central Plains into the Ohio Valley. That moisture is a mixed blessing: while it’ll help to keep surface air temperatures a notch lower than they’d otherwise be, it will also help pump up the amount of water vapor in the air. In addition, the Midwest’s vast corn crops are at a stage where they add moisture to the air through evapotranspiration, a process dubbed “corn sweat”. As a result, heat index values will soar to uncomfortable and even dangerous levels as this week progresses over large parts of the central and eastern U.S., especially toward the south. The atmospheric moisture will also help boost nighttime lows, which exacerbates the potential risk to human and animal health from a multi-day heat wave. (Climate Central recently documented the long-term rise in summertime atmospheric moisture across the U.S. as measured by surface dew points.)

Figure 2. In this 4-day outlook from Monday, July 18, 2016, daily heat indices on Friday, July 22, are projected to exceed 105°F (magenta) over large parts of the central and eastern U.S. The highest indices will shift toward the Southeast by early next week. Image credit: NOAA/NWS Weather Prediction Center.
Watching for flash drought

It wasn’t too long ago--in 2012--that a promising-looking spring morphed into a terrible summer for the U.S. Midwest. A long-term drought that began in late 2010 had intensified over the Southern Plains in 2011, punishing farmers and ranchers and facilitating the loss of roughly 10% of all trees in Texas. The real shocker was how quickly drought conditions took hold further north across the Midwest in the summer of 2012, leading to the most widespread U.S. drought conditions since the 1930s. “Nobody called that [in advance],” said Mark Svoboda (National Drought Mitigation Center). Even NOAA’s 30-day and seasonal drought outlooks from June 2012 failed to predict that month’s emergence of drought in the Midwest, according to Svoboda. It appears this wasn’t a simple northward extension of the ongoing drought further south, but something else--a classic case of what’s increasingly known as a flash drought, a rapid-onset drying of the landscape (there is not yet a standard definition). Svoboda first brought the term to a general audience when he used it in 2000 in a USA TODAY interview. but it was the 2012 Midwest event that gave the flash-drought concept much more prominence.

While long-term drought can emerge simply through a lack of precipitation, a flash drought is closely linked to hot summer weather. The type of flash drought most often observed in the Midwest develops as a torrid air mass sweeps in for a period of a few days to several weeks. At first, the landscape may not be particularly dry, in which case large amounts of water vapor flow from vegetation and soils into the scorching surface air (as is expected later this week). If the heat is strong and sustained enough, the landscape quickly dries out and a flash drought takes hold. 

Figure 3. Frequency of occurrence of five-day periods of two types of flash drought--heat wave flash drought (top) and precipitation-deficit flash drought (bottom)--in an ensemble of four models employed to replicate land-surface conditions for the climate from 1916 to 2012. Image credit: Figs. 4a and 4b from Kingtse C. Mo and Dennis P. Lettenmaier, “Precipitation Deficit Flash Droughts over the United States,” Journal of Hydrometeorology 2016 17:4, 1169-1184, ©American Meteorological Society.  Used with permission.

In a paper published this spring in the Journal of Hydrometeorology, Kingtse Mo (NOAA Climate Prediction Center) and Dennis Lettenmaier (University of California, Los Angeles) label the above sequence of events a “heat wave flash drought.” The scientists also identified another, more common flavor--what they call a “precipitation deficit flash drought”--that’s most frequent in the Southern Plains. In this type of flash drought, a lack of precipitation allows the local landscape to heat up quickly under the strong summer sun, which leads to additional drying and further heating. Both types of flash drought are dangerous, the scientists warn: “Even though flash droughts tend not to persist, they initiate the depletion of soil moisture, and the persisting deficits can cause large damages to the agriculture community.”

Better late than sooner

If there’s any potential for a heat wave flash drought over the next few days, at least the timing is on our side. The heat is arriving several weeks later than the worst conditions in 2012, which means that major crops are further along and better able to handle the heat. In addition, the nation’s corn and soybean crops have been maturing more quickly than usual this year (see Figure 4 below). According to USDA meteorologist Brad Rippey, this is largely due to the unusual warmth of June (the warmest on record for the 48 contiguous states) together with early planting in some areas. “Corn typically has a very short period of time--ten days or so--when it is acutely sensitive to air temperature and moisture availability,” Rippey told me. Temperatures above 95°F and/or a lack of moisture are detrimental to the pollination process. However, Rippey said, “by the time the heat wave hits the western and southern Corn Belt this week, a decent amount of the corn will have passed through its most sensitive stage.” Further to the northeast, in Wisconsin, Indiana, Ohio, and Michigan, corn planting was delayed due to the cool, wet May. However, temperatures there should stay below critical thresholds during the coming week.

One concern is the potential impact of very warm nights in the 70s and 80s, which can lower corn yields by depleting sugars produced by daytime photosynthesis. “Today’s drought-tolerant varieties are really quite good at overcoming weather challenges, although nighttime heat seems to be a particular problem,” Rippey said. He adds that soybeans--another hugely important U.S. crop–are more flexible in dealing with flash droughts, since they have a wider reproductive window and can slow or halt their growth processes as needed to conserve energy and moisture.

Figure 4. The nation’s corn and soybean crops have grown more quickly than usual in 2016. Left: percent of corn crops silking in each state by July 17 (top numbers), together with the percentage difference (bottom numbers) from the five-year average for this date. Right: percent of soybean crops blooming by state. Image credit: USDA, courtesy Brad Rippey.
Two new tools for tracking flash drought

A burst of research has led to the emergence of new monitoring efforts that may help identify and even predict flash droughts. The National Drought Mitigation Center has been leading a multiyear effort to develop the Quick Drought Response Index (QuickDRI), which will monitor changes in vegetation over periods of a week or two. QuickDRI is building on VegDRI, an operational product that uses satellite and climate data to map vegetation change at the seasonal scale. Funded by NASA, QuickDRI is being tested this year and may become “pseudo-operational” as soon as 2017, according to NDMC’s Mark Svoboda.

Meanwhile, a group led by Michael Hobbins (NOAA/ESRL Physical Science Division) and Dan McEvoy and Justin Huntington (Desert Research Institute) has developed the Evaporative Demand Drought Index (EDDI), which focuses on the weather that drives both flash drought and long-term drought. Rather than assessing the landscape itself, or recent rainfall, EDDI looks solely at evaporative demand--the impact of atmospheric temperature, humidity, wind, and solar radiation over a particular time period--and how it compares to climatology. A positive EDDI indicates drier-than-average conditions. In evaluations thus far, EDDI appears to work well in providing advance notice of drought development, often ahead of other commonly used indexes. The EDDI is spotlighted in the June issue of the Journal of Hydrometeorology, where a pair of papers explains the rationale for the index and a U.S.-based evaluation of its skill. The EDDI will be moving toward operational use at NOAA’s National Water Center over the next two to three years, according to Hobbins. “The three sectors or drought types for which we think EDDI holds out the greatest hope for warning capabilities are flash drought, sustained long-term agricultural and hydrologic drought, and wildfire risk,” he added. In the meantime, regularly updated EDDI maps for various time frames are available for download. The most recent 4-week EDDI map suggests that the eastern Great Lakes, New York, and western New England have experienced drought-favorable weather over the last month. Patches of moderate drought in these areas are now showing up in the weekly U.S. Drought Monitor.

We’ll be back with our next post by Tuesday afternoon.

Bob Henson

Figure 5. EDDI values for the four weeks ending on July 13, 2016. EDDI categories are similar to those used in the U.S. Drought Monitor, although EDDI is designed to monitor the weather that leads to drought rather than to diagnose existing drought. EDDI does not incorporate precipitation or soil conditions--only atmospheric and solar variables. Image credit: Courtesy Michael Hobbins, NOAA/ESRL/PSD.