An ESSIC scientist is creating a product to predict heat waves weeks in advance

By Teri West

ESSIC assistant research scientist Augustin Vintzileos has developed a new experimental product that provides global excessive heat outlooks two and three weeks in advance.

Although still a pilot program, the outlook system could eventually be a vital early warning tool for relief agencies who respond to critical weather emergencies and disasters. For example, relief agencies could begin planning for extended operation of cooling centers and the reinforcement of drinking water supplies well prior to the onset of a projected heatwave.

“Heat kills a lot of people,” Vintzileos said.

Elderly populations can be especially vulnerable to excessive heat, he said.

One of the objectives for scientists who want to track excessive heat has been defining it. There have been more than 100 definitions posited, Vintzileos said, and adopting a common one remains a challenge.

Excessive heat is more than just a hot temperature; definitions include factors such as the length of the heat wave, the climate preceding that period and the humidity.

“105 Fahrenheit for one day is bearable, but if it’s three or four days in a row it’s not,” Vintzileos said.

Vintzileos’ product determines the mean daily temperature for three consecutive days then compares that temperature with the 95th percentile of conditions during the 30 previous days.  From there, it determines the probability of the warmest day within the second and third week exceeding the 50th and 80th percentiles. It also considers acclimatization of the human body to the projected conditions by again contrasting the intensity of a forecasted heat event to the current mean. 

“If it was very warm [consistently] then your body’s acclimated,” Vintzileos said. “If it was colder than normal and then boom you have a heat wave, people suffer more.”

The product also functions on a global scale and provides the ability to examine specific regions considered to be at risk for heat waves.

One way that Vintzileos has checked the validity of the system is by searching the news. Excessive heat should be an abnormality and therefore newsworthy, so if the system predicts a heat wave that the local news then reports on, Vintzileos knows he is on target. For predictions from Warsaw, Poland to Rajasthan, India, a quick search on Google News has been able to confirm his findings.

Vintzileos presented the excessive heat model in numerous venues including a recent visit to the Tunisian Department of Agriculture. He prepares an experimental real time forecast bulletin that is distributed to climate and health scientists and decision makers bi-weekly.

Future work includes the extension of the system to identify marine heatwaves.