The first attempt by scientists to assess the role of climate change in the deadly European heat waves of 2003 has attributed about half of the 1,050 deaths in Paris and London that summer to the effects of global warming.
Tens of thousands of people, many of them elderly, died in France and other European nations in one of the most notable extreme weather events since the turn of the century. The implication of the study, published Thursday in the journal Environmental Research Letters, is that manmade climate change was largely responsible for the disaster.
"Other studies have looked at how humans have changed the frequencies of heat waves, or they've looked at how heat affects mortality," said Daniel Mitchell, a researcher with the Environmental Change Institute and lead author on the study. "But no studies looked directly at how human-induced climate change has impacted mortality in these extreme heat waves. Our study was the first to directly model that and put a number on how many deaths there were in these two cities for the 2003 heat wave, or in fact for any heat wave. "
Researchers from the University of Oxford, Public Health England, the UK's Center for Ecology and Hydrology, as well as Reading University and the Union of Concerned Scientists, merged climate data with health data to determine the likelihood that a warming climate had influenced the death toll.
The researchers began by running climate models. The first, "the natural scenario," allowed them to estimate what the climate of 2003 would have been like without any climate change due to human factors, like deforestation and industrial emissions of greenhouse gases.
The second, "the observed scenario," included all known climatic forces under which the 2003 heat wave occurred.
One novel element in the analysis was its crowd-sourcing of computer power. Researchers in this study were able to run simulations of more than 2,000 years by partnering with Weather@home, a civic science site. The site allowed the researchers to borrow computing time and run simulations while volunteers' computers were idling. Often researchers are limited to a handful of simulations.
Comparing the two scenarios, the researchers discerned how climate change had affected that summer's temperatures. In London, the likelihood of a summer as scorching as 2003 has shifted from a 1-in-10-year event to a 1-in-3-year event. In Paris, such soaring temperatures had moved from a 1-in-92-year event to a 1-in-30-year event.
This is an example of a climate attribution study, a method of showing how broad trends in the climate have influenced individual extreme weather events. Attribution is gaining increased attention among scientists and policy makers who search for climate ties as year after year of record heat brings on one catastrophic episode after another.
With the calculation of the weather odds in hand, the researchers applied a health impact assessment model. That model takes factors known to trigger heat-related deaths and correlates them with the temperatures in the simulation scenarios. In London, which experienced lower extreme heat, climate change increased the likelihood of heat-related deaths by 20 percent. In Paris, it increased the likelihood by 70 percent. In raw numbers, climate change caused 64 of the 315 deaths that London experienced during the heat wave. In Paris, where it was much warmer, 506 out of 735 deaths were attributable to global warming.
"It's what we call an indirect attribution or secondary attribution," said Michael F. Wehner, a senior staff scientist in the Computational Research Division at the Lawrence Berkeley National Laboratory, who did not contribute to the study. "Which is estimating in a probabilistic way the effect of climate change on either the likelihood of an extreme event of some given magnitude, or how much did climate change affect the magnitude of an event.
"For the 2003 heat wave, climate change increased the probability of that temperature being achieved by at least a factor of 2, and that's a fairly conservative estimate. What this paper does is take us a little farther, looking at apparent temperature, a function of temperature and humidity, and relates it to how many people died, attributable to this magnitude."
The researchers focused on London and Paris, in part because of available health data, but the heat wave of 2003 was not just limited to those two cities. The heat spread across the Europe. Portugal, the Netherlands, Spain, Italy, Germany, Switzerland along with the France and the UK all suffered ill effects. More than 70,000 people died across Europe.
Locations such as Auxerre, France, where summer temperatures typically peak at 66 degrees Fahrenheit, recorded temperatures of 104 degrees. In France, so many people died, roughly 15,000, that there was a shortage of space in mortuaries. Municipalities converted refrigerator trucks into temporary morgues. It's likely that climate change impacted them in ways similar to Paris and London.
Both Wehner and Mitchell said differentiating between natural weather events and those influenced by climate change is critical to adapting to the changing climate. Many Parisians, for example, headed to the countryside or the seashore for traditional vacations in August when the heat wave was at its peak. They unwittingly left behind elderly relatives who were particularly susceptible to the heat effects. Many people had never faced temperatures this high and didn't know how to change their behaviors to cope. The structure of the city also conspired against its inhabitants.
"The block formation of American cities has the benefit of allowing air to move around, cooling things down, Mitchell said, "Whereas all the European cities are built around stopping the other European countries invading us. In Europe our cities have buildings in the way so the air can't transfer anywhere; the air is a lot more stagnant."
Most Parisian homes also lacked air conditioning. Fortunately, Europe learned from the experience. The study authors noted that a heatwave in 2012 was almost as hot as 2003, but not as deadly because of better emergency plans.
Yet, there's a limit to how much people and cities can adapt. "We will get used to temperatures of 2003," said Wehner, "the problem is that an extreme event of similar rarity will be so much larger that it will be even more deadly."
Under business-as-usual carbon emissions scenarios, one-day heat waves that now happen once every 20 years will happen every year. The difference between 100 degrees and 115 degrees is extremely dangerous for agricultural workers, highway workers, even vacationers who spend a long time outside.
"Until recently, whenever we talked about climate change we talked about the globally averaged increase in temperature of 1 degrees and people just don't really know or frankly care about that," said Mitchell. "But now we're telling you that 500-odd people died in Paris in this one event because of climate change. People can really start to understand that these are impacts we're seeing now, not in the future."
"There are a lot of people who lost family in the 2003 event, mostly grandparents," said Wehner. "I think that it would be a pretty awful feeling to know that the odds of your granny dying during that heat wave were increased because of climate change. That would make it difficult for many people to sleep of at night. And that's what this paper does."