The great heat wave of 1936 produced the largest seasonal swing in North American temperatures ever recorded: 181 degrees Fahrenheit. In the town of Steele, N.D., the U.S. Weather Bureau took a winter reading of minus 60 F (minus 51.1 Celsius) in mid-February and a summer reading of 121 F (49.4 C) in mid-July.
The heat of '36 was catastrophic. The devastating "dust bowl" windstorms moved topsoil through the air from the U.S. Midwest and from Saskatchewan to the Atlantic Ocean, literally raining dirt on Toronto and Detroit and Cleveland and Chicago. Soil temperatures at a depth of four inches reached 200 F (93 C). Many people thought the continent's breadbasket had turned permanently into desert.
Environment Canada notes in its history of extreme weather in this country that the heat wave of '36 curled railway lines, twisted steel beams, buckled sidewalks - and baked fruit right on the trees. With temperatures routinely in excess of 104 F (40 C), 5,000 people died in Canada and the U.S. - mostly, as always in heat waves, the very old and the very young. Of the 730 Canadians who died, 225 of them lived in Toronto. It was North America's worst heat wave of the 20th century. Many records set in 1936 still stand.
But people adapt to changes in climate. In the Dirty Thirties, people delivered blocks of ice to the poor, slept in basements, wore wet headbands under their hats, went to air-conditioned movies - and took it easy.
By the most recent calculations of the U.S. National Oceanic and Atmospheric Administration, the past decade was 0.6 C warmer than the 1960s decade and 0.2 C warmer than the 1990s decade. These slight increases suggest that, at this rate, it'll be a long time before people will be able to distinguish a global-warming heat wave from an ordinary heat wave. NASA's Goddard Institute for Space Studies, incidentally, cited 2005 as the hottest year since the 1800s - by a few hundredths of one Celsius degree.
All weather, in the final analysis, is local - and all adaptation is local. In Canada, the mean temperature last year was 0.8 C above the historic norm, tying 1988 as the 14th warmest year since 1948 - a retrospectively inconclusive observation. In Toronto, the average daily high in July is 27 C. But on a particular July day in a particular year, the temperature can reach 41 C, a swing of 14 C from the norm (as happened for three days in a row in 1936). On another particular July day, it can reach 38 C (as happened in 1916) or 37 C (as happened in 1932, 1937 and 1988).
And so on. July hasn't produced a record high in Toronto for 15 years (since a 36 C day in 1995). Indeed, of the 31 record-setting hot days in July, 18 occurred in 1936 or earlier. When you add up all of the record-setting highs for July, and divide by 31 days, you get an average record high of 34.7 C - remarkably close to the high in July of 1878: 35 C.
Thus, in a sense, it's no surprise that two eminent economists concluded, in a research paper published three years ago by the Massachusetts Institute of Technology, that Americans will adapt to global-warming temperature increases with relative ease; extrapolating, Canadians can expect to do so with greater ease.
University of California economist Olivier Deschênes and MIT environmental economist Michael Greenstone analyzed heat-wave mortality rates from the 20th century to assess global-warming risks to health in the 21st century. They determined that the impact on people's health in the U.S. would be marginal - "and a zero impact," they said, "cannot be rejected."
Heat that kills people is obviously not the only negative consequence of global warming. Floods, droughts and other forms of mayhem kill people, too. To paraphrase the TV drug ads, results will vary. Yet, you might think, from our summertime angst, that hot North American days are getting more lethal. The fact is, they aren't. Profs. Deschênes and Greenstone note that heat waves do kill people - but that they don't increase mortality rates. People who die in heat waves are almost always people who will otherwise die, within days or weeks, from other causes.
In their study, the economists assumed that nothing (such as carbon taxes) would be done to reduce carbon emissions through 2100. On this worst-case assumption, they concluded, climate change will lead to an increase in the U.S. mortality rate "statistically undistinguishable from zero." Whatever the precise increase in temperature - perhaps a Celsius degree or two - people will adapt. They will rely more, in the next 90 years, on air conditioning (which itself will become relatively less expensive). But then that's what people always do when it gets hot. They adapt.