Robert Watts

Tulane University, Department of Mechanical Engineering

Have we really already detected a global warming signal in the temperature record? Is there a deep ocean surprise in the future? Watts defined climatic surprise as a relatively dramatic and unexpected change, perhaps rapid, at least in geological terms. He then offered what he called plausible and interesting explanations for three past climatic surprises which occurred on short geologic time scales:

1. Roughly 100 million years ago in the Cretaceous period it was warmer and the temperature difference between the poles and the equator was smaller. Antarctica was a polar desert, with no ice cap, and then it cooled quickly and the ice cap formed, some 60 to 80 million years ago. That was a surprise.

2. From 2 million years ago to about 1 million years ago, as ice sheets were formed periodically on the North American continent, an abrupt change brought glacial cycles with periods of 30,000 to 40,000 years, and another abrupt change brought glacial cycles with longer periods of 100,000 to 120,000 years.

3. Ice core data show that at the end of the last glacial cycle, at the start of the present warm period, extreme events, such as the Younger-Dryas cooling in midst of warming, could have occurred on time scales as short as 20 years.

The ocean contains a mixed layer 50 to 100 meters deep. The more ocean and less land there is, the smaller the variability from summer to winter (land has only a small heat capacity compared to ocean).

Watts posed one plausible explanation of why the seasonality, the difference between peak summer and winter temperatures, is larger for larger continents than for smaller continents. When Australia was joined to Antarctica, 100 million years ago, the polar land mass was so large that mid-summer temperatures were higher than freezing, preventing the formation and growth of a permanent ice sheet. When Australia "drifted" away, the polar continent became small enough that mid-summer temperatures remained below freezing so that an ice cap could form.

A similar event might explain the transition in North America from small 30,000 to 40,000 year glacial advances to large advances with 100,000 to 200,000 year periods. As long as the northernmost points on the continent are far enough from the poles, the continental ice sheet will grow and recede in response to variations in the Earthıs orbital cycles of the obliquity and the latitude of the perihelion, which affect the distribution and seasonality of the received solar radiation. The summers are warm enough for all of the glacial ice to melt during high latitude warm periods. As the continent "drifts" closer to the pole, some ice remains at the end of the polar warm periods, and glaciers continue to grow until some other mechanism, yet to be explained, causes them to retreat after about 100,000 years.

The Younger Dryas cooling event may have been caused by a temporary change in the formation of North Atlantic deep water. Upwelling of deep water in the mid-latitudes traps heat in the upper part of the ocean. If the rate of deep water formation (and consequently the upwelling speed) decreases, heat can diffuse downward, cooling the surface and warming the water at intermediate depths.

Watts modeled temperature change due to a 50% reduction in Atlantic thermohaline strength after 5, 10, 15, 20, and 25 years. A series of isotherm plots shows a rapid decrease in temperature in the North Atlantic (3° in 5 years). A maximum decrease is reached in about ten years (very rapid), then recovers very slowly.

What might be the effect of variations in the thermohaline circulation on recent climate? A variety of global surface temperature data sets shows an apparent 0.5° C warming over the past 100 years. However, the Northern Hemisphere actually experienced a cooling trend from the 1940s to the 1970s. Moreover, singular spectrum analysis of these data, as well as longer, local data sets, identifies consistent periodic temperature variations with periods ranging from a decade to more than a century. Watts speculates that these variations may be responsible for the absence of a surface warming from the 1940s to the 1970s. A decrease in upwelling in the North Atlantic ocean of just 10% could easily lead to a cooling of the upper ocean large enough to temporarily cancel the effects of greenhouse warming. The recent detection of a warming of water at intermediate depths in the North Atlantic lends credence to this hypothesis. Watts emphasizes that detection of global warming is a three dimensional problem.