in conveying the results of their studies to decision- and policy-makers. The use of data with a high degree of uncertainty, or the results of models based on questionable assumptions, places the credibility of our science at risk.
RIND: I find myself in the strange position of having to defend Steve Schneider. I'm not sure it's reasonable to compare the glacial-to-interglacial warming over 5000 years with the projected warming over the next 100 years or so, when the time spans are so different.
REIFSNYDER: Oh, I admit that's comparing apples with oranges. If you extrapolate the IPCC values for 1000 years they become absurd. You do have to be careful how you choose the periods for comparison. The longer the time period, the slower the rate of change.
MOREL: Allow me in turn to defend the IPCC report. For a U.N.-sponsored six-month exercise, I would say it's remarkably respect-worthy. We were trying to give the reader a reasonable idea of what could happen, working from uncertainties quite as large as one finds in nature. After all, we can't even link some substantial ecosystem changes, such as the desertification in Africa, with any changes in global climate indicators.
REIFSNYDER: I quite agree. However, that report has been used by non-scientific people as the basis for so many outrageous suggestions that I think it's important to call attention to its inconsistencies and limitations.
LEHMAN: I hesitate to add fuel to the fire, but we did find that during the deglaciation sea-surface temperatures in the North Atlantic changed by as much as 5°C in less than 40 years, in both negative and positive directions. Ice cores show warming of 7° in 50 years. These regional modulations probably relate to some fairly significant global changes.
TRENBERTH: It seems to me that the models essentially represent the anthropogenic change. Why should they be relevant to natural climate change? Conversely, how can one deduce the rate at which man can change the climate from that of natural change?
REIFSNYDER: My point was more a matter of what plants have been exposed to, whatever the origin of the change. The slow secular temperature increase ascribable to increased carbon dioxide in the atmosphere is several orders of magnitude less than that of the climate to which a plant is exposed in the course of its life.
JONES: I think we should keep in mind that most of the points to the right of the 100-year band on your Figure 1 are regional. A 10°C change in Vostok, or even the entire Antarctic region, probably means a global change on the order of 4° or 5°. Also, I'd like to emphasize that small changes can make a significant difference. In the 1920s and 1930s there were some very warm summers in northern Scandinavia, meaning 1° or 2°C warmer on average. As a result, the tree line moved many kilometers further north in a short space of time.
REIFSNYDER: That's not what's at issue. My question is how much an increase due to global warming will affect such natural changes.
GROISMAN: I'd just like to add another reminder that the rate of mean temperature changes at high latitudes is not the same as the global rate, either in the last hundred years' observations or in the model results, whatever the IPCC report may say.
KEELING: I think this is just the kind of healthy challenge the IPCC report needs to counter the word-bite publicity it's been getting. I'd like to make a few points here. First, I don't see how we can be sure that the reason for the temperature increase you cited is global warming. Even worse, it may be that there are several anthropogenic signals whose effects are partially canceling each other out; Mike Schlesinger brought that possibility up at a recent meeting.
Second, those fluxes bother me. The fluxes involved in the CO2 increase over the last 100 years are dwarfed by the natural fluxes in the carbon cycle, and I don't think they have anything to do with the accumulation of CO2 in the atmosphere. Third, I don't think we need to worry so much about what level of temperature we reach as about how long the increased temperature will persist. Some trees can handle a passing fire; that doesn't mean they could survive 5 years of heat and drought, followed by another 5. That sort of possible anthropogenic effect is what we need to know about.
REIFSNYDER: Indeed. But in response to your first point, let me say that we really have no better hypothesis then a CO2 increase for what caused the warming over the last 100 years. My position is that it's our best estimate.
GROOTES: A propose of climate's societal impacts, let us not forget the Norse colonies in Greenland. They lived there happily for 500 years before being driven out by the effects of the Little Ice Age. But I have a question for you, Bill. I understood you to say that the concern about global warming should and probably would disappear, and we should concentrate our efforts elsewhere. What are your suggestions?
REIFSNYDER: Oh, there's a great deal more we need to know about anthropogenic global warming. I just think it will die down as a social-policy issue, since we can't yet relate it to the warming during this century. As for Greenland. remember that the Little