More from Tim Ball

Tim Ball writes:

One of the sequences I followed with increasing alarm through my career was the intrusion of computer modelers into climate science. I had worked with mathematicians and statisticians such as Alex Basilevsky on using long term climate records as a basis for math and stats theory. For example, Basilevsky was studying Markov chains that consider probability of future events based on historic data - his concern wasn't weather forecasting, but the implications are there. The complexity of climate offered a great challenge for mathematicians and computer modelers. Andrew Weaver is a mathematician who used ocean /atmosphere modeling for his doctoral thesis and now lists himself as a climatologist. In my first interview with him I realized he actually knew very little about climate even though he was a member of the IPCC. He now edits a climate journal. It reached a point when modelers were the keynote speakers at most conferences and then dominated the conference. They did argue among themselves, but it was more about the size and speed of their computers than about the efficacy of what they were doing. For example, it gave one group, ( I believe it was GISS) an advantage when they got Cray computers.

My first experience with Schlesinger was a conference in Edmonton in 1987 titled "The Impact of Climate Variability and Change on the Canadian Prairies." (Proceeding were published by the Canadian Climate Centre.) The basic issue was government planning based on future climate conditions. For example, in one discussion I recall the Alberta Deputy Minister of Resources (Most senior bureaucrat in a department) asking about forestry. His question was, "Your models tell us that southern Alberta will be a desert in 50 years. How sure are you of that prediction, because we are planning to plant trees and you are telling us they won't survive and we can't move them?" It was the first time I saw the models confronted with the realities of policy. As I recall Schlesinger thought for a while and then said "About 50% sure." To which the Deputy replied, "My Minister wants 98%."

Schlesinger's paper took the same data and put it into five of the major models of that time, including, 1. the GDFL of Weatherald and Manabe, 2. the GISS of Hansen et al, 3. the NCAR of Washington and Meehl, 4. the OSU of Schlesinger and Zhoa and, 5. the UKMO of Wilson and Mitchell.

A major thrust of the debate was because Schlesinger had put the same data into the five models and each produced results that he claimed were meaningful. Somebody pointed out that the results differed considerably from model to model. For example, they differed by 180° in their predictions for large areas. Schlesinger's reply was the models were not accurate for small regions. The person pointed out that the small regions were continental in their dimension. Much laughter at this point. Schlesinger then said the models were not quantitatively correct but they were qualitatively correct. When asked to explain what he meant he said well they all showed global warming with increased CO2. It was quickly pointed out that if you program them to have temperature increase with a CO2 increase (ceteris paribus) then that was an inevitable result of the programming not the reality. The noise volume increased at which point a bizarre incident occurred.

During Schlesinger's presentation, titled, "Model projections of the Equilibrium and Transient Climatic Changes induced by Increased Atmospheric CO2" there were general rumblings in the audience about the nature and assertiveness of the presentation, something that I had come to know as normal for modelers. This erupted in the question period. However, prior to that there were strange noises coming from behind me. I did not want to look around based on my experience at english soccer matches. In the question period voices were raised and frustrations expressed about the inadequacy of the models. Suddenly at the height of the din a shoe flew upon to the platform from behind me. There was shocked silence and a strange voice said, "I didn't have a towel." He then asked permission to go onto the platform. It turned out the strange noises were from the shoe thrower who had a voice box. He explained he had two Ph.Ds one in Atmospheric Physics and proceeded to put a formula on the blackboard. Schlesinger agreed it was the formula for the atmosphere at the basis of his models. The man then eliminated variables one at a time, each time having Schlesinger agree he eliminated them from the final model. The man then said what you have left no longer represents the atmosphere and any results from such as model were meaningless.

It was my first public experience with the concern I have about the boundary and difference between scientific responsibility in the laboratory, which unfortunately too many scientists are not meeting, and the completely different social and economic responsibilities when you push your ideas as viable to the public.