Raw data, please

I started reading Marc S. Reisch's article on coolants (C&EN, Oct. 3, 2005, page 23) with great interest. However, the deeper I got into the article, the higher my frustration level rose. There were a lot of forceful conclusions in there, but very little data were presented to support them.

For example, it would have been useful to present the data that show how "the Montreal protocol has led to some recovery of the ozone layer." Also, data that show the relationship between the reduction in CFCs/HCFCs released to the environment and the corresponding changes in the ozone layer would be important for readers to get a feel for what the data are really saying. Is the corresponding change measured in the ozone layer significant?

On global warming, Reisch indicates that in 1990 the amount of CFCs/HCFCs released in the atmosphere had the equivalent global warming potential of one-third that of the CO₂ released that year. By 2000, that relationship had changed to one-tenth. Now, to make sense out of this information, one would need to plot the change in CFCs/HCFCs versus the change in CO₂ over that period of time on the same plot as the corresponding global warming levels measured. From such a viewing of the data, one can get a feel for relative contributions of these two sets of compounds to the observed global warming and derive a more objective conclusion that is supported by the data.

From the content of this article, I find it hard to conceptualize how Reisch's sources came to the conclusion that a lowering of CFCs/HCFCs between 1990 and 2000, at the same time that CO₂ levels were rising, resulted in a significant decrease in global warming due to the observed decrease in CFCs/HCFCs. If this is so, there should be a decrease in the slope of the global warming curve over this same period of time. My suspicion is that such a plot may not show this, and therefore models would be used to extract the stated conclusion from the baseline noise.

My main point here is that most of the people who read articles such as these have had some degree of scientific training and know how to interpret raw data when it is presented to them. In fact, some of us do not feel comfortable with declared conclusions unless we can work the supporting raw data in a way that makes it easier to understand what the data are saying.

Allen A. Aradi
Richmond, Va.