Figure 2. Area-weighted temperatures from the ground based (solid circle) and satellite (open circle) records.

4. The most likely explanation for the failure of the models is in their forecast of temperature in the zone from 5,000 to 50,000 feet.

All transient climate models, including those modified with sulfate aerosol, predict a rather smooth and consistent warming of the entire troposphere, or bottom 50,000 feet of the atmosphere. It is apparent that the atmosphere has cooled above 30,000 feet and that there is no net change in the last 20 years in the region between 5,000 and 30,000 feet. There was a sudden jump in this layer temperature in 1977; no change was noted prior to then, either.

The implications of the behavior in these layers is profound. Everything else being equal, the cooling from 30,000 to 50,000 feet increases the vertical transport of moisture in the atmosphere and could be responsible for observed increases in large-area cloudiness. In turn, the increases in cloudiness could explain the overall lack of warming and its distribution primarily into the winter and night; in the region of the planet that has shown the most warming--Siberia--the ratio of winter to summer (which is, essentially, night (or low sun)-to-day) is approximately 4.2 to 1. The primary climatic effect would be two slightly lessen the severity of the coldest airmasses in the Northern Hemisphere.

Several scientists, including Dr. Lindzen on this Panel, have hypothesized that model calculations of the vertical distribution of heat in the atmosphere are likely to be wrong or at least unreliable; the failure of the models to even simulate the sign of observed changes in the last two decades bears strong witness to this hypothesis.

Data: While there are many hypotheses concerning the inaccuracy of the prediction of vertical temperature change, data required to resolve this will come from atmospheric sounding devices such as weather balloons and satellite profilers. These should therefore by high priorities for support.

5. LACK of data can cause embarrassing errors.

In early January we were treated to a number of stories stating that ground-based thermometers indicated that 1995 was the warmest year in the instrumental record. In fact, the temperatures used to make this calculation only included data through November, 1995, and an assumption that December temperature departures from normal would be the same as they were for the rest of the year. This temperature history was complied by Phil Jones of the University of East Anglia.

As noted in the satellite data (Figure 1), December temperature departures from the mean in fact reflected the largest single one-month drop in the entire record for the Northern Hemisphere, declining by 0.72° from November.

These types of errors could be expected to propagate throughout the climate measurement system if surface data collection and analysis is allowed to degrade.

6. Increases in strong summer rainfall are not explainable by climate models, and are more likely to have been beneficial rather than "dangerous".

In his March 17, 1995 "Earth Day " address at George Washington University, the VicePresident noted that, with regard to global climate change, "torrential rains have increased in the summer during agricultural growing regions".

The source for his statement was a then-unpublished manuscript by Thomas Karl et al. that showed a statistically significant, but very small increase in the percent of United States rainfall that occurred from rainstorms that produced at least two inches of rain in 24 hours. The equivalent change is that, on the average, there is now one more day in every 730 days in which it rains more than two inches.

Examination of Karl's own graphics indicates that the largest increase in this parameter took place between 1930 and 1950, or prior to 70% of the human-induced changes in the greenhouse effect. Ascribing this change to the greenhouse effect is therefore dubious at best.

Perhaps as important, the vast majority (70%, nationally) of summer 24-hour rainfall that is greater than two inches is less than three inches. It is very hard to entertain the notion that a 2-3 inch rain in summer is “torrential”, and, in toto hardly a “dangerous” change in climate. At that time of the year, all major agricultural regions of the U.S. are, on the average, losing more moisture to evaporation than they gain from rainfall, and this rain is much more likely to be welcomed by farmers than feared.

Data: The data for this study came from the cooperative-observer network of raingauges, maintained by the National Climatic Data Center. Maintenance of this high-quality record is a very high priority. Recent attempts to augment this history with satellite measurements have met with limited success, but it is nonetheless important to develop a system that reliably provides global monitoring of rainfall, especially over the ocean and in unpopulated regions. This should also be a very high priority.

7. Model calculations have made large errors in projections of jet stream positions.

These areas have been discussed in the testimony of Robert Davis.

Data: The data required to examine these aspects of model performance include weather balloon soundings and ground-based rainfall measurements. As noted above, these can and should be augmented with an improved remotely sensed rainfall network as well as satellite-based vertical profilers of atmospheric temperature, pressure, and moisture.

Figure 3. Plot of the increasing percentage of rain falling from storms of two inches or more. The large increase was before the greenhouse effect had changed very much.

Figure 4. Percent of rain from storms of greater than two inches falls from storms that are more than three inches. Approximately 70% of two inch storms produce less than three inches, making them hardly "torrential" [see text].

26-244 0-96-9

Thoughts on funding

It is clear that we live in an era of declining federal financial support of many aspects of the nation's life. One cannot expect that federal research budgets will be immune to this trend.

In the area of global climate change, there is little doubt that several public servants feel very strongly that research indicates the "dangerous" view of climate change is more likely than the newer, "moderate" synthesis. Thus the federal research presence in this area will always raise suspicions--real and imagined--that the funding process has become politicized.

One can change neither this perception nor the ambitions of those who champion any point of view; rather, the best tactic may be to admit to the reality of the problem and purposefully attempt to broaden the research base to explicitly include as providers some of the communities that are especially concerned with the issue of climate change.

Might it not be appropriate, in this era of declining funding, for interested parties other than government to begin to assume some of the research burden? I am referring specifically to two groups with considerable resources: industry and the environmental community. Perhaps you can develop a mechanism where both of these groups explicitly demonstrate financial support for research on climate change, and then the federal outlay is reduced an equivalent amount. I do not know much of these matters, but I suspect there are some incentives that can aid in this process.

This proposal would have the effect of maintaining financial support for research on climate change while broadening the base of support. There is no known constitutional fiat that I know of that requires that the federal government be the sole provider of funding for climate change research, even though that is virtually the case today. And further, there is demonstrable evidence that removal of the “monopoly” provider status of the federal government may in fact result in a more diverse, and therefore healthy, research culture on this important issue.

REFERENCES

Intergovernmental Panel on Climate Change (IPCC) 1990. U. N. Environment Programme. 200pp.

1992. Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment. pp

Karl, T.R., et al. 1995. Nature 337, 217-220.

Michaels, P. J. and D. E. Stooksbury, 1992. Bull. Amer. Met. Soc.

Michaels, P. J. et al., 1994. Technology: J. Franklin Inst. 331A, 123-133.

Mitchell, J.F.B. et al, 1995. J. Climate 8, 2364-2385.

Wigley, T.M.L., 1987. Climate Monitor 16, 14-28.