With southerly flow, no additional precipitation was detected when the seeded target areas were compared with the control areas. With westerly flow, the cold clouds, that is, those clouds with a temperature less than 23° F., gave 32 percent more precipitation, and the chance of obtaining such an increase purely as a result of statistical fluctuations is about one in a hundred. On the other hand, the westerly warm clouds, when they were seeded, gave 23 percent less precipitation.

I next turn to the seeding of cumulus clouds. The University of Chicago Whitetop project has studied the precipitation from showerproducing summertime cumulus in southeast Missouri over a period of 5 years. Preliminary analysis of the experiments which were properly randomized in the statistical sense showed an increase of 5 to 10 percent of the precipitation in the seeded clouds.

At the Berkeley conference a new and important study from Australia was reported. In this study, individual cumulus clouds have been seeded. The results are striking. Sixty clouds in all have been investigated. Of these, 23 involved clouds with temperatures less than 14° F at the cloud tops. In these, the precipitation was seven times greater in the seeded cases than in the unseeded. The warm clouds with a temperature greater than 14° F showed less precipitation in the seeded cases. These results have led the Australian workers to reinvestigate earlier long-term projects for which cloud-top temperatures were available.

Cold clouds show positive anomalies ranging from 10 to 40 percent as a result of seeding while warm clouds show decreases in precipitation ranging from 4 to 25 percent. These last results illustrate the complexity of the physical processes within a cloud. Furthermore, at the present time there is no known physical reason which would explain why seeding is more effective in cold clouds.

There have been fewer experiments on extratropical cyclonic storm systems. The most complete study has been carried out in Israel, where randomized silver iodide seeding of winter cloud systems conducted over 412 years shows 15 percent more rain on seeded days than might have been expected.

During the course of its investigation, the National Academy of Sciences Panel obtained over a hundred reports of commercial rainmaking operations. The Panel then selected certain of these reports for an independent evaluation. Fourteen of the selected reports were concerned with projects in the Eastern and Middle Atlantic States and covered a wide range of meteorological conditions. The projects ran from 19 days to 5 months and in all but one of the 14 cases there was an increase in precipitation, with an average increase of about 15 to 20 percent. This result was so striking that the Panel asked for and obtained additional studies by the Rand Corp. and the U.S. Weather Bureau. These independent groups substantiated the Panel's results.

The Panel also studied four long-term projects in the westerly United States. Here increases averaged about 10 percent in the seeded areas as compared with the control regions.

The analysis of these operational results combined with a close study of the experiments led the Panel to conclude that there is increasing

but still somewhat ambiguous statistical evidence that precipitation from some types of cloud systems can be increased or redistributed by seeding techniques. The evidence is complex and in part contradictory. The Panel thus concludes that the enhancement of precipitation is in the research and development stage and that one should approach large-scale operation with caution, since there are conditions where seeding may lead to a decrease rather than an increase of precipitation.

The Panel also investigated the question as to whether or not the local increase of precipitation would result in a decrease in precipitation downwind of the area of increase. The data are insufficient for any firm conclusion, but what data are available show no rain shadow effect.

I will now turn from the questions of precipitation enhancement and redistribution to other aspects of weather modification. It has been postulated that hail can be suppressed by the introduction of nuclei such as silver iodide. The physical basis for this suggestion is that the formation of many ice crystals will prevent the growth of any one ice particle to a very large dimension as is required for the formation of hail.

Extensive efforts aimed at reducing hail have been carried out in Argentina, Switzerland, and the Soviet Union.

In Switzerland the preliminary analysis of the data suggests that seeding increases the frequency of hail rather than decreases it, while in Argentina it appears that a small reduction in the frequency of hail follows from seeding. The workers in the Soviet Union are most enthusiastic about hail reduction and have mounted sophisticated operational programs, but hard data on the results of their projects are not available.

The introduction of many ice crystals or of chaff into lightningproducing clouds may reduce the frequency of fire-producing ground strokes. Preliminary experiments carried out in the Western United States indicate that, indeed, substantial reduction in lightning strokes can be produced in this way.

In certain regions of the world fogs at temperatures below freezing present a hazard for aircraft operations. The seeding of these fogs has been shown to be effective in cutting lanes for landing and takeoff. On the other hand, there is no known way of dispersing warm fogs, that is fogs whose temperature lie above 32° F.

Preliminary experiments have been carried out by a team of U.S. Weather Bureau and U.S. Navy scientists in attempting to modify hurricanes. The variability of hurricanes is so great and their dynamics and energetics so little understood that the Panel concludes that it is impossible to determine whether or not changes were brought about in the hurricanes.

In the above discussion I have emphasized the deliberate change of our atmospheric environment. As civilization grows more complex, society places greater burdens on the atmosphere. We are just beginning to appreciate that the atmosphere is not a dump of unlimited capacity, but we do not yet know what the critical capacity of the atmosphere is or by what effects it should be measured. For example, we can consider carbon dioxide introduced into the atmosphere as a result of the Industrial Revolution.

It is generally agreed that the amount of carbon dioxide has increased by 10 to 15 percent in this century and that this increase is due to the burning of fossil fuels. Calculation of the effects of this increased carbon dioxide are still tentative and crude, but they indicate that the temperature averaged over the surface of the earth has increased by two- or three-tenths of 1° Fahrenheit, and that this increase will continue as a result of further introduction of carbon dioxide into the atmosphere. An increase of temperature of similar magnitude might also result from the water vapor introduced into the high atmosphere by the planned supersonic transports. However, the Panel has not come to any firm estimate because the residence time of the water vapor in this part of the atmosphere is still unknown.

In a sense each city is an experiment in weather modification. It is known that urban regions have climates that differ significantly from the surrounding rural regions. We know very little of the causes of these differences. For example, there are data that suggest that in cities the temperature and rainfall are greater than in the surrounding areas. What is not known is whether these differences are due to the artificial energy sources within the city, to the changing of the heat absorption characteristics by construction or to changing the roughness of the ground.

In summary, I believe that it is now time for an increased effort on a national scale in the fields of weather and climate modification. The emphasis in these efforts should be on research and development. In the precipitation enhancement there are a multitude of unanswered questions. For example, why should the seeding of cold clouds be more effective than the seeding of warm clouds?

Why is there an apparent decrease in the effectiveness of seeding with the passage of time in a given locality?

Is this due to a carryover of the nuclei or is it due to a feedback effect resulting from changing the thermal characteristics of the ground by increased moisture?

Too many questions remain unanswered for the Nation to embark on very large operation projects in rainmaking without the most careful planning based on experimentation. However, it is important that the advances in research and development go forward with operational efforts and that new understanding of the atmosphere be quickly translated into operational methods which may benefit the whole of society. That is my statement, Mr. Chairman.

Senator DOMINICK. Thank you, Doctor.

I think this has been very helpful to our understanding of some of the problems that we are faced with.

Senator Hart?

Senator HART. The chairman by his questions has indicated a much greater understanding of some of these problems than the other member of the committee that is here this morning.

How many students are in the planetary and space science department at UCLA?

Dr. MACDONALD. It is a graduate department. About 50 students working toward a doctorate degree in the general field of planetary sciences.

Senator HART. What will the majority of them do upon graduation?

Dr. MACDONALD. Some will work in atmospheric sciences, some will work on problems dealing with the earth's interior, some will work in space-related areas.

Senator HART. There is a broad employment opportunity available? Dr. MACDONALD. Yes, indeed. There is an area for which the demand far outstrips the supply.

Senator DOMINICK. Doctor, I have a few questions. You referred to hail problems in Switzerland, the Soviet Union, and Argentina, where experiments have been going on. Did you do any investigation of commercial operations in Canada?

Dr. MACDONALD. Yes; we have looked at commercial operations in Canada, some in this country. None of them have been as extensive programs as the ones that I mentioned, and, again, we find complexity in the results. There are individual cases where the data apparently indicate a reduction in hail. There are other cases where again the data indicate an increase in frequency in hail.

Senator DOMINICK. Did these situations vary according to geographic location where the applications were going on?

Dr. MACDONALD. We have attempted to make a crude stratification of the data on the basis of geographic locality, and on general meteorological conditions, and have not been able to get a clear separation between when seeding would be effective and when it would not be.

Senator DOMINICK. I was interested in your statement regarding hail because earlier you had said that cold clouds, if we may call them that in nonscientific terms, produced more precipitation.

For example, Switzerland, which is undoubtedly colder than the Argentine except in the Andes area, is producing more hail, than the Argentine, which would seem, offhand, to comply with this cold-warm theory.

Dr. MACDONALD. Indeed, in Switberland there was an increase in frequency of hail and also an increase in total amount of precipitation associated with the seeding operations, so this kind of correlation apparently holds up. Unfortunately, we do not have good data on the cloud top temperatures either in Switzerland or in Argentina to make clear comparison.

Senator DOMINICK. You also refer to the introduction of additional amounts of carbon dioxide over the last 60 years.

Dr. MACDONALD. Since 1900 we would estimate the 10- to 15-percent increase.

Senator DOMINICK. In previous conversations with Dr. Roberts, he has referred from time to time to ozone. What is ozone?

Dr. MACDONALD. Ozone is a form of oxygen. It plays an important role in our atmosphere in that it is formed as a result of ultraviolet radiation striking the atmosphere at high levels and it plays an important role in controlling the thermal balance within the atmosphere that is, where the atmosphere is warm and where it is cold. And there have been suggestions in the past of changing the amounts of ozone.

We did not look specifically at this problem. The industrial contamination, the introduction of ozone by industrialization is completely negligible, but the big production is just due to the ultraviolet coming in from the sun and striking the atmosphere.

Senator DOMINICK. And contamination in the air, as a result of industrialization, does not increase the degree of saturation of ozone? Dr. MACDONALD. Not at these high levels. Locally, of course, it is very much a problem, and I am quite aware of it, living in Los Angeles. Whether or not it has an influence on the climate, the long-term climatic situation in a local area, based on the situation as in Los Angeles, is not certain. This is one of the problems that we point to. We do not know the extent to which the climate the long-term weather conditions within a limited area-is influenced by the pollutants, by the local energy sources, by the construction, and so forth.

Senator DOMINICK. Am I correct in understanding that ozone is poisonous?

Dr. MACDONALD. I am certain if you were breathing nothing but pure ozone you would not survive. I don't know how one would define it in a medical terminology. In low concentration it is just a high irritant.

Senator DOMINICK. Have you heard of any reports that the socalled ozone layer is coming down in terms of height?

Dr. MACDONALD. No. Observations and fairly careful monitoring of the ozone layer have been underway for the last 15 years. It fluctuates in its intensity and somewhat in its position with the solar cycle. That is, there is long-term variability, but it is small and there is no noticeable long-term change of the total character of the ozone layer.

Senator DOMINICK. You would be surprised, then, if another scientist should dispute this?

Dr. MACDONALD. Yes, and we would have an interesting argument. Senator DOMINICK. We deferred before on my questions to Dr. Malone on the need for additional tools. Do you have any comment on that?

Dr. MACDONALD. Yes. In order to carry out the kind of studies that are needed to determine whether or not seeding actually is increasing precipitation, and identifying the mechanism by which this is happening, we have to improve our ability to describe the cloud.

Now, this improvement can come about in many ways. Dr. Malone referred, let's say, to the kind of instrumentation that you carry aboard airplanes and fly through a cloud and determine essential instantaneous distribution of cloud droplet size. This is an important factor because it determines how the cloud droplets themselves grow and reach a critical size.

Another kind of instrument that one would wish to carry aboard an airplane and needs very marked improvement are instruments designed to count the number of nuclei. Dr. Malone referred to the question of following the path of the nuclei. A problem that we have is that there is an apparent difference in the effectiveness of seeding, depending on whether the silver iodide is introduced at ground level or in airplanes, and is a major question that still remains unanswered. What is the path that the nuclei follow as they enter into the cloud? Are they indeed carried up by the internal drafts, the upgoing drafts within the center of the cloud, or do they follow on a much more complex pattern? Is it better to seed at ground level than to seed within the cloud itself? These are the kind of questions that would be an