Dr. SINGER. No, I do not.

Mr. RUMSFELD. This would not be the case.

Do you know that this is not the plan or are you just-
Dr. SINGER. This is not the plan at the present time.

Now, plans can be changed; but I think there would have to be a good reason for changing them.

Mr. RUMSFELD. We have so many problems of accumulating information and disseminating information. As each department of the Government comes upon new things, it strikes me they have a very real responsibility to raise these new possible problems for resolution with other related problems within Government.

I assume you are doing that.

Dr. SINGER. Yes, we are.

I would like to support this, if I may by reading from my prepared

statement.

There are important unforseen aspects of weather satellites; namely, the possibility of extracting information which is not strictly meteorological, from a system designed primarily for weather forecasting.

This is an example of what you have just mentioned.

I discussed last September the many existing applications such as forest fire detection, snow reconnaissance, ice reconnaissance, and the possible detection of locust clouds.

We feel that it is appropriate for us to take the initiative in pointing out such applications wherever we see them, and in establishing their feasibility; and then, of course, in turning over the application to the appropriate Government agency to pursue.

Mr. RUMSFELD. I am glad to hear that.

Dr. SINGER. For the future development of an operational satellite system, our major concern must be a low operating cost. We have completed the systems analysis, which has convinced us that a long operating life, perhaps several years, is one of the most important goals for an operational satellite. We feel that the design of a spacecraft capable of operating for many years should be initiated immediately. By the very nature of its mission, the research aspects will have to be subordinated, and it may indeed turn out that research satellites and operational satellites are not compatible in the sense that they cannot be based on the same spacecraft design.

As I indicated in my opening remarks, I want to describe, in this fourth and final part of my prepared statement, our most significant NOMSS achievements since my last appearance before this committee.

These achievements have been many, most of them individually small, but collectively of major importance to the national space effort. They are, in essence, the data processing achievements which will insure Weather Bureau readiness to receive, handle, and disseminate, in real-time, the more than 5 billion bits of weather data transmitted to Earth by Nimbus every day.

To do these things we are establishing an advanced processing capability for weather satellite data at our Suitland, Md., location. Here we will process satellite meteorological data in two ways:

In the first, the signals received from the satelliteborne TV camera will be converted on the ground into the cloud pictures as they originally appeared to the camera. Several preliminary editing steps must be carired out, however.

The pictures must be located geographically, using precomputed orbital tracks, the exact time of each picture frame to the nearest second, and the attitude of the satellite. Latitude and longitude grids will then be computed for each picture and converted into electrical signals. These, in turn, will be combined with the incoming picture signals to produce photographic copies of the cloud pictures with the grids superimposed upon them. In making these computations we shall take into account distortions produced by the satellite camera systems and correct for them in preparing the gridding signals. Provision will also be made for precise determination of picture locations through rapid preanalysis of the data to locate landmarks and other visible clues to exact position.

In the second processing method, all data from the satellite, including pictorial cloud images and infrared scan signals, will be converted into numerical values for precise evaluation and correlation by computer. This conversion takes place at upward of 1 million values per second. All the data from a given orbit will be stored on magnetic tapes and disk files and will be withdrawn to the central computer unit in batches for evaluation.

This second processing technique, like the first, will locate each data element geographically and remove system distortion. The elements will then be grouped in accordance with a fine-meshed global network based on standard map projections. By comparision with a file of surface brightness, maintained in the computer, clouds will be separated from background and will be characterized in each grid square as to coverage, height, distribution, and brightness.

These values will be then combined in the computer to produce global mosaics of cloud conditions in a form for direct weathher analysis and forecast applications. These mosaics will be available pictorially and in numerical form to the world meteorological community. Infrared measurements will be processed in similar fashion to provide heat balance data for numerical forecasting and for providing information needed to know cloud heights, surface temperatures, and other parameters in the global atmosphere.

To do these things we are establishing an advanced processing capability for weather satellite data at our Suitland (Md.) location. We have already constructed the necessary addition to the building and are prepared to move the equipment in within the next month and be in operation by this summer.

In the first processing techinque, the signals are received from the satellite-borne camera and converted on the ground. Now I have discussed here in a quite detailed fashion this technique to indicate the complexity of the data processing job. It is not a simple job. It requires a great deal of forethought.

We have taken care to make sure it is just as flexible as we can make it to be able to take care of any new developments in weather satellites.

We have three different ways of processing the data for three different types of users.

We can, by processing the data, extract and provide such information as clouds and in turn, the distribution of clouds, their brightness, and even their heights; and these cloud maps will go out to the users because they are an immense aid in forecasting.

The activity of this data processing center is described in detail in

these last few pages, but I would like to turn now, finally, to the importance of the work of our Meteorological Satellite Laboratory.

I have mentioned that this is the largest and preeminent scientific laboratory in the field of satellite meteorology. Such research, carried out on the basis of cloud pictures and infrated measurements, is very exciting to research scientists. Without the knowledge derived from such research, the Nation's weather satellite program, our data processing complex, and the communications channels by which meteorological information can be globally distributed would fall far short of achieving its potential for usefulness.

Through research, we increase our knowledge of the nature of the weather data provided by satellites, and constantly learn new things about such data and how to apply them to the maximum benefit of mankind.

Now, I've brought with me some slides I think will illustrate much better than these words, just what the Laboratory does.

With your permission, Mr. Chairman, I'd like to run through just seven of these slides to indicate how the Laboratory operates.

This is a picture of the Caribbean (fig. 5). You will recognize the island of Cuba, as well as some of the Bahama islands.

What this picture illustrates, I think better than I could portray it in words, is (in the upper left-hand corner) a cloudband-a cold front which is sweeping through the Caribbean. You can tell the position of this cloudband with an accuracy of about 3 to 4 miles which means you can tell the time that the weather is going to get bad to a time accuracy of a few minutes. This type of information, in my opinion, has tremendous operational importance for many different purposes.

This (fig. 6) shows, on the left-hand side, the Florida Peninsula. Stretching out over the Gulf of Mexico, you see a bright band of clouds

in a nearly straight line. That is a squall line. It produces severe weather especially in the air but also on the ground, on the ocean. It gives rise to extreme air turbulence and, to severe thunderstorms below. This sort of information is of extreme importance to aviation and to smaller ships. On the right-hand side you see an example of an extra-tropical cyclone in the Atlantic. This type of picture has, for the first time, confirmed the textbook ideas of how such a cloud system should look. Now we see that some of the cloud pictures really turn out to look like the textbooks have pictured them. Others do not.

Now, of course, our laboratories are trying to discover the reasons such differences exist.

On the right side is a picture (fig. 7) taken of the Central United States. You can see Texas in the lower right-hand corner. Oklahoma just above, the Panhandle sticking out.

What I want to call to your attention are the three large cloud masses on the right-hand side of the picture. The scientific examination of this picture showed these cloud masses were extremely bright compared to the clouds in the left side of the picture. Also, they showed these cloud masses had very sharp borders on the western edge and "fuzzy" borders on the eastern edge.

The ground observations with which we compared the satellite observation showed each one of these three clouds contained the most

violent types of weather that you can imagine. One of them had a tornado and all of them had hail. In one case the hail was as large as a baseball. It was a very violent weather situation.

Chairman MILLER. How high was that taken?

Dr. SINGER. This was taken from 450 miles.

Chairman MILLER. How high are the tops of those clouds?

Dr. SINGER. This is a very important question which we can answer now only by inference because of the brightness of the clouds and because we know something about the nature of the cloud.

We say that the tops are about 50,000 feet high.

With Nimbus and infrared data, we will be able to measure the temperature of the cloud tops. Once we know this temperature, we will also know the altitude of the clouds.

This is one of the purposes of our Research Laboratory-to establish techniques by which we can obtain, from satellite pictures, further information about clouds.

Chairman MILLER. This is off the record.

Mr. Rousн. Doctor are you able to detect seasonal changes or are you doing any research so as to more definitely pinpoint seasonal changes in various sections of the country?