Chairman WALKER. Thank you very much. Dr. Castruccio? Dr. CASTRUCCIO. Castruccio. Chairman WALKER. Castruccio. Dr. CASTRUCCIO. Right. Chairman WALKER. Okay. I'm sorry. Thank you very much. We'll recognize you next. STATEMENT OF DR. PETER CASTRUCCIO, ECOSYSTEMS Dr. CASTRUCCIO. Thank you, Mr. Chairman. It's a pleasure to be here, despite the meteorology and the parking situation, the combination of the two. Chairman WALKER. We're going to solve that with satellites some day. Dr. CASTRUCCIO. Yes, we're going to solve that. [Laughter.] The name of the game here, the way I interpret your charges, Mr. Chairman, is how can we make it cheaper? I will omit discussion of cheaper sensors because my colleague on the right is covering that. But I want to talk about simplifications and cost savings from methodologies. Three points. One, we might consider, or should consider, the operational use of surface data. I know surface data are now being gathered for calibration purposes. However, world-wide, we have on the surface approximately 250,000 sensors which work all the time in different countries and they are ground, sea, water, and air. Some of them have fairly long data streams gathered in the past decades. The Nile is quoted at 5000 years. The others have less. These sensors now exist and the beauty is they are paid for and the data is paid for by other countries. I don't want to confuse this type of sensor with gathering calibration data. They're two different things. Now, these remote sensors of this type, such as buoys in the ocean, can be connected to existing space data collection systems like the GOES and the ARGOS. The ARGOS, for example, will charge 40 cents an hour-that's cents of a dollar-40 cents an hour to gather these data and relay them back to the user or the scientist. The surface data thus gathered is robust. That means it measures really what happens-salinity, presence of algae, and so forth. The space data rely on radiometric information, so they have to translate that information into what really happens. The two should be combined, in my opinion. If you look at how data are gathered in the modern world, by institutions such as like USDA, USGS, National Weather Service, you'll find they're really all surface data, the satellite being as an adjunct to them. Potential economic trade-offs-take the Mission TRMM, for example, Tropical Rainfall Measuring Mission. The cost of that mission could purchase 33,000 sophisticated ocean-going buoys-because that mission does mostly ocean, or 150,000 existing, less sophisticated buoys. The questions are to what extent would these economical data satisfy science objectives? Our experience shows a considerable number of these ground systems can satisfy a lot of science objectives. Now, what do they cost and what international exchange arrangements ought to be considered? The second point is use of sampling techniques. At the present, the data gathering by agencies is practically all done by sampling. Sampling means simply we take a few points-a few may be 10, 20, 100 to represent a population. For example, opinion sampling is not done over the entire population. It's done on a few thousand people using statistical techniques. For example, the crop computation in the United States is done over less than one percent of the territory, little segments carefully chosen. Likewise, the weather is done by sampling. And it's interesting to note that the sampling technique works everywhere. It works in Pennsylvania as well as in Albania, where recently, a survey was done to determine the crops using exporting sampling techniques developed by USDA. Now, the space sensor-and then I'll be finished-space sensors really sample in time. In other words, they take a snapshot and return perhaps two days later. That's called a time sampling. Surface sensors are fixed and sample in space. Can the two be combined to further the objectives of EOS? The third point-utilization of scientists. Long experience shows one important fact. No scientists, with very few exceptions, no scientists will work on this program unless he gets paid. There's no such thing as a free lunch. And we must then take into account the considerable number of scientists which will be around until the end of the program. It is in my experience fallacious, as was done in the past, to say, oh, well, once we got the data base, the world will beat a path to our door. That will not happen. We are all in favor of science, but we're also all in favor of making it affordable, making it reasonable. Thank you. [The prepared statement of Dr. Castruccio follows:] Report on Mission to Planet Earth (MTPE) Earth Observation System (EOS) Prepared at the request of: The Committee on Science by Dr. Peter A. Castruccio Panel I Members of Panel I were charged with "discussing different means of collecting data about the earth and its processes, with an eye towards exploring innovative and cost-effective methods". Analysis of pertinent documentation, discussions with scientific and managerial personnel within and outside the EOS program, supported by this panelist's 30 year's involvement in remote sensing and earth resources programs, leads to certain observations and suggestions, aimed at reducing the program's costs and enhancing its effectiveness: Augment the operational use of surface-derived data. These are relatively inexpensive to obtain, and frequently more robust than space-derived data. Make as much as possible use of sampling techniques. It is not practically feasible, nor desirable, to "measure everything everywhere". Optimize the usage of scientific personnel involved in the program 1. Greater operational use of surface-derived data Consider the operational exploitation of the archival data available from existing surface-based sensors (on land, oceans, atmosphere). These amount to approximately 250,000 installations that measure various phenomena: rain, temperature, streamflow, etc. Some of the data sequences generated by these sensor stretch backwards in time for decades (the Nile is quoted as having been gaged for the last 5,000 years). These data represent considerable historical record, whose wise exploitation might reduce the need and/or costs for measurements from space. a In addition to archival data, continuous use is being made of this instrumentation in many nations. An important point is that the data are paid for by each host country. Current operational technology allows remotely sited ground-based instruments, e.g., raingages, streamflow gages, temperature sensors, to be outfitted with transmitters communicating with data-gathering satellites, whence the data can be relayed to earth-based collecting stations, thence to users. One such data collection and dissemination system is ARGOS, that charges on the order of 40 cents per hour, or $3,500 per year per sensor operating continuously full time. Another is GOES. These considerations apply as well to open-sea measurements where instrumented buoys are currently used to measure oceanic parameters (temperature, salinity, seacurrents, waveheights etc.), and which transmit the collected data to spaceborne data collection and dissemination systems such as ARGOS. Similar arguments apply to the numerous aircraft and balloonbased sensing programs, both past and ongoing in several countries, for example the massive program conducted in the US by USDA/ASCS and DOI/USGS, or NASA's excellent aircraft color infrared imagery program of the early '70s. Last but not least, consider the use of recently perfected UAVS (unmanned aerospace vehicles). Good meteorological data are being generated by weather radars, in the US and in several foreign countries. Surface measurements are by their nature more "close to the phenomena" than their space-borne counterparts, which generally rely on inferring the underlying mechanisms from intermediate parameters, mainly the characteristics of electromagnetic radiation emitted by or reflected from, the objects being looked at. Moreover, there are phenomena that cannot be sensed from space: for example, oceanic subsurface temperature. 26-244 0-96-7 |