Introduction

Of all the techniques available for study of the planet Earth, spacecraft are recognized as the most valuable. Today's spacecraft are far more sophisticated than the first weather satellite, Tiros 1, which was launched by the United States more than 25 years ago. They can monitor such diverse resources as crops, minerals, and fisheries, as well as atmospheric pollution. Moreover, according to Dr. Moustafa T. Chahine, manager of the division of earth and space sciences at the California Institute of Technology's Jet Propulsion Laboratory, the remote sensing tools now on the drawing board show enormous promise for the next 25 years of viewing the earth from space.

In describing how space science and remote sensing technology are applied to crop production, agro-climatology, and mineral exploration, Chahine pointed out that researchers are now on the verge of establishing a remote sensing system that can provide the long-term, global, synoptic measurements needed to understand the earth's interactive system of oceans, atmosphere, continents, cryosphere (snow, ice, glaciers, and permafrost), and biosphere. Such knowledge is essential to understanding how nature will respond to the increased amounts of atmospheric carbon dioxide, sulfur, and methane that humans are pumping into the air they breathe. According to Chahine, the present rate of fossil fuel consumption could double the amount of atmospheric carbon dioxide by the year 2020, elevating the global surface temperature by more than two degrees centigrade. Such an increase would produce flooding, alter precipitation patterns, and have a considerable impact on regional fresh water and agriculture. Sulfuric acid rain is the result of the additional atmospheric sulfur (almost twice that produced by natural processes) generated by human activities.

Today, the lack of accurate global data on life-sustaining carbon, nitrogen, oxygen, and sulfur greatly impedes our ability to study the movements of these l;ey chemical constituents of the earth's system. Global variations of the hydrological cycle in all its phases - evaporation, precipitation, water flow - need to be monitored.

The two most important elements of a remote sensing system that gathers data on the earth's interactive system are in situ measurements and satellite observations.

Deep ocean and underground processes are not generally amenable to observation from satellites, explained Chahine. Land-based biological processes and some important tropospheric chemical reactions require in situ measurements as well. Long-term in situ observations are also needed to ensure the stability and accuracy of the various space-borne sensors. A worldwide effort is required to gather such data, suggested Chahine, perhaps in the form of a broad composite data system that uses ground stations, ocean buoys, aircraft, and balloons, as well as satellites, which have a special role to play in the collection of in situ data and its subsequent transmission to users.

Space-borne sensors are presently operating at many wavelengths and observing a variety of geophysical phenomena. According to Chahine, space-acquired data fall into four major categories of study: (1) sun-atmosphere interactions that involve the composition, chemistry, and photochemistry of the stratosphere and variations in the output of the sun; (2) land-atmosphere interactions that involve the exchange of water, gases, aerosols, and heat; (3) ocean-atmosphere interactions that include the transfer of water and gases, heat, and momentum; and (4) the ozone budget, the global hydrological cycle, and the other biogeochemical cycles of carbon, sulfur, nitrogen, and phosphorus that regulate the earth's metabolic system.

Using the latest in supercomputers, researchers can run interactive numerical models able to simulate to a limited degree the earth's entire system. Chahine warned, however, that such models must be tied to global synoptic observations if our understanding of the causes and effects of global change are to advance.

Today, a large number of geostationary and polar-orbiting satellites launched by the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), and the Department of Defense are measuring daily the variability of such important parameters as solar radiance, ultraviolet flux, volcanic aerosols, total ozone, earth radiation budget, atmospheric temperature, winds in the tropics, atmospheric water vapor, sea surface temperatures, land-use changes, and vegetation cover. Better knowledge of some of these parameters is vital to the economic development of Third World countries.

"The greatest single unmet need at present is for acquiring global information on the biosphere and on land biology," emphasized Chahine. This need as well as that for other types of data are being met by the development of several improved and more sophisticated experimental sensors.

The ultimate objective of the research under way in earth science is to obtain simultaneous measurements of selected sets of parameters over long periods of time for use in analyzing and synthesizing global change. During this measurement phase, emphasis will shift to the use of the new space platforms in polar and geosynchronous orbits, as envisaged by NASA's proposed Earth Observing System (EOS).

"Major advances in earth science will come from the synthesis of ideas drawn from the study of such global data," said Chahine. For example, the accuracy of weather predictions has increased significantly as a result of the present understanding of the dynamics of the large-scale circulation of the atmosphere, which is derived from global observations. It is clear, he concluded, that "we are on the verge of being able to comprehend the coupling between the ocean and the atmosphere."

Applications in Crop Production and Agroclimatology

Data from global satellite remote sensing systems can be used to inventory crops, estimate crop production, and provide information on which to base agricultural management decisions. These data must, however, be delivered rapidly to farmers and to others who make management decisions affecting agricultural production if they are to be of value. If these data are monitored over a growing season, it might be possible to predict crop production on a regional scale.

Unfortunately, the management promise of space remote sensing ' has not been fully realized (even in the United States) for two reasons. First, the relatively poor spatial resolution of the instruments used has limited their utility, particularly in developing countries. For example, the ground area examined by the first generation of Landsat sensors was about the size of a football field, which is considerably larger than farmers' fields in most developing countries. While the most advanced Landsat sensor can look at an area about the size of a basketball court, it is often inadequate for dealing with the small, irregularly shaped fields found in developing countries. Second, it has been difficult to deliver the products of satellite remote sensing quickly. The revisit frequency of the Landsat satellites, which is approximately two weeks, does not provide timely enough data for monitoring crops during the growing season. Thus, the data are virtually useless for crop management decisions.

Despite these problems, which were cited by Dr. Charles F. Hutchinson, director of the Arizona Remote Sensing Center at the University of Arizona, remote sensing has been used by foreign assistance agencies and others to monitor agriculture worldwide. Efforts to apply the general remote sensing techniques developed in the United States to Third World countries have had only mixed results, however. Difficulties have resulted from the problem with field size just mentioned; gaps in important geographic areas that result from spacecraft malfunctions, cloud cover, and limitations of the ground receiving station network; and the unreliability of the historical crop production data needed to determine how the current crops are performing relative to past years. The most important problem' however, is the difficulty in establishing developing country institutions that can carry on the assessments after donor assistance is removed.

Agricultural climatology deals with the processes that occur at the earth's surface, particularly the transfer of mass and energy between the surface and the atmosphere. The fields of meteorology' crop physiology, and soil science are therefore of interest to agroclimatologists. Using remote sensing and other technologies, these scientists have been able to determine rainfall and temperature patterns in the Sahel region of Africa.

According to Dr. Edward Kanemasu, professor of agronomy at Kansas State University, researchers have discovered a very strong correlation between the onset of the rainy period and its duration. Early onset of the rainy season was found to be typical of a long growing season. Similarly, they also saw a strong correlation between the tote, amount of rainfall and when the rainy period began. For the last 16 years in the Sahel, however, there has been a change in this correlation, explained Kanemasu; the amount of rainfall has decreased. This is consistent with the climatic analysis undertaken by others that suggests a change in the Sahel's weather pattern.

Knowing the length of the growing season, farmers can determine what kind of crop management practices should be cons,idered. Crop yields can be predicted using some of the more recent crop simulation mode,s, such as those developed under IBSNAT. (See p. 50) These models require minimum biophysical data for a particular site.

Unfortunately, these models have had only mixed success, particularly when used in the stressful environment of the Sahel. One problem is trying to determine the amount of leaf area generated by each crop - that is, trying to simulate the growth of the leaves. The Landsat and Spot satellites are able to estimate the amount of leaf area, but such an estimate is affected by a number of things that the modes cannot easily simulate, for example, the amount of water available to the crop. The amount of precipitation may be known, but how much of it is lost to runoff? In determining fertility, how much fertilizer does a particular farmer have? In terms of disease, when was the onset and what was the severity of damage caused by a disease or insect?

The ability of remote sensing to estimate leaf numbers is based on the unique spectral distribution of reflectance of a green, healthy leaf. A green leaf absorbs strongly in visible light and reflects highly in infrared light. An index that uses that ratio allows one to tell the difference between a plant and the background of the soil. While satellites can estimate global and large regional changes in vegetation, explained Kanemasu, ,imitations are attached to those kinds of data because the satellite is in a fixed orbit, and, as also pointed out by Hutchinson, has a rather low spatial resolution. Moreover, there is the delayed turnaround of data.

Video remote sensing aboard an aircraft is one possibility for obtaining more current data. The requirements for instruments and equipment are relatively simple, and data are available as soon as the aircraft lands. More important, data can be obtained when needed. For a sate,'lite in fixed orbit this is not possible.

Forecasting Food Emergencies Using LANDSAT Data

During the 1985 growing season, the U.S. Agency for International Development used Landsat data to forecast sorghum and millet crop failures in the Sudan and to support planning food relief operations. The forecast permitted the timely delivery of U.S. food grains to over 1 million starving people in western Sudan. The photo maps made from the satellite data were a,so distributed to U.S. helicopter pilots who used them to navigate the uncharted terrain and to get the critical food supplies to those most badly in need. Other U.S. federal agencies (NASA, NOAA, and USDA) provided assistance in carrying out the forecasting effort and the Sudanese air force supplied the aerial photography necessary to confirm the satellite data interpretations.

The successful use of Landsat data for Sudan emergency relief efforts led to the development of the Agency's Food and Early Warning System (FEWS), in which eight African states historically subject to famine are being monitored for vegetation browning indicative of crop and pasture failure. The USAID mission in Sudan replicates this successful procedure every year in its Emergency and Rehabilitation Information Surveillance System.

Applications in Mineral Exploration

Dr. Ronald J. P. Lyon, professor of mineral exploration at Stanford University, was not optimistic about the suitability of satellite technology for mineral exploration - in particular, lead, zinc, copper, uranium, and gold - in developing countries. Better spatial resolution is needed, he believes, before satellite-based remote sensing can be practicably applied to the search for minerals.

According to Lyon, the multispectral scanners designed specifically for the mineralologies of deposits such as copper, lead, and zinc are used more appropriately in aircraft than in spacecraft. Deeply buried oil is more effectively discovered using various geophysical and seismic techniques, but Lyon admitted that the Spot satellite is "extremely attractive" for evaluating potential fracture systems for oil and coal basins. Also, data acquired by the Landsat Thematic Mapper are used by multinational firms to explore for hydrocarbons in the developing countries. Lyon added though that aircraft camera and scanner systems should not be dismissed. Any search for building materials - phosphates and the materials needed for cement, for example - lends itself more to black-and-white aerial photography followed by seismic testing.

Ten years ago, the U.S. Agency for International Development and the U.S. Geological Survey assisted the government of Bolivia in the discovery of some of the world's richest lithium deposits in the Salar de Uyuni. Landsat data were used to identify salt brines with extremely high concentrations of lithium (2,600-3,000 parts per million), and the subsequent ground samples taken by the investigators confirmed the high concentrations. Following these discoveries, which were covered by the major news media' an American-based multinational corporation invested $137 million in Bolivia's economy to extract some of the lithium.

USAID's application of remote sensing to mineral exploration is currently limited to using Landsat data to construct photogeological atlases that are combined with seismic and aeromagnetic data to locate and map petroleum, mineral, and groundwater resources in Egypt.

The Future

In predicting future developments in this field, Hutchinson predicted new satellites (Spot), new sensors (video), new platforms (high flyer), and inexpensive powerful computers, but no major technical or scientific breakthroughs. "Techniques for monitoring crop production have evolved to the point where their regular and reliable application in developing countries is possible now," declared Hutchinson. The U.S. Department of Agriculture (USDA) and USAID are using satellite remote sensing to monitor crop production both in the United States (USDA) and in other countries (USDA, USAID).

The need for information on crop production, particularly in Africa, has not lessened as the pictures of the drought-burdened land and people of Sahelian Africa have disappeared from the pages of newspapers and magazines. According to Hutchinson, the two most prominent features of arid climates are low rainfall amounts and extreme interannual variability. Thus, monitoring in times of relative plenty is much more important than monitoring in times of drought. Remote sensing offers the opportunity to forecast crop failures in the making, so that steps can be taken beforehand to prepare for food emergencies.