[Illustration: The Landsat simulated true color mosaic (left) shows the Selima Sand Sheet covering all but rocky areas of the Sahara Desert in Sudan. On the right, a 50-kilometer-wide strip of Shuttle Imaging Radar, SIR-A, is placed over the Landsat mosaic to reveal old stream channels and geologic structures like these. Structures that are otherwise invisible under the surface sands are potential sources of water, placer minerals, ancient artifacts, and information on changes of climate in arid areas (courtesy of USGS Image Processing Facility, Flagstaff).] The scarcity of vegetation makes spectral remote sensing especially effective in arid lands. Rocks containing limonite, a hydrous iron oxide, may be identified readily from Landsat Multispectral Scanner data. The Landsat Thematic Mapper (TM) has increased our ability to detect and map the distribution of minerals in volcanic rocks and related mineral deposits in arid and semiarid lands. More than a million images of Earth have been acquired by the Landsat satellites. A Landsat image may be viewed as a single band in black-and-white, or as a combination represented by three colors, called a color composite. The most widely used Landsat color image is called a false-color composite because it reproduces the infrared band (invisible to the naked eye) as red, the red band as green, and the green band as blue. Healthy vegetation in a false-color composite is red. Desert studies still are hampered in many regions by lack of accurate climate data. Most desert weather stations are in oases surrounded by trees and buildings and have been subjected to many location and elevation changes throughout the life of the station. Data from oases do not reflect conditions from the surrounding desert. A wide variety of instruments has been used to record measurements over varying lengths of time and in different formats, making data difficult to interpret and compare. To overcome some of these problems in deserts of the American Southwest, the U.S. Geological Survey (USGS) established its Desert Winds Project to measure in a standard format several key meteorologic characteristics of arid lands. Project scientists have successfully established instrument stations to measure wind-speed, including peak gusts, which alter the landforms the most. A station recorded a windstorm near Vicksburg, Arizona, for example, with peak gusts of almost 150 kilometers per hour. Using low-maintenance, automatic, solar-powered sensors, the stations also measure wind direction, precipitation, humidity, soil and air temperatures, and barometric pressure at specific heights above the surface. Data are sampled at 6-minute intervals and transmitted every 30 minutes to a Geostationary Operational Environmental Satellite (GOES). From GOES, the data are transmitted to the USGS laboratory in Flagstaff, Arizona.