three data-collecting systems: satellites, aircraft, and ground observers, 
in that sequence. Each of these systems would provide progressively closer 
looks at progressively smaller areas, and would provide more detailed informa 
tion about those areas. Then, the more detailed information would be applied 
to a much larger area for which the limited sample appeared to be represen 
tative, as evidenced by the similarity of the small areas to certain surround 
ing areas as recognized in the space and aircraft data. The first tests of 
this three stage approach for multispectral scanners have come from Earth 
Resources Technology Satellite (ERTS) experiments. 
But to achieve a practical tool, the sensors and automatic processors 
must replace a sufficient number of ground observers for the costs to be 
reasonable (a few U.S. dollars per square kilometer is the goal). 
Earth Resources Survey Systems are used to gather information to help 
government, private or foreign users with statutory or other requirements to 
solve problems or manage resources. These systems consist of remote sensors 
in aircraft or spacecraft, data formatting and telemetry links, preprocessors, 
extractive processors, and user applications models. Remote sensors register 
attributes of the terrain being covered. The sensor data may be relayed to 
ground facilities through telemetry links. At the facilities, preprocessing 
and extractive processing are performed, to extract information from data. 
Finally, user applications models are used to relate information outputs from 
the extractive processing, and ancillary data, to the user in terms most 
meaningful to him. 
With the launch of the ERTS-1 satellite in 1972, the ability of remote 
sensors to collect periodic multispectral data was significantly increased 
over the previous and continuing airborne sensor capability. The potential of 
coverage of a given site on the earth every 18 days facilitates change detec 
tion and opens up possibilities of analysis techniques using the time signa 
ture of remote sensible terrain attributes. Finally, the ability of the ERTS 
system to synoptically view 100 n mi 2 "frames" (with 7.2 x 10 6 picture elements 
or pixels per frame) facilitates surveys of large areas, and because of this 
capability, more users of Earth Resources Systems data are actively interested. 
At the same time however, considerable strain is being put on the exist 
ing preprocessing and extractive processing hardware and methodology because 
of the potential of ERTS for spectral-spatial-temporal analyses, because of 
the high data rates involved, because of the need to correlate data with 
existing airborne sensors for the high spatial resolution view often required, 
and because of user-imposed timeliness requirements. Additionally, considerable 
development of the user applications model area is required if the full 
potential of Earth Resources Information Systems to deliver cost-beneficial 
results is to be achieved. 
At ERIM, we have been exploring for ten years the utility of remote sensing 
data to various natural resources problems, to developing the system method 
ology and hardware necessary to rapidly provide accurate information to users 
at low cost, and developing user applications models to enhance the value of