Disaster monitoring 
Because monitoring programs involve systematic observation, they can often 
give warning of disasters and developing critical situations. This alarm-giving 
function can be the main role of a monitoring program, but it usually is not. For 
example, the World's meteorological observation and forecasting systems, in which 
the familiar weather satellites play a vital role, involve routine forecasts and 
observation of the current meteorological situation. The capability to warn of 
tornadoes, hurricanes, and snowstorms is important, but only part of one of the best 
monitoring systems in existence. 
The role of aerial photography in damage surveys is well documented. 
Conventional and large-scale photography are the main approaches. Some of the 
relevant works have already been mentioned in the discussions of forestry 
applications. One remarkable approach to disaster warning is the low aerial 
photography system developed for the study of the breeding sites and distribution 
of desert locust, (Rosetti, 1971). On-board processing, flying heights below 30 m 
and scales as large as 1:100 are irwolved. 
Last year Robinove (1975) wrote an interesting paper in which he speculated 
about the capability of Landsats to provide disaster warning and assessment. He 
concluded that research already completed had shown excellent possibilities for the 
assessment of damage from floods, drought, fire and glacier movements and 
possibilities for the assessment of earthquakes, volcanic eruptions, crop disasters, 
water pollution and hail storms. He also discusses the sometimes promising 
possibilities of early warning systems for many of these disasters. 
Among the operational applications of Landsat, damage assessment already 
ranks high. Most other potential applications of Landsat in, for example, land 
classification, vegetation mapping, and the monitoring of human activity, are only 
now emerging from the research stages, but the mapping of boundaries of major 
floods is sufficiently reliable to be operational, provided the satellite is overhead 
during or soon after the crest of a flood, and cloud presents no obstacle. Major 
wind damage on forested land can also be clearly seen (Moore, 1974) and the 
assessment of recent fires on Landsat imagery, is today part of the operational 
program of the Ontario Remote Sensing Centre (Anon., 1975). Their technique 
involves imagery obtained prior to, during and after each fire season: the 
boundaries of burns are delineated on post-fire imagery and transferred to forest 
inventory maps, to provide estimates of area and timber volume destroyed. It has 
also been demonstrated that the rate of spread of large fires can be estimated from 
Landsat imagery. Fires in the savannah of southern Africa, which were included in 
the overlap of coverage on successive days, were the object of this experiment 
(Wightman, 1973). In a subsequent related paper it was demonstrated that during 
each dry season there is a seasonal progression of huge grass fires over vast tracts 
of savannah from central Nigeria east to the Ethiopian highlands. Within an area 
ranging from 800 to 1400 km, north to south, more than half the land surface is 
burned (Deshler, 1974). The ecological consequences of this annual "disaster", so 
readily monitored by satellite remote sensing, cannot be underestimated. 
CONCLUSIONS 
Effective management of natural resources and of our surroundings at large 
requires first, a clear understanding of what society expects and demands and 
secondly, a well thought out strategy for best use of the resources. Information is