International Archives of Photogrammetry and Remote Sensing. Vol. XXXII Part 7C2, UNISPACE HI, Vienna, 1999 
91 
I5PR5 
V^y 
UNISPACE III - ISPRS/EARSeL Workshop on 
“Remote Sensing for the Detection, Monitoring 
and Mitigation of Natural Disasters” 
2:30-5:30 pm, 22 July 1999, VIC Room B 
Vienna, Austria 
connection with natural disasters. I feel that in spite of the many 
demonstrations - often carried out post festum - useful opera 
tional systems designed for the purpose have not yet appeared 
and the techniques are applied on an ad hoc basis. To obtain the 
full benefit of remote sensing it shall be integrated in a manage 
ment system that is in place and tested before a disaster occurs. 
This requires full co-operation by authorities and remote sensing 
specialists as well as communications people. 
System for flood monitoring 
I shall attempt to describe a number of components of an inte 
grated system for flood monitoring that are important for opti 
mum use of remote sensing data when a flood may occur within a 
regioa 
(a) Short-term predictions of major events of rain in a 
catchment area is an essential part of the system which will alert 
the people involved - resulting in false alarms in some cases. It 
involves meteorological institutes exploiting data from 
METEOSAT, for instance, and modelling. The meteorologist my 
take advantage of hindcasts describing previous cases of cloud 
formation that created floods. With METEOSAT data archived 
for more than 20 years such analyses seem feasible - although 
time consuming. 
(b) Meteorological studies may extended to include ‘his 
torical data’ even if the meteorological data are less precise. At 
any rate these data and true historical records may be used for 
defining disaster-prone areas that shall be included in the prepa 
rations. 
(c) With the availability of a DEM of most populated 
areas in Europe it is possible to predict what will be the conse 
quences of a major rain event in disaster-prone areas. Model 
simulations including the development with time of events will 
be useful in describing likely flood scenarios. 
(d) An analysis centre shall be established with the re 
sponsible authorities and in co-operation with remote sensing 
specialists. The centre shall include the software needed for im 
age processing and analysis. This includes also the relevant 
DEM’s. A Geographical Information System (GIS) may prove 
useful to assist classification work. 
(e) Based on flood simulations a proxy remote sensing 
model may be worked out for training purposes exploiting actual 
images covering the area in question. Models should be devel 
oped for different seasons to cover different vegetation condi 
tions. Remote sensing data from previous cases might be avail 
able and used for training. 
(f) The infrastructure for data transfer to the analysis cen 
tre is very essential. Links to data acquisition functions shall be 
established, tested and regularly up-dated to encompass the de 
velopment of the remote sensing and telecommunications tech 
niques. Satellite communications may be of advantage to avoid 
break-down due to the disasters that shall be monitored. This 
applies to all types of data and reporting which are part of the 
management system. 
(g) With the same token the infrastructure between the 
analysis centre and the authorities responsible for rescue and 
relief shall be prepared to cover various levels of disasters. This 
includes also the communication to rescue teams in the field 
where ordinary means of communications may have broken 
down. 
The above points deal with actions that are needed to ensure an 
efficient and timely use of remote sensing data. They may appear 
obvious and one might wonder why such system has not been 
implemented already. Probably w'e are faced with the same prob 
lem as before when attempting to transfer remote sensing tech 
niques to various geophysical disciplines. Funds may be obtained 
relatively easy for research and initial developments but it seems 
more difficult to secure funds for technology transfer. The so- 
called value-added industry that lias appeared the last years may 
offer a solution to this schism. 
A number of demonstrations liave been made and described 
widely but I like to suggest that to obtain full benefit they should 
be carried out with full participation of the potential end user. In 
the process of transfer of know-how it is a must that human and 
organisational aspects are taken seriously into account. 
Flooding may occur in a region that is part of several countries. 
Therefore, the system should be organised to encompass authori 
ties from all countries to form a true regional system. This will 
most likely complicate the matter due to different ‘cultures’ and 
languages. On the other hand, with the present progress of inte 
gration of a number of functions in European countries this may 
not be an insurmountable task. 
The EUR-OPA Major Risks Agreement initiated by the Council 
of Europe may be instrumental in this respect. 1/ seeks a close 
dialogue w'ith the civil protection authorities in order to promote 
the use of remote sensing data for rapid assessment of disasters 
and assisting rescue teams. 
Insurance companies and public planning 
I cannot help presenting a controversial aspect of a plan as the 
one described. The simulations suggested will be very useful for 
evaluation of risks to people and in money and might interest 
research departments of the larger insurance companies. They 
may also interest public departments for land development and 
lead to new policies as to housing and industrial developments in 
areas/regions that are likely to be hit by major flooding. Authori 
ties may seriously consider the result of the simulations for the