EA STE Ty 
"e 
EET 
aa 
of the CEO is substantial. Especially, since there is a lot of 
stimulation and information required to demonstrate to these 
organisations the advantages of using satellite EO data. 
Different users are working in different phases and types of 
disasters. Disaster management comprehends the aspects of 
prevention (assessing vulnerability or hazard analysis), 
monitoring (disaster warning or early warning) and disaster 
relief (rescue). Remote sensing has made significant 
contributions to these phases, notably in areas such as 
integrating land-use and hazard maps. 
2.1 Communication 
Several ways of communication have been used throughout the 
study to inform potential users about EOFIND. The 
INTERNET was used as basic information tool as much as 
possible. 
The results of the catastrophes that are analyzed in the EOFIND 
study (storms, floods, forest fires) are published on 
INTERNET. They can be observed in the CEO Enabling 
Service - European Wide Service Exchange 
(http://ewse.ceo.org/) - and in the Kayser-Threde-homepage 
(http://www.kayser-threde.de/eofind). 
During the course of the EOFIND study the EWSE system at 
the Joint Research Centre in ISPRA was used to place 
information about Kayser-Threde into the exchange database. 
The major advantage of the EWSE system is that companies 
willing to provide information are able to change the 
information from a web browser anywhere without the need to 
contact responsible persons at the JRC. The EWSE system may 
prove as a valuable tool for information retrieval. This is due to 
its decentralized approach and the fact that more and more 
people in general and thus more potential users of EWSE are 
using INTERNET as an information resource. 
2.2 User requirements 
Satellite EO data can be a useful tool in all types and phases of 
a disaster. The typical users are attracted by this tool but there 
are still some obstacles that have to be overcome in the future: 
m Insufficient geometric and/or radiometric resolution 
m Frequent cloud coverage (for sensors working in the optical 
domain of the electromagnetic spectrum) 
m Data availability in terms of time lag or data acquisition, 
processing and delivery 
m Insufficient expertise on data interpretation 
m Insufficient technical facilities for data evaluation under 
disaster conditions. 
Thus, EOFIND investigates the possibilities and obstacles of 
modern remote sensing technology and electronic networking 
for data dissemination in detail. The requirements on the 
temporal, spatial and spectral resolution as well as on the 
coverage area differ too much from one type of disaster to 
another to specify them in general. 
They are addressed in detail in the case studies for the three 
different types of disasters. 
208 
In general the main user needs that have been identified in this 
study can be summarized as follows: 
m Improvement of geographical cartography of exposed areas 
m Improvement of vulnerability data availability 
m Improvement of routine risk monitoring 
m Improvement of monitoring during crisis 
m Development of post-crisis monitoring 
m Development of forecasting models 
m Reinforcement and extension of monitoring networks. 
3 CASE STUDIES 
The case studies concentrate on the three major natural risks 
(fire, storm, flood) mentioned already above. Their impacts and 
the needs of managers (users) for prevention, emergency 
management and relief are described. This is followed by a 
discussion of the possibilities of modern remote sensing and the 
role of CEO in the framework of an European Network 
concerning natural hazards. During the EOFIND study users 
have been directly involved. Their needs and current barriers in 
using EO data for the three disaster types have been analysed. 
The first case study concentrated on forest fires in Greece. 
Landsat TM data were used to assess the current vegetation 
cover and the related fuels for risk assessment and the damage 
caused by wild fires on the Sithonia Peninsula, Halkidiki, 
Greece. 
The second case study dealt with storm damages caused by a 
hurricane in Central Europe, using multitemporal Landsat TM 
data. The satellite images provided important preliminary 
information about the areas damaged on wide scale. This 
information is needed to start direct response actions on 
regional and national level and to get an overview over of the 
economic influence on the European timber market. 
In the third study ERS-1 SAR images were used to interpret the 
damage caused by a flood in 1995 in Central Europe. The 
Rapid Information Dissemination System (RAIDS) of MATRA 
was employed to provide EO-data as fast as possible. The 
produced flood map was used by naive end users as well as 
commercial users for damage and risk assessment. In the flood 
study special emphasis was put on timeliness and operationality 
to demonstrate fast information delivery after a disaster for 
immediate help actions. The respective data flux from a data 
provider to a value adding (VA) company and back to potential 
customers has been analysed. 
In the following the experiences made in the three case studies 
are outlined in short. 
3.1 Forest Fires 
As the reasons for the occurrence of forest fires are manifold 
and their impacts very complex, depending on the local 
situation the issues of fire management are very complex, too. 
Thus EO data will play only a minor, but nevertheless 
important part of any Fire Management System (FMS). 
In the following cases EO data can be used on a semi- 
operational level as was proven by several pilot projects: 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B2. Vienna 1996 
ml 
As 
mu 
wel