The ENVIREF partners work on defining user-needs, developing new
and improved products and methods, and demonstrating satellite EO
products for the relief community in order to make relief operations
more efficient. To ensure input from end-users, several international
relief agencies are involved in ENVIREF, such as United Nations
High Commissioner for Refugees (UNHCR), International Federation
of Red Cross and Red Crescent Societies (IFRC), Medecins Sans Frontieres
(MSF) and Oxfam UK.



Satellite imagery of various types were acquired and analyzed for
three different study areas, including several refugee camp settlements
in the Kosovo-Albania-Macedonia border region in Europe, the southeastern
part of Nepal in Asia, and the eastern part of Kenya in Africa.
Each area represents different climatic, environmental and refugee-related
conditions.



Medium-resolution satellite sensors (spatial resolution from 6 to
30m) including LANDSAT TM/ETM+, SPOT, and IRS-1D were used to produce
maps of infrastructure, water resources and vegetation cover at
an overview level ranging from scale 1:15.000 to 1:250.000. Images
from all these sensors can be used to accurately measure the location
and extent of a refugee camp, although the spatial resolution is
not good enough to reveal any details within a camp area, e.g. to
identify independent residences, roads, and so forth.

Water bodies derived from the satellite imagery were combined with
digital elevation data in order to determine river flow direction.
Vegetation cover mapping was performed both with and without additional
information, e.g. from field observations or available vegetation
maps. Without any supporting information about the vegetation in
an area, only very general assumptions can be made, such as the
location and distribution of forest. These maps can, on the other
hand, be prepared very quickly, and also be used to plan where in
situ measurements should be performed. With additional information
about the vegetation, more detailed maps can be prepared, showing,
for example, the distribution of various vegetation types.

Satellite images with only 1m spatial resolution from the IKONOS
satellite were used to perform mapping of details such as roads,
footpaths, separate buildings, tents, and water sites inside the
refugee camp area. This is the first satellite to deliver very high-resolution
satellite imagery on the commercial market. An especially important
result of the IKONOS-data analysis is the potential for identifying
single tents and buildings inside a camp area. For the investigated
camp settlement in Nepal, close to 100% of all present tents and
buildings were identified and could be counted in the image. This
information can be used to make better estimates of the number of
people living in a camp, which is important for determining the
amount of food and medicine needed. IKONOS images also have the
potential to be used for detailed planning of refugee camps, including
expansions, and assessments of environmental changes in the immediate
surroundings.

We find that by providing fast access to general thematic information
over a larger region the LANDSAT-7 ETM+ sensor could be a valuable
data source for the relief community. The sensor could be especially
helpful due to both the global data archive and the low price of
only US$600 per image. For performing more detailed analysis, which
is especially needed in the later stages of a crisis situation,
images at a resolution available from the IKONOS sensor are needed.

The data set for the Nepal study includes one IKONOS scene (pan+ms),
one LANDSAT-7 ETM+ scene, two SPOT XS scenes, two ERS SAR scenes,
as well as several topographic maps at scale 1:25,000, including
a DTM for the area surrounding the Beldangi camp area. Some aerial
photographs over the Beldangi camp area could also be obtained.

In addition to the above mentioned, the analysis of the IKONOS
image from the Nepal case study includes land use classification,
automatic detection of buildings and shelters within the camp, as
well as determination of best-camp-location using GIS.