You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Proceedings, XXth congress

u 2004

Esra Erten* , Vedat Kurgun" , Nebiye Musaoglu®
* ITU, Institute of Informatics, 34469 Maslak Istanbul, TURKEY —esra@nik.com.tr
? ITU, Civil Engineering Faculty, Remote Sensing Division, 34469 Maslak Istabul, TURKEY- vkurgun@hotmail.com
“ITU, Civil Engineering Faculty, Remote Sensing Division, 34469 Maslak Istabul, TURKEY-
KEY WORDS: Landsat, GIS, Forest Fire, Analysis, Disaster
Forest fires are one of the major natural risk in the Mediterranean countries. In such areas, fires occur frequently and there is a need
for supranational approaches that analyse wide scenarios of factors involved and global fire effects. It is impossible to control nature,
but is possible to map forest fire risk zone and thereby minimise the frequency of fire. Fire risk must involve both ignition and
spreading risk. It is necessary to be able to estimate the spread of fire starting in any stand of a forest, giving the burning conditions.
Spreading of forest fire can impose a threat to the natural coverage of land and safety of population. Early detection of forest fires is
essential in reduction of fire damage. Satellite data plays a vital role in identifying and mapping forest fire and in recording the
frequency at which different vegetation types/zones are effected. A geographic information system (GIS) can be used effectively to
combine different forest-fire causing factors for demanding the forest fire risk zone map. A risk model for fire spreading is set up
for Gallipoli Peninsula as a pilot zone, because it continually faces a forest fire problem. It is based upon a combination of remote
sensing and GIS data. In this study, the LANDSAT satellite images were used. In order to search the effects of the fire occurred in
the region on the 25" July 1994, the satellite images both before the fire and after the fire were used. Parameters that effect the fire
such as topography and vegetation with the other land use information including population, settlements, forest fire towers, fire
stations, intervention places, the characteristics of the staff that will intervene and transportation were integrated within GIS. The
shortest way of intervention during the disaster and the areas to be emptied were questioned. For these analyses ARCGIS software
was used. Land use information were obtained from the satellite images in this study. In this phase the distinction of spices in the
forest was determined using supervised classification. The lands that have priorities in case of fire were decided by combining the
moisture of the land and slope classes that were determined by conventional approaches with the satellite images. The results of the
analysis were shown by reports and graphics. The test region results should be applied all over Turkey.
1. INTRODUCTION fires have also been effectively used in this study.
Understanding the behaviour of forest fire, the factors that
It is possible both to control nature and also to map risk zone contribute to making an environment fire prone, and the factors
and thereby to minimise the frequency of fire and avert damage. that influence fire behaviour are essential for forest fire
Forest fire risk zones are locations where a fire is likely to start, (Chuvieco and Congalton, 1989).
and from where it can easily spread to other areas. A precise
evaluation of forest fire problems and decision on solutions can In the present study, an attempt was made to prepare a forest
only be satisfactory when a fire risk zone mapping is available fire risk zone map by integrating a satellite image,
(Jaiswal et al, 2002). topographical and other ancillary data from a geographic
information system (GIS) for the Gallipoli where is the most
In Turkey, forest fire is still one of the greatest natural hazard forest fire sensitive area in Turkey. This study is also an attempt
problems. According to the study, last six decades, 1.504.245 to exploit the capabilities of remote sensing and GIS techniques
ha of the forests has been affected by fire. In other words, it and to suggest an appropriate methodology for forest fire risk
shows annual affecting forest areas over 24.000 ha. In addition zone mapping. Such maps will help forest department officials
to this, along the cost line from Antalya to Istanbul, prevent or minimize fire risk activities within the forest and
approximately 12 billion hectares forests are in the first order take proper action when fire breaks out (Chuvieco and Sales,
fire sensitive region. 1996).
For these reasons, 133627 km roads and 17219 km fire safety
bands have been constructed to prevent forest. fire; moreover,
forests are watched by 779 fire watchtowers and 17561 unit
various radios are used to communicate among these
watchtowers. Fire Operation Centre was established in Ankara
in 1997. This centre serves 24 hours to monitor forest fire and
to coordinate logical support or to communicate.
Satellite remote sensing has opened up opportunities for
qualitative analyses of forest and other ecosystems at all
geographic and spatial scales. Monitoring and detecting forest