A Fire Information System for East Kalimantan, Indonesia
(IFFN No. 23 – December 2000, p. 29-32)
After the severe fire episodes during the El Niño-Southern Oscillation (ENSO) events of 1982-83, 1991 and 1994 a prolonged and extremely severe fire season occurred during the last ENSO of 1997-98 (Goldammer et al. 1996, Goldammer 1999). The Indonesian province of East Kalimantan was the area worst affected by the fires. The daily NOAA-AVHRR (National Oceanic Atmosphere Administration – Advanced Very High Resolution Radiometer) data received and processed by the Integrated Forest Fire Management (IFFM) Project of East Kalimantan show that by the time the rain started at the beginning of May 1998 almost the whole basin area in the district Kutai had been burned. Between February and March 1999 prevailing strong winds, combined with the effects of drought, allowed the rapid spread of wildfires into primary, logged-over and secondary forests, degraded vegetation, plantations and other land use systems. By mid March it became obvious that efforts being made to put out these fires had had only marginal effects. At that time the IFFM NOAA-AVHRR receiving station recorded more than 2000 High-Temperature Events (HTE) (Siegert and Hoffmann, in press). The two MoFEC-GTZ projects located East Kalimantan, the Sustainable Forest Management Project (SFMP) and IFFM conducted a study using ERS-SAR space borne radar to disclose the size of the fire affected area for the entire province according to land uses. The investigations revealed that a total of 5.2 million hectares (ha) of different vegetation types had been affected by fire (Hoffmann et al., 1999).
The aim of the IFFM project is to support and build up a fire management centre which will create and command the structures that are needed to overcome the fire problem in the province of East Kalimantan. Jointly with fire prevention and suppression the IFFM project currently develops a computer-supported Fire Information System (FIS) in an attempt to prevent such fire disasters in the future. The FIS integrates fire-related data and fire information such as monitoring and detection, to determine fire danger criteria as well as data analysis and information dissemination to support fire management, prevention, suppression and policy decisions. The proposed Fire Information System for East Kalimantan contains three major parts: (1) input data (2) output data and (3) information dissemination.
Design of the Fire Information System (FIS)
The Fire Information System (FIS) is a system, that manages spatial fire-related data and information in an integrated manner. Fire information components are fire monitoring and detection, determining fire danger criteria, data analysis and information dissemination.
Static and dynamic information is used as FIS input data. Static data are exclusively those that change over a long time such as road infrastructure or topographic data. Also data which does not change weekly or monthly such as forest concession or plantation boundaries, transmigration/settlement data and additionally vegetation data belong to the static data. Dynamic data are those that change continuously like climate data, fuel conditions and fire distribution.
Some of the static and dynamic data are derived from satellite data, such as the HTE depicted by the NOAA-AVHRR and processed by IFFM for near-real time fire information and distribution as well as historical fire data (Hoffmann 1998). The Normalized Difference Vegetation Index (NDVI) derived from AVHRR data as a component of the FIS monitors the seasonal changes in living vegetation moisture and can be used to forecast the long-term seasonal trends in fire potential in East Kalimantan. Vegetation and land use system data can be obtained from Landsat TM 5 (Thematic Mapper (Roy et al. 1991, Saxena et al. 1994, Goldammer et al. 1997) as well as complemented and expanded by the ERS-2 SAR (European Radar Satellite-2-Synthetic Aperture Radar) sensor which is able to penetrate clouds and haze and provides a high spatial resolution (25 m) (Kuntz and Siegert 1999, Siegert and Hoffmann 1999). Furthermore with ERS-SAR data, burned areas, which are future fire prone areas, can be reliably discerned at a scale of 1:200,000 (Siegert and Ruecker 1999, Siegert and Hoffmann 1999, Hoffmann et al. 1999).
Static data such as land use (e.g. concession boundaries) as well as transmigration/settlement data have to be taken from various agencies responsible for mapping. All data must be available in digital format to integrate them in the FIS. Simple topographic data for East Kalimantan available can be used to add information on different elevations and the resulting fire conditions.
One important aspect for preventing future fire disasters is the level of awareness which can be gained by an early warning system such as Fire Danger Rating (FDR), based on the Keetch-Byram Drought Index (KBDI) (Keetch and Byram 1968). IFFM has been working with FDR in East-Kalimantan since 1995 (Deeming, 1995). The Fire Danger Rating indicates the fire danger or dryness for a variety of meteorological conditions and is based on only a few meteorological data (see Buchholz and Weidemann, this volume).
In order to support optimal resource allocation and pre-positioning of equipment and human resources such information will be integrated to sustain mobilization planning of the local fire management centres.
The most important output from the FIS is the early warning information which at first stage comes in form of fire danger rating maps. Weather data are interpolated to generate FDR maps for the whole province of East Kalimantan (see Buchholz and Weidemann, this volume). It will be the basis information for development of fire hazard and fire risk maps.
In addition, land use as well as settlement and road infrastructure information are included to indicate fire risk zones. To a certain extent, the 1997-98 fires were started due to land speculation and large scale forest conversion into plantations. Nevertheless, also in settlement / transmigration areas fires started caused by land use conflicts, ethnic conflicts or hunting of deer and turtles as well as carelessness or agricultural activities. The AVHRR HTE distribution of 1997 clearly demonstrates that the fires from the first period were mostly along rivers and streets and thus set by farmers and accidentally through carelessness. This shows that settlements and transmigration areas play an important role and hence must be considered in the FIS to reveal fire risk areas. Meanwhile the 1998 fire disaster went out of control additionally because of large scale land clearing and forest conversion, needs this a revision of the spatial regional planning. However, towards prevention campaigns and awareness messages transmigration/settlement information assist by identification of sensitive areas were prevention measures should be focused. This will make prevention activities more effective and valuable. Beside that, land status data allows for detection of AVHRR HTE in forest concessions and plantations and is a first step towards more precise locating of fires as well as knowing where to take suppression action.
For strategic planning, having the current information about equipment and human resources at the district centres and the surrounding concessions available would optimise the resource allocation, hence mobilization of suppression activities can be efficiently and economically planned and prepared. Together with information about fire risk areas ensures this effective distribution of resources for fire management where it is needed.
The most crucial and critical issue of the FIS is rapid and efficient information dissemination. Detection and monitoring information and decisions made based on the FIS should be distributed via telecommunication technology like radio, fax machine or e-mail to the district fire centres. Meanwhile monitoring results (e.g. weather data or fire occurrence) from local level can contribute to the FIS at the provincial level. This presumes a commando structure of responsibility for fire management at the provincial level. It is imperative to know who the involved parties and target groups, including number of concession holders, to build up a communication network for dissemination of fire relevant information. Figure 1 shows the possible information flow and values of fire relevant data of the proposed Fire Information System of East Kalimantan.
Fig. 1. Possible information flow of the FIS
The supposed involved parties at the provincial and the local level as well as data quality and quantity are shown with different arrow/box structures. Since the IFFM project is implemented under the structure of the provincial forestry agencies these are the major involved parties at the provincial level. The two provincial forestry agencies (DINAS and Kanwil Kehutanan) in conjunction with IFFM processes, analyses and provides fire relevant data to the local government and more detailed to district forest fire centres (Cabang Dinas Kehutanan). This not only contains reports with e.g., fire locations derived from NOAA-AVHRR but also detailed maps of where the fire risk zone in the district are located and what the suppression capabilities are.
In combination with the FDR information as well as the vegetation conditions fire risk information will gradually change over the year in relation to the annual natural conditions. The district fire centres disseminate the information to the local community to support the fire village crews and awareness campaigns (Abberger 1999). Additionally the surrounding timber concession and plantation companies should be involved since almost the whole province is covered by forest concessions and forest and estate crop plantations, thus having an important role to play within the fire problem. They are encouraged to calculate their own fire danger rating and to build up communication and action networks with the neighbouring companies and the local villages. Vice versa the information should flow back to the provincial fire management centre in type of weather data or rather FDR index data and fire reports. These, in conjunction with the fire management centre, are the important parties where a communication plan is needed. A communication plan contains first what kind of information is needed and secondly what kind of communication network is available to distribute the information. In consequence a proficient information network from the provincial to the local level/local communities and from there to the concessionaires and vice versa must be build up to guarantee the data flow in all direction both horizontal and vertical.
Further involved parties are the Meteorological and Geophysical Service of Indonesia (BMG) which is supposed to calculate the FDR index for all existing BMG weather stations in East Kalimantan. Furthermore e.g., BMG should transfer general awareness messages to the local TV or radio in form of simple fire danger fore castings telling messages in which areas are medium and high fire danger to point out the dangerous use of fire to the people in these areas. Moreover, the provincial environmental agency (Bapedalda) has to be informed about the monitoring results of fire locations and early warning.
Despite the fact that fire management programs have been conducted in Indonesia since the mid-1980s, there were no country-wide operational fire information and management systems available to cope with the fires during the 1997-98 ENSO. The development of a Fire Information System as is currently underway in IFFM project in East Kalimantan is a move in the right direction. All IFFM modules are providers for and users of information of a FIS. However, without a strong IFFM implementation program the integrated use of information within a FIS as a precursors to the occurrence and effects of fire is of limited value. Most important is fire prevention based on a participatory approach which includes all stakeholders potentially affected by fire, efficient law enforcement, improvement of fire management skills at all levels of fire users. This means support for efforts towards Integrated Forest Fire Management through all involved forestry and regional departments. Finally, establish a consistent land-use policy which includes considerations of fire and smoke management.
Abberger, H. 1999. Fire prevention activities from February – July 1998. IFFM Homepage http://www.iffm.or.id
Deeming, J.E. 1995. Development of a Fire Danger Rating System for East Kalimantan. IFFM Report (mimeo).
Goldammer, J.G. 1999. Environmental problems arising from land use, climate variability, fire and smog in Indonesia: Development of policies and strategies for land use and fire management. In: WMO Workshop on Regional Transboundary Smoke and Haze in Southeast Asia, Singapore, 2-5 June 1998, Vol. 2 (G.R.Carmichael, comp.), 13-88. World Meteorological Organization, Global Atmosphere Watch Report Ser. No. 131, WMO TD No. 948, Geneva, 346 p.
Goldammer, J.G., and B. Seibert. 1990. The impact of droughts and forest fires on tropical lowland rain forest of Eastern Borneo. In: Fire in the tropical biota. Ecosystem processes and global challenges (J.G. Goldammer, ed.), 11-31. Ecological Studies 84, Springer-Verlag, Berlin-Heidelberg-New York, 497 p.
Goldammer, J.G., B. Seibert, and W. Schindele. 1996. Fire in dipterocarp forests. In: Dipterocarp forest ecosystems: Towards sustainable management (A.Schulte and D.Schöne, eds.), 155-185. World Scientific Publ., Singapore-New Jersey-London-Hongkong, 666 pp.
Hoffmann, A.A., A. Hinrichs, and F. Siegert, 1999. Fire damage in East Kalimantan in 1997/98 related to land use and vegetation classes: Satellite radar inventory and proposals for further actions. With contributions from Ruecker G., Christy L. & Yuliarso B.
Keetch, J.J., and G.M. Byram. 1968. A drought index for forest fire control. USDA Forest Service, South Eastern Forest Exp. Sta. Res. Pap. SE-38.
Kuntz S., and F. Siegert. 1999. Monitoring of deforestation and land use in Indonesia with multitemporal ERS data. Int. J. Remote Sensing, 14, 2835-2853.
Roy, P.S., B.K. Rangaanath, P.G. Diwakar, T.P.S. Vohra, S.K. Bhan, I.J. Singh, and V.C. Pandian. 1991. Tropical forest type mapping and monitoring using remote sensing. Int. J. Remote Sensing, 12, 2205-2225.
Saxena, K.G., A.K. Tiwari, M.C. Porwal, and A.R.R. Menon. 1992. Vegetation maps, mapping needs and scopes of digital processing of Landsat Thematic Mapper data in tropical region of South West India. Int. J. Remote Sensing, 13, 2017-2037.
Siegert F., and A.A. Hoffmann. 2000. The 1998 Forest fires in East Kalimantan (Indonesia): A quantitative evaluation using high resolution, multitemporal ERS-2 SAR Images and NOAA-AVHRR hotspot data. Remote Sensing of the Environment 72, 64-77.
Siegert F. , and G. Rücker. 1999. Use of multitemporal ERS-2 SAR images for identification of burned scars in South East Asian tropical rainforest. Int. J. Remote Sensing 21, 831-837.
Anja A. Hoffmann
Integrated Forest Fire Management IFFMGTZ,
Jln Harmonika, Perkantoran Dinas Kehutanan,