Utilization of READY-ARL NOAA data and CMORPH for land and forest fire risk model development in Central Kalimantan
Pemanfaatan data READY-ARL NOAA dan CMORPH untuk pengembangan model risiko kebakaran hutan dan Lahan di Kalimantan Tengah
View/ Open
Date
2012Author
Prasasti, Indah
Boer, Rizaldi
Ardiansyah, Muhammad
Buono, Agus
Syaufina, Lailan
Metadata
Show full item recordAbstract
Land and forest fires are one of the many causes of land degradation in Central Kalimantan. The utilization of remote sensing data, particularly READY-ARL NOAA and CMORPH data, is helpful in providing climate observation data. The objectives of this study are: 1) to analyze the relationship between the surface observation data and the READY-ARL NOAA and CMORPH (CPC Morphing) data by using Partial Least Square (PLS) Method to extract climate data from the satellite, 2) to develop the FDRS (Fire Danger Rating System) indices by using READY-ARL NOAA, CMORPH and hotspot data derived from the satellite data, 3) to develop an estimation model for burned area from hotspot, rainfall condition, and FDRS indices, and 4) to develop fire risk prediction model. The result of this study indicates that the READY-ARL NOAA and CMORPH data have the potential to make climate data estimation and are relatively good as FDRS (SPBK) data input. The use of PLS method is much better in generating a model estimation than simple regression. Precipitation accumulation for two months prior to fire occurrence and drought condition have correlation with the burned area. There is a correlation between the total number of fire hotspot and a series of days without rain around one to two months prior to fire occurrence. In addition, this study also found that the burned area in Central Kalimantan will increase if the drought code exceeds 500 point. Burned area can be estimated by using the following formulas: Burned Area (Ha) = 5.13 – 21.7 (CH2bl – 93) (R-sq = 67.2%) and this formula: Burned Area (Ha) = -62.9 + 5.14 (DC – 500) (R-sq = 58%) where CH2bl = precipitation accumulation for two months prior to fire occurrence and DC = drought code. The forecasts of fire occurrence probability can be determined by using a precipitation accumulation for two months prior to fire occurrence and Monte Carlo simulation. Efforts to anticipate and address the fire risk should be carried out as early as possible, i.e. two months in advance if the probability of fire risk has exceeded the value of 40%.