| dc.description.abstract | The occurrence of drought in Indonesia is almost an event that occurs every year. This drought affects many sectors and tends to occur more frequently. Indonesia is in 4th position (behind India, Japan and China) with potential losses of up to US 50 billion from the top 10 countries in the Asia Pacific region that have the potential to lose due to natural disasters and drought. Drought tends not to have a universal definition. The impact of drought is generally non-structural in nature so that the area of impact spreads wider than the damage caused by other natural hazards. The potential for drought can also increase if regional scale climate phenomena such as El Nino and positive Indian Ocean Dipole Mode appear. Drought conditions that are getting worse have resulted in a large number of forest and land fires occurring. The incidence of forest and land fires has had a very significant impact. The total losses due to forest and land fires in 2019 borne by Indonesia reached US5.2 billion, equivalent to Rp 72.95 trillion. At least 900,000 residents have had their respiratory health affected, and have had operational problems at 12 national airports. It is estimated that 620,201 hectares of forest and land were burned with the impact of losses of up to US157 million in damage to assets and US5 billion due to lost potential from economic activity. Forest fires can be identified by hotspots which can be obtained by utilizing MODIS (Moderate Resolution Imaging Spectro-radiometer) remote sensing measurements on the Terra-Aqua satellite and / or by utilizing Himawari satellite imagery. Observing the sequence of events as described above, apart from anthropogenic factors, from a climatic perspective it is very important to be able to know the conditions of climate dynamics that exist in Indonesia (and the conditions of the dynamics of the global climate surrounding Indonesian territory which affect the dynamics of the atmosphere in Indonesia) as one trigger point in the process of starting a drought which could have an impact on the potential for forest and land fires. By utilizing the numerical weather prediction model output from the ECMWF (European Center for Medium-Range Weather Forecasts) it can be used to identify potential drought. To reduce the potential bias from the model output, validation and correction are necessary. The output of the model was validated and corrected using the Piani method and the Brier Skill Score to the climate observation data in the field from 774 selected rain observation posts. This step produces corrected ECMWF rainfall data which has a high correlation of 0.7844 at the significant level α = 0.05 with an RMSE value of 87.6 on the observed rainfall data. This validation and correction results in an increase in the output performance of the corrected model. This is indicated by the monthly variability of rainfall, which is initially not detected to be detected, and the bias correction value is smaller. The output of the corrected ECMWF model is used to calculate the drought index using the SPI (The Standardized Precipitation Index) method to describe meteorological drought conditions in Sumatra. The drought index values in Sumatra were in the range of -3.9 - 3.4. The SPI drought index value is close to positive, generally observed in the central part of Sumatra towards the north and the negative SPI drought index value is generally observed in the central eastern part of Sumatra to the south. Temporally, the concentration distribution of the SPI drought index occurs in the mid-year months for the central to northern Sumatra region; for the central eastern part of Sumatra to the south generally appears in the months of February-March and will reappear in the months of July-September. In conditions of the El Nino climate phenomenon (seen from the Oceanic Nino Index - ONI) and the Indian Ocean Dipole Mode (seen from the Dipole Mode Index - DMI), the drought index is more correlated with the emergence of the El Nino phenomenon than the Indian Ocean Dipole Mode phenomenon. In the appearance of El Nino, the correlation value between SPI and ONI is the most negative -0.673 at the significant level α = 0.05. This can also result in the emergence of more hotspots (> 50 events per month). At the appearance of the Indian Ocean Dipole Mode, the correlation value between SPI and DMI is the most negative of -0.327 at the significant level α = 0.05. The distribution of the concentration of the correlation value between ONI and SPI is similar to the distribution pattern with the correlation value between DMI and SPI, but the distribution of the correlation value between ONI and SPI is stronger. Noting that climate dynamics play a role in the emergence of drought events, furthermore, the derivative drought index from rainfall observations is used to identify potential forest and land fires. This potential is viewed from climatic factors, land cover factors, and peat land distribution factors. The results of calculations using fuzzy logic and ArcGIS software FuzzyOverlay tools, the results of superposition analysis of the impact of climatic conditions, distribution of land cover, and distribution of peatlands produce a representative map of potential forest and land fires. Areas with increasing potential levels generally occur in central eastern Sumatra extending from Riau, Jambi, South Sumatra, and parts of northern Lampung. From the color gradations on the map, it can be indicated that the areas with medium, medium, and high potential levels. Utilizing corrected predictive ECMWF output data, these data can be used to estimate drought indices which in turn can be used to produce maps of potential forest and land fires. It is hoped that the information on this map can be used to anticipate potential forest and land fires. | id |
