ABSTRACT ANALYSIS OF HOTSPOT SPREAD IN THERMAL CHANNELS OF SATELLITE HIMAWARI-8 USING PSEUDO COLORING METHOD (STUDY CASE : FOREST FIRE ON SUMATERA ISLAND IN 2015)

• Main Author: RAHAYU, NURHAYATI
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/29787
• Institution: Institut Teknologi Bandung

Forest and land fires are one type of massive environmental degradation threat to species in the surrounding ecosystem. Damage also affects the economy, health, education and infrastructure. Events of forest fires that occurred in the Year 2015 on the island of Sumatera recorded as the largest fire disaster in Indonesia because it hit the forest and land area of 1,658,584 hectares. In this regard, efforts to identify the increase in Land Surface Temperature (LST) indicated as hotspots should be done so that fires can be detected earlier. <br /> <br /> <br /> <br /> Information on hot spots at present, including data processing from TERRA/AQUA Satellite and NOAA18 Satellite. The TERRA/AQUA satellite is a polar orbital pattern satellite that has a spatial resolution of 1,1 km and a temporal 12 hour resolution utilizing thermal channel 21/22(3,929-3,989 ?m), 31(10,78-11,28 ?m) and 32(11,77-12,27 ?m). The thermal day temperature threshold at more than 46,85°C and nighttime is more than 36,85°C. The NOAA18 satellites generally have similarities with TERRA/AQUA satellites in the orbital pattern, spatial resolution and temporal resolution by utilizing thermal channel 3B (3,55-3,93 ?m), 4(10,3-11,3 ?m) and 5 (11,5-12,5 ?m). Temperature threshold during daytime is more than 41,85°C and nighttime is more than 36,85°C. <br /> <br /> <br /> <br /> Due to the importance of the detection of hotspots, weather sensors that have high temporal resolution and adequate spatial resolution in static and continuous observation areas, such as the Himawari-8 Satellite. To detect hotspots, IR4 thermal channel channels (3,5-4,0 ?m) and IR1 channels (10,3-11,3 ?m) are used. The thermal day temperature threshold at more than 41,85°C and nighttime is more than 36,85°C. Methods for identifying hotspots and their location are used Pseudo Coloring Methods by involving IR4 channels as red component, IR4-IR1 (Split Window) as a component of green color and IR1 as a blue color component. In order to use a hotspot color image to process a further analysis or dissemination to various media without degrading the image quality, it is necessary to transform the colored image into another color space tailored to the needs of the user. Color spaces include RGB fictitious color space (CIE XYZ) for image analysis and other multimedia products, National Televisions System Committee (NTSC) color space for news broadcast on digital or video television, Cyan Magenta Yellow Black (CMYK) color space for printing, The Hue Saturation Intensity (HSI) color space for advanced image analysis and color chamberance Blue Chrominance Red (YCbCr) luminance for digital video processing. <br /> <br /> <br /> <br /> From the data processing of Satellite Himawari-8 shows that on the island of Sumatra in 2015 there has been 21.434 hotspots where 47% occurred in South Sumatra Province, 42% occurred in Lampung Province and 11% occurred in other provinces. A comparison analysis of hotspot identification data from both satellites to the AAWS weather sensor shows that in September and October 2015 the absolute value of the Himawari-8 Satellite error is lower than that of the TERRA/AQUA Satellite against field observation data from the AAWS weather sensor. This means that for SPT thresholds in the field of more than 38°C, can make the information from Himawari-8 as the basic reference of hotspot determination. While information from the TERRA/AQUA Satellite indicates that the SPT expressed as the hotspot temperature becomes irrelevant because the temperature range of 23,79°C - 36.83°C is not a hotspot temperature. <br /> <br /> <br /> <br /> Comparative analysis of hotspot identification from the Himawari-8 Satellite data to TERRA/AQUA satellites shows that RMSE measurements of the Himawari-8 Satellite to the TERRA/AQUA satellite are quite high. In determining the accuracy of the hotspots temperature according to the day and night limits on both satellites, the day time temperature threshold for the Himawari-8 Satellite is 41,85°C while the day temperature threshold for the TERRA/AQUA is 36,85°C. The Himawari-8 satellite shows a temperature of 42,04°C – 44,29°C, and the TERRA/AQUA satellite shows a temperature of 21,37°C – 40,86°C is the hotspots temperature range with the highest appearance. While the hotspots temperature threshold at night time for the Himawari-8 Satellite and TERRA/AQUA Satellite is 36,85°C. Each hotspots information source has a different temperature distribution of hotspots. The Himawari-8 satellite shows a temperature of 37,24°C – 46,43°C and the TERRA/AQUA satellite shows the temperature of 6,21°C – 30,86°C is the hotspots temperature range with the highest appearance. <br /> <br /> <br /> <br /> The result of the measurement of the color image quality of hotspot as much as 630 image data during the study period on color image transformed by color space with RMSE, PSNR and MAE shows 100% data with YCbCr color space has superior quality compared to CIE XYZ, NTSC, CMYK, And HSI. Meanwhile, based on the measurement of Universal Objective Image Quality Index (UOIQI) shows the CIE XYZ color space is at the lowest position in July of 0% and the highest in September as much as 45,22%, NTSC color space is at the lowest position in September of 8,92% and the highest In December as much as 63,16%, while the color space YCbCr was in the lowest position in December as much as 36,84% and the highest in July as much as 59,38%. CIE XYZ is best used when cloud cover is below 40% and SPT is more than 36,85°C, NTSC is the opposite of CIE XYZ which is very well used when cloud cover is over 40% and SPT is less than 36,85°C and YCbCr is highly adaptive to changing conditions any environment.