Author(s): Grace Puyang Ng; Yoshiya Touge; So Kazama
Linked Author(s): So Kazama
Keywords: No Keywords
Abstract: A large forest fire broke out in Kamaishi, Iwate Prefecture on 8 May 2017 with a burnt area of 413 ha which is larger than the total burnt area for the whole of Japan in 2016. The extensive fire was due to strong presence of wind and low accessibility of vehicles into the burnt area. Historically, many research use satellite analysis to detect burnt area by forest fire but most research focus on estimation of burnt area to detect affected and nonaffected area. However, the fire severity in forest varies based on fire intensity, type of vegetation, topography, weather conditions and others. Since forest fire affects terrestrial hydrological and material cycle, the variation of fire severity cause various effects on ecosystem both in the forest and whole basin. Therefore in this research, the Normalized Difference Vegetation Index (NDVI) based phenology in different fire severity and type of trees in Kamaishi were compared to examine the possibility of using NDVI from Terra and Landsat 8. Firstly, five points in Kamaishi’s burnt area with different fire severity classified into three fire severity classes (low, moderate and high severity) were identified and used in the first time series analysis to assess the applicability of NDVI in detecting changes in different fire severity classes. Both satellites can detect changes of NDVI in these three classes of fire severity. Next, using the same fire severity classification, eight points with different fire severity and three types of tree (cedar, red pine and broadleaf trees) were identified and used as ground truth in the second time series to analyze the changes of NDVI in different type of trees. The result shows the time series using two weeks average NDVI from Landsat 8 was better than Terra because Terra could only describe the changes of NDVI for two different type of trees (cedar and red pine), but the changes does not characterize the low severity fire well.
Year: 2018