Spatial distribution patterns of hotspot and relationship between hotspot and vegetation indices in Chiang Mai Province, Thailand

Abstract

This work focused on Chiang Mai Province, Thailand, had 2 targets which were 1) to analyse spatial distribution patterns of hotspot and 2) to analyse a relationship between hotspot and vegetation indices in the area. The hotspots data of 2016 - 2020 which had a significant level > 70% were gathered from MODIS satellite images, was provided by Fire Information for Resource Management System (FIRMS). An analyse method was performed by Nearest Neighbour Index (NNI) with Moran? s I to present spatial distribution patterns and density of hotspot. Analysis of Getis - Ord Gi* statistic was for identify heat of hotspot comparing with surrounding area. Moreover, vegetation indices values (Normalized Difference Vegetation Index: NDVI, Soil Adjustment Vegetation Index: SAVI and Normalized Difference Water Index: NDWI) was examined by satellite images of the same period from Landsat 8 OLI to analyse a relationship between hotspot and each vegetation index. The results illustrated that there were different number of hotspots over 5 studying years, especially in 2016 which had the most hotspot. The spatial distribution of hotspot patterns was classified as clustered type (Getis - Ord Gi* statistic with Z-Score > 1.96) with different hotspot density in each year. The area which had high heat was found in upper and west area with medium to high hotspots density. The hotspot and NDVI had relationship in contrast by a correlation coefficient value at -.887 (r = -.887) with a significant level at .05. However, SAVI and NDWI had no relationship with hotspot.