Mathematical and statistical analysis for LST trend and variation by using MODIS data in the tropical rainforest of Hala-Bala in Narathiwat, Southern Thailand during 2002-2017

Abstract

For Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data, Land Surface Temperature (LST) is a radiative skin temperature that is measured in the direction of the atmosphere to the ground. This data has been used several times in climatological and environmental studies. The tropical rainforest is one of valuable area on this earth that is need to be protected. It exhibits high levels of biodiversity and habitat of animal and plant species on the planet. In Thailand, a tropical rainforest can be found in Hala-Bala where is situated at the boundary of Thai-Malaysia. In mathematical analysis, cubic spline is an approximation function. It gives smooth and flexible way of fitting non linear models, while Generalized Estimate Equation (GEE) is statistical method of parameter estimation for the correlated data from longitudinal studies at different points in time. The purpose of this study was to analyze the data of LST by using cubic spline and GEE to perceive trend and variation at Hala-Bala in Narathiwat, Southern Thailand during 2002-2017. LST data were recorded by MODIS, an important data resource provided efficiency information relating to the Earth, between 2002 and 2017. The result of cubic spline showed that it was fitted quite well on pattern of LST data. The LST has increased statistically significant around the center pixel. GEE showed that LST trends had changed with statistically significant over 15 years.