REMOTE SENSING BASED PHENOLOGY ANALYSIS OF WESTERN HIMALAYAN FOREST REGIONS

Abstract

Monitoring of forest regions of the Himalayan mountains is quite a challenging task hence remote sensing induced matrices such as vegetation indices are widely used for the estimation of phenological attributes and ecosystem study. Many previous researches have taken the effect of topography and vegetation into account for the phenological analysis, where they have been typically focusing on the comparison of various indices with respect to different landuse/landcover. The Himalayan ecosystem has been considered distinguished as well as diverse encompassing different diversities in flora and fauna, where these regions are often characterized by heavy precipitation and variant temperatures compared to the flat terrain. This research mainly aims to assess the phenological behavior of the forest types of Doon valley in Shivalik range during pre-monsoon and post-monsoon seasons. The primary objective of this study is to evaluate the use of remote sensing data in the process of producing spatio-temporal phenological maps of these mountainous regions on annual basis. This study aims to map the distribution and validation of phenological variables for different forest types of Doon valley in Shivalik range over duration from November 2015 to March 2018 using LANDSAT-8 and SENTINEL-2 datasets. However the success rate in achieving such an objective is mainly a function of availability of cloud free data for this region throughout the study duration, with which more detailed phenological trends across these areas could be explained. Results indicate that LANDSAT-8 and SENTINEL-2 data offers new possibilities in trend analysis of health and productivity of these study areas. Thus with the help of these data the objectives of this study were attained and the phenological changes in the forest area were observed and understood, their different phenological stages were also identified within the time duration selected. In conclusion the detailed ground validation was not possible as it would require proper resourced techniques thus the cross validation was done using the reference studies, where the leaf flush and leaf fall duration for different forest types were compared with the obtained results of this study. However, a limited number of literature reviews were used to validate the current result and in most cases they revealed a similar trend. Thus, this study has evaluated the phenological behavior of these forest types and modeled their phenolgical changes with the help of NDVI and TNPI values