Abstract

Mangroves are imperative to coastal systems, providing security against extreme weather events by acting as natural barriers. These halophytic plants play a crucial role in mitigating global warming and act as an invaluable resource for consumption. Despite proving to be resilient, mangroves exhibit sensitivity to climatic (e.g., Land Surface Temperature, Salinity, etc.) and man-made factors (e.g., Land Use/Land Cover Changes). Numerous past studies recording the relationship between mangrove growth & development with the aforesaid constituents, but those were mostly restricted to visual observation/pattern recognition and single type of regression analysis. Also, the evaluation of simultaneous exploration of multiple aspects influencing mangrove evolution was not prominent. Therefore, the main objective of this study was to focus on the impact of both salinity and land surface temperature on mangrove biomass by the joint-venture of remote sensing, geographic information system and several machine learning algorithms. The study considered appropriate mangrove site selections with pre-processing of the acquired satellite images. Also, mathematical computations were performed on the raster layers to determine the previously mentioned natural aspects. Finally, several types of regression analysis were conducted to delineate potential patterns, governing mangrove vegetation health by virtue of temperature and salinity. Mangroves’ relationship with temperature and salinity showed insignificant coefficient of determination. However, the generated response curves postulated that high mangrove biomass could be achieved for a specific temperature window (30-33◦C) and vegetation health could deteriorate at increasing salinity. Overall, combined effects of surface temperature and salinity on mangrove vegetation were significantly more (i.e., Maximum coefficient of determination of 0.31) than individual component alone.