In recent years, cloud computing and Earth Observation data has significantly grown to allow for large-scale calculation of vital statistics, such as land cover and biomass, to facilitate for the monitoring of the health and development of ecosystems in light of climate change and the United Nations Sustainable Development Goals. With the advancement of cloud computing platforms, such as Google Earth Engine, and the data availability of remotely sensed data on said platforms, can machine learning by utilized in conjunction with high resolution data to estimate mangrove land covers as well as above ground biomass in these threatened areas?

Previous studies estimating mangrove biomass (Simard et al. 2019), utilize data collected at lower resolutions than what is currently available, and/or also collected during a period which is no longer applicable to the conditions of the present. With global datasets, such as Sentinel-2's global imagery at 10m resolution, and active sensors such as NASA's ICESat-2 and GEDI LIDAR platforms sampling the elevation of key features on the Earth's surface at a fine resolution, derivation of high resolution mangrove health statistics, such as area, height, and biomass, should be achievable utilizing GEE.

The workflow developed in this study, within a cloud computing environment, can easily be deployed in any region of the globe (at a country-wide scale) where mangroves may be present with minimal changes in order to generate high resolution mangrove health statistics to facilitate the reporting of these ecosystems and improving stewardship and development of the UN SDGs. Cloud computing and actively sensed data such as GEDI canopy height returns in concert with machine learning techniques have proven to accurately predict key mangrove characteristics, height and above ground biomass, at a large scale (country-wide), allowing for rapid and detailed reporting and management of these ecosystems, proving that they are powerful tools in the fight against future climate change. This study intends to build and develop upon the workflow and techniques derived and utilized in the previous semester's study by Alexander Boest-Petersen in Fall 2021 with Aalborg University Copenhagen and DHI Group.