Development of cloud-native and scalable algorithms to estimate seagrass composition and related carbon stocks in support of the Nationally Determined Contributions of the Paris Agreement

Kurzfassung

Seagrasses are one of the world’s most productive ecosystems, playing an important role in climate change mitigation and adaptation. They are vast natural carbon sinks which have important, yet largely overlooked and underestimated implications into national climate agendas like the Nationally Determined Contributions of the Paris Agreement. Precise knowledge of spatially-explicit seagrass distribution and country-specific in-situ blue carbon data is crucial for the ten countries which currently recognise this ecosystem within their Nationally Determined Contributions. This thesis combines open Sentinel-2 multi-temporal data with the open cloud computing platform Google Earth Engine to quantify country-scale seagrass extents and associated carbon stocks. The limited availability of reference data restricted the implementation of the created cloud-native mapping approach to only one country - The Bahamas. The mapped Bahamian seagrass covers an area between 11,779.44 and 27,629.32 km2, which can store 181,610,083.57 to 455,509,862.63 Mg carbon, and sequesters between 31.02 and 72.75 Mt CO2 per year. This equals 17 to 40 times the amount of CO2 emitted by The Bahamas in 2018, causing a carbon-neutral state and underlining the importance of the seagrass ecosystem for the Bahamian Nationally Determined Contributions. The generated data inventories could support interdisciplinary scientific research and management efforts within a regional and global climate action context.