Published in Scientific Papers. Series E. Land Reclamation, Earth Observation & Surveying, Environmental Engineering, Vol. XIV
Written by Tamara MYSLYVA, Christiaan Max HUISDEN, Marek MROZ, Nataliia TSUMAN, Yurii BILYAVSKYI
Land use changes monitoring and predicting, as well as assessing their impact on carbon storage dynamics, play a pivotal role in addressing environmental challenges and ensuring effective land use management. This study aims to identify land use changes and their impact on carbon storage in the Marowijne district of Suriname from 2017 to 2024 and predict changes for 2034. Sentinel-2 images were used to analyze land change patterns and predict future trends. A hybrid approach combining Markov chain analysis, cellular automata, multilayer perceptron, support vector machines, and logistic regression was used to forecast future land use dynamics, while InVEST and YASSO models were utilized for carbon storage and sequestration predictions. The support vector machine-Markov chain hybrid model achieved an impressive accuracy of over 97%, outperforming other hybrid models. This model is recommended for generating land use change prediction maps, providing a crucial baseline for sustainable land use management. During the subsequent decade (2024-2034), the net loss of high-carbon areas is expected to intensify, affecting 15-20% of the district's territory. The identified spatiotemporal distribution of carbon storage provides valuable insights that will play a key role in achieving the objectives of Suriname’s national green development strategy.
[Read full article] [Citation]
