Nia Ambarsari, Erfan Rohadi, Faqih Hamami, Cahya Rahmad, Sinung Suakanto, Muhammad Auvi Amril
Mangrove forests are vital for coastal defense, biodiversity preservation, and carbon storage. However, in urban areas like Greater Malang Region, the growth and sustainability of these forests are increasingly threatened by rapid urbanization and climate change. This study explores the growth patterns of mangrove trees in Malang by applying machine learning techniques, specifically clustering, to classify areas based on their urgency for intervention. By analyzing a dataset that includes various environmental and geographic characteristics of mangrove habitats in Malang, we employed the K-means algorithm to segment the data into 5 distinct clusters. These clusters not only reflect varying levels of urgency but also indicate that 2 out of the 5 clusters represent coastal areas requiring immediate and special intervention. The identification of these critical zones allows for targeted conservation efforts and the allocation of resources to the most affected regions. The findings demonstrate that clustering can effectively distinguish between areas with different levels of degradation and restoration needs, highlighting the utility of data-driven approaches in environmental management. This model offers a scalable solution for other urban regions facing similar ecological challenges. © 2024 IEEE.
Telkom University, School of Industrial and System Engineering, Bandung, Indonesia; Jurusan Teknologi Informasi, Politeknik Negeri Malang, Malang, Indonesia