The Use of Object-Based Image Analysis for Monitoring 2021 Marine Mucilage Bloom in the Sea of Marmara

Global warming threatens ecosystems through rising temperatures, increasing sea levels, drought, and extreme weather conditions. The natural balance of seas oceans is also at stake with recent outbreaks mucilage events all over the world. phenomenon, which has been frequently observed in Adriatic Tyrrhenian seas, taken place second time Sea Marmara Spring 2021. dividing Asian European parts Turkey an important inland heavy maritime traffic, hosting many industrial zones surrounded by highly populated cities. This study aims to determine formations that were intensely around Marmara, focusing on coasts Istanbul, Kocaeli, Yalova, Bursa classifying Sentinel-2A images dated 19 24 May 2021, when peak period bloom, using a new paradigm object-based image analysis (OBIA) approach. To create representative homogenous objects, multi-resolution segmentation was applied, its result inputted into classification process Random Forest (RF) classifier generate thematic maps. produced results compared pixel-based high correlation estimated. Object-based found effective for determination mucilage-covered areas (> 90% overall accuracy) both considered dates. More specifically, covered aggregates computed as 56.15 km² 67.51 respectively, indicating rapid growth only 5-day period. resulting maps revealed heavily distributed gulfs Gemlik Izmit along Darica, Tuzla Pendik.