POSTER SESSION 3
15:30–17:30, Wednesday, September 16
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
ABSTRACT 1268 | POSTER 47
UNTANGLING DRIVERS OF SUSPENDED PARTICULATE MATTER ON THE BLACK SEA NORTH WESTERN SHELF
The North Western Shelf of the Black Sea is a highly productive, shallow region subject to massive riverine inputs from the Danube, Dnepr, and Dnestr, and intense winter storms. These forcings drive the resuspension of suspended particulate matter (SPM), with significant consequences for biogeochemical cycles and benthic-pelagic coupling. Yet the shelf lacks in situ sediment measurements, making it difficult to parameterize sediment dynamics in numerical models such as NEMO-BAMHBI.
This exploratory study investigates whether remote sensing paired with machine learning can identify and quantify the drivers of surface SPM variability. We use high-resolution Sentinel-3 OLCI data processed via the regional Black Sea algorithm of Constantin et al. (2023), alongside CMEMS GlobColour L4 SPM products. To untangle non-linear relationships between SPM and its drivers, we adapt the data-driven frameworks of Subirade (2025) and Tavora et al (2025). Self-Organizing Maps (SOM) extract spatial patterns to delineate distinct hydro-sedimentological zones across the shelf; Random Forest (RF) algorithms are then trained within each zone to rank environmental forcings. Predictive variables include bottom wave stress from CMEMS, ERA5 wind data, river discharge, and NEMO-BAMHBI biogeochemical outputs.
By mapping the relative contributions of waves, currents, wind, and river plumes, this approach yields observational benchmarks for resuspension parameterization, ultimately improving the representation of sediment-driven light attenuation and benthic dynamics for simulations of phenomena like shelf hypoxia.
Mathurin Choblet*, ULiege, [email protected], https://orcid.org/0000-0002-5179-9276
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
