HYDROPT is an open-source framework for forward and inverse modelling of multi- and hyperspectral observations from oceans, coastal and inland waters. Our framework is based on radiative transfer principles and is sensor agnostic .
A PyTorch implementation of the Multi-Mode CNN to reconstruct Chlorophyll-a time series in the global ocean from oceanic and atmospheric physical drivers
Combination Of Neural Network models for Estimating Chlorophyll-a over Turbid and clear waters
assess the impact of 2024 Central American fires on surface ocean properties in the Gulf of Mexico
Satellite-based causal attribution of coastal water clarity degradation to nickel smelting expansion at Indonesia's Morowali Industrial Park using Bayesian structural time series, multi-algorithm changepoint detection, and Sentinel-2 land cover intensity analysis.
[In Development] PACE Hackweek 2025 participants passionate about applied science and applying PACE data to understand water quality, coastal resilience, and post-disaster recovery. Team: Katie Miller, Nathan Tesfayi, and Anna Valcarcel
Create Level-2 satellite maps with data from ESA.
Distributory materials for "ML for Aquatic Remote Sensing" sessions at SCOR 2025
ML pipeline for detecting pelagic Sargassum from Sentinel-2/MODIS/Landsat-8 satellite imagery. Models: XGBoost, 1D-CNN, MLP. Outputs fractional cover maps at 10m resolution. Website: https://doi.org/10.5281/zenodo.17246345
Add a description, image, and links to the ocean-color topic page so that developers can more easily learn about it.
To associate your repository with the ocean-color topic, visit your repo's landing page and select "manage topics."