OceanTensorCube forecasts seaweed bleaching with transformer system
OceanTensorCube and ST-MMT: A Multimodal Data and Transformer System for Seven-Day Forecasting of Pyropia Bleaching
Machine Learning
Summary
Bleaching damages seaweed farms in Korea, causing financial losses. The authors created a system that combines multiple types of ocean and weather data to predict bleaching events up to seven days ahead for specific farm clusters. Their system uses a sophisticated machine learning model called ST-MMT that analyzes two weeks of data to forecast bleaching risk. The predictions are fairly accurate but become less reliable when key nutrient data become outdated. The system runs on a public website providing operational forecasts to help farmers manage risks.
What this means in practice
- •For seaweed farm managers: Receive up-to-seven-day bleaching risk forecasts for clusters to improve farm management decisions and reduce crop losses.
- •For coastal environmental monitoring teams: Integrate combined marine, meteorological, and satellite data with transformer models to track and predict ecosystem health events.
Authors
Gwangjin Kim, Cheolyoung Jang, Yosep Jo
Abstract
Bleaching of cultured Pyropia (laver) causes losses in Korean seaweed farming. We present a system that forecasts bleaching seven days ahead for farm clusters. OceanTensorCube aligns 32 channels from public marine, meteorological, ocean-model and satellite sources on a 30 m grid, and ST-MMT, a 9.2-million-parameter spatiotemporal transformer with hub-style multi-head cross-attention between variable groups, maps a 14-day input window to a seven-day-ahead bleaching probability. On the validation set, a six-seed ensemble detects every reported positive at a false-alarm rate of at most 3.5%, an upper bound because the report-based labels are positive-unlabeled. A same-date control on separately trained models indicates that the architecture discriminates between clusters when the calendar is held fixed. We also report the system's weaknesses: skill falls steeply as the quarterly nutrient survey ages (ROC-AUC 0.94 at 30 days, 0.53 at 129 days), and a single LSTM on cluster-average series outperforms a single ST-MMT, with only the ensemble leading at the operating point. We describe the labeling procedure, the evaluation protocol and the operational gate that withholds forecasts when inputs are stale. The forecasting service is operated at https://oceantensor.ai.kr.