Page 248 - AI for Good Innovate for Impact
P. 248
AI for Good Innovate for Impact
Use Case 16: AI-Driven Early Warning System for Oceanic Dead
Zones
Organization: Vellore Institute of Technology, Chennai Campus
Country: India
Contact Person(s): Dr. J Deepika Roselind, deepikarooselind.j@vit.ac.in
1 Use Case Summary Table
Item Details
Category Climate Change/Natural Disaster
Problem Addressed Increasing prevalence of oceanic dead zones due to climate change and
nutrient runoff, causing marine biodiversity loss and impacting coastal
economies.
Key Aspects of Solu- AI(Artificial Intelligence)-based early warning system using satellite and
tion sonar data.
Deep learning models to predict low-oxygen zones 2–4 weeks in
advance.
Real-time alert system for proactive mitigation.
Technology Keywords AI, machine learning, oceanography, satellite data, sonar, hypoxia
prediction, climate analytics, early warning systems
Data Availability Public and Private
National Aeronautics and Space Administration(NASA) OceanColor: [1]
National Oceanic and Atmospheric Administration National Centers for
Environmental Information(NOAA NCEI): [2]
Copernicus Marine Environment Monitoring Service(CMEMS): [3]
Metadata (Type of Images (satellite imagery), environmental sensor data (temperature,
Data) dissolved oxygen, chlorophyll), spatiotemporal marine measurements
Model Training and Deep learning (Convolutional Neural Networks(CNNs)), multivariate
Fine-Tuning regression and anomaly detection models trained on historical hypoxia
data from global marine datasets
Testbeds or Pilot Government agencies such as National Institute of Oceanography (NIO)
Deployments and Earth System Science Organization (ESSO) – Indian National Centre
for Ocean Information Services (INCOIS)
212
