KOFAP - Korea FAO Sustainable and Innovative Fisheries and Aquaculture Programme

Big Data Platform for Smart Aquaculture 

What is a Big Data Platform for Smart Aquaculture?

A Big Data Platform for Smart Aquaculture is a centralized, cloud-based digital system that collects, integrates, manages and analyses data from aquaculture farming operations, regulatory bodies, biosecurity monitoring activities and the wider environment. Data are sourced from sensors, manually entered field observations and farming operations, laboratory results, and remote sensing. The platform integrates both conventional analytics and advanced approaches, including artificial intelligence and machine learning (AI/ML). Its aim is to provide services that support multi-stakeholder aquaculture management.

Explore the key elements of a Big Data Platform

Automated collection, often in real time, on water quality (dissolved oxygen, temperature, pH, salinity, ammonia), aeration status, environmental conditions, and the status and location of farm equipment.


Biosecurity observations, including mortality counts, feeding logs, and disease sampling results entered via mobile apps or web interfaces.


Laboratory results (bacteria, LAMP, water chemistry), growth samples, and hatchery records.


External data on animal and human health, including PCR diagnostics, disease surveillance systems, national biosecurity alerts, and regulatory reporting platforms.


External environmental monitoring data, including remote sensing (satellite imagery, UAV surveys), pond mapping, and national datasets on biodiversity, weather, hydrology and extreme events.


National farm registries, movement records, SPF seed certification, and inter-farm communication channels.


Critical for ensuring data integrity, enhancing security and confidentiality across connected (IoT) devices, and streamlining regulatory compliance.


Statistical forecasting, predictive models for water quality and disease risk, anomaly detection from images, geospatial pond detection, and natural language interfaces for non-technical users.


What services do Big Data Platforms provide?

A Big Data Platform can provide services tailored to the needs of different types of producers, as well as technical data management services that underpin the entire system.

Extensive producers typically operate with limited digital infrastructure and primarily need early warning services and connection to national communication and data systems.

Key platform services at this level include:

  • National and regional early warning systems for extreme weather events (floods, storms, unusual temperature events), delivered to farmers via SMS or mobile app notifications.
  • National farm registries and digital maps provide spatial context for management authorities and designated agencies on farm locations relative to environmental risks and disease outbreak areas.
  • Support for nationally coordinated sampling networks: uploading field observations, disease samples, and manual water quality readings through mobile apps.
  • Basic disease risk advisories based on national surveillance data, disseminated through mobile applications to assist farmers with practical advice.
  • Integration with One Health and Blue Transformation national reporting frameworks.

For extensive producers, the platform requires minimal on-farm infrastructure. A smartphone with (intermittent) mobile connectivity is sufficient to access most services.

Extension services and their agents play a key role in facilitating data entry and interpreting alerts and will be key in a successful Smart Data Platform.

 

Semi-intensive producers benefit from a broader range of services that integrate sensor data with AI-assisted analytics. In addition to all services for extensive producers, a typical smart platform provides:

  • Integration of in-pond water quality sensor data (dissolved oxygen, temperature, pH, salinity, ammonia) with automated alerts when values fall outside predefined thresholds.
  • Statistical and AI-based forecasting of pond conditions, including probabilistic forecasts of dissolved oxygen levels with user-configurable alert thresholds and forecast horizons. The SAB project demonstrated cost-effective hybrid statistical/ML forecasting for DO and other parameters in Peruvian ponds.
  • Generate semi-automatically data for passive and active surveillance programs to improve health, environment and biosecurity.
  • Remote sensing-assisted water quality assessment for larger ponds using satellite imagery (e.g. Copernicus Sentinel-2 based on water colour classification).
  • Integration with Specific Pathogen Free (SPF) certification data for traceability.
  • Data sharing across farms in a cluster or region, enabling pattern detection and joint risk assessments, potentially with AI and remote sensing / GIS Systems.
  • Demand-side integration with feed and supply chain partners for automated restocking and traceability.
  • Dashboards for field extension officers and biosecurity inspectors with farm-level status and alert histories for tailored advice on production and risk management.

Extension services and their agents play a key role in ensuring consistent data quality, even if staff turnover is high, if the Smart Data Platform enforces quality data entry by e.g. geofencing, photographic evidence and calendar driven reporting. This enhanced data quality also improves the possibilities and quality of subsequent analysis.

Keeping an up-to-date inventory of farm inputs is a tedious and repetitive task that can be much simplified with mobile assisted data collection. Inventories are extremely useful in the case of disaster recovery support, maintenance of feed-stock, and to understand farm losses due to feed quality deterioration and misaligned feeding patterns.

Intensive producers require comprehensive, near-real-time operational data management. In addition to all semi-intensive services, the platform provides:

  • Continuous high-frequency sensor monitoring with automated controls for feeding, aeration, and water exchange based on real-time analytics.
  • Advanced AI/ML forecasting of multiple water quality parameters simultaneously, including phytoplankton composition and mineral balance indicators.
  • Innovative AI-driven image analysis for disease risk classification, shrimp growth estimation (achieving over 90% accuracy in pilot programmes), and behaviour monitoring.
  • Integration of Recirculating Aquaculture System (RAS) or Biofloc Technology (BFT) data streams in monitoring and control applications.
  • On-farm or mobile laboratory results (PCR, qPCR, LAMP, LFSB field kits) fed directly into the platform’s disease surveillance module.
  • Comprehensive food safety traceability from hatchery to harvest, supporting export compliance and audit documentation.
  • Risk Index (RI) scoring combining environmental, health, and economic indicators in real time.
  • Integration with suppliers and retailers' data platforms to synchronize production schedules with procurement and e-commerce platforms.

Who is needed: Key roles and team composition

A Big Data Platform brings together expertise from different sectors. It is not a fixed or standardized structure; rather, it adapts to local production systems and governance arrangements. Suggested roles that can be considered include:

Platform Manager

Stakeholder coordination, governance, reporting

Required Skills:

  • Aquaculture/fisheries background, project management
Data Engineer

Cloud infrastructure, APIs, data pipelines, security

Required Skills:

  • Cloud platforms (AWS/GCP), Python/SQL, networking
Data Scientist

AI/ML model development, statistical analysis, reports

Required Skills:

  • Python/R, time-series analysis, machine learning
Aquaculture Specialist

Domain expertise, model validation, extension support

Required Skills:

  • Shrimp biology, biosecurity, water quality management
GIS Specialist

Pond mapping, satellite imagery, geospatial services

Required Skills:

  • GIS software, remote sensing, Python/QGIS
Field Coordinator

Farmer training, data quality, alert interpretation

Required Skills:

  • Field extension, communication, mobile apps