AI Plays a Part in Sustainable Forestry

Client Background

NeXT FOREST Inc., a joint venture between IHI Corporation and Sumitomo Forestry, aims to protect tropical forests while balancing environmental conservation and economic development. The project seeks to ensure the sustainable management of tropical peatlands, particularly in West Kalimantan, Indonesia, using advanced science and technology.

Challenge

Proper management of the groundwater levels in tropical peatlands is of utmost importance. If levels drop excessively, drying out can occur, turning the peatlands into highly flammable areas prone to forest fires. On the other hand, overly high groundwater levels can inhibit tree growth, underscoring the necessity of continually maintaining an optimal balance.

However, predicting and managing groundwater levels has traditionally depended on experienced technicians, which is time-consuming and inefficient. Conventional methods require manual surveying and the creation of topographic & contour maps as detailed as 50 centimeters, which typically takes up to 5 years.

Recursive's Solution

Recursive developed a hybrid AI-physics model to predict and manage groundwater levels efficiently. This innovative model integrates machine learning with physical hydraulic simulations to deliver actionable predictions.

Solution Components:

• Current Groundwater Level Estimation: Combines ground sensors, vegetation data, and topographic maps to estimate accurate groundwater levels in real time.
• Future Groundwater Level Prediction: Processes the current groundwater level estimates, weather forecast data, irrigation canal information, hypothetical dam operations, and physical system knowledge to predict groundwater changes up to 7 days in advance with a precision of 6 cm.
• Risk Notifications: The system provides early alerts when groundwater levels approach thresholds that increase the risk of fires, enabling stakeholders to take preventive measures swiftly.

Results

The implementation of Recursive's AI-powered groundwater management solution delivered significant, measurable outcomes:

• 1,200 km² Coverage: Enabled efficient monitoring and management of extensive tropical peatland areas.
• CO₂ Reduction: Proactively prevented forest fires, reducing carbon emissions and supporting climate change mitigation efforts.
• Operational Efficiency: Reduced the need for time-consuming manual surveys and minimized reliance on technician-dependent processes.
• Optimized Resource Management: Enabled better placement of dams and irrigation channels through AI-driven simulations.