Coupled hydrodynamic, water-quality and fish bioenergetics models run continuously against live sensor data — producing a real-time, defensible estimate of fish survival. The defining capability: test how any operating decision would change survival before you commit.
The physical system informs the model. Model predictions feed back to inform physical-world decisions. BlueGrid's twin is calibrated to your specific facility, stocks and life stages — not regional averages.
Three-dimensional reservoir and tailrace hydrodynamics updated continuously from flow, spill and gate-position data. Temperature stratification, current patterns and residence time computed in real time.
Total dissolved gas, temperature, dissolved oxygen and turbidity modelled through the reservoir and downstream reach. Each parameter is a direct driver of salmon mortality; BlueGrid tracks all continuously.
Individual-based and population-level models calibrated to local stocks and life stages, validated against historical acoustic telemetry. Survival prediction is facility-specific — not regional average.
The twin ingests live feeds from SCADA, acoustic telemetry, fixed sensors and BlueGrid AquaTwin field assets. The model state updates continuously — operators see right now, not last week's field report.
"Three converging developments have made ecological digital twins feasible: proliferation of cheap IoT sensors, advances in machine learning for data fusion and surrogate modelling, and scalable cloud computing supporting real-time simulation."
Rasheed et al. (NTNU) — Digital Twins for Aquatic SystemsModels are tuned against multi-year historical datasets from your facility — telemetry, temperature records, survival studies.
Calibration performance is validated against held-out data and reviewed by your scientific team before operational deployment.
Each year's monitoring data is incorporated in a structured re-calibration cycle, keeping the twin current with environmental change and population shifts.