A hydraulic model is only as good as its ability to reflect the real world. That’s where model calibration comes in. It’s the process of fine-tuning your model—adjusting parameters like pipe roughness, demand patterns, or operational settings—so that outputs like pressure, flow, and tank levels match actual measurements in the network.

Calibrating your model helps ensure:

• Confidence in results—A calibrated model demonstrates it can reproduce real system behavior under known conditions.
• Better planning—Simulations become a reliable tool for future developments or emergencies.
• Data validation—Calibration often uncovers missing or incorrect inputs.
• Deeper insights—You'll have a profound understanding of the system behaviour.
• Smarter operations—A calibrated model supports proactive decision-making to reduce losses, manage pressure, and optimimze system performance.

What data do you need

Good calibration depends on solid groundwork. You’ll need:

• A robust model build: Correct network layout, asset data, and operating rules.
• Reliable field data: Pressures, flows, tank levels, and pump/valve states.
• Accurate demand inputs: Reflecting actual usage patterns in your network.
• A GIS tool of your choice: To investigate and solve issues.

To compare your model with reality, you’ll need field data such as:

• Pressure at different locations
• Flow data for the inlets and outlets of zones or DMAs; and for any large consumer within those (i.e.: airports, university, hospitals, etc.)
• Tank water levels over time
• Pump and valve status logs (controls)

Make sure your data is complete, synchronized, and reflects standard operating conditions—not unusual events or ongoing maintenance.

How calibration works

Model calibration is a cycle of testing, comparing, and adjusting. Here’s a basic flow:

1. Run the model with best-guess parameters.
2. Compare simulated results to field data.
3. Adjust parameters like pipe roughness, demands, or pump curves.
4. Repeat until results match within acceptable limits.

This step-by-step refinement may also involve checking for unexpected network issues, like closed valves or misconfigured assets. A sensitivity analysis can help identify which parameters matter most.

Once adjustments are made, you can measure calibration quality using Qatium tools:

• Check the graphs within your assets' information to compare simulated vs. real values.
• Use the Model Deviation panel to see statistical indicators, such as:
  • The overall deviation for flow, pressure, and tank volume values.
  • The average error per each value.
  • Your network elements highlighted on the map in different colors, depending on the level of deviation.

Perfect matches are rare, but good calibration is about consistency and reducing uncertainty.

As best practices, we recommend you to:

• Use both pressure logging and flow testing for richer insights.
• Document your process—data used, assumptions made, changes applied.
• Recalibrate regularly, especially after infrastructure or operational changes.
• Keep the model up to date with GIS data to avoid discrepancies with the real word.