Skip to content

Continuous simulation

A design storm answers “what happens in a 5-year event”. Continuous simulation answers “how often does this happen at all”, which is the question a regulator and an investment case usually ask.

The best input is your own gauge record. Import it and the run uses measured rainfall at its true resolution.

Where no local record exists, a historical record can be fetched for the model’s location. That data is hourly, which matters: hourly averages under-represent the short intense bursts that drive pipe surcharge. When the application disaggregates hourly data into shorter steps it says so explicitly, and the result is labelled as derived rather than measured.

Frequency statistics rather than a single event: how many times a node flooded, how many spills occurred and their total volume, and how much of the record the system spent surcharged.

That converts “it floods in a 5-year storm” into “it flooded eleven times in five years”, which is a far stronger basis for a decision.

This is a full hydraulic simulation over the whole record. A year of dynamic-wave routing takes minutes to tens of minutes depending on network size and your machine. That is the engine doing the work, not a delay.

Start with a shorter horizon to confirm the model behaves, then commit to the long run.