France’s flood emergency in February 2026 isn’t primarily a story about extreme rainfall. It’s a story about what happens when you apply that rainfall to an already-saturated landscape — and it has direct implications for how Australian hydrologists should be thinking about initial loss parameters.

According to official Météo-France data, France endured 37 consecutive days of rain as of February 20, 2026 — the longest unbroken wet streak since records began in 1959. But the volume of water is only half the story.

“We are facing a generalised flood situation across the entire country because all the soils are saturated everywhere and have lost their infiltration capacity. Soils are now so saturated that as little as 20–30 mm of additional rainfall can trigger further flooding — the highest soil-moisture levels recorded since 1959.” — Lucie Chadourne-Facon, Director of Vigicrues

The Antecedent Moisture Problem

For Australian practitioners, this mechanism is immediately familiar. The Bureau of Meteorology’s Special Climate Statement 76, analysing the 2022 East Coast floods, described the identical process: rainfall falling on “catchments that were already wet so water storages and river levels were high and catchments quickly became saturated.”

The question isn’t whether the rainfall was extreme. It’s whether the catchment had any capacity left to absorb it.

What This Means for Computational Hydrology

In standard 1D/2D hydraulic modelling practice, Initial Loss (IL) and Continuing Loss (CL) values are typically set based on historical flood calibration or regional parameters. These are baked assumptions — and they fail in exactly this scenario.

When soils are fully saturated or have become hydrophobic following a drought, standard runoff coefficients underpredict peak flows dramatically. To model these volatile environments accurately, we need to integrate real-time soil moisture data to dynamically adjust loss parameters before and during a flood event.

The Bureau of Meteorology’s Australian Water Resources Assessment Landscape model (AWRA-L) provides the soil moisture framework to do this — but uptake in operational hydraulic modelling practice is still limited.

The Broader Pattern: Hydro-Climatic Whiplash

The most dangerous scenario in modern water management isn’t a drought or a flood in isolation. It’s the violent swing between the two — drought-baked catchments hit by intense rainfall before any recovery in soil moisture.

France moved from managing a prolonged dry period to a nationwide flood emergency across a single weather sequence. The transition was not gradual.

Australia is familiar with this pattern. What’s changing is the amplitude and speed of the oscillation.

The historical baseline has shifted. We can no longer design for the climate we had; we must engineer for the climate we have.

Satellite imagery: Copernicus Sentinel-1 multi-temporal radar, 6–18 February 2026. © ESA. Originally shared on LinkedIn — 4,631 impressions.