Parametric flood trigger design, with basis risk measured

The challenge

Parametric cover pays when a measured index crosses an agreed level, so the policy is only as good as its trigger. Where there is no river gauge, the index has to come from a global flood model, and few programmes check how often that model says flood in the years the river actually flooded.

A trigger written in cubic metres per second straight from a global model inherits the model's bias. On 14 Spanish river gauges, GloFAS 4.0 saw floods at about a fifth of their measured size, so a trigger set in m³/s would have paid in 6 of 60 flood years. GloFAS 5.0 removes most of that bias (26 of 60), which shows how much the model version and the way the trigger is written decide the payout record.

Our solution

KAI-Hydro moves your site onto the river by catchment size, pulls the GloFAS v5 reanalysis (1980 to today), and fits the flood-frequency curve with honest uncertainty bands. It then designs the trigger four ways, as a fixed flow, a percentile, a return period or a share of a reference flood, with attachment and exhaustion levels, and back-tests each one year by year: hits, misses and false payouts.

Where a gauge, satellite flood maps or local records exist, the trigger is calibrated on them, because a global model alone carries too much basis risk to sell. Engineers get the same study as design floods: the 1-in-10 to 1-in-200-year flow with its band.

Who it is for: Insurers and reinsurers, parametric brokers, risk pools, lenders and agricultural insurers, and engineers who need design floods