Multi-Hazard Risk Assessment

Several single-hazard studies produce separate reports in incompatible units that cannot legitimately be added, and they often assume different exposed assets. Multi-hazard assessment forces a common exposure inventory and a common loss scale, then aggregates while explicitly modeling how hazards interact. Kappes and colleagues stress that this last point — handling cascades and coincidences instead of summing independent risks — is what makes the assessment genuinely multi-hazard rather than a stapled set of single-hazard analyses.

Because hazards are rarely independent. One event can trigger another, as when an earthquake sets off landslides; two events can strike the same already-weakened assets, compounding damage; and protection against one hazard can aggravate another. Naively adding single-hazard risks misses triggered losses and double-counts shared ones. The interaction term in the aggregation step is where the method captures these dependencies, and omitting it is the most common way multi-hazard assessments go wrong.

Vulnerability functions are the bridge. A flood is measured in inundation depth and an earthquake in ground acceleration, which cannot be compared directly, but each can be converted through its own fragility curve into a damage ratio or monetized loss for the same asset. Once every hazard is expressed as expected loss on a shared exposure inventory, the results live on a common scale and can be aggregated and ranked. This is why building credible vulnerability functions for every asset-hazard pair is the most consequential step.