California can build a levee, restore a wetland or elevate a structure and still struggle to show an insurer what changed. A report released Sept. 9 by California Ocean Science Trust and UC San Diego's Scripps Institution of Oceanography identifies that gap between physical protection and financial recognition, then proposes eight steps to narrow it.
Catastrophe models estimate the probability and scale of losses across many possible disasters. Insurers use those estimates in decisions about price and availability. The report finds that flood models do not consistently or transparently represent mitigation. Conventional defenses such as levees and pump stations are constrained by incomplete public data, while nature-based projects such as wetlands and living shorelines are often represented indirectly or with limited documentation.
That is a systems problem for coastal counties. Water moves through watersheds, streets, storm drains, rivers and shorelines; a defense in one place can alter exposure elsewhere. California's flood processes can also overlap. Atmospheric rivers bring intense precipitation, high tides and waves push from the coast, and saturated ground reduces the landscape's ability to absorb another storm. A model built for only one mechanism can miss the compound event produced by several.
The authors organize their recommendations around four needs. First, strengthen baseline measurements, including an assessment of uninsured and underinsured exposure and access to property-level National Flood Insurance Program loss data for model testing. Second, improve inputs through statewide inventories of structures and defenses, a flood-event archive and benchmark scenarios that allow models to be compared against the same conditions.
Third, make the path from risk reduction to insurance outcome visible. The report recommends greater reporting about California's private flood-insurance market and closer evaluation of community-scale incentives in both private coverage and the federal Community Rating System. If a verified project lowers expected damage but a resident cannot see any resulting change in price or availability, the financial signal to maintain or replicate that investment is weak.
The fourth proposal is a public California flood catastrophe model. Its purpose would be transparency rather than a promise of perfect prediction: researchers could test assumptions, regulators could examine market models, and planners could compare adaptation choices. A public model would still depend on data quality, maintenance and clear uncertainty ranges. Publishing an algorithm does not remove the physical unknowns in a rare storm.
Scale explains the urgency but should not be confused with a forecast. Ocean Science Trust cites an estimate of $8 billion in average annual flood-related property damage in California. It also references a U.S. Geological Survey scenario in which an exceptional 1-in-1,000-year event could produce $725 billion in property and business losses in 2011 dollars, more than $1 trillion today. The latter is a stress scenario, not a prediction that such a flood is imminent.
For San Diego County, the report's value lies in connecting projects that sit in different budgets. A restored marsh may be funded as habitat, a pump as infrastructure and an insurance premium by a household, yet all respond to the same water. The recommendations ask the data and markets to follow those connections. Whether premiums ultimately change will require regulators, model vendors, insurers and public agencies to implement the proposals and then measure what happens.