Learn more about First Street Foundation's Version 2.0 updates.
First Street Foundation's Flood Model now expands to include Alaska, Hawaii, and Puerto Rico. The model covers cities in Alaska with >1,000 people, plus the coast of the Alaskan Peninsula. The model also covers the 8 major islands of Hawaii (Hawaii, Maui, Kahoolawe, Lanai, Molokai, Oahu, Kauai, and Niihau) and all of Puerto Rico.
First Street Foundation Flood Model Version 2.0 Data Available
The value of Flood Factor is the ability to be adaptable and responsive to new and improved data and information. We are committed to transparency, and to using the best data, peer-reviewed science, and available information in our model. As such, we have been working on expanding and improving our current flood model. Our Version 2.0 data is now available. You can learn more about this data update here and you can also find additional information below. Please reach out if you would like updated data.
Key findings
Alaska
- More than 66,000 properties in Alaska face flood risk today
- This grows by nearly 5% in 2052 to more than 69,000 due to the effects of a changing climate.
- Annualized total economic damage from flooding in the state grows from approximately $74.5 million today to more than $82 million in 2052, a significant increase of more than 10% owing to climate-related changes in flood risk.
- Highly populated Fairbanks City and Anchorage Municipality are among the areas those facing the most, and fastest growing economic risk:
- Anchorage’s Municipality’s total economic risk grows from over $10 million on an annualized basis today to over $12 million in 2052, an increase of 10.7%.
- Fairbanks city’s economic risk grows from $13 million today to more than $17 million in 2052, an increase of 24.62%.
Hawaii
- More than 30% of all properties on the Hawaiian Islands face flood risk today.
- Approximately 122,581 properties across all islands face flood risk in 2022. This grows to 123,370 by 2052.
- Total potential economic damage from flood risk in Hawaii is approximately $70 million this year, growing to more than $72 million in 2052 due to the effects of a changing climate.
- Honolulu County, a major residential and tourist destination, faces a nearly 7% increase in economic risk to its beaches, high rises, and hotels, growing from over $28 million in 2022 to more than $30 million on an annualized basis by 2052.
Puerto Rico
- More than one in five properties in Puerto Rico face flood risk in 2022, a significant share of the more than 1.2 million properties on the island.
- Nearly 20% (18.31%) of all properties in the city of San Juan face flood risk today.
Model updates
Statistics from building footprints
We’ve improved our property and building footprints for previously shared outlines such as condos as well as making other improvements to how a property and address are defined and connected.
Building footprints will now be clipped to parcel boundaries prior to running stats, improving model accuracy in dense areas and land uses other than single-family residential. Stats will be drawn from all footprints over a certain area threshold that lies on a parcel polygon.
Change from non-linear regression to linear interpolation
Our previous method calculated flood depth at all probabilities and years based on a best-fit line from the modeled data. Using this method, however, sometimes depths could be significantly different in stats than what was displayed in the hazard layer.
The new method uses the flood depths from hazard layers and interpolates only in between. Ultimately, this reduction in complexity only affected about 3% of total pixels in our hazards as about 97% of all pixels modeled increased linearly over the short 30-year time period associated with our models.
Resolve Flood Factor Score sensitivity
The previous method used the mode (most common) of 6 scores at specific depth thresholds. This makes the final score very sensitive to changes in depths, as that can change the six scores and therefore which number occurs most frequently. In V2.0, this has changed to use a combination of annualized depth and frequency of flooding which will create a more stable score in the event of changing flood data. This method also aligns directly with the approach used in the development of our AALs.
Adaptation methodology updates
We have made some changes to how we classify adaptation protection zones and how they apply to properties. Changes to our adaptation methodology include:
- A refinement to the hydraulic representation of features that remove some flooding but not all (such as pump stations), will decrease the removal of flooding.
- Flooding will be fully removed at the center of the shape and removed by a decay function toward the edge of the shape