Bernice, OK
Bernice, OK's wildfire risk, in USFS's own numbers
Bernice sits at the 84th percentile nationally for wildfire risk to structures — in USFS's highest wildfire-risk band nationally — per USFS's Wildfire Risk to Communities model, built from its 482 counted buildings, not vegetation cover alone. (Source: USFS's Wildfire Risk to Communities methodology.)
USFS's separate burn-probability score — fire likelihood alone, before counting what's built there — puts Bernice at the 87th percentile, close to its 84th-percentile risk score.
What "at risk" means for the buildings here
Of Bernice's 482 counted buildings, 49.4% carry Direct exposure and only 5.2% carry Minimal — a lopsided split that puts defensible-space clearing ahead of vents or roofing as the intervention worth doing first.
Bernice against the rest of the country
Bernice's 84th national percentile looks worse in isolation than its 29th ranking inside Oklahoma does — this place is on the milder end for its own state, by 55 points. Among the 31,521 US communities USFS scores, Bernice ranks 5,020 for wildfire risk (1 is highest) and 17,672 by building count (1 is largest). Within Oklahoma alone, it ranks 586 of 834 places by risk. See the full county-by-county picture for Oklahoma on its state page.
What this risk score means for insurance
Bernice's very high rating (84th percentile) puts it among the places where the national carrier-pullback trend is most relevant — not a guarantee of a coverage problem, but a reason to shop rather than assume.
What would actually reduce this score
With 49.4% of Bernice in Direct exposure, defensible-space clearing and ember-resistant construction (vents, Class-A roofing) address the exposure this page's own numbers describe, not a generic checklist. See the home-hardening guide.
Where Bernice's figures come from
The methodology guide shows exactly how USFS turned 482 counted buildings into the percentiles shown above for Bernice. The exposure-zones guide covers what Bernice's dominant direct exposure actually means, with real examples from across the dataset.