How CoastErosionCheck Calculates a Coast's Erosion Rate

One row of USGS data is not one page

USGS's National Assessment of Shoreline Change publishes 86,539 individual transects, survey lines a few hundred feet apart, each with its own long-term rate. That's finer-grained than useful and carries no place name at all, so CoastErosionCheck groups nearby transects into ~3.5-mile segments (a 0.05-degree grid, chosen because it produces roughly the page count the underlying search demand actually supports) and reports one rate per segment.

The rate itself: LRR and WLR

Most regions use USGS's Linear Regression Rate (LRR) method, fitting a trend line through every historical shoreline position on file for a transect; Hawaii's assessment instead uses a Weighted Linear Regression (WLR) that favors more reliable shoreline dates. Both produce feet-per-year, and both come with a published confidence interval, which matters more than the rate itself - see the guide to what "no significant trend" means.

Why some readings are simply missing

USGS flags a transect where the rate could not be reliably computed with a sentinel value (±9,999 ft/yr in this dataset) rather than a real number. CoastErosionCheck treats those as missing data, not as the fastest erosion on record - the alternative would make the site's own worst-looking pages the least trustworthy ones.

Naming a segment that has no name

USGS's transects carry latitude and longitude, nothing else. Each segment here is matched to its nearest named place from the Census Gazetteer, with the real distance always stated, because a stretch of remote coast can be ten or more miles from the nearest town with a name at all - the Golden Meadow, LA page states its nearest-town distance for exactly this reason.

Where the raw numbers come from

Shoreline-change rates are USGS's regional Open-File Reports (Gulf of Mexico, New England and Mid-Atlantic, Pacific Northwest, California, Hawaii). Place names and coordinates are the Census Bureau's 2025 place gazetteer. Every segment page cites its own source report.

Where to go next

See the range this plays out across: Golden Meadow, LA, Dauphin Island, AL, Ocean Shores, WA.

What "No Significant Trend" Actually Means · Why USGS's Coastal Surveys Are a Different Age in Every Region