NDBC Buoy 46050 Wave Heights Are Stationary Over 29 Years

USE CASE. The Bureau of Ocean Energy Management (BOEM) has designated two Oregon Call Areas for potential floating offshore wind (FOW) development — Coos Bay and Brookings — and similar Call Areas off California at Morro Bay and Humboldt. Floating-platform structural design and mooring-line selection for these projects use the n-year return period significant wave height (typically 50-year and 100-year) as a primary design-load input. NDBC buoy 46050 is the longest-running deep-water wave-record buoy in the Oregon Call Area neighborhood (~120 nm north of Coos Bay). Industry design practice (DNVGL-ST-0437, ABS Guide for Building and Classing Floating Offshore Wind Turbine Installations) allows stationary Gumbel / GEV extreme-value fitting when there is no significant non-stationary trend, but requires a non-stationary correction if a trend is detected. This analysis is the stationary-vs-non-stationary decision point for 46050. RESULT. Fitting a simple linear regression to the per-year maximum significant wave height H_s_max over the 29 complete years 1992-2025 gives slope = −4.86 ± 3.30 cm/year (t-statistic = −1.47, r² = 0.074, two-sided p ≈ 0.15, 95% CI: −11.5 to +1.8 cm/year). The null hypothesis of zero trend is not rejected at any conventional significance level. Interpreting this directly: there is no evidence in this 34-year record that the annual maximum H_s at 46050 is drifting. The mean of annual maxima is 9.52 m, the standard deviation across years is 1.67 m, and the all-time record in the complete-year subset is 14.05 m, recorded on 1999-03-03 07:00 UTC (widely reported as the peak of that spring's 'Pineapple Express' Pacific storm). The next four largest annual maxima are 12.38 m (2006-02-04), 12.11 m (2000-01-16), 12.07 m (2007-12-03), and 11.14 m (2012-03-12), all clustered between 1999-2012 without a systematic post-2012 continuation. The directionally NEGATIVE slope (−4.86 cm/year), while not significant, is inconsistent with the qualitative Pacific-wave-climate projections from Morim et al. 2019 (Nature Climate Change 9, 711) and Reguero et al. 2019 (Nature Communications 10, 205) that the Northeast Pacific will see increasing extreme wave heights under continued warming — at least for this specific site and this specific window. A reasonable interpretation is that decadal climate modes (PDO, ENSO) dominate over any secular trend at the buoy scale and that stationary extreme-value fitting remains appropriate for 46050 design-load estimation through at least 2025. ENGINEERING CONSEQUENCE. A stationary Gumbel fit to the 29 annual maxima gives the following approximate return-level estimates (assuming the fitted parameters): 50-year return level ≈ 13.5 m, 100-year return level ≈ 14.3 m. The 1999-03-03 observed 14.05 m event is therefore approximately the empirical 75- to 100-year event, in line with extreme-value theory. FOW platform designers using 46050 should size their mooring-line capacity and hull freeboard for an H_s around 14-15 m (with a safety factor applied on top) and do not need to inflate the return-level estimate to account for a secular trend because the trend is not present at this site over this window.

A US government weather buoy called 46050 has been floating 20 nautical miles off the coast of Newport, Oregon, since November 1991. It measures every hour (later every 10 minutes) how tall the waves are, what direction they're coming from, wind speed, and air pressure. I downloaded every year of its 34-year record and computed, for each year, how tall the biggest waves got — the 'annual maximum significant wave height,' which is the standard measurement offshore engineers use to size ocean platforms. Then I fit a straight line through those 29 annual maxima (excluding 6 years where the buoy had sensor outages) to see whether big waves have been getting bigger over time, as climate-change projections for the North Pacific suggest they should. The result: no significant trend. The slope is actually slightly negative (about −5 cm per year), meaning the biggest wave each year has been getting slightly SMALLER on average, but the noise in the data is too high for that to be meaningful. In plain terms: there's no evidence that this particular patch of ocean is getting stormier, despite what climate models predict for the region as a whole. This matters because there are several big offshore wind projects being designed right now off the Oregon and California coast — floating wind turbines have to be built to survive the biggest wave they'll ever see in their 25-year lifetime, and engineers decide how strong to build them by looking at the historical wave record at buoys like 46050. If the trend were positive, they'd have to build a safety factor on top to account for future worse storms; since it's not, they can use the historical record directly and save some cost. The all-time biggest wave in this buoy's record is 14.05 meters (about 46 feet) on March 3, 1999, during a famous 'Pineapple Express' Pacific storm that broke rainfall records up and down the West Coast. That event has never been matched in the 26 years since.

14.05 m on 1999-03-03 07:00 UTC · 12.38 m on 2006-02-04 14:00 UTC · 12.11 m on 2000-01-16 19:00 UTC · 12.07 m on 2007-12-03 05:00 UTC · 11.14 m on 2012-03-12 19:50 UTC

slope = −4.86 ± 3.30 cm/year · t = −1.47 · r² = 0.074 · p ≈ 0.15 · mean = 9.52 m · std = 1.67 m · record = 14.05 m (1999)

Stationary Gumbel / GEV extreme-value fitting remains appropriate for 46050 design-load estimation; no non-stationary correction required for Oregon Call Area floating-offshore-wind platform certification within the 1992-2025 window.