Wave intelligence · NDBC 46229 · Pacific Northwest
Umpqua, OR Offshore
43°46.3′N 124°32.9′W · Station page · All stations
Reading the buoy's directional spectra from NDBC and taking them apart…
The sea at —
Directional spectrum
Energy by the direction waves come from (north up) and period, long periods at the centre. Colours are the systems.
Energy by period
Each system's share of the spectrum, stacked. The dashed line is NDBC's swell / wind-sea boundary.
Systems through time
The spectrogram, energy by period and time, with each system's peak traced over it. A swell from a distant storm arrives long periods first, so its trace climbs toward shorter periods; a dotted line marks the fit that locates the storm.
Where the swell came from
Distances from the rate each swell's period shortened as it arrived, along the direction it came from. Rings every 1000 nautical miles; the map keeps true distances and bearings from the buoy.
Crest length
Directional spread of each system: under 25° the crests run long and orderly, as in groundswell; over 40° the sea is short-crested and confused.
Crossing seas
When the two largest systems meet at 40° or more with comparable energy.
How this is worked out
NDBC's buoys report, for each frequency band, the energy and four numbers describing how it spreads over direction. From these Bluegraph estimates the full directional spectrum with the maximum entropy method (Lygre and Krogstad, 1986), which reproduces the buoy's measurements exactly and is the usual choice for these buoys.
The spectrum is split into systems by following each cell uphill to its peak; systems separated by shallow valleys are joined, and ones holding under 5% of the energy, or under 0.15 m, are absorbed. A system is wind sea when the wind, raised to 10 m, outruns its waves (1.5·U10·cos Δθ above the peak phase speed); otherwise swell, and groundswell at 13 s or longer.
Systems are followed from spectrum to spectrum by period and direction. Swell spreads out as it travels: waves of frequency f move at g/(4πf), so a swell from a storm D away arrives with its frequency rising at g/(4πD). Fitting that rise gives the distance and the time the waves left, with the fit's uncertainty; the storm lies along the direction the first waves came from. Positions are estimates, good to a few hundred miles.
Spectra are NDBC's realtime files, provisional and not quality-controlled. The reference gives every threshold.