Loading map…
Hover the map for PRISM values
Storms off the Pacific run into the Olympic Mountains and drop most of their water on the way up. Forks, on the west side, gets about 115 inches of rain a year, and the high peaks get more than 200. On the far side of the mountains, Sequim gets about 17 — less than half of what falls on Seattle. Click anywhere on the map, or drag the pin, to see how much falls there.
PRISM annual normal along the dashed line on the map, from the Pacific coast eastward. Gaps are open water. Click the profile to move the pin along the line.
PRISM 1991–2020 monthly normals. The shaded June–September band is the dry season, when the rain shadow matters most for gardens and wells.
1. PRISM at the point. The 1991–2020 annual normal from the PRISM 800 m grid, read two ways: the cell the point falls in, and a bilinear blend of the four nearest cell centers. The 1 km and 3 km buffers show how steep the local gradient is. If the 3 km range is wide, a few hundred meters of position error matter.
2. Gauge normals. Daily GHCN records (NWS co-op, Port Angeles airport, SNOTEL and CoCoRaHS volunteers) are totaled by year. A year counts only if at most 10% of days are missing; CoCoRaHS multi-day totals count toward coverage. Stations with 25+ complete years inside 1991–2020 use their own mean. Shorter records are scaled to 1991–2020 by the ratio at the nearest long-record station over the same years.
3. Gauge adjustment. At each gauge, the ratio of gauge normal to PRISM measures PRISM’s local bias. The point estimate is PRISM multiplied by the inverse-distance-squared average of that ratio at the 3 nearest usable gauges within 25 km (40 km if fewer than 3).
4. Error range. The same adjustment was run at every calibration gauge with that gauge left out. The root-mean-square of those misses is the ± range on every estimate. It is a regional figure: in the foothill transition zone, where gauges disagree with PRISM by more, treat it as optimistic.
Confidence is moderate when a usable gauge is within 8 km and the 3 km range is under 25% of the value, low when the nearest gauge is over 15 km away or the 3 km range exceeds 40%, and low–moderate otherwise. No point is rated high: PRISM is a model, and it already uses most of these gauges as inputs, so agreement at gauges overstates its accuracy elsewhere.
Each dot is one gauge within 40 km of Sequim. Points above the dashed 1:1 line are places where PRISM is wetter than the gauge. Log scales, so the dry lowlands are not squeezed into a corner.