Rain totals, how far apart
Widest gap 5.5 in at 20.0N 117.25W · unshaded ground is where all 4 land within 0.1 in of each other
Widest gap 12.45 in at 20.75N 121.75W · unshaded ground is where all 4 land within 0.25 in of each other
Widest gap 12.54 in at 20.75N 122.0W · unshaded ground is where all 4 land within 0.25 in of each other
ECMWF's heaviest 24 h totals (in): 2.98, 40 mi W of West Palm Beach, Florida · 2.15 near Baie-Comeau, Québec · 1.92, 40 mi W of Ironwood, Michigan · 1.89 near Coos Bay, Oregon · 1.85, 20 mi ENE of Erie, Pennsylvania · 1.85, 25 mi SW of Marquette, Michigan
ECMWF's heaviest 72 h totals (in): 3.05, 40 mi W of West Palm Beach, Florida · 2.83 at 37.25N 68.5W · 2.81, 35 mi SSE of Weyburn, Saskatchewan · 2.68 near Coos Bay, Oregon · 2.57, 100 mi ENE of St. Augustine, Florida · 2.56, 20 mi ENE of Erie, Pennsylvania
ECMWF's heaviest 96 h totals (in): 10.51 at 22.75N 123.5W · 4.86, 25 mi SW of Colón, Cuba · 4.09, 20 mi N of Lethbridge, Alberta · 3.5, 85 mi NW of Red Deer, Alberta · 3.29, 55 mi NE of Havre, Montana · 3.26, 40 mi W of West Palm Beach, Florida
Thin lines are ECMWF's own total, every rung from 0.25 in up — one model inside the argument, not a blend of the four. Dots mark where that model puts its heaviest.
Thin lines are ECMWF's own total, every rung from 0.5 in up — one model inside the argument, not a blend of the four. Dots mark where that model puts its heaviest.
Thin lines are ECMWF's own total, every rung from 0.5 in up — one model inside the argument, not a blend of the four. Dots mark where that model puts its heaviest.
Methodology
The shaded field is the gap between the wettest and driest of GFS, ECMWF, ECMWF AIFS, GEM — four models and three centres, named because a roster is a fact about a chart: two of these are ECMWF's, and where those two agree they agree partly for family reasons. Each model's total is the difference between two of its own accumulations from the same run, so this is what falls inside the window rather than since the run began. Raw model output on every side: nothing here is bias-corrected.
Three windows, and the reason is measured rather than tidy. Over six to twenty-four hours these models disagree by a FACTOR — a median 16 to 20 times — and hardly at all by an amount, so a map of the amount is flat almost everywhere and blown out on a few cells; that short window is what the coverage map counts instead. By three to four days the ratio has fallen to four or five times and the gap itself spans a quarter inch to three. The scale changes with the window for the same reason, so the key is worth a second look after stepping.
The contours and dots are ECMWF alone, chosen with the chips above the map — one model's own answer inside the argument. Deliberately not an average of the four: these models agree closely on how much water falls over the map as a whole and disagree about where it lands, so an average would draw rain in a place no model forecast.
A dot is a local maximum standing at least 0.5 in high over these 24 hours and clearly above its own surroundings, far enough inside the frame for that comparison to be even-handed; up to ten are kept and the heaviest few are named under the map.
A dot is a local maximum standing at least 1 in high over these 72 hours and clearly above its own surroundings, far enough inside the frame for that comparison to be even-handed; up to ten are kept and the heaviest few are named under the map.
A dot is a local maximum standing at least 1 in high over these 96 hours and clearly above its own surroundings, far enough inside the frame for that comparison to be even-handed; up to ten are kept and the heaviest few are named under the map.
Liquid-equivalent, which is what the models publish: frozen precipitation appears here as the water it would melt to. Every window ends at 18Z, the hour the other two precip and temperature maps share, so charts of one moment compare — and each one starts at the same time, so the three are nested rather than three unrelated slices.
Shaded ground is a disagreement rather than an error — none of these models is the truth here, and the map cannot say which is right. For who has actually been closest, the model leaderboards are the place.
Every map on this page is drawn from model output we store ourselves, not from anyone's picture of it. The per-city read of the same day — which cities the models are fighting about, and what the discussion makes of it — is on the national page.