Model

FourCastNet v2 (GFS init)

fourcastnet_gfs

source aiwp · product FOUR · initial conditions from GFS · steps every 6 h to 240 h · initializations 0, 12 UTC. Scores use the latest model version segment, FOUR_v200, starting 2026-08-28. Earlier days ran a different version of the model and are excluded rather than silently pooled into the same average.

aiwp GFS initFOUR_v200
Window
all available history
Init
12Z
Interpolation
bilinear
Data through
2026-08-30
Updated
2026-08-31T06:29:35+00:00
Next update
2026-08-31T11:00:00+00:00

Instantaneous temperature — MAE °F by stationt2

Scored against the observed 2 m temperature at the same instant (ASOS routine METAR). This is the headline variable: the same four instants on the forecast and the observation side. These pages are always bilinear interpolation, the headline method. The nearest-node variant, and every other window/initialization/interpolation/variable combination, is published on each permanent link below — follow any number.

all available history, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same row across lead days 1–9 on a scale shared within this table.
Station d0d1d2d3d4d5d6d7d8d9lead 1–9
KATL 2.16−1.33 · n=2 2.34−2.34 · n=1 KATL MAE by lead day 1 to 9, t2
KAUS 3.22+1.61 · n=2 2.68+0.33 · n=1 KAUS MAE by lead day 1 to 9, t2
KBOS 2.00+0.74 · n=1 KBOS MAE by lead day 1 to 9, t2
KDCA 1.83+0.07 · n=2 1.83+0.14 · n=1 KDCA MAE by lead day 1 to 9, t2
KDEN 3.13+1.40 · n=1 KDEN MAE by lead day 1 to 9, t2
KDFW 2.25−2.24 · n=2 2.42−2.42 · n=1 KDFW MAE by lead day 1 to 9, t2
KEWR 1.56+0.08 · n=2 3.29+0.04 · n=1 KEWR MAE by lead day 1 to 9, t2
KIAH 1.89+0.85 · n=2 1.15+0.36 · n=1 KIAH MAE by lead day 1 to 9, t2
KLAS 6.97−6.97 · n=1 KLAS MAE by lead day 1 to 9, t2
KLAX 2.36+1.71 · n=1 KLAX MAE by lead day 1 to 9, t2
KMIA 2.74+0.04 · n=2 2.16−0.88 · n=1 KMIA MAE by lead day 1 to 9, t2
KMSP 2.95−2.95 · n=2 3.02−3.02 · n=1 KMSP MAE by lead day 1 to 9, t2
KMSY 2.19−2.19 · n=2 1.63−1.63 · n=1 KMSY MAE by lead day 1 to 9, t2
KNYC 2.12+0.97 · n=2 1.80+0.10 · n=1 KNYC MAE by lead day 1 to 9, t2
KOKC 2.45−1.29 · n=2 2.11−0.76 · n=1 KOKC MAE by lead day 1 to 9, t2
KORD 1.77+1.04 · n=2 2.37+0.55 · n=1 KORD MAE by lead day 1 to 9, t2
KPHL 1.88+0.31 · n=2 3.44−0.01 · n=1 KPHL MAE by lead day 1 to 9, t2
KPHX 6.50−6.50 · n=1 KPHX MAE by lead day 1 to 9, t2
KSAN 2.43−1.46 · n=1 KSAN MAE by lead day 1 to 9, t2
KSAT 2.27+0.36 · n=2 2.09−0.73 · n=1 KSAT MAE by lead day 1 to 9, t2
KSEA 1.71−0.16 · n=1 KSEA MAE by lead day 1 to 9, t2
KSFO 1.63−1.21 · n=1 KSFO MAE by lead day 1 to 9, t2
KTTN 4.41+4.20 · n=2 5.63+4.73 · n=1 KTTN MAE by lead day 1 to 9, t2

+ model too warm − model too cold bias interval includes zero ★ lowest MAE = not distinguishable from the leader ▼ worse than the leader ▲ better than the leader — all three Holm-corrected within this table, over the family of comparisons against the leader; the uncorrected verdict for every pair is published in the pairwise table of each permanent link and in pairwise_latest.csv.gzn < 30 greyed and unranked CI is a 95 % moving-block bootstrap interval on the group's own days, and reads — when it could not be computed

Sampled daily maximum — MAE °F by stationtmax_s

Scored against the maximum of the four *observed* samples on the same day.

all available history, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same row across lead days 1–9 on a scale shared within this table.
Station d0d1d2d3d4d5d6d7d8d9lead 1–9
KATL 0.67+0.67 · n=2 2.07−2.07 · n=1 KATL MAE by lead day 1 to 9, tmax_s
KAUS 2.91−2.91 · n=2 3.40−3.40 · n=1 KAUS MAE by lead day 1 to 9, tmax_s
KBOS 2.56+2.56 · n=1 KBOS MAE by lead day 1 to 9, tmax_s
KDCA 2.42+2.42 · n=2 2.74+2.74 · n=1 KDCA MAE by lead day 1 to 9, tmax_s
KDEN 0.17−0.17 · n=1 KDEN MAE by lead day 1 to 9, tmax_s
KDFW 2.51−2.51 · n=2 2.36−2.36 · n=1 KDFW MAE by lead day 1 to 9, tmax_s
KEWR 2.09+0.20 · n=2 2.64−2.64 · n=1 KEWR MAE by lead day 1 to 9, tmax_s
KIAH 0.31+0.31 · n=2 0.81+0.81 · n=1 KIAH MAE by lead day 1 to 9, tmax_s
KLAS 5.60−5.60 · n=1 KLAS MAE by lead day 1 to 9, tmax_s
KLAX 0.60−0.60 · n=1 KLAX MAE by lead day 1 to 9, tmax_s
KMIA 3.57−2.57 · n=2 4.90−4.90 · n=1 KMIA MAE by lead day 1 to 9, tmax_s
KMSP 4.18−4.18 · n=2 5.65−5.65 · n=1 KMSP MAE by lead day 1 to 9, tmax_s
KMSY 2.06−2.06 · n=2 1.22−1.22 · n=1 KMSY MAE by lead day 1 to 9, tmax_s
KNYC 2.00+2.00 · n=2 0.62−0.62 · n=1 KNYC MAE by lead day 1 to 9, tmax_s
KOKC 2.64−2.64 · n=2 2.57−2.57 · n=1 KOKC MAE by lead day 1 to 9, tmax_s
KORD 1.07+0.69 · n=2 1.13−1.13 · n=1 KORD MAE by lead day 1 to 9, tmax_s
KPHL 3.21+0.69 · n=2 3.03−3.03 · n=1 KPHL MAE by lead day 1 to 9, tmax_s
KPHX 8.30−8.30 · n=1 KPHX MAE by lead day 1 to 9, tmax_s
KSAN 5.64−5.64 · n=1 KSAN MAE by lead day 1 to 9, tmax_s
KSAT 2.52−2.52 · n=2 2.67−2.67 · n=1 KSAT MAE by lead day 1 to 9, tmax_s
KSEA 2.96−2.96 · n=1 KSEA MAE by lead day 1 to 9, tmax_s
KSFO 0.84+0.84 · n=1 KSFO MAE by lead day 1 to 9, tmax_s
KTTN 3.86+3.86 · n=2 1.22+1.22 · n=1 KTTN MAE by lead day 1 to 9, tmax_s

Sampled daily minimum — MAE °F by stationtmin_s

Scored against the minimum of the four *observed* samples on the same day.

all available history, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same row across lead days 1–9 on a scale shared within this table.
Station d0d1d2d3d4d5d6d7d8d9lead 1–9
KATL 1.52−1.52 · n=2 0.96−0.96 · n=1 KATL MAE by lead day 1 to 9, tmin_s
KAUS 3.99+3.99 · n=2 3.93+3.93 · n=1 KAUS MAE by lead day 1 to 9, tmin_s
KBOS 0.57−0.57 · n=1 KBOS MAE by lead day 1 to 9, tmin_s
KDCA 1.34−1.34 · n=2 1.09−1.09 · n=1 KDCA MAE by lead day 1 to 9, tmin_s
KDEN 2.71+2.71 · n=1 KDEN MAE by lead day 1 to 9, tmin_s
KDFW 0.41−0.36 · n=2 1.36−1.36 · n=1 KDFW MAE by lead day 1 to 9, tmin_s
KEWR 0.63+0.63 · n=2 2.67+2.67 · n=1 KEWR MAE by lead day 1 to 9, tmin_s
KIAH 2.02+2.02 · n=2 2.20+2.20 · n=1 KIAH MAE by lead day 1 to 9, tmin_s
KLAS 7.64−7.64 · n=1 KLAS MAE by lead day 1 to 9, tmin_s
KLAX 2.03+2.03 · n=1 KLAX MAE by lead day 1 to 9, tmin_s
KMIA 1.46+1.46 · n=2 1.16+1.16 · n=1 KMIA MAE by lead day 1 to 9, tmin_s
KMSP 2.66−2.66 · n=2 2.49−2.49 · n=1 KMSP MAE by lead day 1 to 9, tmin_s
KMSY 1.47−1.47 · n=2 2.44−2.44 · n=1 KMSY MAE by lead day 1 to 9, tmin_s
KNYC 2.31+0.68 · n=2 0.50+0.50 · n=1 KNYC MAE by lead day 1 to 9, tmin_s
KOKC 1.41+1.41 · n=2 0.73+0.73 · n=1 KOKC MAE by lead day 1 to 9, tmin_s
KORD 1.44+1.44 · n=2 0.16+0.16 · n=1 KORD MAE by lead day 1 to 9, tmin_s
KPHL 1.64+1.64 · n=2 4.01+4.01 · n=1 KPHL MAE by lead day 1 to 9, tmin_s
KPHX 5.95−5.95 · n=1 KPHX MAE by lead day 1 to 9, tmin_s
KSAN 0.98+0.98 · n=1 KSAN MAE by lead day 1 to 9, tmin_s
KSAT 1.21−0.31 · n=2 1.56−1.56 · n=1 KSAT MAE by lead day 1 to 9, tmin_s
KSEA 0.50−0.50 · n=1 KSEA MAE by lead day 1 to 9, tmin_s
KSFO 1.46−1.46 · n=1 KSFO MAE by lead day 1 to 9, tmin_s
KTTN 5.93+5.93 · n=2 10.68+10.68 · n=1 KTTN MAE by lead day 1 to 9, tmin_s

Against the NWS daily extremes (secondary)

This comparison scores four 6-hourly forecast samples against the true daily extreme, so it includes a sampling penalty. The size of that penalty depends on each model's own diurnal amplitude, which is one of the things this site measures, so these numbers are published for operational relevance and are never used for ranking.

Daily maximum vs NWS CLI

all available history, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same row across lead days 1–9 on a scale shared within this table.
Station d0d1d2d3d4d5d6d7d8d9lead 1–9
KATL 2.33−2.33 · n=2 4.07−4.07 · n=1 KATL MAE by lead day 1 to 9, tmax_cli
KAUS 5.11−5.11 · n=2 4.80−4.80 · n=1 KAUS MAE by lead day 1 to 9, tmax_cli
KBOS 4.27−2.70 · n=2 7.02−7.02 · n=1 KBOS MAE by lead day 1 to 9, tmax_cli
KDCA 2.48−2.48 · n=2 3.06−3.06 · n=1 KDCA MAE by lead day 1 to 9, tmax_cli
KDEN 4.28−4.28 · n=2 2.76−2.76 · n=1 KDEN MAE by lead day 1 to 9, tmax_cli
KDFW 5.51−5.51 · n=2 4.36−4.36 · n=1 KDFW MAE by lead day 1 to 9, tmax_cli
KEWR 1.50−1.50 · n=2 3.04−3.04 · n=1 KEWR MAE by lead day 1 to 9, tmax_cli
KIAH 3.69−3.69 · n=2 3.19−3.19 · n=1 KIAH MAE by lead day 1 to 9, tmax_cli
KLAS 5.66−5.66 · n=2 5.11−5.11 · n=1 KLAS MAE by lead day 1 to 9, tmax_cli
KLAX 3.47−3.47 · n=2 2.76−2.76 · n=1 KLAX MAE by lead day 1 to 9, tmax_cli
KMIA 5.07−5.07 · n=2 4.90−4.90 · n=1 KMIA MAE by lead day 1 to 9, tmax_cli
KMSP 6.68−6.68 · n=2 5.65−5.65 · n=1 KMSP MAE by lead day 1 to 9, tmax_cli
KMSY 3.56−3.56 · n=2 2.22−2.22 · n=1 KMSY MAE by lead day 1 to 9, tmax_cli
KNYC 1.54+1.54 · n=2 0.54−0.54 · n=1 KNYC MAE by lead day 1 to 9, tmax_cli
KOKC 7.14−7.14 · n=2 6.57−6.57 · n=1 KOKC MAE by lead day 1 to 9, tmax_cli
KORD 0.81−0.81 · n=2 1.13−1.13 · n=1 KORD MAE by lead day 1 to 9, tmax_cli
KPHL 2.31−2.31 · n=2 5.03−5.03 · n=1 KPHL MAE by lead day 1 to 9, tmax_cli
KPHX 9.95−9.95 · n=2 12.20−12.20 · n=1 KPHX MAE by lead day 1 to 9, tmax_cli
KSAN 2.89−2.89 · n=2 0.58−0.58 · n=1 KSAN MAE by lead day 1 to 9, tmax_cli
KSAT 3.62−3.62 · n=2 2.87−2.87 · n=1 KSAT MAE by lead day 1 to 9, tmax_cli
KSEA 3.95−3.95 · n=2 0.75−0.75 · n=1 KSEA MAE by lead day 1 to 9, tmax_cli
KSFO 2.01−2.01 · n=2 0.18−0.18 · n=1 KSFO MAE by lead day 1 to 9, tmax_cli
KTTN 2.11+0.86 · n=2 1.78−1.78 · n=1 KTTN MAE by lead day 1 to 9, tmax_cli

Daily minimum vs NWS CLI

all available history, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same row across lead days 1–9 on a scale shared within this table.
Station d0d1d2d3d4d5d6d7d8d9lead 1–9
KATL 1.02−1.02 · n=2 0.04+0.04 · n=1 KATL MAE by lead day 1 to 9, tmin_cli
KAUS 4.49+4.49 · n=2 3.93+3.93 · n=1 KAUS MAE by lead day 1 to 9, tmin_cli
KBOS 2.18−0.74 · n=2 2.00−2.00 · n=1 KBOS MAE by lead day 1 to 9, tmin_cli
KDCA 0.75+0.66 · n=2 0.91+0.91 · n=1 KDCA MAE by lead day 1 to 9, tmin_cli
KDEN 2.98+2.98 · n=2 1.64+1.64 · n=1 KDEN MAE by lead day 1 to 9, tmin_cli
KDFW 0.41−0.36 · n=2 1.36−1.36 · n=1 KDFW MAE by lead day 1 to 9, tmin_cli
KEWR 3.13+3.13 · n=2 4.67+4.67 · n=1 KEWR MAE by lead day 1 to 9, tmin_cli
KIAH 2.02+2.02 · n=2 2.20+2.20 · n=1 KIAH MAE by lead day 1 to 9, tmin_cli
KLAS 5.69−5.69 · n=2 6.48−6.48 · n=1 KLAS MAE by lead day 1 to 9, tmin_cli
KLAX 1.63+1.63 · n=2 1.98+1.98 · n=1 KLAX MAE by lead day 1 to 9, tmin_cli
KMIA 3.26+3.26 · n=2 0.76+0.76 · n=1 KMIA MAE by lead day 1 to 9, tmin_cli
KMSP 2.16−2.16 · n=2 2.49−2.49 · n=1 KMSP MAE by lead day 1 to 9, tmin_cli
KMSY 1.77−0.47 · n=2 2.44−2.44 · n=1 KMSY MAE by lead day 1 to 9, tmin_cli
KNYC 1.81+1.18 · n=2 1.50+1.50 · n=1 KNYC MAE by lead day 1 to 9, tmin_cli
KOKC 2.91+2.91 · n=2 0.73+0.73 · n=1 KOKC MAE by lead day 1 to 9, tmin_cli
KORD 2.44+2.44 · n=2 2.16+2.16 · n=1 KORD MAE by lead day 1 to 9, tmin_cli
KPHL 3.64+3.64 · n=2 6.01+6.01 · n=1 KPHL MAE by lead day 1 to 9, tmin_cli
KPHX 4.00−4.00 · n=2 6.18−6.18 · n=1 KPHX MAE by lead day 1 to 9, tmin_cli
KSAN 0.76+0.22 · n=2 0.57−0.57 · n=1 KSAN MAE by lead day 1 to 9, tmin_cli
KSAT 1.21−0.31 · n=2 1.56−1.56 · n=1 KSAT MAE by lead day 1 to 9, tmin_cli
KSEA 1.79+0.71 · n=2 0.32+0.32 · n=1 KSEA MAE by lead day 1 to 9, tmin_cli
KSFO 1.27−1.27 · n=2 0.24−0.24 · n=1 KSFO MAE by lead day 1 to 9, tmin_cli
KTTN 8.43+8.43 · n=2 11.68+11.68 · n=1 KTTN MAE by lead day 1 to 9, tmin_cli

Month-by-month slices live on the canonical page: they do not depend on the window, initialization or interpolation selected here.

Source and licence

Where the numbers come from, what may be done with them, and what was — deliberately — not done to the forecast before scoring it.

Where it comes from

aiwp · FOUR · GFS initial conditions

Steps every 6 h to 240 h; sampled at 00/06/12/18 UTC only.

Licence

NOAA/CIRA AIWP, U.S. Government work, public domain. Cite Radford et al. (2025), BAMS, doi:10.1175/BAMS-D-24-0057.1.

Post-processing

None. Raw grid values, bilinear (headline) or nearest-node interpolation, no MOS, no bias correction, no downscaling.