Model

FourCastNet v2 (IFS init)

fourcastnet_ifs

source aiwp · product FOUR · initial conditions from IFS · 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 IFS initFOUR_v200
Window
the last 365 days
Init
12Z
Interpolation
bilinear
Data through
2026-08-30
Updated
Next update

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.

the last 365 days, 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.
Stationd0d1d2d3d4d5d6d7d8d9lead 1–9
KATL2.12−1.13 · n=22.05−2.05 · n=1KATL MAE by lead day 1 to 9, t2
KAUS3.14+1.02 · n=22.68+0.34 · n=1KAUS MAE by lead day 1 to 9, t2
KBOS2.33+1.02 · n=1KBOS MAE by lead day 1 to 9, t2
KDCA2.07−0.37 · n=21.84+0.61 · n=1KDCA MAE by lead day 1 to 9, t2
KDEN2.71+0.81 · n=1KDEN MAE by lead day 1 to 9, t2
KDFW3.44−3.44 · n=23.16−3.16 · n=1KDFW MAE by lead day 1 to 9, t2
KEWR1.57−0.80 · n=22.27+0.28 · n=1KEWR MAE by lead day 1 to 9, t2
KIAH1.83+0.65 · n=21.10+0.22 · n=1KIAH MAE by lead day 1 to 9, t2
KLAS7.27−7.27 · n=1KLAS MAE by lead day 1 to 9, t2
KLAX2.66+2.25 · n=1KLAX MAE by lead day 1 to 9, t2
KMIA2.85−0.23 · n=22.31−1.03 · n=1KMIA MAE by lead day 1 to 9, t2
KMSP2.60−2.60 · n=22.93−2.93 · n=1KMSP MAE by lead day 1 to 9, t2
KMSY1.57−1.56 · n=21.10−1.10 · n=1KMSY MAE by lead day 1 to 9, t2
KNYC2.20+0.15 · n=21.51+0.35 · n=1KNYC MAE by lead day 1 to 9, t2
KOKC2.98−1.95 · n=22.28−1.58 · n=1KOKC MAE by lead day 1 to 9, t2
KORD1.94+0.32 · n=22.37+0.36 · n=1KORD MAE by lead day 1 to 9, t2
KPHL1.70−0.75 · n=22.38+0.14 · n=1KPHL MAE by lead day 1 to 9, t2
KPHX5.87−5.87 · n=1KPHX MAE by lead day 1 to 9, t2
KSAN2.32−0.57 · n=1KSAN MAE by lead day 1 to 9, t2
KSAT2.43+0.15 · n=22.04−0.61 · n=1KSAT MAE by lead day 1 to 9, t2
KSEA1.50+0.75 · n=1KSEA MAE by lead day 1 to 9, t2
KSFO1.87−1.63 · n=1KSFO MAE by lead day 1 to 9, t2
KTTN3.73+3.38 · n=25.10+5.00 · n=1KTTN 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 skill reads — when fewer than 10 days are common to the model and the baseline: the ratio of two means over a handful of shared days is not a number worth printing

Sampled daily maximum — MAE °F by stationtmax_s

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

the last 365 days, 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.
Stationd0d1d2d3d4d5d6d7d8d9lead 1–9
KATL0.96+0.96 · n=21.84−1.84 · n=1KATL MAE by lead day 1 to 9, tmax_s
KAUS3.20−3.20 · n=23.38−3.38 · n=1KAUS MAE by lead day 1 to 9, tmax_s
KBOS3.16+3.16 · n=1KBOS MAE by lead day 1 to 9, tmax_s
KDCA2.27+2.27 · n=23.71+3.71 · n=1KDCA MAE by lead day 1 to 9, tmax_s
KDEN0.73−0.73 · n=1KDEN MAE by lead day 1 to 9, tmax_s
KDFW3.61−3.61 · n=23.05−3.05 · n=1KDFW MAE by lead day 1 to 9, tmax_s
KEWR2.45−0.52 · n=21.61−1.61 · n=1KEWR MAE by lead day 1 to 9, tmax_s
KIAH0.29+0.29 · n=20.54+0.54 · n=1KIAH MAE by lead day 1 to 9, tmax_s
KLAS5.60−5.60 · n=1KLAS MAE by lead day 1 to 9, tmax_s
KLAX0.15−0.15 · n=1KLAX MAE by lead day 1 to 9, tmax_s
KMIA3.35−2.76 · n=25.24−5.24 · n=1KMIA MAE by lead day 1 to 9, tmax_s
KMSP3.61−3.61 · n=26.66−6.66 · n=1KMSP MAE by lead day 1 to 9, tmax_s
KMSY1.23−1.23 · n=20.49−0.49 · n=1KMSY MAE by lead day 1 to 9, tmax_s
KNYC2.23+1.36 · n=20.36+0.36 · n=1KNYC MAE by lead day 1 to 9, tmax_s
KOKC3.73−3.73 · n=23.67−3.67 · n=1KOKC MAE by lead day 1 to 9, tmax_s
KORD1.44−0.52 · n=21.55−1.55 · n=1KORD MAE by lead day 1 to 9, tmax_s
KPHL3.24−0.09 · n=22.05−2.05 · n=1KPHL MAE by lead day 1 to 9, tmax_s
KPHX7.27−7.27 · n=1KPHX MAE by lead day 1 to 9, tmax_s
KSAN4.58−4.58 · n=1KSAN MAE by lead day 1 to 9, tmax_s
KSAT2.62−2.62 · n=22.48−2.48 · n=1KSAT MAE by lead day 1 to 9, tmax_s
KSEA1.50−1.50 · n=1KSEA MAE by lead day 1 to 9, tmax_s
KSFO0.48+0.48 · n=1KSFO MAE by lead day 1 to 9, tmax_s
KTTN3.21+3.21 · n=22.39+2.39 · n=1KTTN 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.

the last 365 days, 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.
Stationd0d1d2d3d4d5d6d7d8d9lead 1–9
KATL1.37−1.37 · n=20.56−0.56 · n=1KATL MAE by lead day 1 to 9, tmin_s
KAUS3.61+3.61 · n=23.98+3.98 · n=1KAUS MAE by lead day 1 to 9, tmin_s
KBOS0.46−0.46 · n=1KBOS MAE by lead day 1 to 9, tmin_s
KDCA2.31−2.31 · n=21.13−1.13 · n=1KDCA MAE by lead day 1 to 9, tmin_s
KDEN2.93+2.93 · n=1KDEN MAE by lead day 1 to 9, tmin_s
KDFW1.30−1.30 · n=21.83−1.83 · n=1KDFW MAE by lead day 1 to 9, tmin_s
KEWR0.39−0.34 · n=21.68+1.68 · n=1KEWR MAE by lead day 1 to 9, tmin_s
KIAH1.71+1.71 · n=22.10+2.10 · n=1KIAH MAE by lead day 1 to 9, tmin_s
KLAS7.85−7.85 · n=1KLAS MAE by lead day 1 to 9, tmin_s
KLAX2.88+2.88 · n=1KLAX MAE by lead day 1 to 9, tmin_s
KMIA1.02+0.98 · n=21.09+1.09 · n=1KMIA MAE by lead day 1 to 9, tmin_s
KMSP2.39−2.39 · n=21.60−1.60 · n=1KMSP MAE by lead day 1 to 9, tmin_s
KMSY1.11−1.11 · n=22.14−2.14 · n=1KMSY MAE by lead day 1 to 9, tmin_s
KNYC2.54−0.19 · n=20.37−0.37 · n=1KNYC MAE by lead day 1 to 9, tmin_s
KOKC1.13+1.13 · n=20.33+0.33 · n=1KOKC MAE by lead day 1 to 9, tmin_s
KORD1.54+1.07 · n=20.32+0.32 · n=1KORD MAE by lead day 1 to 9, tmin_s
KPHL0.14+0.14 · n=22.78+2.78 · n=1KPHL MAE by lead day 1 to 9, tmin_s
KPHX5.89−5.89 · n=1KPHX MAE by lead day 1 to 9, tmin_s
KSAN1.87+1.87 · n=1KSAN MAE by lead day 1 to 9, tmin_s
KSAT1.54−0.20 · n=21.48−1.48 · n=1KSAT MAE by lead day 1 to 9, tmin_s
KSEA0.24+0.24 · n=1KSEA MAE by lead day 1 to 9, tmin_s
KSFO1.78−1.78 · n=1KSFO MAE by lead day 1 to 9, tmin_s
KTTN4.94+4.94 · n=29.54+9.54 · n=1KTTN 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

the last 365 days, 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.
Stationd0d1d2d3d4d5d6d7d8d9lead 1–9
KATL2.04−2.04 · n=23.84−3.84 · n=1KATL MAE by lead day 1 to 9, tmax_cli
KAUS5.40−5.40 · n=24.78−4.78 · n=1KAUS MAE by lead day 1 to 9, tmax_cli
KBOS4.49−2.33 · n=26.85−6.85 · n=1KBOS MAE by lead day 1 to 9, tmax_cli
KDCA2.63−2.63 · n=22.09−2.09 · n=1KDCA MAE by lead day 1 to 9, tmax_cli
KDEN5.09−5.09 · n=23.61−3.61 · n=1KDEN MAE by lead day 1 to 9, tmax_cli
KDFW6.61−6.61 · n=25.05−5.05 · n=1KDFW MAE by lead day 1 to 9, tmax_cli
KEWR2.22−2.22 · n=22.01−2.01 · n=1KEWR MAE by lead day 1 to 9, tmax_cli
KIAH3.71−3.71 · n=23.46−3.46 · n=1KIAH MAE by lead day 1 to 9, tmax_cli
KLAS5.51−5.51 · n=23.23−3.23 · n=1KLAS MAE by lead day 1 to 9, tmax_cli
KLAX3.51−3.51 · n=22.68−2.68 · n=1KLAX MAE by lead day 1 to 9, tmax_cli
KMIA5.26−5.26 · n=25.24−5.24 · n=1KMIA MAE by lead day 1 to 9, tmax_cli
KMSP6.11−6.11 · n=26.66−6.66 · n=1KMSP MAE by lead day 1 to 9, tmax_cli
KMSY2.73−2.73 · n=21.49−1.49 · n=1KMSY MAE by lead day 1 to 9, tmax_cli
KNYC1.69+0.90 · n=20.44+0.44 · n=1KNYC MAE by lead day 1 to 9, tmax_cli
KOKC8.23−8.23 · n=27.67−7.67 · n=1KOKC MAE by lead day 1 to 9, tmax_cli
KORD2.02−2.02 · n=21.55−1.55 · n=1KORD MAE by lead day 1 to 9, tmax_cli
KPHL3.09−3.09 · n=24.05−4.05 · n=1KPHL MAE by lead day 1 to 9, tmax_cli
KPHX9.86−9.86 · n=210.13−10.13 · n=1KPHX MAE by lead day 1 to 9, tmax_cli
KSAN2.57−2.57 · n=20.03−0.03 · n=1KSAN MAE by lead day 1 to 9, tmax_cli
KSAT3.72−3.72 · n=22.68−2.68 · n=1KSAT MAE by lead day 1 to 9, tmax_cli
KSEA2.65−2.65 · n=20.07−0.07 · n=1KSEA MAE by lead day 1 to 9, tmax_cli
KSFO2.34−2.34 · n=20.68−0.68 · n=1KSFO MAE by lead day 1 to 9, tmax_cli
KTTN2.39+0.21 · n=20.61−0.61 · n=1KTTN MAE by lead day 1 to 9, tmax_cli

Daily minimum vs NWS CLI

the last 365 days, 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.
Stationd0d1d2d3d4d5d6d7d8d9lead 1–9
KATL0.99−0.87 · n=20.44+0.44 · n=1KATL MAE by lead day 1 to 9, tmin_cli
KAUS4.11+4.11 · n=23.98+3.98 · n=1KAUS MAE by lead day 1 to 9, tmin_cli
KBOS2.54−1.00 · n=21.64−1.64 · n=1KBOS MAE by lead day 1 to 9, tmin_cli
KDCA0.31−0.31 · n=20.87+0.87 · n=1KDCA MAE by lead day 1 to 9, tmin_cli
KDEN2.63+2.63 · n=20.96+0.96 · n=1KDEN MAE by lead day 1 to 9, tmin_cli
KDFW1.30−1.30 · n=21.83−1.83 · n=1KDFW MAE by lead day 1 to 9, tmin_cli
KEWR2.16+2.16 · n=23.68+3.68 · n=1KEWR MAE by lead day 1 to 9, tmin_cli
KIAH1.71+1.71 · n=22.10+2.10 · n=1KIAH MAE by lead day 1 to 9, tmin_cli
KLAS5.68−5.68 · n=26.66−6.66 · n=1KLAS MAE by lead day 1 to 9, tmin_cli
KLAX1.88+1.88 · n=21.66+1.66 · n=1KLAX MAE by lead day 1 to 9, tmin_cli
KMIA3.22+2.78 · n=20.69+0.69 · n=1KMIA MAE by lead day 1 to 9, tmin_cli
KMSP1.89−1.89 · n=21.60−1.60 · n=1KMSP MAE by lead day 1 to 9, tmin_cli
KMSY1.98−0.11 · n=22.14−2.14 · n=1KMSY MAE by lead day 1 to 9, tmin_cli
KNYC2.04+0.31 · n=20.63+0.63 · n=1KNYC MAE by lead day 1 to 9, tmin_cli
KOKC2.63+2.63 · n=20.33+0.33 · n=1KOKC MAE by lead day 1 to 9, tmin_cli
KORD2.07+2.07 · n=22.32+2.32 · n=1KORD MAE by lead day 1 to 9, tmin_cli
KPHL2.14+2.14 · n=24.78+4.78 · n=1KPHL MAE by lead day 1 to 9, tmin_cli
KPHX3.82−3.82 · n=25.16−5.16 · n=1KPHX MAE by lead day 1 to 9, tmin_cli
KSAN1.34+0.53 · n=20.73−0.73 · n=1KSAN MAE by lead day 1 to 9, tmin_cli
KSAT1.54−0.20 · n=21.48−1.48 · n=1KSAT MAE by lead day 1 to 9, tmin_cli
KSEA1.80+1.44 · n=20.47+0.47 · n=1KSEA MAE by lead day 1 to 9, tmin_cli
KSFO1.22−1.22 · n=20.81−0.81 · n=1KSFO MAE by lead day 1 to 9, tmin_cli
KTTN7.44+7.44 · n=210.54+10.54 · n=1KTTN 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 · IFS 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.