Raw model output on the native 0.25° grid — no MOS, no bias correction, no post-processing. Scores understate operational forecast quality. Read the fairness statement →

CastCheck methodology v0.2 · data through 2026-08-30

KEWR — Newark Liberty

40.6825, -74.1694 · 5 m· standard offset UTC-5 h · truth product CLIEWR

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.

Daily maximum temperature — MAE °F

the last 90 days, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same model across lead days 1–9 on a scale shared within this table.
Modeld0d1d2d3d4d5d6d7d8d9lead 1–9
ECMWF AIFS Singleaifs_single 2.48-2.46 · n=29 3.18-3.12 · n=28 3.07-2.86 · n=27 2.93-2.65 · n=26 3.17-2.95 · n=24 3.47-3.12 · n=24 3.11-3.07 · n=23 4.19-4.04 · n=22 4.99-4.92 · n=21 ECMWF AIFS Single MAE by lead day 1 to 9, tmax
Aurora (GFS init)aurora_gfs 4.36-4.36 · n=2 3.20-3.20 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmax
Aurora (IFS init)aurora_ifs 4.07-4.07 · n=2 1.92-1.92 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (GFS init)fourcastnet_gfs 1.50-1.50 · n=2 3.04-3.04 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (IFS init)fourcastnet_ifs 2.22-2.22 · n=2 2.01-2.01 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmax
NCEP GFSgfs 1.99-1.02 · n=29 2.37-1.40 · n=28 2.87-1.67 · n=27 3.52-2.11 · n=26 2.88-1.20 · n=25 3.43-1.33 · n=24 3.64-0.36 · n=23 3.31-0.55 · n=22 4.97-0.46 · n=21 NCEP GFS MAE by lead day 1 to 9, tmax
GraphCast (GFS init)graphcast_gfs 3.95-3.95 · n=2 5.26-5.26 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmax
GraphCast (IFS init)graphcast_ifs 4.69-4.69 · n=27 4.93-4.93 · n=26 5.04-5.04 · n=25 4.38-4.38 · n=25 4.84-4.75 · n=25 4.26-4.23 · n=24 4.35-4.35 · n=23 5.75-5.40 · n=22 5.65-5.27 · n=21 GraphCast (IFS init) MAE by lead day 1 to 9, tmax
ECMWF IFS HRESifs_hres 3.97-3.61 · n=24 4.30-4.11 · n=22 3.90-3.66 · n=22 3.85-2.61 · n=21 3.35-2.24 · n=20 3.61-2.38 · n=19 4.83-4.22 · n=17 4.15-3.26 · n=17 5.29-4.30 · n=15 ECMWF IFS HRES MAE by lead day 1 to 9, tmax
Pangu-Weather (GFS init)pangu_gfs 1.54+0.72 · n=2 0.96+0.96 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmax
Pangu-Weather (IFS init)pangu_ifs 1.41+0.39 · n=2 1.19+1.19 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmax
Persistence (baseline)persistence 4.44-0.09 · n=90 6.12-0.10 · n=90 7.21-0.30 · n=90 7.66-0.30 · n=90 7.14-0.34 · n=90 7.17-0.30 · n=90 6.81-0.30 · n=90 7.49-0.36 · n=90 7.76-0.60 · n=90 Persistence (baseline) MAE by lead day 1 to 9, tmax

Up to 1 of the scored days in this table carry a QC flag on the observation (a CLI value that disagreed with the hourly observations, or a fallback source). They are kept in the scores and counted here rather than dropped — see methodology.

+ model too warm · − model too cold · bias interval includes zero · ★ lowest MAE · = not distinguishable from the leader · ▼ significantly worse · n < 30 greyed and unranked · CI is a 95 % moving-block bootstrap interval.

Daily minimum temperature — MAE °F

the last 90 days, 12Z, bilinear. MAE with the bias and n underneath; every cell links to its permanent page. The sparkline shows the same model across lead days 1–9 on a scale shared within this table.
Modeld0d1d2d3d4d5d6d7d8d9lead 1–9
ECMWF AIFS Singleaifs_single 2.02+1.39 · n=29 2.30+1.84 · n=28 2.33+1.88 · n=27 2.60+1.38 · n=26 2.78+1.20 · n=24 2.80+1.43 · n=24 2.59+1.56 · n=23 2.72+1.68 · n=22 3.55+1.11 · n=21 ECMWF AIFS Single MAE by lead day 1 to 9, tmin
Aurora (GFS init)aurora_gfs 1.97+1.97 · n=2 2.33+2.33 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmin
Aurora (IFS init)aurora_ifs 0.53+0.53 · n=2 0.38+0.38 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (GFS init)fourcastnet_gfs 3.13+3.13 · n=2 4.67+4.67 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (IFS init)fourcastnet_ifs 2.16+2.16 · n=2 3.68+3.68 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmin
NCEP GFSgfs 3.35+2.97 · n=29 3.47+3.05 · n=28 4.01+3.62 · n=27 3.56+3.00 · n=26 3.89+2.90 · n=25 4.23+3.63 · n=24 4.57+3.46 · n=23 4.64+3.53 · n=22 4.38+3.02 · n=21 NCEP GFS MAE by lead day 1 to 9, tmin
GraphCast (GFS init)graphcast_gfs 2.48+2.48 · n=2 2.63+2.63 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmin
GraphCast (IFS init)graphcast_ifs 2.02+0.61 · n=27 2.11+0.93 · n=26 2.21+1.07 · n=25 2.39+1.10 · n=25 2.82+0.91 · n=25 2.89+1.18 · n=24 2.22+1.04 · n=23 3.44+1.00 · n=22 3.51-0.22 · n=21 GraphCast (IFS init) MAE by lead day 1 to 9, tmin
ECMWF IFS HRESifs_hres 2.21+1.02 · n=24 2.12+1.57 · n=22 2.50+1.80 · n=22 3.10+1.96 · n=21 2.52+1.34 · n=20 2.77+2.07 · n=19 2.89+2.40 · n=17 3.91+2.50 · n=17 3.40+1.48 · n=15 ECMWF IFS HRES MAE by lead day 1 to 9, tmin
Pangu-Weather (GFS init)pangu_gfs 1.25+1.25 · n=2 1.03+1.03 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmin
Pangu-Weather (IFS init)pangu_ifs 0.99+0.99 · n=2 1.31+1.31 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmin
Persistence (baseline)persistence 3.27-0.09 · n=90 4.83-0.28 · n=90 5.23-0.46 · n=90 4.97-0.59 · n=90 4.82-0.62 · n=90 5.16-0.60 · n=90 5.32-0.66 · n=90 5.69-0.78 · n=90 5.59-0.97 · n=90 Persistence (baseline) MAE by lead day 1 to 9, tmin

Up to 1 of the scored days in this table carry a QC flag on the observation (a CLI value that disagreed with the hourly observations, or a fallback source). They are kept in the scores and counted here rather than dropped — see methodology.

Month-by-month slices, model availability and the observed-truth table live on the canonical page: they do not depend on the window, initialization or interpolation selected here.

All scores as JSON scores_latest.csv daily_errors.csv.gz data & licences