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

KLAS — Las Vegas Harry Reid

36.0719, -115.1634 · 664 m· standard offset UTC-8 h · truth product CLILAS

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 365 days, 00Z, 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 5.08-4.82 · n=30 4.47-4.34 · n=29 4.95-4.82 · n=28 5.07-5.02 · n=27 5.79-5.28 · n=26 6.05-5.34 · n=25 7.07-6.53 · n=24 7.03-6.78 · n=23 8.25-8.08 · n=22 ECMWF AIFS Single MAE by lead day 1 to 9, tmax
Aurora (GFS init)aurora_gfs 5.38-5.38 · n=2 5.12-5.12 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmax
Aurora (IFS init)aurora_ifs 3.97-3.97 · n=3 3.98-3.98 · n=2 3.92-3.92 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (GFS init)fourcastnet_gfs 5.26-5.26 · n=2 4.45-4.45 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (IFS init)fourcastnet_ifs 5.93-5.93 · n=3 6.07-6.07 · n=2 5.92-5.92 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmax
NCEP GFSgfs 5.08-4.52 · n=30 5.14-4.42 · n=29 4.60-3.75 · n=28 4.69-3.69 · n=27 4.46-2.80 · n=26 4.72-3.16 · n=25 5.23-3.15 · n=24 6.75-3.40 · n=23 5.99-3.48 · n=22 NCEP GFS MAE by lead day 1 to 9, tmax
GraphCast (GFS init)graphcast_gfs 7.94-7.94 · n=2 7.64-7.64 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmax
GraphCast (IFS init)graphcast_ifs 5.08-4.89 · n=34 4.89-4.75 · n=33 5.51-5.24 · n=32 5.56-5.38 · n=31 6.26-5.74 · n=31 6.07-5.68 · n=31 6.98-6.35 · n=30 7.33-6.99 · n=29 8.82-8.26 · n=28 GraphCast (IFS init) MAE by lead day 1 to 9, tmax
ECMWF IFS HRESifs_hres 3.89-3.66 · n=25 3.83-2.58 · n=22 3.14-2.37 · n=23 3.46-1.89 · n=22 3.31-2.52 · n=20 3.33-3.02 · n=18 3.92-2.74 · n=19 4.68-3.37 · n=18 6.30-4.38 · n=17 ECMWF IFS HRES MAE by lead day 1 to 9, tmax
Pangu-Weather (GFS init)pangu_gfs 5.16-5.16 · n=2 4.89-4.89 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmax
Pangu-Weather (IFS init)pangu_ifs 5.32-5.32 · n=3 5.80-5.80 · n=2 5.51-5.51 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmax
Persistence (baseline)persistence 3.32+0.00 · n=365 4.87-0.03 · n=365 5.95-0.07 · n=365 6.61-0.10 · n=365 7.22-0.15 · n=365 7.68-0.20 · n=365 7.79-0.21 · n=365 7.76-0.22 · n=365 7.83-0.24 · n=365 Persistence (baseline) MAE by lead day 1 to 9, tmax

Up to 2 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 365 days, 00Z, 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 4.99-4.92 · n=30 5.34-5.29 · n=29 5.29-5.01 · n=28 5.11-4.94 · n=27 5.21-4.70 · n=26 5.36-4.85 · n=25 5.31-5.10 · n=24 6.17-6.13 · n=23 6.71-6.37 · n=22 ECMWF AIFS Single MAE by lead day 1 to 9, tmin
Aurora (GFS init)aurora_gfs 6.88-6.88 · n=2 8.05-8.05 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmin
Aurora (IFS init)aurora_ifs 4.29-4.29 · n=3 5.96-5.96 · n=2 7.33-7.33 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (GFS init)fourcastnet_gfs 6.12-6.12 · n=2 6.18-6.18 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (IFS init)fourcastnet_ifs 4.94-4.94 · n=3 6.12-6.12 · n=2 7.64-7.64 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmin
NCEP GFSgfs 3.19-1.22 · n=30 3.46-1.39 · n=29 3.71-0.94 · n=28 3.39-0.12 · n=27 3.59-0.20 · n=26 3.37+0.44 · n=25 3.81+0.15 · n=24 4.54+0.22 · n=23 4.93-0.42 · n=22 NCEP GFS MAE by lead day 1 to 9, tmin
GraphCast (GFS init)graphcast_gfs 6.58-6.58 · n=2 9.42-9.42 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmin
GraphCast (IFS init)graphcast_ifs 2.26-1.44 · n=34 2.70-2.10 · n=33 2.85-2.30 · n=32 2.89-2.24 · n=31 3.45-2.26 · n=31 4.27-2.43 · n=31 4.65-2.35 · n=30 5.46-2.43 · n=29 5.91-2.66 · n=28 GraphCast (IFS init) MAE by lead day 1 to 9, tmin
ECMWF IFS HRESifs_hres 3.09-2.61 · n=25 3.61-2.76 · n=22 3.82-3.06 · n=23 4.25-2.54 · n=22 4.58-3.15 · n=20 4.56-2.50 · n=18 5.00-2.76 · n=19 5.43-3.68 · n=18 5.68-4.43 · n=17 ECMWF IFS HRES MAE by lead day 1 to 9, tmin
Pangu-Weather (GFS init)pangu_gfs 7.76-7.76 · n=2 6.67-6.67 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmin
Pangu-Weather (IFS init)pangu_ifs 4.85-4.85 · n=3 6.21-6.21 · n=2 7.28-7.28 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmin
Persistence (baseline)persistence 2.71-0.02 · n=365 4.00-0.06 · n=365 4.86-0.09 · n=365 5.40-0.14 · n=365 5.89-0.16 · n=365 6.13-0.17 · n=365 6.12-0.18 · n=365 6.15-0.18 · n=365 6.27-0.18 · n=365 Persistence (baseline) MAE by lead day 1 to 9, tmin

Up to 2 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