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

KDCA — Washington Reagan

38.8483, -77.0342 · 4 m· standard offset UTC-5 h · truth product CLIDCA

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 2.17-1.87 · n=30 2.39-2.14 · n=29 2.57-2.11 · n=28 3.02-2.09 · n=27 2.83-2.52 · n=26 3.54-2.33 · n=25 3.79-3.15 · n=24 3.26-3.06 · n=23 4.53-3.71 · n=22 4.55-3.42 · n=21 ECMWF AIFS Single MAE by lead day 1 to 9, tmax
Aurora (GFS init)aurora_gfs 6.64-6.64 · n=2 8.08-8.08 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmax
Aurora (IFS init)aurora_ifs 6.61-6.61 · n=3 6.08-6.08 · n=2 7.76-7.76 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (GFS init)fourcastnet_gfs 4.14-4.14 · n=2 3.63-3.63 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (IFS init)fourcastnet_ifs 5.71-5.71 · n=3 2.44-2.44 · n=2 2.89-2.89 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmax
NCEP GFSgfs 2.33-1.23 · n=30 3.74-2.71 · n=29 3.00-1.60 · n=28 3.58-1.08 · n=27 3.69-1.49 · n=26 4.34-2.18 · n=25 3.59-0.47 · n=24 4.47-2.82 · n=23 5.61-2.04 · n=22 7.28-1.61 · n=21 NCEP GFS MAE by lead day 1 to 9, tmax
GraphCast (GFS init)graphcast_gfs 5.94-5.94 · n=2 5.13-5.13 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmax
GraphCast (IFS init)graphcast_ifs 4.33-4.33 · n=34 4.62-4.62 · n=33 4.53-4.53 · n=32 4.57-4.27 · n=31 4.57-3.72 · n=31 4.96-3.06 · n=31 5.37-3.37 · n=30 5.13-3.35 · n=29 6.13-4.04 · n=28 6.78-4.29 · n=27 GraphCast (IFS init) MAE by lead day 1 to 9, tmax
ECMWF IFS HRESifs_hres 2.93-1.91 · n=24 3.61-2.80 · n=23 4.10-2.96 · n=22 3.91-2.37 · n=22 3.47-1.84 · n=20 3.63-2.75 · n=18 4.71-2.00 · n=19 3.86-1.92 · n=18 3.17-1.74 · n=17 4.35-1.09 · n=16 ECMWF IFS HRES MAE by lead day 1 to 9, tmax
Pangu-Weather (GFS init)pangu_gfs 2.21-2.21 · n=2 2.52-2.52 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmax
Pangu-Weather (IFS init)pangu_ifs 2.80-2.80 · n=3 1.96-1.96 · n=2 3.70-3.70 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmax
Persistence (baseline)persistence 6.02-0.03 · n=363 8.31-0.04 · n=363 8.84+0.01 · n=363 8.90-0.09 · n=363 9.64-0.11 · n=363 9.64-0.10 · n=363 9.42-0.14 · n=363 9.56-0.13 · n=363 9.69-0.11 · n=363 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 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 1.36+0.33 · n=30 1.66+0.55 · n=29 1.89+0.47 · n=28 2.00+0.45 · n=27 2.53+0.43 · n=26 2.86+0.42 · n=25 3.03+0.48 · n=24 2.54+0.58 · n=23 2.73+0.30 · n=22 4.09+0.79 · n=21 ECMWF AIFS Single MAE by lead day 1 to 9, tmin
Aurora (GFS init)aurora_gfs 1.70-1.70 · n=2 1.54-1.54 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmin
Aurora (IFS init)aurora_ifs 2.81-2.81 · n=3 1.48-1.48 · n=2 2.60-2.60 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (GFS init)fourcastnet_gfs 2.53-2.53 · n=2 0.28-0.28 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (IFS init)fourcastnet_ifs 3.84-3.84 · n=3 0.51-0.04 · n=2 0.76+0.76 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmin
NCEP GFSgfs 1.72-0.42 · n=30 2.44-0.70 · n=29 2.83-1.22 · n=28 2.40-0.88 · n=27 2.98-1.23 · n=26 2.98-1.00 · n=25 3.16-0.18 · n=24 3.76-1.00 · n=23 4.44-1.54 · n=22 4.95-0.57 · n=21 NCEP GFS MAE by lead day 1 to 9, tmin
GraphCast (GFS init)graphcast_gfs 2.09-2.09 · n=2 1.22-1.22 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmin
GraphCast (IFS init)graphcast_ifs 1.42+0.07 · n=34 1.72+0.17 · n=33 2.16+0.25 · n=32 2.24+0.44 · n=31 2.66+0.65 · n=31 3.24+1.10 · n=31 3.47+1.40 · n=30 3.55+1.04 · n=29 3.45+0.06 · n=28 4.37-1.01 · n=27 GraphCast (IFS init) MAE by lead day 1 to 9, tmin
ECMWF IFS HRESifs_hres 1.60+0.06 · n=24 2.03+0.35 · n=23 2.14-0.18 · n=22 2.79+0.66 · n=22 2.44+0.48 · n=20 2.51-0.20 · n=18 2.98+0.35 · n=19 2.96+0.98 · n=18 3.07+0.89 · n=17 3.36+1.45 · n=16 ECMWF IFS HRES MAE by lead day 1 to 9, tmin
Pangu-Weather (GFS init)pangu_gfs 2.04-1.33 · n=2 1.92-1.92 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmin
Pangu-Weather (IFS init)pangu_ifs 1.80-1.80 · n=3 2.14-1.07 · n=2 1.47-1.47 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmin
Persistence (baseline)persistence 4.43-0.05 · n=363 6.11-0.05 · n=363 6.46-0.06 · n=363 6.44-0.13 · n=363 6.68-0.13 · n=363 6.99-0.15 · n=363 7.09-0.13 · n=363 7.33-0.16 · n=363 7.58-0.12 · n=363 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