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

KMIA — Miami Intl

25.7906, -80.3164 · 3 m· standard offset UTC-5 h · truth product CLIMIA

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 4.38-4.38 · n=29 4.53-4.53 · n=28 4.51-4.51 · n=27 4.46-4.46 · n=26 4.82-4.82 · n=24 4.72-4.72 · n=24 4.72-4.72 · n=23 4.68-4.68 · n=22 4.54-4.54 · n=21 ECMWF AIFS Single MAE by lead day 1 to 9, tmax
Aurora (GFS init)aurora_gfs 6.63-6.63 · n=2 6.92-6.92 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmax
Aurora (IFS init)aurora_ifs 7.01-7.01 · n=2 6.81-6.81 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (GFS init)fourcastnet_gfs 5.07-5.07 · n=2 4.90-4.90 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmax
FourCastNet v2 (IFS init)fourcastnet_ifs 5.26-5.26 · n=2 5.24-5.24 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmax
NCEP GFSgfs 1.90-1.63 · n=29 1.74-1.41 · n=28 2.09-1.88 · n=27 2.08-1.76 · n=26 2.38-2.08 · n=25 2.58-2.51 · n=24 2.79-2.73 · n=23 2.82-2.73 · n=22 3.40-3.40 · n=21 NCEP GFS MAE by lead day 1 to 9, tmax
GraphCast (GFS init)graphcast_gfs 6.20-6.20 · n=2 5.95-5.95 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmax
GraphCast (IFS init)graphcast_ifs 4.11-4.11 · n=27 4.41-4.41 · n=26 4.56-4.56 · n=25 4.73-4.73 · n=25 5.26-5.26 · n=25 5.32-5.32 · n=24 5.23-5.23 · n=23 4.92-4.92 · n=22 5.06-5.06 · n=21 GraphCast (IFS init) MAE by lead day 1 to 9, tmax
ECMWF IFS HRESifs_hres 3.42-3.42 · n=24 3.64-3.64 · n=22 3.87-3.87 · n=22 4.03-4.00 · n=21 4.20-4.07 · n=20 3.05-3.05 · n=19 3.77-3.71 · n=17 4.06-3.99 · n=17 3.92-3.80 · n=15 ECMWF IFS HRES MAE by lead day 1 to 9, tmax
Pangu-Weather (GFS init)pangu_gfs 5.94-5.94 · n=2 6.01-6.01 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmax
Pangu-Weather (IFS init)pangu_ifs 5.99-5.99 · n=2 6.18-6.18 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmax
Persistence (baseline)persistence 1.41+0.03 · n=90 1.89+0.09 · n=90 2.18+0.02 · n=90 2.23+0.01 · n=90 2.28-0.06 · n=90 2.36-0.09 · n=90 2.48-0.12 · n=90 2.53-0.18 · n=90 2.61-0.23 · n=90 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 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 1.82+1.50 · n=29 1.80+1.45 · n=28 1.75+1.37 · n=27 1.77+1.39 · n=26 1.81+1.47 · n=24 1.78+1.23 · n=24 1.93+1.33 · n=23 1.95+1.42 · n=22 2.06+1.57 · n=21 ECMWF AIFS Single MAE by lead day 1 to 9, tmin
Aurora (GFS init)aurora_gfs 2.94+1.06 · n=2 2.35-2.35 · n=1 Aurora (GFS init) MAE by lead day 1 to 9, tmin
Aurora (IFS init)aurora_ifs 3.13+1.28 · n=2 1.56-1.56 · n=1 Aurora (IFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (GFS init)fourcastnet_gfs 3.26+3.26 · n=2 0.76+0.76 · n=1 FourCastNet v2 (GFS init) MAE by lead day 1 to 9, tmin
FourCastNet v2 (IFS init)fourcastnet_ifs 3.22+2.78 · n=2 0.69+0.69 · n=1 FourCastNet v2 (IFS init) MAE by lead day 1 to 9, tmin
NCEP GFSgfs 4.63+4.63 · n=29 4.72+4.72 · n=28 4.49+4.49 · n=27 4.45+4.44 · n=26 4.45+4.45 · n=25 4.30+4.30 · n=24 4.31+4.31 · n=23 4.60+3.97 · n=22 4.40+4.40 · n=21 NCEP GFS MAE by lead day 1 to 9, tmin
GraphCast (GFS init)graphcast_gfs 2.92+2.92 · n=2 0.56-0.56 · n=1 GraphCast (GFS init) MAE by lead day 1 to 9, tmin
GraphCast (IFS init)graphcast_ifs 1.58+0.20 · n=27 1.61+0.00 · n=26 1.49-0.19 · n=25 1.52+0.01 · n=25 1.73-0.16 · n=25 1.66-0.37 · n=24 1.69-0.49 · n=23 1.64-0.49 · n=22 1.65-0.33 · n=21 GraphCast (IFS init) MAE by lead day 1 to 9, tmin
ECMWF IFS HRESifs_hres 1.46-0.31 · n=24 1.71-0.07 · n=22 1.74-0.23 · n=22 1.56-0.02 · n=21 1.82+0.53 · n=20 1.40+0.65 · n=19 1.76+0.50 · n=17 1.76+0.91 · n=17 2.13+0.35 · n=15 ECMWF IFS HRES MAE by lead day 1 to 9, tmin
Pangu-Weather (GFS init)pangu_gfs 2.71+1.46 · n=2 1.01-1.01 · n=1 Pangu-Weather (GFS init) MAE by lead day 1 to 9, tmin
Pangu-Weather (IFS init)pangu_ifs 1.97-0.17 · n=2 1.69-1.69 · n=1 Pangu-Weather (IFS init) MAE by lead day 1 to 9, tmin
Persistence (baseline)persistence 2.40-0.04 · n=90 2.90-0.06 · n=90 2.98-0.09 · n=90 2.99-0.12 · n=90 3.01-0.14 · n=90 3.14-0.14 · n=90 3.16-0.11 · n=90 3.36-0.13 · n=90 3.47-0.16 · n=90 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