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

Miami Intl · NCEP GFS · lead day 2

KMIA · gfs · lead 2

Target day = initialization date + 2 days. This page is a permanent link: the numbers change as history accumulates, the address and the methodology version do not.

Scores use the latest model version segment, gfs-0p25, starting 2026-08-01 — days before that ran a different version of the model and are excluded rather than silently pooled. Of the scored days, 2 carry a QC flag on the observation.

JSON for this page CSV of the daily errors Cite this page

Every number for this combination

All four windows × both initializations × both interpolations × both variables. Errors are computed in °C and shown in °F; CI is a 95 % moving-block bootstrap interval. MAE debiased and skill, debiased remove the constant offset of the window first, so a station that the model always reads a degree too warm is not scored as if it had no skill — compare the two skill columns to see how much of the error is a fixed offset. n QC next to n counts days whose observation carries a QC flag.

VariableInitMethod Windown MAE °F MAE debiased Bias °F RMSE °F ±1 °F±2 °F±3 °F Skill Skill, debiased Period
tmax00Zbilinear30d28 2 QC1.69 [1.08, 2.22]1.02-1.46 [-2.07, -0.76]1.95 [1.38, 2.39]25% [7%, 46%]64%93%+0.15+0.502026-08-03 → 2026-08-30
tmax00Zbilinear90d28 2 QC1.69 [1.02, 2.16]1.02-1.46 [-2.05, -0.62]1.95 [1.28, 2.37]25% [10%, 48%]64%93%+0.15+0.502026-08-03 → 2026-08-30
tmax00Zbilinear365d28 2 QC1.69 [1.02, 2.22]1.02-1.46 [-2.10, -0.66]1.95 [1.29, 2.39]25% [7%, 47%]64%93%+0.15+0.502026-08-03 → 2026-08-30
tmax00Zbilinearall28 2 QC1.69 [0.92, 2.20]1.02-1.46 [-2.08, -0.58]1.95 [1.21, 2.38]25% [8%, 50%]64%93%+0.15+0.502026-08-03 → 2026-08-30
tmax00Znearest30d28 2 QC3.26 [2.49, 4.03]1.07-3.26 [-4.03, -2.49]3.53 [2.76, 4.20]11% [3%, 18%]21%32%-0.63+0.472026-08-03 → 2026-08-30
tmax00Znearest90d28 2 QC3.26 [2.33, 3.99]1.07-3.26 [-3.99, -2.33]3.53 [2.61, 4.17]11% [0%, 20%]21%32%-0.63+0.472026-08-03 → 2026-08-30
tmax00Znearest365d28 2 QC3.26 [2.35, 4.06]1.07-3.26 [-4.06, -2.35]3.53 [2.60, 4.22]11% [0%, 20%]21%32%-0.63+0.472026-08-03 → 2026-08-30
tmax00Znearestall28 2 QC3.26 [2.24, 4.02]1.07-3.26 [-4.02, -2.24]3.53 [2.48, 4.20]11% [0%, 20%]21%32%-0.63+0.472026-08-03 → 2026-08-30
tmax12Zbilinear30d28 2 QC1.74 [1.23, 2.21]1.04-1.41 [-2.11, -0.55]1.96 [1.43, 2.43]21% [4%, 46%]64%93%+0.13+0.492026-08-03 → 2026-08-30
tmax12Zbilinear90d28 2 QC1.74 [1.18, 2.21]1.04-1.41 [-2.09, -0.39]1.96 [1.34, 2.42]21% [7%, 50%]64%93%+0.13+0.492026-08-03 → 2026-08-30
tmax12Zbilinear365d28 2 QC1.74 [1.19, 2.22]1.04-1.41 [-2.14, -0.38]1.96 [1.36, 2.44]21% [5%, 50%]64%93%+0.13+0.492026-08-03 → 2026-08-30
tmax12Zbilinearall28 2 QC1.74 [1.13, 2.21]1.04-1.41 [-2.14, -0.29]1.96 [1.33, 2.41]21% [5%, 55%]64%93%+0.13+0.492026-08-03 → 2026-08-30
tmax12Znearest30d28 2 QC3.36 [2.40, 4.19]1.16-3.33 [-4.18, -2.29]3.64 [2.72, 4.35]7% [0%, 16%]21%36%-0.68+0.422026-08-03 → 2026-08-30
tmax12Znearest90d28 2 QC3.36 [2.26, 4.16]1.16-3.33 [-4.16, -2.13]3.64 [2.51, 4.33]7% [0%, 19%]21%36%-0.68+0.422026-08-03 → 2026-08-30
tmax12Znearest365d28 2 QC3.36 [2.26, 4.22]1.16-3.33 [-4.22, -2.16]3.64 [2.59, 4.35]7% [0%, 19%]21%36%-0.68+0.422026-08-03 → 2026-08-30
tmax12Znearestall28 2 QC3.36 [2.07, 4.21]1.16-3.33 [-4.21, -1.99]3.64 [2.39, 4.35]7% [0%, 20%]21%36%-0.68+0.422026-08-03 → 2026-08-30
tmin00Zbilinear30d28 2 QC4.86 [3.95, 5.85]1.64+4.86 [3.95, 5.85]5.29 [4.44, 6.11]4% [0%, 10%]7%21%-1.23+0.262026-08-03 → 2026-08-30
tmin00Zbilinear90d28 2 QC4.86 [3.87, 6.02]1.64+4.86 [3.87, 6.02]5.29 [4.33, 6.25]4% [0%, 11%]7%21%-1.23+0.262026-08-03 → 2026-08-30
tmin00Zbilinear365d28 2 QC4.86 [3.84, 6.27]1.64+4.86 [3.84, 6.27]5.29 [4.29, 6.52]4% [0%, 11%]7%21%-1.23+0.262026-08-03 → 2026-08-30
tmin00Zbilinearall28 2 QC4.86 [3.85, 5.98]1.64+4.86 [3.85, 5.98]5.29 [4.27, 6.27]4% [0%, 11%]7%21%-1.23+0.262026-08-03 → 2026-08-30
tmin00Znearest30d28 2 QC5.21 [4.28, 6.18]1.65+5.21 [4.28, 6.18]5.63 [4.82, 6.42]4% [0%, 10%]4%14%-1.39+0.252026-08-03 → 2026-08-30
tmin00Znearest90d28 2 QC5.21 [4.19, 6.32]1.65+5.21 [4.19, 6.32]5.63 [4.71, 6.56]4% [0%, 11%]4%14%-1.39+0.252026-08-03 → 2026-08-30
tmin00Znearest365d28 2 QC5.21 [4.14, 6.56]1.65+5.21 [4.14, 6.56]5.63 [4.68, 6.78]4% [0%, 11%]4%14%-1.39+0.252026-08-03 → 2026-08-30
tmin00Znearestall28 2 QC5.21 [4.12, 6.36]1.65+5.21 [4.12, 6.36]5.63 [4.69, 6.58]4% [0%, 11%]4%14%-1.39+0.252026-08-03 → 2026-08-30
tmin12Zbilinear30d28 2 QC4.72 [3.60, 5.92]1.97+4.72 [3.60, 5.92]5.30 [4.13, 6.33]4% [0%, 10%]14%32%-1.17+0.102026-08-03 → 2026-08-30
tmin12Zbilinear90d28 2 QC4.72 [3.55, 6.15]1.97+4.72 [3.55, 6.15]5.30 [4.12, 6.52]4% [0%, 11%]14%32%-1.17+0.102026-08-03 → 2026-08-30
tmin12Zbilinear365d28 2 QC4.72 [3.48, 6.26]1.97+4.72 [3.48, 6.26]5.30 [3.99, 6.64]4% [0%, 11%]14%32%-1.17+0.102026-08-03 → 2026-08-30
tmin12Zbilinearall28 2 QC4.72 [3.47, 6.20]1.97+4.72 [3.47, 6.20]5.30 [4.03, 6.60]4% [0%, 11%]14%32%-1.17+0.102026-08-03 → 2026-08-30
tmin12Znearest30d28 2 QC5.06 [3.98, 6.18]2.04+5.06 [3.98, 6.18]5.62 [4.56, 6.59]4% [0%, 10%]11%29%-1.32+0.082026-08-03 → 2026-08-30
tmin12Znearest90d28 2 QC5.06 [3.92, 6.40]2.04+5.06 [3.92, 6.40]5.62 [4.51, 6.77]4% [0%, 11%]11%29%-1.32+0.082026-08-03 → 2026-08-30
tmin12Znearest365d28 2 QC5.06 [3.87, 6.51]2.04+5.06 [3.87, 6.51]5.62 [4.43, 6.90]4% [0%, 11%]11%29%-1.32+0.082026-08-03 → 2026-08-30
tmin12Znearestall28 2 QC5.06 [3.86, 6.43]2.04+5.06 [3.86, 6.43]5.62 [4.41, 6.84]4% [0%, 11%]11%29%-1.32+0.082026-08-03 → 2026-08-30

+ 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 error, last 90 days

Daily forecast minus observed error in °F for NCEP GFS at KMIA, lead day 2, 00Z bilinear tmax+5+2.50-2.5-52026-08-03: -0.21 °F2026-08-04: -1.01 °F2026-08-05: +0.15 °F2026-08-06: +0.16 °F2026-08-07: -2.58 °F2026-08-08: -1.49 °F2026-08-09: -1.84 °F2026-08-10: -2.97 °F2026-08-11: -2.4 °F2026-08-12: -1.9 °F2026-08-13: -1.35 °F2026-08-14: -4.03 °F2026-08-15: -2.49 °F2026-08-16: -2.71 °F2026-08-17: -1.78 °F2026-08-18: -3.27 °F2026-08-19: -0.7 °F2026-08-20: +1.5 °F2026-08-21: -2.45 °F2026-08-22: -2.16 °F2026-08-23: -1.45 °F2026-08-24: -0.9 °F2026-08-25: -0.9 °F2026-08-26: -1.81 °F2026-08-27: -2.16 °F2026-08-28: -1.58 °F2026-08-29: +1.18 °F2026-08-30: +0.17 °F2026-08-032026-08-30°F, forecast − observed
Forecast − observed, °F, 00Z bilinear daily maximum. The dashed line is zero; the thick line is a centred 7-day mean, drawn only where the whole window is present. Gaps are days with no score, never interpolation. Hover a point for its date and value.
The same series as a table (28 days, newest first) · download CSV
Climatological dayError °F
2026-08-30+0.17
2026-08-29+1.18
2026-08-28-1.58
2026-08-27-2.16
2026-08-26-1.81
2026-08-25-0.90
2026-08-24-0.90
2026-08-23-1.45
2026-08-22-2.16
2026-08-21-2.45
2026-08-20+1.50
2026-08-19-0.70
2026-08-18-3.27
2026-08-17-1.78
2026-08-16-2.71
2026-08-15-2.49
2026-08-14-4.03
2026-08-13-1.35
2026-08-12-1.90
2026-08-11-2.40
2026-08-10-2.97
2026-08-09-1.84
2026-08-08-1.49
2026-08-07-2.58
2026-08-06+0.16
2026-08-05+0.15
2026-08-04-1.01
2026-08-03-0.21

Error distribution

Distribution of the signed daily error in °F for NCEP GFS at KMIA, lead day 2-5 to -4.5 °F: 0 days-4.5 to -4 °F: 1 day-4 to -3.5 °F: 0 days-3.5 to -3 °F: 1 day-3 to -2.5 °F: 3 days-2.5 to -2 °F: 5 days-2 to -1.5 °F: 5 days-1.5 to -1 °F: 4 days-1 to -0.5 °F: 3 days-0.5 to 0 °F: 1 day0 to 0.5 °F: 3 days0.5 to 1 °F: 0 days1 to 1.5 °F: 1 day1.5 to 2 °F: 1 day2 to 2.5 °F: 0 days2.5 to 3 °F: 0 days3 to 3.5 °F: 0 days3.5 to 4 °F: 0 days4 to 4.5 °F: 0 days4.5 to 5 °F: 0 daysdashed: P50 |e| 1.68 · P90 |e| 2.79 °F-50+5 °Fn = 28 days
All scored days for 00Z bilinear daily maximum, 20 equal bins. The dashed marks are the 50th and 90th percentiles of the absolute error, mirrored on both sides. Hover a bar for its bin and count.
Bin counts as a table
From °FTo °FDays
-5.00-4.500
-4.50-4.001
-4.00-3.500
-3.50-3.001
-3.00-2.503
-2.50-2.005
-2.00-1.505
-1.50-1.004
-1.00-0.503
-0.50+0.001
+0.00+0.503
+0.50+1.000
+1.00+1.501
+1.50+2.001
+2.00+2.500
+2.50+3.000
+3.00+3.500
+3.50+4.000
+4.00+4.500
+4.50+5.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08281.69 -1.46

Availability

ModelScored period DaysCoverage
NCEP GFS2026-08-03 → 2026-08-30 28
NCEP GFS2026-08-03 → 2026-08-30 28
NCEP GFS2026-08-03 → 2026-08-30 28
NCEP GFS2026-08-03 → 2026-08-30 28

Compared with other models

Paired bootstrap on the days both models cover (90d, 00Z, bilinear, daily maximum). Negative means NCEP GFS has the lower MAE. "Distinguishable" means the 95 % interval excludes zero.

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 28 -2.71 [-4.39, -1.74] yes
Aurora (IFS init) 1 -6.68 [-6.68, -6.68] yes
FourCastNet v2 (IFS init) 1 -5.09 [-5.09, -5.09] yes
GraphCast (IFS init) 26 -2.93 [-5.68, -1.51] yes
ECMWF IFS HRES 22 -2.20 [-3.76, -1.26] yes
Pangu-Weather (IFS init) 1 -6.94 [-6.94, -6.94] yes
Persistence (baseline) 28 -0.31 [-1.12, 0.49] no

Cite this page

ShortCastCheck, KMIA gfs lead 2, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/gfs/lead/2/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model gfs, lead day 2. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KMIA/model/gfs/lead/2/

Accessed2026-08-31 (data through 2026-08-30; page built 2026-08-31T03:52:06+00:00)

JSON CSV all scores methodology v0.2 licence