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 9

KMIA · gfs · lead 9

Target day = initialization date + 9 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
tmax00Zbilinear30d21 2 QC2.64 [1.28, 3.93]1.58-2.52 [-3.93, -0.96]3.25 [1.49, 4.36]19% [0%, 38%]48%62%+0.10+0.472026-08-10 → 2026-08-30
tmax00Zbilinear90d21 2 QC2.64 [1.19, 3.94]1.58-2.52 [-3.94, -0.66]3.25 [1.30, 4.36]19% [0%, 39%]48%62%+0.10+0.472026-08-10 → 2026-08-30
tmax00Zbilinear365d21 2 QC2.64 [1.16, 4.04]1.58-2.52 [-4.04, -0.63]3.25 [1.27, 4.45]19% [0%, 39%]48%62%+0.10+0.472026-08-10 → 2026-08-30
tmax00Zbilinearall21 2 QC2.64 [1.16, 4.02]1.58-2.52 [-4.02, -0.61]3.25 [1.26, 4.43]19% [0%, 43%]48%62%+0.10+0.472026-08-10 → 2026-08-30
tmax00Znearest30d21 2 QC4.45 [2.87, 5.84]1.63-4.45 [-5.84, -2.87]4.97 [3.12, 6.33]5% [0%, 17%]10%29%-0.51+0.462026-08-10 → 2026-08-30
tmax00Znearest90d21 2 QC4.45 [2.59, 5.88]1.63-4.45 [-5.88, -2.59]4.97 [2.76, 6.27]5% [0%, 20%]10%29%-0.51+0.462026-08-10 → 2026-08-30
tmax00Znearest365d21 2 QC4.45 [2.57, 6.00]1.63-4.45 [-6.00, -2.57]4.97 [2.76, 6.45]5% [0%, 20%]10%29%-0.51+0.462026-08-10 → 2026-08-30
tmax00Znearestall21 2 QC4.45 [2.46, 6.02]1.63-4.45 [-6.02, -2.46]4.97 [2.69, 6.41]5% [0%, 22%]10%29%-0.51+0.462026-08-10 → 2026-08-30
tmax12Zbilinear30d21 2 QC3.40 [2.40, 4.35]1.48-3.40 [-4.35, -2.40]3.93 [2.80, 4.89]14% [0%, 29%]24%43%-0.15+0.512026-08-10 → 2026-08-30
tmax12Zbilinear90d21 2 QC3.40 [2.40, 4.34]1.48-3.40 [-4.34, -2.40]3.93 [2.75, 4.92]14% [0%, 30%]24%43%-0.15+0.512026-08-10 → 2026-08-30
tmax12Zbilinear365d21 2 QC3.40 [2.36, 4.45]1.48-3.40 [-4.45, -2.36]3.93 [2.70, 5.03]14% [0%, 33%]24%43%-0.15+0.512026-08-10 → 2026-08-30
tmax12Zbilinearall21 2 QC3.40 [2.29, 4.44]1.48-3.40 [-4.44, -2.29]3.93 [2.66, 5.00]14% [0%, 33%]24%43%-0.15+0.512026-08-10 → 2026-08-30
tmax12Znearest30d21 2 QC5.14 [3.99, 6.20]1.38-5.14 [-6.20, -3.99]5.47 [4.22, 6.54]0% [0%, 0%]0%14%-0.74+0.542026-08-10 → 2026-08-30
tmax12Znearest90d21 2 QC5.14 [3.94, 6.19]1.38-5.14 [-6.19, -3.94]5.47 [4.18, 6.53]0% [0%, 0%]0%14%-0.74+0.542026-08-10 → 2026-08-30
tmax12Znearest365d21 2 QC5.14 [3.92, 6.33]1.38-5.14 [-6.33, -3.92]5.47 [4.17, 6.68]0% [0%, 0%]0%14%-0.74+0.542026-08-10 → 2026-08-30
tmax12Znearestall21 2 QC5.14 [3.91, 6.30]1.38-5.14 [-6.30, -3.91]5.47 [4.13, 6.66]0% [0%, 0%]0%14%-0.74+0.542026-08-10 → 2026-08-30
tmin00Zbilinear30d21 2 QC4.60 [3.35, 5.72]1.81+4.60 [3.35, 5.72]5.10 [3.94, 5.95]5% [0%, 19%]10%24%-0.29+0.502026-08-10 → 2026-08-30
tmin00Zbilinear90d21 2 QC4.60 [3.14, 5.79]1.81+4.60 [3.14, 5.79]5.10 [3.75, 6.03]5% [0%, 25%]10%24%-0.29+0.502026-08-10 → 2026-08-30
tmin00Zbilinear365d21 2 QC4.60 [3.19, 5.86]1.81+4.60 [3.19, 5.86]5.10 [3.71, 6.04]5% [0%, 20%]10%24%-0.29+0.502026-08-10 → 2026-08-30
tmin00Zbilinearall21 2 QC4.60 [3.25, 5.81]1.81+4.60 [3.25, 5.81]5.10 [3.74, 6.05]5% [0%, 22%]10%24%-0.29+0.502026-08-10 → 2026-08-30
tmin00Znearest30d21 2 QC4.85 [3.59, 5.97]1.80+4.85 [3.59, 5.97]5.32 [4.08, 6.19]0% [0%, 0%]10%24%-0.36+0.502026-08-10 → 2026-08-30
tmin00Znearest90d21 2 QC4.85 [3.40, 6.06]1.80+4.85 [3.40, 6.06]5.32 [3.86, 6.25]0% [0%, 0%]10%24%-0.36+0.502026-08-10 → 2026-08-30
tmin00Znearest365d21 2 QC4.85 [3.43, 6.08]1.80+4.85 [3.43, 6.08]5.32 [3.84, 6.26]0% [0%, 0%]10%24%-0.36+0.502026-08-10 → 2026-08-30
tmin00Znearestall21 2 QC4.85 [3.46, 6.07]1.80+4.85 [3.46, 6.07]5.32 [3.86, 6.29]0% [0%, 0%]10%24%-0.36+0.502026-08-10 → 2026-08-30
tmin12Zbilinear30d21 2 QC4.40 [3.18, 5.41]1.47+4.40 [3.18, 5.41]4.81 [3.67, 5.59]5% [0%, 19%]19%24%-0.23+0.592026-08-10 → 2026-08-30
tmin12Zbilinear90d21 2 QC4.40 [2.96, 5.51]1.47+4.40 [2.96, 5.51]4.81 [3.42, 5.67]5% [0%, 25%]19%24%-0.23+0.592026-08-10 → 2026-08-30
tmin12Zbilinear365d21 2 QC4.40 [2.94, 5.53]1.47+4.40 [2.94, 5.53]4.81 [3.41, 5.69]5% [0%, 20%]19%24%-0.23+0.592026-08-10 → 2026-08-30
tmin12Zbilinearall21 2 QC4.40 [3.01, 5.54]1.47+4.40 [3.01, 5.54]4.81 [3.45, 5.73]5% [0%, 22%]19%24%-0.23+0.592026-08-10 → 2026-08-30
tmin12Znearest30d21 2 QC4.63 [3.40, 5.65]1.48+4.63 [3.40, 5.65]5.01 [3.83, 5.81]0% [0%, 0%]14%24%-0.30+0.592026-08-10 → 2026-08-30
tmin12Znearest90d21 2 QC4.63 [3.13, 5.77]1.48+4.63 [3.13, 5.77]5.01 [3.57, 5.89]0% [0%, 0%]14%24%-0.30+0.592026-08-10 → 2026-08-30
tmin12Znearest365d21 2 QC4.63 [3.11, 5.79]1.48+4.63 [3.11, 5.79]5.01 [3.56, 5.91]0% [0%, 0%]14%24%-0.30+0.592026-08-10 → 2026-08-30
tmin12Znearestall21 2 QC4.63 [3.17, 5.79]1.48+4.63 [3.17, 5.79]5.01 [3.59, 5.92]0% [0%, 0%]14%24%-0.30+0.592026-08-10 → 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 9, 00Z bilinear tmax+10+50-5-102026-08-10: -2.68 °F2026-08-11: -3.43 °F2026-08-12: -3.81 °F2026-08-13: -3.51 °F2026-08-14: -5.82 °F2026-08-15: -2.58 °F2026-08-16: -3.77 °F2026-08-17: -3.74 °F2026-08-18: -8.28 °F2026-08-19: -3.74 °F2026-08-20: -0.34 °F2026-08-21: -1.98 °F2026-08-22: -2.97 °F2026-08-23: -1.58 °F2026-08-24: -0.07 °F2026-08-25: -1.45 °F2026-08-26: -0.65 °F2026-08-27: -1.55 °F2026-08-28: -1.75 °F2026-08-29: +1.32 °F2026-08-30: -0.51 °F2026-08-102026-08-30°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 (21 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-0.51
2026-08-29+1.32
2026-08-28-1.75
2026-08-27-1.55
2026-08-26-0.65
2026-08-25-1.45
2026-08-24-0.07
2026-08-23-1.58
2026-08-22-2.97
2026-08-21-1.98
2026-08-20-0.34
2026-08-19-3.74
2026-08-18-8.28
2026-08-17-3.74
2026-08-16-3.77
2026-08-15-2.58
2026-08-14-5.82
2026-08-13-3.51
2026-08-12-3.81
2026-08-11-3.43
2026-08-10-2.68

Error distribution

Distribution of the signed daily error in °F for NCEP GFS at KMIA, lead day 9-10 to -9 °F: 0 days-9 to -8 °F: 1 day-8 to -7 °F: 0 days-7 to -6 °F: 0 days-6 to -5 °F: 1 day-5 to -4 °F: 0 days-4 to -3 °F: 6 days-3 to -2 °F: 3 days-2 to -1 °F: 5 days-1 to 0 °F: 4 days0 to 1 °F: 0 days1 to 2 °F: 1 day2 to 3 °F: 0 days3 to 4 °F: 0 days4 to 5 °F: 0 days5 to 6 °F: 0 days6 to 7 °F: 0 days7 to 8 °F: 0 days8 to 9 °F: 0 days9 to 10 °F: 0 daysdashed: P50 |e| 2.58 · P90 |e| 3.81 °F-100+10 °Fn = 21 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
-10.00-9.000
-9.00-8.001
-8.00-7.000
-7.00-6.000
-6.00-5.001
-5.00-4.000
-4.00-3.006
-3.00-2.003
-2.00-1.005
-1.00+0.004
+0.00+1.000
+1.00+2.001
+2.00+3.000
+3.00+4.000
+4.00+5.000
+5.00+6.000
+6.00+7.000
+7.00+8.000
+8.00+9.000
+9.00+10.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08212.64 -2.52

Availability

ModelScored period DaysCoverage
NCEP GFS2026-08-10 → 2026-08-30 21
NCEP GFS2026-08-10 → 2026-08-30 21
NCEP GFS2026-08-10 → 2026-08-30 21
NCEP GFS2026-08-10 → 2026-08-30 21

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 21 -1.84 [-3.12, -0.76] yes
GraphCast (IFS init) 21 -2.36 [-4.45, -0.54] yes
ECMWF IFS HRES 16 -1.58 [-4.16, 1.09] no
Persistence (baseline) 21 -0.31 [-2.13, 1.03] no

Cite this page

ShortCastCheck, KMIA gfs lead 9, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/gfs/lead/9/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model gfs, lead day 9. 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/9/

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