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 0

KMIA · gfs · lead 0

Target day = initialization date + 0 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
tmax00Zbilinear30d30 2 QC1.52 [1.16, 1.88]0.90-1.25 [-1.75, -0.75]1.73 [1.31, 2.09]30% [10%, 50%]77%93%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d30 2 QC1.52 [1.12, 1.90]0.90-1.25 [-1.74, -0.67]1.73 [1.26, 2.10]30% [11%, 50%]77%93%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d30 2 QC1.52 [1.08, 1.91]0.90-1.25 [-1.77, -0.67]1.73 [1.20, 2.11]30% [10%, 53%]77%93%——2026-08-01 → 2026-08-30
tmax00Zbilinearall30 2 QC1.52 [1.04, 1.90]0.90-1.25 [-1.77, -0.63]1.73 [1.16, 2.11]30% [10%, 56%]77%93%——2026-08-01 → 2026-08-30
tmax00Znearest30d30 2 QC3.11 [2.58, 3.73]1.03-3.11 [-3.73, -2.58]3.40 [2.87, 3.96]10% [3%, 17%]20%37%——2026-08-01 → 2026-08-30
tmax00Znearest90d30 2 QC3.11 [2.50, 3.74]1.03-3.11 [-3.74, -2.50]3.40 [2.80, 3.95]10% [2%, 19%]20%37%——2026-08-01 → 2026-08-30
tmax00Znearest365d30 2 QC3.11 [2.49, 3.78]1.03-3.11 [-3.78, -2.48]3.40 [2.80, 4.02]10% [0%, 19%]20%37%——2026-08-01 → 2026-08-30
tmax00Znearestall30 2 QC3.11 [2.48, 3.77]1.03-3.11 [-3.77, -2.47]3.40 [2.76, 3.98]10% [0%, 18%]20%37%——2026-08-01 → 2026-08-30
tmin00Zbilinear30d30 2 QC4.51 [3.72, 5.39]2.00+4.50 [3.71, 5.38]5.13 [4.19, 5.99]3% [0%, 10%]13%30%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d30 2 QC4.51 [3.72, 5.48]2.00+4.50 [3.71, 5.48]5.13 [4.22, 6.10]3% [0%, 13%]13%30%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d30 2 QC4.51 [3.70, 5.76]2.00+4.50 [3.68, 5.75]5.13 [4.16, 6.27]3% [0%, 12%]13%30%——2026-08-01 → 2026-08-30
tmin00Zbilinearall30 2 QC4.51 [3.69, 5.55]2.00+4.50 [3.68, 5.55]5.13 [4.20, 6.14]3% [0%, 14%]13%30%——2026-08-01 → 2026-08-30
tmin00Znearest30d30 2 QC4.96 [4.00, 5.95]2.02+4.96 [4.00, 5.95]5.52 [4.48, 6.41]0% [0%, 0%]10%20%——2026-08-01 → 2026-08-30
tmin00Znearest90d30 2 QC4.96 [4.04, 6.08]2.02+4.96 [4.04, 6.08]5.52 [4.52, 6.58]0% [0%, 0%]10%20%——2026-08-01 → 2026-08-30
tmin00Znearest365d30 2 QC4.96 [3.95, 6.32]2.02+4.96 [3.95, 6.32]5.52 [4.45, 6.69]0% [0%, 0%]10%20%——2026-08-01 → 2026-08-30
tmin00Znearestall30 2 QC4.96 [3.94, 6.14]2.02+4.96 [3.94, 6.14]5.52 [4.47, 6.68]0% [0%, 0%]10%20%——2026-08-01 → 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 0, 00Z bilinear tmax+5+2.50-2.5-52026-08-01: -1.26 °F2026-08-02: +0.32 °F2026-08-03: -0.89 °F2026-08-04: -0.35 °F2026-08-05: -2 °F2026-08-06: +0.97 °F2026-08-07: -1.59 °F2026-08-08: -1.74 °F2026-08-09: -1.25 °F2026-08-10: -1.76 °F2026-08-11: -2.09 °F2026-08-12: -2.73 °F2026-08-13: -1.55 °F2026-08-14: -3.96 °F2026-08-15: -1.41 °F2026-08-16: -2.02 °F2026-08-17: -1.17 °F2026-08-18: -1.72 °F2026-08-19: +0.79 °F2026-08-20: -0.54 °F2026-08-21: -1.56 °F2026-08-22: -1.69 °F2026-08-23: -0.75 °F2026-08-24: -1.07 °F2026-08-25: -0.76 °F2026-08-26: -1.88 °F2026-08-27: -3.36 °F2026-08-28: -2.35 °F2026-08-29: +1.47 °F2026-08-30: +0.49 °F2026-08-012026-08-30°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 (30 days, newest first) · download CSV
Climatological dayError °F
2026-08-30+0.49
2026-08-29+1.47
2026-08-28-2.35
2026-08-27-3.36
2026-08-26-1.88
2026-08-25-0.76
2026-08-24-1.07
2026-08-23-0.75
2026-08-22-1.69
2026-08-21-1.56
2026-08-20-0.54
2026-08-19+0.79
2026-08-18-1.72
2026-08-17-1.17
2026-08-16-2.02
2026-08-15-1.41
2026-08-14-3.96
2026-08-13-1.55
2026-08-12-2.73
2026-08-11-2.09
2026-08-10-1.76
2026-08-09-1.25
2026-08-08-1.74
2026-08-07-1.59
2026-08-06+0.97
2026-08-05-2.00
2026-08-04-0.35
2026-08-03-0.89
2026-08-02+0.32
2026-08-01-1.26

Error distribution

Distribution of the signed daily error in °F for NCEP GFS at KMIA, lead day 0-5 to -4.5 °F: 0 days-4.5 to -4 °F: 0 days-4 to -3.5 °F: 1 day-3.5 to -3 °F: 1 day-3 to -2.5 °F: 1 day-2.5 to -2 °F: 4 days-2 to -1.5 °F: 8 days-1.5 to -1 °F: 5 days-1 to -0.5 °F: 4 days-0.5 to 0 °F: 1 day0 to 0.5 °F: 2 days0.5 to 1 °F: 2 days1 to 1.5 °F: 1 day1.5 to 2 °F: 0 days2 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| 1.51 · P90 |e| 2.39 °F-50+5 °Fn = 30 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
-4.50-4.000
-4.00-3.501
-3.50-3.001
-3.00-2.501
-2.50-2.004
-2.00-1.508
-1.50-1.005
-1.00-0.504
-0.50+0.001
+0.00+0.502
+0.50+1.002
+1.00+1.501
+1.50+2.000
+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
2026-08301.52 -1.25

Availability

ModelScored period DaysCoverage
NCEP GFS2026-08-01 → 2026-08-30 30
NCEP GFS2026-08-01 → 2026-08-30 30

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 30 -2.54 [-3.40, -1.96] yes
Aurora (GFS init) 2 -5.98 [-6.96, -4.99] yes
Aurora (IFS init) 3 -5.50 [-6.09, -4.31] yes
FourCastNet v2 (GFS init) 2 -3.56 [-4.47, -2.66] yes
FourCastNet v2 (IFS init) 3 -3.65 [-4.65, -3.14] yes
GraphCast (GFS init) 2 -5.07 [-6.03, -4.11] yes
GraphCast (IFS init) 28 -2.22 [-3.40, -1.34] yes
ECMWF IFS HRES 24 -1.63 [-2.86, -0.97] yes
Pangu-Weather (GFS init) 2 -4.85 [-5.92, -3.79] yes
Pangu-Weather (IFS init) 3 -3.21 [-3.77, -2.10] yes

Cite this page

ShortCastCheck, KMIA gfs lead 0, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/gfs/lead/0/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model gfs, lead day 0. 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/0/

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