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 4

KMIA · gfs · lead 4

Target day = initialization date + 4 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
tmax00Zbilinear30d26 2 QC2.04 [1.44, 2.52]1.17-1.91 [-2.51, -1.07]2.42 [1.73, 2.90]19% [7%, 33%]58%81%+0.07+0.472026-08-05 → 2026-08-30
tmax00Zbilinear90d26 2 QC2.04 [1.25, 2.49]1.17-1.91 [-2.49, -0.75]2.42 [1.49, 2.89]19% [7%, 33%]58%81%+0.07+0.472026-08-05 → 2026-08-30
tmax00Zbilinear365d26 2 QC2.04 [1.20, 2.56]1.17-1.91 [-2.56, -0.68]2.42 [1.34, 2.93]19% [6%, 33%]58%81%+0.07+0.472026-08-05 → 2026-08-30
tmax00Zbilinearall26 2 QC2.04 [1.19, 2.52]1.17-1.91 [-2.52, -0.66]2.42 [1.38, 2.90]19% [5%, 36%]58%81%+0.07+0.472026-08-05 → 2026-08-30
tmax00Znearest30d26 2 QC3.60 [2.50, 4.41]1.29-3.60 [-4.41, -2.50]3.95 [2.88, 4.63]8% [0%, 20%]15%31%-0.64+0.412026-08-05 → 2026-08-30
tmax00Znearest90d26 2 QC3.60 [2.12, 4.40]1.29-3.60 [-4.40, -2.12]3.95 [2.48, 4.61]8% [0%, 24%]15%31%-0.64+0.412026-08-05 → 2026-08-30
tmax00Znearest365d26 2 QC3.60 [2.07, 4.48]1.29-3.60 [-4.48, -2.07]3.95 [2.45, 4.69]8% [0%, 24%]15%31%-0.64+0.412026-08-05 → 2026-08-30
tmax00Znearestall26 2 QC3.60 [2.04, 4.42]1.29-3.60 [-4.42, -2.04]3.95 [2.37, 4.66]8% [0%, 25%]15%31%-0.64+0.412026-08-05 → 2026-08-30
tmax12Zbilinear30d26 2 QC2.08 [1.43, 2.61]1.28-1.76 [-2.51, -0.78]2.40 [1.71, 2.86]19% [7%, 38%]42%85%+0.05+0.412026-08-05 → 2026-08-30
tmax12Zbilinear90d26 2 QC2.08 [1.25, 2.60]1.28-1.76 [-2.52, -0.49]2.40 [1.50, 2.88]19% [7%, 40%]42%85%+0.05+0.412026-08-05 → 2026-08-30
tmax12Zbilinear365d26 2 QC2.08 [1.27, 2.66]1.28-1.76 [-2.57, -0.56]2.40 [1.50, 2.91]19% [6%, 40%]42%85%+0.05+0.412026-08-05 → 2026-08-30
tmax12Zbilinearall26 2 QC2.08 [1.16, 2.63]1.28-1.76 [-2.54, -0.20]2.40 [1.46, 2.90]19% [6%, 47%]42%85%+0.05+0.412026-08-05 → 2026-08-30
tmax12Znearest30d26 2 QC3.78 [2.74, 4.57]1.41-3.70 [-4.56, -2.51]4.11 [3.10, 4.75]4% [0%, 12%]19%27%-0.72+0.362026-08-05 → 2026-08-30
tmax12Znearest90d26 2 QC3.78 [2.43, 4.55]1.41-3.70 [-4.55, -2.23]4.11 [2.74, 4.76]4% [0%, 15%]19%27%-0.72+0.362026-08-05 → 2026-08-30
tmax12Znearest365d26 2 QC3.78 [2.43, 4.61]1.41-3.70 [-4.61, -2.23]4.11 [2.78, 4.81]4% [0%, 15%]19%27%-0.72+0.362026-08-05 → 2026-08-30
tmax12Znearestall26 2 QC3.78 [2.17, 4.59]1.41-3.70 [-4.59, -1.88]4.11 [2.48, 4.76]4% [0%, 14%]19%27%-0.72+0.362026-08-05 → 2026-08-30
tmin00Zbilinear30d26 2 QC4.64 [3.61, 5.89]1.91+4.64 [3.61, 5.89]5.22 [4.22, 6.26]4% [0%, 11%]12%27%-0.49+0.402026-08-05 → 2026-08-30
tmin00Zbilinear90d26 2 QC4.64 [3.49, 6.07]1.91+4.64 [3.49, 6.07]5.22 [4.13, 6.51]4% [0%, 12%]12%27%-0.49+0.402026-08-05 → 2026-08-30
tmin00Zbilinear365d26 2 QC4.64 [3.49, 6.38]1.91+4.64 [3.49, 6.38]5.22 [4.09, 6.71]4% [0%, 11%]12%27%-0.49+0.402026-08-05 → 2026-08-30
tmin00Zbilinearall26 2 QC4.64 [3.40, 6.13]1.91+4.64 [3.40, 6.13]5.22 [4.07, 6.57]4% [0%, 12%]12%27%-0.49+0.402026-08-05 → 2026-08-30
tmin00Znearest30d26 2 QC4.95 [3.86, 6.19]1.92+4.95 [3.86, 6.19]5.52 [4.51, 6.58]4% [0%, 11%]12%23%-0.59+0.392026-08-05 → 2026-08-30
tmin00Znearest90d26 2 QC4.95 [3.73, 6.35]1.92+4.95 [3.73, 6.35]5.52 [4.40, 6.80]4% [0%, 12%]12%23%-0.59+0.392026-08-05 → 2026-08-30
tmin00Znearest365d26 2 QC4.95 [3.75, 6.67]1.92+4.95 [3.75, 6.67]5.52 [4.36, 6.96]4% [0%, 11%]12%23%-0.59+0.392026-08-05 → 2026-08-30
tmin00Znearestall26 2 QC4.95 [3.64, 6.43]1.92+4.95 [3.64, 6.43]5.52 [4.35, 6.86]4% [0%, 12%]12%23%-0.59+0.392026-08-05 → 2026-08-30
tmin12Zbilinear30d26 2 QC4.45 [3.43, 5.75]2.04+4.44 [3.41, 5.75]5.10 [4.12, 6.13]8% [0%, 14%]23%35%-0.43+0.352026-08-05 → 2026-08-30
tmin12Zbilinear90d26 2 QC4.45 [3.31, 6.04]2.04+4.44 [3.29, 6.04]5.10 [4.02, 6.42]8% [0%, 14%]23%35%-0.43+0.352026-08-05 → 2026-08-30
tmin12Zbilinear365d26 2 QC4.45 [3.31, 6.26]2.04+4.44 [3.29, 6.26]5.10 [3.96, 6.59]8% [0%, 14%]23%35%-0.43+0.352026-08-05 → 2026-08-30
tmin12Zbilinearall26 2 QC4.45 [3.26, 6.19]2.04+4.44 [3.24, 6.19]5.10 [3.95, 6.50]8% [0%, 14%]23%35%-0.43+0.352026-08-05 → 2026-08-30
tmin12Znearest30d26 2 QC4.67 [3.62, 5.98]2.03+4.67 [3.62, 5.98]5.32 [4.33, 6.35]8% [0%, 14%]15%31%-0.50+0.362026-08-05 → 2026-08-30
tmin12Znearest90d26 2 QC4.67 [3.50, 6.23]2.03+4.67 [3.50, 6.23]5.32 [4.23, 6.57]8% [0%, 14%]15%31%-0.50+0.362026-08-05 → 2026-08-30
tmin12Znearest365d26 2 QC4.67 [3.51, 6.44]2.03+4.67 [3.51, 6.44]5.32 [4.16, 6.77]8% [0%, 14%]15%31%-0.50+0.362026-08-05 → 2026-08-30
tmin12Znearestall26 2 QC4.67 [3.46, 6.35]2.03+4.67 [3.46, 6.35]5.32 [4.12, 6.63]8% [0%, 14%]15%31%-0.50+0.362026-08-05 → 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 4, 00Z bilinear tmax+10+50-5-102026-08-05: -1.19 °F2026-08-06: -0.03 °F2026-08-07: -3.15 °F2026-08-08: -1.83 °F2026-08-09: -1.6 °F2026-08-10: -2.69 °F2026-08-11: -3.38 °F2026-08-12: -2.89 °F2026-08-13: -1.07 °F2026-08-14: -4.51 °F2026-08-15: -2.1 °F2026-08-16: -2.52 °F2026-08-17: -1.64 °F2026-08-18: -5.57 °F2026-08-19: -0.7 °F2026-08-20: -0.69 °F2026-08-21: -2.95 °F2026-08-22: -2.34 °F2026-08-23: -4.09 °F2026-08-24: -1 °F2026-08-25: -1.53 °F2026-08-26: -0.61 °F2026-08-27: -1.33 °F2026-08-28: -1.92 °F2026-08-29: +1.21 °F2026-08-30: +0.43 °F2026-08-052026-08-30°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 (26 days, newest first) · download CSV
Climatological dayError °F
2026-08-30+0.43
2026-08-29+1.21
2026-08-28-1.92
2026-08-27-1.33
2026-08-26-0.61
2026-08-25-1.53
2026-08-24-1.00
2026-08-23-4.09
2026-08-22-2.34
2026-08-21-2.95
2026-08-20-0.69
2026-08-19-0.70
2026-08-18-5.57
2026-08-17-1.64
2026-08-16-2.52
2026-08-15-2.10
2026-08-14-4.51
2026-08-13-1.07
2026-08-12-2.89
2026-08-11-3.38
2026-08-10-2.69
2026-08-09-1.60
2026-08-08-1.83
2026-08-07-3.15
2026-08-06-0.03
2026-08-05-1.19

Error distribution

Distribution of the signed daily error in °F for NCEP GFS at KMIA, lead day 4-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 0 days-7 to -6 °F: 0 days-6 to -5 °F: 1 day-5 to -4 °F: 2 days-4 to -3 °F: 2 days-3 to -2 °F: 6 days-2 to -1 °F: 9 days-1 to 0 °F: 4 days0 to 1 °F: 1 day1 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| 1.73 · P90 |e| 3.73 °F-100+10 °Fn = 26 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
-9.00-8.000
-8.00-7.000
-7.00-6.000
-6.00-5.001
-5.00-4.002
-4.00-3.002
-3.00-2.006
-2.00-1.009
-1.00+0.004
+0.00+1.001
+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
2026-08262.04 -1.91

Availability

ModelScored period DaysCoverage
NCEP GFS2026-08-05 → 2026-08-30 26
NCEP GFS2026-08-05 → 2026-08-30 26
NCEP GFS2026-08-05 → 2026-08-30 26
NCEP GFS2026-08-05 → 2026-08-30 26

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 26 -2.28 [-3.44, -1.64] yes
GraphCast (IFS init) 25 -2.47 [-4.25, -1.41] yes
ECMWF IFS HRES 20 -1.49 [-3.45, -0.43] yes
Persistence (baseline) 26 -0.16 [-0.90, 0.41] no

Cite this page

ShortCastCheck, KMIA gfs lead 4, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/gfs/lead/4/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model gfs, lead day 4. 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/4/

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