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 · GraphCast (IFS init) · lead day 1

KMIA · graphcast_ifs · lead 1

Target day = initialization date + 1 day. 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, GRAP_v100, starting 2026-01-01 — 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
tmax00Zbilinear30d27 2 QC4.43 [3.55, 5.55]1.84-4.43 [-5.55, -3.55]4.89 [3.94, 5.88]0% [0%, 0%]7%30%-1.72-0.102026-08-02 → 2026-08-30
tmax00Zbilinear90d27 2 QC4.43 [3.58, 5.87]1.84-4.43 [-5.87, -3.58]4.89 [3.98, 6.16]0% [0%, 0%]7%30%-1.72-0.102026-08-02 → 2026-08-30
tmax00Zbilinear365d33 2 QC4.18 [3.41, 5.23]1.70-4.18 [-5.23, -3.41]4.62 [3.72, 5.61]0% [0%, 0%]9%33%-1.38+0.032026-01-02 → 2026-08-30
tmax00Zbilinearall33 2 QC4.18 [3.37, 5.29]1.70-4.18 [-5.29, -3.37]4.62 [3.66, 5.65]0% [0%, 0%]9%33%-1.38+0.032026-01-02 → 2026-08-30
tmax00Znearest30d27 2 QC4.81 [4.11, 5.72]1.72-4.81 [-5.72, -4.11]5.18 [4.41, 6.04]0% [0%, 0%]4%19%-1.95-0.032026-08-02 → 2026-08-30
tmax00Znearest90d27 2 QC4.81 [4.08, 5.99]1.72-4.81 [-5.99, -4.08]5.18 [4.40, 6.28]0% [0%, 0%]4%19%-1.95-0.032026-08-02 → 2026-08-30
tmax00Znearest365d33 2 QC4.56 [3.88, 5.45]1.59-4.56 [-5.45, -3.88]4.91 [4.13, 5.79]0% [0%, 0%]3%18%-1.59+0.092026-01-02 → 2026-08-30
tmax00Znearestall33 2 QC4.56 [3.85, 5.49]1.59-4.56 [-5.49, -3.85]4.91 [4.10, 5.83]0% [0%, 0%]3%18%-1.59+0.092026-01-02 → 2026-08-30
tmax12Zbilinear30d27 2 QC4.11 [3.40, 5.07]1.68-4.11 [-5.07, -3.40]4.51 [3.73, 5.41]0% [0%, 0%]11%37%-1.52-0.012026-08-02 → 2026-08-30
tmax12Zbilinear90d27 2 QC4.11 [3.44, 5.30]1.68-4.11 [-5.30, -3.44]4.51 [3.77, 5.61]0% [0%, 0%]11%37%-1.52-0.012026-08-02 → 2026-08-30
tmax12Zbilinear365d32 2 QC3.98 [3.38, 4.84]1.50-3.98 [-4.84, -3.38]4.35 [3.63, 5.19]0% [0%, 0%]9%34%-1.24+0.162026-01-02 → 2026-08-30
tmax12Zbilinearall32 2 QC3.98 [3.37, 4.86]1.50-3.98 [-4.86, -3.37]4.35 [3.62, 5.21]0% [0%, 0%]9%34%-1.24+0.162026-01-02 → 2026-08-30
tmax12Znearest30d27 2 QC4.54 [4.01, 5.25]1.53-4.54 [-5.25, -4.01]4.85 [4.25, 5.55]0% [0%, 0%]4%19%-1.79+0.092026-08-02 → 2026-08-30
tmax12Znearest90d27 2 QC4.54 [3.99, 5.42]1.53-4.54 [-5.42, -3.99]4.85 [4.25, 5.72]0% [0%, 0%]4%19%-1.79+0.092026-08-02 → 2026-08-30
tmax12Znearest365d32 2 QC4.39 [3.88, 5.06]1.39-4.39 [-5.06, -3.88]4.68 [4.08, 5.37]0% [0%, 0%]3%19%-1.47+0.222026-01-02 → 2026-08-30
tmax12Znearestall32 2 QC4.39 [3.87, 5.10]1.39-4.39 [-5.10, -3.87]4.68 [4.06, 5.37]0% [0%, 0%]3%19%-1.47+0.222026-01-02 → 2026-08-30
tmin00Zbilinear30d27 2 QC1.60 [1.18, 2.14]1.60+0.28 [-0.38, 1.26]2.11 [1.46, 2.86]37% [21%, 56%]74%85%+0.40+0.402026-08-02 → 2026-08-30
tmin00Zbilinear90d27 2 QC1.60 [1.20, 2.28]1.60+0.28 [-0.43, 1.49]2.11 [1.48, 2.97]37% [19%, 58%]74%85%+0.40+0.402026-08-02 → 2026-08-30
tmin00Zbilinear365d33 2 QC1.71 [1.29, 2.25]1.71+0.27 [-0.36, 1.12]2.17 [1.60, 2.80]36% [20%, 54%]67%82%+0.41+0.402026-01-02 → 2026-08-30
tmin00Zbilinearall33 2 QC1.71 [1.26, 2.28]1.71+0.27 [-0.37, 1.19]2.17 [1.57, 2.83]36% [21%, 55%]67%82%+0.41+0.402026-01-02 → 2026-08-30
tmin00Znearest30d27 2 QC1.60 [1.13, 2.29]1.56+0.79 [0.11, 1.79]2.20 [1.44, 3.08]37% [21%, 52%]74%85%+0.40+0.422026-08-02 → 2026-08-30
tmin00Znearest90d27 2 QC1.60 [1.13, 2.49]1.56+0.79 [0.08, 2.02]2.20 [1.42, 3.27]37% [17%, 52%]74%85%+0.40+0.422026-08-02 → 2026-08-30
tmin00Znearest365d33 2 QC1.74 [1.23, 2.41]1.70+0.75 [0.12, 1.60]2.28 [1.54, 3.01]30% [11%, 47%]70%82%+0.40+0.412026-01-02 → 2026-08-30
tmin00Znearestall33 2 QC1.74 [1.24, 2.48]1.70+0.75 [0.13, 1.72]2.28 [1.56, 3.07]30% [10%, 47%]70%82%+0.40+0.412026-01-02 → 2026-08-30
tmin12Zbilinear30d27 2 QC1.58 [1.10, 2.24]1.60+0.20 [-0.40, 1.05]2.16 [1.38, 3.03]37% [24%, 50%]74%85%+0.41+0.402026-08-02 → 2026-08-30
tmin12Zbilinear90d27 2 QC1.58 [1.13, 2.41]1.60+0.20 [-0.49, 1.24]2.16 [1.43, 3.17]37% [20%, 50%]74%85%+0.41+0.402026-08-02 → 2026-08-30
tmin12Zbilinear365d32 2 QC1.65 [1.20, 2.28]1.67+0.20 [-0.43, 0.99]2.17 [1.51, 2.89]34% [18%, 47%]66%88%+0.43+0.422026-01-02 → 2026-08-30
tmin12Zbilinearall32 2 QC1.65 [1.20, 2.31]1.67+0.20 [-0.48, 1.14]2.17 [1.51, 2.99]34% [20%, 47%]66%88%+0.43+0.422026-01-02 → 2026-08-30
tmin12Znearest30d27 2 QC1.62 [1.06, 2.47]1.58+0.74 [0.17, 1.62]2.26 [1.36, 3.28]41% [24%, 53%]70%81%+0.39+0.412026-08-02 → 2026-08-30
tmin12Znearest90d27 2 QC1.62 [1.08, 2.72]1.58+0.74 [0.07, 1.84]2.26 [1.40, 3.47]41% [20%, 52%]70%81%+0.39+0.412026-08-02 → 2026-08-30
tmin12Znearest365d32 2 QC1.70 [1.16, 2.47]1.70+0.71 [0.11, 1.53]2.30 [1.48, 3.19]38% [19%, 50%]69%78%+0.41+0.412026-01-02 → 2026-08-30
tmin12Znearestall32 2 QC1.70 [1.16, 2.62]1.70+0.71 [0.07, 1.66]2.30 [1.52, 3.27]38% [19%, 50%]69%78%+0.41+0.412026-01-02 → 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 GraphCast (IFS init) at KMIA, lead day 1, 00Z bilinear tmax+10+50-5-102026-08-02: -3.01 °F2026-08-03: -7.53 °F2026-08-04: -6.74 °F2026-08-05: -7.23 °F2026-08-06: -4.51 °F2026-08-07: -4.32 °F2026-08-08: -1.59 °F2026-08-09: -2.38 °F2026-08-10: -2.07 °F2026-08-11: -3.34 °F2026-08-12: -2.79 °F2026-08-13: -5.18 °F2026-08-14: -6.17 °F2026-08-15: -1.6 °F2026-08-16: -3.07 °F2026-08-17: -2.12 °F2026-08-18: -5.17 °F2026-08-19: -2.04 °F2026-08-20: -2.78 °F2026-08-21: -7.92 °F2026-08-22: -3.18 °F2026-08-23: -5.84 °F2026-08-24: -3.23 °F2026-08-25: -4.15 °F2026-08-26: -7.56 °F2026-08-29: -6.72 °F2026-08-30: -7.47 °F2026-08-022026-08-30°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 (29 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-7.47
2026-08-29-6.72
2026-08-28—
2026-08-27—
2026-08-26-7.56
2026-08-25-4.15
2026-08-24-3.23
2026-08-23-5.84
2026-08-22-3.18
2026-08-21-7.92
2026-08-20-2.78
2026-08-19-2.04
2026-08-18-5.17
2026-08-17-2.12
2026-08-16-3.07
2026-08-15-1.60
2026-08-14-6.17
2026-08-13-5.18
2026-08-12-2.79
2026-08-11-3.34
2026-08-10-2.07
2026-08-09-2.38
2026-08-08-1.59
2026-08-07-4.32
2026-08-06-4.51
2026-08-05-7.23
2026-08-04-6.74
2026-08-03-7.53
2026-08-02-3.01

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMIA, lead day 1-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 5 days-7 to -6 °F: 3 days-6 to -5 °F: 3 days-5 to -4 °F: 3 days-4 to -3 °F: 8 days-3 to -2 °F: 8 days-2 to -1 °F: 3 days-1 to 0 °F: 0 days0 to 1 °F: 0 days1 to 2 °F: 0 days2 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| 3.34 · P90 |e| 7.42 °F-100+10 °Fn = 33 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
-8.00-7.005
-7.00-6.003
-6.00-5.003
-5.00-4.003
-4.00-3.008
-3.00-2.008
-2.00-1.003
-1.00+0.000
+0.00+1.000
+1.00+2.000
+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
2026-08274.43 -4.43
2026-0163.06 -3.06

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-02 → 2026-08-30 33
GraphCast (IFS init)2026-01-02 → 2026-08-30 33
GraphCast (IFS init)2026-01-02 → 2026-08-30 32
GraphCast (IFS init)2026-01-02 → 2026-08-30 32

Compared with other models

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

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 27 +0.20 [-0.32, 0.94] no
Aurora (GFS init) 1 +0.54 [0.54, 0.54] yes
Aurora (IFS init) 2 +0.22 [0.10, 0.33] yes
FourCastNet v2 (GFS init) 1 +2.14 [2.14, 2.14] yes
FourCastNet v2 (IFS init) 2 +2.11 [1.46, 2.77] yes
NCEP GFS 27 +2.80 [1.55, 4.59] yes
GraphCast (GFS init) 1 +1.39 [1.39, 1.39] yes
ECMWF IFS HRES 21 +0.19 [-0.86, 0.93] no
Pangu-Weather (GFS init) 1 +1.47 [1.47, 1.47] yes
Pangu-Weather (IFS init) 2 +0.55 [0.03, 1.06] yes
Persistence (baseline) 27 +2.80 [1.82, 4.34] yes

Cite this page

ShortCastCheck, KMIA graphcast_ifs lead 1, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/graphcast_ifs/lead/1/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model graphcast_ifs, lead day 1. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KMIA/model/graphcast_ifs/lead/1/

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