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 9

KMIA · graphcast_ifs · 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, 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, 3 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 QC5.01 [4.29, 5.84]1.84-5.01 [-5.84, -4.29]5.44 [4.48, 6.28]0% [0%, 0%]0%24%-0.70+0.392026-08-10 → 2026-08-30
tmax00Zbilinear90d21 2 QC5.01 [4.11, 5.88]1.84-5.01 [-5.88, -4.11]5.44 [4.34, 6.30]0% [0%, 0%]0%24%-0.70+0.392026-08-10 → 2026-08-30
tmax00Zbilinear365d27 3 QC5.45 [4.43, 6.72]2.47-5.12 [-5.98, -4.13]6.02 [4.66, 7.39]0% [0%, 0%]0%19%-0.39+0.392026-01-10 → 2026-08-30
tmax00Zbilinearall27 3 QC5.45 [4.52, 6.69]2.47-5.12 [-5.93, -4.22]6.02 [4.82, 7.38]0% [0%, 0%]0%19%-0.39+0.392026-01-10 → 2026-08-30
tmax00Znearest30d21 2 QC5.44 [4.90, 6.07]1.70-5.44 [-6.07, -4.90]5.79 [5.08, 6.46]0% [0%, 0%]0%10%-0.84+0.432026-08-10 → 2026-08-30
tmax00Znearest90d21 2 QC5.44 [4.80, 6.10]1.70-5.44 [-6.10, -4.80]5.79 [4.96, 6.48]0% [0%, 0%]0%10%-0.84+0.432026-08-10 → 2026-08-30
tmax00Znearest365d27 3 QC5.81 [5.00, 6.85]2.31-5.50 [-6.17, -4.67]6.30 [5.23, 7.48]0% [0%, 0%]0%7%-0.48+0.422026-01-10 → 2026-08-30
tmax00Znearestall27 3 QC5.81 [5.04, 6.86]2.31-5.50 [-6.20, -4.68]6.30 [5.31, 7.49]0% [0%, 0%]0%7%-0.48+0.422026-01-10 → 2026-08-30
tmax12Zbilinear30d21 2 QC5.06 [4.18, 5.86]2.43-5.06 [-5.86, -4.18]5.88 [4.63, 6.97]0% [0%, 0%]5%29%-0.71+0.192026-08-10 → 2026-08-30
tmax12Zbilinear90d21 2 QC5.06 [4.16, 5.82]2.43-5.06 [-5.82, -4.16]5.88 [4.40, 6.96]0% [0%, 0%]5%29%-0.71+0.192026-08-10 → 2026-08-30
tmax12Zbilinear365d26 3 QC5.52 [4.38, 6.94]2.87-5.52 [-6.94, -4.38]6.58 [4.85, 8.33]0% [0%, 0%]4%27%-0.48+0.262026-01-10 → 2026-08-30
tmax12Zbilinearall26 3 QC5.52 [4.44, 6.99]2.87-5.52 [-6.99, -4.44]6.58 [4.85, 8.35]0% [0%, 0%]4%27%-0.48+0.262026-01-10 → 2026-08-30
tmax12Znearest30d21 2 QC5.46 [4.73, 6.22]2.22-5.46 [-6.22, -4.73]6.13 [5.05, 7.18]0% [0%, 0%]5%14%-0.85+0.262026-08-10 → 2026-08-30
tmax12Znearest90d21 2 QC5.46 [4.69, 6.11]2.22-5.46 [-6.11, -4.69]6.13 [4.90, 7.17]0% [0%, 0%]5%14%-0.85+0.262026-08-10 → 2026-08-30
tmax12Znearest365d26 3 QC5.88 [4.83, 7.20]2.67-5.88 [-7.20, -4.83]6.79 [5.15, 8.42]0% [0%, 0%]4%15%-0.58+0.312026-01-10 → 2026-08-30
tmax12Znearestall26 3 QC5.88 [4.88, 7.22]2.67-5.88 [-7.22, -4.88]6.79 [5.21, 8.47]0% [0%, 0%]4%15%-0.58+0.312026-01-10 → 2026-08-30
tmin00Zbilinear30d21 2 QC1.50 [1.15, 2.19]1.50-0.14 [-1.22, 0.54]1.81 [1.40, 2.50]43% [20%, 57%]76%90%+0.58+0.582026-08-10 → 2026-08-30
tmin00Zbilinear90d21 2 QC1.50 [1.14, 2.34]1.50-0.14 [-1.61, 0.57]1.81 [1.38, 2.70]43% [10%, 57%]76%90%+0.58+0.582026-08-10 → 2026-08-30
tmin00Zbilinear365d27 2 QC3.86 [1.26, 9.69]4.01-1.66 [-5.64, 0.59]6.63 [1.52, 11.97]33% [0%, 52%]59%70%+0.36+0.332026-01-10 → 2026-08-30
tmin00Zbilinearall27 2 QC3.86 [1.30, 10.16]4.01-1.66 [-6.13, 0.69]6.63 [1.59, 12.31]33% [0%, 52%]59%70%+0.36+0.332026-01-10 → 2026-08-30
tmin00Znearest30d21 2 QC1.59 [1.31, 2.10]1.54+0.37 [-0.75, 1.09]1.88 [1.59, 2.38]38% [23%, 46%]67%86%+0.56+0.572026-08-10 → 2026-08-30
tmin00Znearest90d21 2 QC1.59 [1.31, 2.20]1.54+0.37 [-1.11, 1.15]1.88 [1.60, 2.46]38% [20%, 46%]67%86%+0.56+0.572026-08-10 → 2026-08-30
tmin00Znearest365d27 2 QC3.84 [1.42, 9.37]4.04-1.20 [-5.23, 1.12]6.43 [1.70, 11.60]30% [5%, 43%]52%67%+0.36+0.332026-01-10 → 2026-08-30
tmin00Znearestall27 2 QC3.84 [1.43, 9.84]4.04-1.20 [-5.79, 1.16]6.43 [1.73, 11.89]30% [0%, 43%]52%67%+0.36+0.332026-01-10 → 2026-08-30
tmin12Zbilinear30d21 2 QC1.65 [1.26, 2.21]1.63-0.33 [-1.43, 0.39]2.06 [1.53, 2.59]33% [17%, 46%]71%86%+0.54+0.552026-08-10 → 2026-08-30
tmin12Zbilinear90d21 2 QC1.65 [1.20, 2.29]1.63-0.33 [-1.66, 0.44]2.06 [1.48, 2.68]33% [10%, 46%]71%86%+0.54+0.552026-08-10 → 2026-08-30
tmin12Zbilinear365d26 2 QC3.05 [1.41, 6.78]3.04-1.99 [-6.41, 0.17]5.42 [1.74, 9.86]27% [0%, 41%]58%73%+0.46+0.462026-01-10 → 2026-08-30
tmin12Zbilinearall26 2 QC3.05 [1.38, 7.41]3.04-1.99 [-7.30, 0.16]5.42 [1.71, 10.36]27% [0%, 42%]58%73%+0.46+0.462026-01-10 → 2026-08-30
tmin12Znearest30d21 2 QC1.71 [1.37, 2.11]1.69+0.14 [-0.94, 0.87]2.06 [1.60, 2.48]33% [11%, 48%]67%86%+0.52+0.532026-08-10 → 2026-08-30
tmin12Znearest90d21 2 QC1.71 [1.36, 2.18]1.69+0.14 [-1.16, 0.95]2.06 [1.55, 2.52]33% [0%, 50%]67%86%+0.52+0.532026-08-10 → 2026-08-30
tmin12Znearest365d26 2 QC3.03 [1.50, 6.52]3.09-1.54 [-5.96, 0.66]5.26 [1.76, 9.53]27% [0%, 44%]54%73%+0.47+0.452026-01-10 → 2026-08-30
tmin12Znearestall26 2 QC3.03 [1.48, 7.02]3.09-1.54 [-6.91, 0.66]5.26 [1.73, 10.00]27% [0%, 44%]54%73%+0.47+0.452026-01-10 → 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 GraphCast (IFS init) at KMIA, lead day 9, 00Z bilinear tmax+10+50-5-102026-08-10: -3.96 °F2026-08-11: -4.22 °F2026-08-12: -4.2 °F2026-08-13: -4.98 °F2026-08-14: -6.34 °F2026-08-15: -2.26 °F2026-08-16: -3.77 °F2026-08-17: -2.93 °F2026-08-18: -6.95 °F2026-08-19: -6.89 °F2026-08-20: -2.87 °F2026-08-21: -2.86 °F2026-08-22: -9.4 °F2026-08-23: -6.08 °F2026-08-24: -3.25 °F2026-08-25: -3.47 °F2026-08-26: -6.05 °F2026-08-27: -8.4 °F2026-08-28: -8.38 °F2026-08-29: -2.03 °F2026-08-30: -5.93 °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
2026-08-30-5.93
2026-08-29-2.03
2026-08-28-8.38
2026-08-27-8.40
2026-08-26-6.05
2026-08-25-3.47
2026-08-24-3.25
2026-08-23-6.08
2026-08-22-9.40
2026-08-21-2.86
2026-08-20-2.87
2026-08-19-6.89
2026-08-18-6.95
2026-08-17-2.93
2026-08-16-3.77
2026-08-15-2.26
2026-08-14-6.34
2026-08-13-4.98
2026-08-12-4.20
2026-08-11-4.22
2026-08-10-3.96

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMIA, lead day 9-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 0 days-12 to -10 °F: 2 days-10 to -8 °F: 4 days-8 to -6 °F: 5 days-6 to -4 °F: 4 days-4 to -2 °F: 11 days-2 to 0 °F: 0 days0 to 2 °F: 0 days2 to 4 °F: 0 days4 to 6 °F: 1 day6 to 8 °F: 0 days8 to 10 °F: 0 days10 to 12 °F: 0 days12 to 14 °F: 0 days14 to 16 °F: 0 days16 to 18 °F: 0 days18 to 20 °F: 0 daysdashed: P50 |e| 4.46 · P90 |e| 8.8 °F-200+20 °Fn = 27 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
-20.00-18.000
-18.00-16.000
-16.00-14.000
-14.00-12.000
-12.00-10.002
-10.00-8.004
-8.00-6.005
-6.00-4.004
-4.00-2.0011
-2.00+0.000
+0.00+2.000
+2.00+4.000
+4.00+6.001
+6.00+8.000
+8.00+10.000
+10.00+12.000
+12.00+14.000
+14.00+16.000
+16.00+18.000
+18.00+20.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08215.01 -5.01
2026-0166.98 -5.49

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-10 → 2026-08-30 27
GraphCast (IFS init)2026-01-10 → 2026-08-30 27
GraphCast (IFS init)2026-01-10 → 2026-08-30 26
GraphCast (IFS init)2026-01-10 → 2026-08-30 26

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 21 +0.53 [-0.40, 1.42] no
NCEP GFS 21 +2.36 [0.54, 4.45] yes
ECMWF IFS HRES 16 +1.35 [-0.86, 2.31] no
Persistence (baseline) 21 +2.06 [0.88, 3.21] yes

Cite this page

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

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