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 2

KMIA · graphcast_ifs · lead 2

Target day = initialization date + 2 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, 1 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 1 QC4.65 [3.60, 6.19]1.76-4.65 [-6.19, -3.60]5.08 [3.85, 6.50]0% [0%, 0%]0%35%-1.32+0.132026-08-03 → 2026-08-30
tmax00Zbilinear90d26 1 QC4.65 [3.64, 6.60]1.76-4.65 [-6.60, -3.64]5.08 [3.89, 6.88]0% [0%, 0%]0%35%-1.32+0.132026-08-03 → 2026-08-30
tmax00Zbilinear365d32 1 QC4.40 [3.48, 5.69]1.63-4.40 [-5.69, -3.48]4.81 [3.69, 6.07]0% [0%, 0%]0%34%-0.83+0.272026-01-03 → 2026-08-30
tmax00Zbilinearall32 1 QC4.40 [3.47, 5.89]1.63-4.40 [-5.89, -3.47]4.81 [3.64, 6.24]0% [0%, 0%]0%34%-0.83+0.272026-01-03 → 2026-08-30
tmax00Znearest30d26 1 QC5.02 [4.09, 6.38]1.65-5.02 [-6.38, -4.09]5.38 [4.32, 6.65]0% [0%, 0%]0%23%-1.51+0.182026-08-03 → 2026-08-30
tmax00Znearest90d26 1 QC5.02 [4.12, 6.70]1.65-5.02 [-6.70, -4.12]5.38 [4.35, 6.95]0% [0%, 0%]0%23%-1.51+0.182026-08-03 → 2026-08-30
tmax00Znearest365d32 1 QC4.78 [3.99, 5.91]1.53-4.78 [-5.91, -3.99]5.12 [4.17, 6.25]0% [0%, 0%]0%25%-0.99+0.322026-01-03 → 2026-08-30
tmax00Znearestall32 1 QC4.78 [3.97, 6.08]1.53-4.78 [-6.08, -3.97]5.12 [4.14, 6.40]0% [0%, 0%]0%25%-0.99+0.322026-01-03 → 2026-08-30
tmax12Zbilinear30d26 1 QC4.41 [3.50, 5.76]1.65-4.41 [-5.76, -3.50]4.83 [3.79, 6.07]0% [0%, 0%]12%35%-1.21+0.182026-08-03 → 2026-08-30
tmax12Zbilinear90d26 1 QC4.41 [3.49, 6.10]1.65-4.41 [-6.10, -3.49]4.83 [3.76, 6.38]0% [0%, 0%]12%35%-1.21+0.182026-08-03 → 2026-08-30
tmax12Zbilinear365d31 1 QC4.22 [3.42, 5.43]1.56-4.22 [-5.43, -3.42]4.61 [3.65, 5.76]0% [0%, 0%]10%39%-0.74+0.312026-01-03 → 2026-08-30
tmax12Zbilinearall31 1 QC4.22 [3.39, 5.52]1.56-4.22 [-5.52, -3.39]4.61 [3.64, 5.86]0% [0%, 0%]10%39%-0.74+0.312026-01-03 → 2026-08-30
tmax12Znearest30d26 1 QC4.79 [4.04, 5.90]1.51-4.79 [-5.90, -4.04]5.11 [4.29, 6.15]0% [0%, 0%]4%19%-1.39+0.252026-08-03 → 2026-08-30
tmax12Znearest90d26 1 QC4.79 [4.03, 6.20]1.51-4.79 [-6.20, -4.03]5.11 [4.29, 6.41]0% [0%, 0%]4%19%-1.39+0.252026-08-03 → 2026-08-30
tmax12Znearest365d31 1 QC4.60 [3.94, 5.65]1.44-4.60 [-5.65, -3.94]4.91 [4.14, 5.93]0% [0%, 0%]3%16%-0.90+0.372026-01-03 → 2026-08-30
tmax12Znearestall31 1 QC4.60 [3.89, 5.68]1.44-4.60 [-5.68, -3.89]4.91 [4.07, 5.98]0% [0%, 0%]3%16%-0.90+0.372026-01-03 → 2026-08-30
tmin00Zbilinear30d26 1 QC1.71 [1.24, 2.38]1.72+0.06 [-0.66, 1.11]2.19 [1.47, 3.02]35% [23%, 44%]73%88%+0.21+0.212026-08-03 → 2026-08-30
tmin00Zbilinear90d26 1 QC1.71 [1.21, 2.52]1.72+0.06 [-0.75, 1.41]2.19 [1.38, 3.20]35% [23%, 44%]73%88%+0.21+0.212026-08-03 → 2026-08-30
tmin00Zbilinear365d32 1 QC1.71 [1.31, 2.26]1.71+0.11 [-0.61, 1.07]2.12 [1.50, 2.87]34% [25%, 42%]69%91%+0.39+0.402026-01-03 → 2026-08-30
tmin00Zbilinearall32 1 QC1.71 [1.28, 2.28]1.71+0.11 [-0.61, 1.12]2.12 [1.47, 2.90]34% [25%, 43%]69%91%+0.39+0.402026-01-03 → 2026-08-30
tmin00Znearest30d26 1 QC1.64 [1.12, 2.46]1.66+0.59 [-0.13, 1.68]2.24 [1.41, 3.25]38% [17%, 59%]73%88%+0.24+0.242026-08-03 → 2026-08-30
tmin00Znearest90d26 1 QC1.64 [1.10, 2.72]1.66+0.59 [-0.25, 1.99]2.24 [1.33, 3.49]38% [12%, 58%]73%88%+0.24+0.242026-08-03 → 2026-08-30
tmin00Znearest365d32 1 QC1.66 [1.18, 2.39]1.68+0.58 [-0.13, 1.56]2.18 [1.45, 3.11]34% [11%, 53%]72%91%+0.41+0.412026-01-03 → 2026-08-30
tmin00Znearestall32 1 QC1.66 [1.18, 2.45]1.68+0.58 [-0.13, 1.71]2.18 [1.41, 3.14]34% [13%, 54%]72%91%+0.41+0.412026-01-03 → 2026-08-30
tmin12Zbilinear30d26 1 QC1.61 [1.19, 2.26]1.61+0.00 [-0.60, 1.07]2.12 [1.43, 2.95]35% [20%, 52%]69%88%+0.25+0.262026-08-03 → 2026-08-30
tmin12Zbilinear90d26 1 QC1.61 [1.15, 2.34]1.61+0.00 [-0.74, 1.32]2.12 [1.39, 3.11]35% [19%, 52%]69%88%+0.25+0.262026-08-03 → 2026-08-30
tmin12Zbilinear365d31 1 QC1.57 [1.19, 2.14]1.57+0.04 [-0.66, 1.07]2.06 [1.45, 2.80]35% [22%, 50%]68%90%+0.45+0.452026-01-03 → 2026-08-30
tmin12Zbilinearall31 1 QC1.57 [1.17, 2.14]1.57+0.04 [-0.63, 1.13]2.06 [1.43, 2.86]35% [23%, 52%]68%90%+0.45+0.452026-01-03 → 2026-08-30
tmin12Znearest30d26 1 QC1.60 [1.10, 2.44]1.59+0.53 [-0.10, 1.64]2.17 [1.35, 3.17]38% [17%, 55%]73%88%+0.26+0.272026-08-03 → 2026-08-30
tmin12Znearest90d26 1 QC1.60 [1.08, 2.68]1.59+0.53 [-0.24, 1.90]2.17 [1.28, 3.42]38% [12%, 53%]73%88%+0.26+0.272026-08-03 → 2026-08-30
tmin12Znearest365d31 1 QC1.56 [1.10, 2.31]1.57+0.52 [-0.17, 1.55]2.10 [1.35, 3.09]39% [19%, 53%]74%90%+0.46+0.462026-01-03 → 2026-08-30
tmin12Znearestall31 1 QC1.56 [1.10, 2.35]1.57+0.52 [-0.15, 1.66]2.10 [1.33, 3.11]39% [20%, 53%]74%90%+0.46+0.462026-01-03 → 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 2, 00Z bilinear tmax+10+50-5-102026-08-03: -8.9 °F2026-08-04: -7.86 °F2026-08-05: -6.65 °F2026-08-06: -4.17 °F2026-08-07: -4.8 °F2026-08-08: -2.61 °F2026-08-09: -2.43 °F2026-08-10: -2.24 °F2026-08-11: -3.69 °F2026-08-12: -2.99 °F2026-08-13: -4.19 °F2026-08-14: -5.42 °F2026-08-15: -2.04 °F2026-08-16: -2.92 °F2026-08-17: -3.1 °F2026-08-18: -4.73 °F2026-08-19: -2.01 °F2026-08-20: -2.74 °F2026-08-21: -6.71 °F2026-08-22: -4.38 °F2026-08-23: -6.38 °F2026-08-24: -2.8 °F2026-08-25: -4.58 °F2026-08-26: -7.82 °F2026-08-27: -7.47 °F2026-08-30: -7.23 °F2026-08-032026-08-30°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 (28 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-7.23
2026-08-29—
2026-08-28—
2026-08-27-7.47
2026-08-26-7.82
2026-08-25-4.58
2026-08-24-2.80
2026-08-23-6.38
2026-08-22-4.38
2026-08-21-6.71
2026-08-20-2.74
2026-08-19-2.01
2026-08-18-4.73
2026-08-17-3.10
2026-08-16-2.92
2026-08-15-2.04
2026-08-14-5.42
2026-08-13-4.19
2026-08-12-2.99
2026-08-11-3.69
2026-08-10-2.24
2026-08-09-2.43
2026-08-08-2.61
2026-08-07-4.80
2026-08-06-4.17
2026-08-05-6.65
2026-08-04-7.86
2026-08-03-8.90

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMIA, lead day 2-10 to -9 °F: 0 days-9 to -8 °F: 1 day-8 to -7 °F: 4 days-7 to -6 °F: 3 days-6 to -5 °F: 1 day-5 to -4 °F: 7 days-4 to -3 °F: 5 days-3 to -2 °F: 11 days-2 to -1 °F: 0 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| 3.98 · P90 |e| 7.45 °F-100+10 °Fn = 32 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
-8.00-7.004
-7.00-6.003
-6.00-5.001
-5.00-4.007
-4.00-3.005
-3.00-2.0011
-2.00-1.000
-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
2026-08264.65 -4.65
2026-0163.31 -3.31

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-03 → 2026-08-30 32
GraphCast (IFS init)2026-01-03 → 2026-08-30 32
GraphCast (IFS init)2026-01-03 → 2026-08-30 31
GraphCast (IFS init)2026-01-03 → 2026-08-30 31

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 26 +0.28 [-0.37, 1.09] no
Aurora (IFS init) 1 +0.37 [0.37, 0.37] yes
FourCastNet v2 (IFS init) 1 +1.97 [1.97, 1.97] yes
NCEP GFS 26 +2.93 [1.51, 5.68] yes
ECMWF IFS HRES 20 +0.64 [-0.51, 2.01] no
Pangu-Weather (IFS init) 1 +0.12 [0.12, 0.12] yes
Persistence (baseline) 26 +2.65 [1.43, 4.93] yes

Cite this page

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

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

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