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 5

KMIA · graphcast_ifs · lead 5

Target day = initialization date + 5 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
tmax00Zbilinear30d25 2 QC5.03 [4.31, 6.03]2.10-5.03 [-6.03, -4.31]5.57 [4.84, 6.37]0% [0%, 0%]0%20%-1.47-0.032026-08-06 → 2026-08-30
tmax00Zbilinear90d25 2 QC5.03 [4.22, 6.23]2.10-5.03 [-6.23, -4.22]5.57 [4.76, 6.51]0% [0%, 0%]0%20%-1.47-0.032026-08-06 → 2026-08-30
tmax00Zbilinear365d31 3 QC4.74 [3.93, 5.71]1.91-4.74 [-5.71, -3.93]5.24 [4.26, 6.10]0% [0%, 0%]0%23%-0.73+0.352026-01-06 → 2026-08-30
tmax00Zbilinearall31 3 QC4.74 [3.97, 5.78]1.91-4.74 [-5.78, -3.97]5.24 [4.31, 6.16]0% [0%, 0%]0%23%-0.73+0.352026-01-06 → 2026-08-30
tmax00Znearest30d25 2 QC5.40 [4.77, 6.24]1.93-5.40 [-6.24, -4.77]5.83 [5.20, 6.54]0% [0%, 0%]0%4%-1.65+0.062026-08-06 → 2026-08-30
tmax00Znearest90d25 2 QC5.40 [4.72, 6.39]1.93-5.40 [-6.39, -4.72]5.83 [5.12, 6.64]0% [0%, 0%]0%4%-1.65+0.062026-08-06 → 2026-08-30
tmax00Znearest365d31 3 QC5.12 [4.40, 5.93]1.75-5.12 [-5.93, -4.40]5.53 [4.67, 6.31]0% [0%, 0%]0%6%-0.87+0.402026-01-06 → 2026-08-30
tmax00Znearestall31 3 QC5.12 [4.41, 6.03]1.75-5.12 [-6.03, -4.41]5.53 [4.67, 6.38]0% [0%, 0%]0%6%-0.87+0.402026-01-06 → 2026-08-30
tmax12Zbilinear30d25 2 QC5.26 [4.43, 6.37]1.97-5.26 [-6.37, -4.43]5.76 [4.91, 6.69]0% [0%, 0%]0%16%-1.58+0.042026-08-06 → 2026-08-30
tmax12Zbilinear90d25 2 QC5.26 [4.29, 6.60]1.97-5.26 [-6.60, -4.29]5.76 [4.73, 6.73]0% [0%, 0%]0%16%-1.58+0.042026-08-06 → 2026-08-30
tmax12Zbilinear365d30 2 QC4.86 [3.72, 6.14]1.93-4.86 [-6.14, -3.72]5.40 [4.08, 6.47]0% [0%, 0%]0%23%-0.78+0.342026-01-06 → 2026-08-30
tmax12Zbilinearall30 2 QC4.86 [3.74, 6.11]1.93-4.86 [-6.11, -3.74]5.40 [4.04, 6.44]0% [0%, 0%]0%23%-0.78+0.342026-01-06 → 2026-08-30
tmax12Znearest30d25 2 QC5.65 [4.92, 6.62]1.80-5.65 [-6.62, -4.92]6.06 [5.28, 6.86]0% [0%, 0%]0%4%-1.77+0.122026-08-06 → 2026-08-30
tmax12Znearest90d25 2 QC5.65 [4.79, 6.74]1.80-5.65 [-6.74, -4.79]6.06 [5.08, 6.89]0% [0%, 0%]0%4%-1.77+0.122026-08-06 → 2026-08-30
tmax12Znearest365d30 2 QC5.27 [4.23, 6.42]1.76-5.27 [-6.42, -4.23]5.71 [4.50, 6.68]0% [0%, 0%]0%10%-0.93+0.402026-01-06 → 2026-08-30
tmax12Znearestall30 2 QC5.27 [4.20, 6.36]1.76-5.27 [-6.36, -4.20]5.71 [4.44, 6.65]0% [0%, 0%]0%10%-0.93+0.402026-01-06 → 2026-08-30
tmin00Zbilinear30d25 2 QC1.75 [1.33, 2.29]1.74-0.10 [-0.87, 0.66]2.16 [1.67, 2.63]24% [11%, 36%]72%80%+0.34+0.352026-08-06 → 2026-08-30
tmin00Zbilinear90d25 2 QC1.75 [1.31, 2.45]1.74-0.10 [-1.00, 0.76]2.16 [1.64, 2.72]24% [8%, 38%]72%80%+0.34+0.352026-08-06 → 2026-08-30
tmin00Zbilinear365d31 2 QC1.82 [1.42, 2.37]1.76-0.49 [-1.46, 0.48]2.28 [1.81, 2.80]26% [12%, 39%]68%77%+0.59+0.632026-01-06 → 2026-08-30
tmin00Zbilinearall31 2 QC1.82 [1.40, 2.32]1.76-0.49 [-1.48, 0.46]2.28 [1.75, 2.76]26% [14%, 39%]68%77%+0.59+0.632026-01-06 → 2026-08-30
tmin00Znearest30d25 2 QC1.75 [1.41, 2.20]1.76+0.39 [-0.36, 1.21]2.17 [1.68, 2.67]32% [20%, 43%]68%88%+0.34+0.352026-08-06 → 2026-08-30
tmin00Znearest90d25 2 QC1.75 [1.40, 2.29]1.76+0.39 [-0.48, 1.29]2.17 [1.64, 2.77]32% [19%, 43%]68%88%+0.34+0.352026-08-06 → 2026-08-30
tmin00Znearest365d31 2 QC1.80 [1.47, 2.24]1.79-0.03 [-1.02, 1.03]2.24 [1.78, 2.74]32% [20%, 43%]68%84%+0.59+0.622026-01-06 → 2026-08-30
tmin00Znearestall31 2 QC1.80 [1.45, 2.20]1.79-0.03 [-1.03, 0.94]2.24 [1.69, 2.70]32% [21%, 43%]68%84%+0.59+0.622026-01-06 → 2026-08-30
tmin12Zbilinear30d25 2 QC1.73 [1.35, 2.23]1.69-0.16 [-0.83, 0.52]2.15 [1.64, 2.61]28% [17%, 37%]64%84%+0.34+0.372026-08-06 → 2026-08-30
tmin12Zbilinear90d25 2 QC1.73 [1.32, 2.39]1.69-0.16 [-0.95, 0.62]2.15 [1.61, 2.69]28% [16%, 39%]64%84%+0.34+0.372026-08-06 → 2026-08-30
tmin12Zbilinear365d30 2 QC1.78 [1.42, 2.33]1.63-0.47 [-1.34, 0.39]2.22 [1.76, 2.71]30% [18%, 40%]63%83%+0.60+0.672026-01-06 → 2026-08-30
tmin12Zbilinearall30 2 QC1.78 [1.40, 2.27]1.63-0.47 [-1.37, 0.35]2.22 [1.69, 2.69]30% [19%, 41%]63%83%+0.60+0.672026-01-06 → 2026-08-30
tmin12Znearest30d25 2 QC1.66 [1.32, 2.12]1.70+0.37 [-0.28, 1.08]2.14 [1.64, 2.65]40% [24%, 57%]76%88%+0.37+0.372026-08-06 → 2026-08-30
tmin12Znearest90d25 2 QC1.66 [1.32, 2.20]1.70+0.37 [-0.37, 1.19]2.14 [1.60, 2.73]40% [22%, 57%]76%88%+0.37+0.372026-08-06 → 2026-08-30
tmin12Znearest365d30 2 QC1.65 [1.33, 2.13]1.66+0.04 [-0.86, 0.94]2.16 [1.69, 2.69]40% [24%, 57%]73%87%+0.63+0.662026-01-06 → 2026-08-30
tmin12Znearestall30 2 QC1.65 [1.30, 2.07]1.66+0.04 [-0.88, 0.88]2.16 [1.60, 2.65]40% [24%, 57%]73%87%+0.63+0.662026-01-06 → 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 5, 00Z bilinear tmax+20+100-10-202026-08-06: -3.81 °F2026-08-07: -8.52 °F2026-08-08: -4.63 °F2026-08-09: -2.7 °F2026-08-10: -3.5 °F2026-08-11: -3.56 °F2026-08-12: -3.99 °F2026-08-13: -3.2 °F2026-08-14: -10.72 °F2026-08-15: -2.55 °F2026-08-16: -2.77 °F2026-08-17: -2.52 °F2026-08-18: -8.7 °F2026-08-19: -3.12 °F2026-08-20: -2.19 °F2026-08-21: -3.45 °F2026-08-22: -5.1 °F2026-08-23: -6.98 °F2026-08-24: -3.16 °F2026-08-25: -6.47 °F2026-08-26: -7.67 °F2026-08-27: -7.98 °F2026-08-28: -7.3 °F2026-08-29: -4.04 °F2026-08-30: -7.14 °F2026-08-062026-08-30°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 (25 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-7.14
2026-08-29-4.04
2026-08-28-7.30
2026-08-27-7.98
2026-08-26-7.67
2026-08-25-6.47
2026-08-24-3.16
2026-08-23-6.98
2026-08-22-5.10
2026-08-21-3.45
2026-08-20-2.19
2026-08-19-3.12
2026-08-18-8.70
2026-08-17-2.52
2026-08-16-2.77
2026-08-15-2.55
2026-08-14-10.72
2026-08-13-3.20
2026-08-12-3.99
2026-08-11-3.56
2026-08-10-3.50
2026-08-09-2.70
2026-08-08-4.63
2026-08-07-8.52
2026-08-06-3.81

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMIA, lead day 5-20 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: 1 day-10 to -8 °F: 2 days-8 to -6 °F: 6 days-6 to -4 °F: 4 days-4 to -2 °F: 18 days-2 to 0 °F: 0 days0 to 2 °F: 0 days2 to 4 °F: 0 days4 to 6 °F: 0 days6 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| 3.68 · P90 |e| 7.98 °F-200+20 °Fn = 31 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
-12.00-10.001
-10.00-8.002
-8.00-6.006
-6.00-4.004
-4.00-2.0018
-2.00+0.000
+0.00+2.000
+2.00+4.000
+4.00+6.000
+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
2026-08255.03 -5.03
2026-0163.51 -3.51

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 30
GraphCast (IFS init)2026-01-06 → 2026-08-30 30

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 25 +0.51 [-0.25, 1.65] no
NCEP GFS 25 +2.86 [1.61, 5.01] yes
ECMWF IFS HRES 18 +1.17 [-0.05, 1.89] no
Persistence (baseline) 25 +2.99 [1.94, 4.67] yes

Cite this page

ShortCastCheck, KMIA graphcast_ifs lead 5, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/graphcast_ifs/lead/5/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model graphcast_ifs, lead day 5. 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/5/

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