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 4

KMIA · graphcast_ifs · 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, 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
tmax00Zbilinear30d25 1 QC4.57 [3.86, 5.60]1.80-4.57 [-5.60, -3.86]5.09 [4.40, 5.90]0% [0%, 0%]12%32%-1.04+0.202026-08-05 → 2026-08-29
tmax00Zbilinear90d25 1 QC4.57 [3.81, 5.80]1.80-4.57 [-5.80, -3.81]5.09 [4.33, 5.98]0% [0%, 0%]12%32%-1.04+0.202026-08-05 → 2026-08-29
tmax00Zbilinear365d31 1 QC4.31 [3.60, 5.27]1.68-4.31 [-5.27, -3.60]4.80 [3.89, 5.62]0% [0%, 0%]10%29%-0.55+0.392026-01-05 → 2026-08-29
tmax00Zbilinearall31 1 QC4.31 [3.59, 5.34]1.68-4.31 [-5.34, -3.59]4.80 [3.89, 5.66]0% [0%, 0%]10%29%-0.55+0.392026-01-05 → 2026-08-29
tmax00Znearest30d25 1 QC4.93 [4.30, 5.82]1.62-4.93 [-5.82, -4.30]5.36 [4.72, 6.08]0% [0%, 0%]8%12%-1.20+0.282026-08-05 → 2026-08-29
tmax00Znearest90d25 1 QC4.93 [4.23, 5.96]1.62-4.93 [-5.96, -4.23]5.36 [4.63, 6.12]0% [0%, 0%]8%12%-1.20+0.282026-08-05 → 2026-08-29
tmax00Znearest365d31 1 QC4.70 [4.08, 5.54]1.50-4.70 [-5.54, -4.08]5.10 [4.29, 5.83]0% [0%, 0%]6%10%-0.69+0.452026-01-05 → 2026-08-29
tmax00Znearestall31 1 QC4.70 [4.07, 5.59]1.50-4.70 [-5.59, -4.07]5.10 [4.30, 5.89]0% [0%, 0%]6%10%-0.69+0.452026-01-05 → 2026-08-29
tmax12Zbilinear30d25 1 QC4.73 [4.05, 5.68]1.78-4.73 [-5.68, -4.05]5.21 [4.50, 6.07]0% [0%, 0%]4%24%-1.11+0.212026-08-05 → 2026-08-29
tmax12Zbilinear90d25 1 QC4.73 [4.00, 5.87]1.78-4.73 [-5.87, -4.00]5.21 [4.37, 6.10]0% [0%, 0%]4%24%-1.11+0.212026-08-05 → 2026-08-29
tmax12Zbilinear365d30 1 QC4.44 [3.63, 5.44]1.68-4.44 [-5.44, -3.63]4.92 [3.92, 5.79]0% [0%, 0%]3%30%-0.57+0.412026-01-05 → 2026-08-29
tmax12Zbilinearall30 1 QC4.44 [3.63, 5.46]1.68-4.44 [-5.46, -3.63]4.92 [3.88, 5.83]0% [0%, 0%]3%30%-0.57+0.412026-01-05 → 2026-08-29
tmax12Znearest30d25 1 QC5.09 [4.51, 5.91]1.62-5.09 [-5.91, -4.51]5.50 [4.83, 6.25]0% [0%, 0%]0%12%-1.27+0.282026-08-05 → 2026-08-29
tmax12Znearest90d25 1 QC5.09 [4.45, 6.04]1.62-5.09 [-6.04, -4.45]5.50 [4.70, 6.25]0% [0%, 0%]0%12%-1.27+0.282026-08-05 → 2026-08-29
tmax12Znearest365d30 1 QC4.84 [4.11, 5.68]1.52-4.84 [-5.68, -4.11]5.23 [4.31, 5.99]0% [0%, 0%]0%13%-0.71+0.462026-01-05 → 2026-08-29
tmax12Znearestall30 1 QC4.84 [4.10, 5.71]1.52-4.84 [-5.71, -4.10]5.23 [4.28, 6.04]0% [0%, 0%]0%13%-0.71+0.462026-01-05 → 2026-08-29
tmin00Zbilinear30d25 1 QC1.52 [1.19, 1.88]1.50-0.08 [-0.72, 0.76]1.86 [1.44, 2.26]48% [37%, 60%]72%88%+0.51+0.522026-08-05 → 2026-08-29
tmin00Zbilinear90d25 1 QC1.52 [1.16, 1.97]1.50-0.08 [-0.76, 0.97]1.86 [1.41, 2.37]48% [36%, 62%]72%88%+0.51+0.522026-08-05 → 2026-08-29
tmin00Zbilinear365d31 1 QC1.46 [1.16, 1.82]1.36-0.30 [-0.93, 0.52]1.81 [1.46, 2.21]48% [37%, 59%]71%90%+0.68+0.692026-01-05 → 2026-08-29
tmin00Zbilinearall31 1 QC1.46 [1.13, 1.79]1.36-0.30 [-0.95, 0.46]1.81 [1.39, 2.19]48% [38%, 60%]71%90%+0.68+0.692026-01-05 → 2026-08-29
tmin00Znearest30d25 1 QC1.46 [1.14, 1.88]1.50+0.43 [-0.19, 1.32]1.91 [1.48, 2.41]44% [32%, 56%]76%88%+0.53+0.522026-08-05 → 2026-08-29
tmin00Znearest90d25 1 QC1.46 [1.13, 1.97]1.50+0.43 [-0.25, 1.64]1.91 [1.43, 2.55]44% [31%, 57%]76%88%+0.53+0.522026-08-05 → 2026-08-29
tmin00Znearest365d31 1 QC1.35 [1.04, 1.75]1.37+0.18 [-0.48, 1.06]1.80 [1.36, 2.29]48% [35%, 63%]77%90%+0.70+0.692026-01-05 → 2026-08-29
tmin00Znearestall31 1 QC1.35 [1.00, 1.78]1.37+0.18 [-0.46, 1.02]1.80 [1.27, 2.29]48% [33%, 63%]77%90%+0.70+0.692026-01-05 → 2026-08-29
tmin12Zbilinear30d25 1 QC1.52 [1.18, 1.87]1.52+0.01 [-0.63, 0.83]1.91 [1.47, 2.28]40% [25%, 56%]76%88%+0.51+0.512026-08-05 → 2026-08-29
tmin12Zbilinear90d25 1 QC1.52 [1.13, 1.94]1.52+0.01 [-0.67, 1.00]1.91 [1.41, 2.31]40% [23%, 57%]76%88%+0.51+0.512026-08-05 → 2026-08-29
tmin12Zbilinear365d30 1 QC1.36 [0.91, 1.81]1.36-0.02 [-0.66, 0.70]1.77 [1.25, 2.22]47% [29%, 68%]80%90%+0.69+0.692026-01-05 → 2026-08-29
tmin12Zbilinearall30 1 QC1.36 [0.87, 1.76]1.36-0.02 [-0.62, 0.71]1.77 [1.14, 2.16]47% [31%, 70%]80%90%+0.69+0.692026-01-05 → 2026-08-29
tmin12Znearest30d25 1 QC1.52 [1.23, 1.91]1.52+0.55 [-0.10, 1.44]1.96 [1.54, 2.41]36% [22%, 48%]80%88%+0.51+0.512026-08-05 → 2026-08-29
tmin12Znearest90d25 1 QC1.52 [1.22, 1.93]1.52+0.55 [-0.12, 1.66]1.96 [1.45, 2.48]36% [21%, 48%]80%88%+0.51+0.512026-08-05 → 2026-08-29
tmin12Znearest365d30 1 QC1.38 [1.04, 1.78]1.37+0.49 [-0.11, 1.26]1.82 [1.28, 2.31]40% [27%, 52%]83%90%+0.69+0.692026-01-05 → 2026-08-29
tmin12Znearestall30 1 QC1.38 [0.97, 1.76]1.37+0.49 [-0.08, 1.26]1.82 [1.20, 2.28]40% [28%, 53%]83%90%+0.69+0.692026-01-05 → 2026-08-29

+ 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 4, 00Z bilinear tmax+20+100-10-202026-08-05: -5.43 °F2026-08-06: -4.63 °F2026-08-07: -6.93 °F2026-08-08: -2.9 °F2026-08-09: -2.8 °F2026-08-10: -2.68 °F2026-08-11: -3.47 °F2026-08-12: -2.98 °F2026-08-13: -3.33 °F2026-08-14: -10.83 °F2026-08-15: -1.62 °F2026-08-16: -2.87 °F2026-08-17: -3.68 °F2026-08-18: -5.29 °F2026-08-19: -1.33 °F2026-08-20: -1.03 °F2026-08-21: -6.44 °F2026-08-22: -5.18 °F2026-08-23: -5.91 °F2026-08-24: -3.57 °F2026-08-25: -5.13 °F2026-08-26: -7.51 °F2026-08-27: -8.31 °F2026-08-28: -5.26 °F2026-08-29: -5.09 °F2026-08-052026-08-29°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
2026-08-29-5.09
2026-08-28-5.26
2026-08-27-8.31
2026-08-26-7.51
2026-08-25-5.13
2026-08-24-3.57
2026-08-23-5.91
2026-08-22-5.18
2026-08-21-6.44
2026-08-20-1.03
2026-08-19-1.33
2026-08-18-5.29
2026-08-17-3.68
2026-08-16-2.87
2026-08-15-1.62
2026-08-14-10.83
2026-08-13-3.33
2026-08-12-2.98
2026-08-11-3.47
2026-08-10-2.68
2026-08-09-2.80
2026-08-08-2.90
2026-08-07-6.93
2026-08-06-4.63
2026-08-05-5.43

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMIA, lead day 4-20 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: 1 day-8 to -6 °F: 3 days-6 to -4 °F: 9 days-4 to -2 °F: 14 days-2 to 0 °F: 3 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| 3.57 · P90 |e| 6.93 °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
-10.00-8.001
-8.00-6.003
-6.00-4.009
-4.00-2.0014
-2.00+0.003
+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
2026-08254.57 -4.57
2026-0163.23 -3.23

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-05 → 2026-08-29 31
GraphCast (IFS init)2026-01-05 → 2026-08-29 31
GraphCast (IFS init)2026-01-05 → 2026-08-29 30
GraphCast (IFS init)2026-01-05 → 2026-08-29 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 +0.23 [-0.39, 1.14] no
NCEP GFS 25 +2.47 [1.41, 4.25] yes
ECMWF IFS HRES 19 +1.29 [-0.03, 1.92] no
Persistence (baseline) 25 +2.33 [1.41, 3.62] yes

Cite this page

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

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