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 8

KMIA · graphcast_ifs · lead 8

Target day = initialization date + 8 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
tmax00Zbilinear30d22 2 QC5.10 [4.02, 6.19]1.99-5.10 [-6.19, -4.02]5.65 [4.39, 6.61]5% [0%, 12%]5%14%-0.81+0.302026-08-09 → 2026-08-30
tmax00Zbilinear90d22 2 QC5.10 [3.87, 6.29]1.99-5.10 [-6.29, -3.87]5.65 [4.06, 6.63]5% [0%, 12%]5%14%-0.81+0.302026-08-09 → 2026-08-30
tmax00Zbilinear365d28 3 QC5.18 [3.99, 6.13]2.39-5.14 [-6.07, -3.96]5.96 [4.48, 7.05]7% [0%, 17%]7%21%-0.41+0.362026-01-09 → 2026-08-30
tmax00Zbilinearall28 3 QC5.18 [4.09, 6.18]2.39-5.14 [-6.13, -4.04]5.96 [4.69, 7.05]7% [0%, 18%]7%21%-0.41+0.362026-01-09 → 2026-08-30
tmax00Znearest30d22 2 QC5.50 [4.63, 6.40]1.85-5.50 [-6.40, -4.63]5.95 [4.90, 6.76]0% [0%, 0%]5%9%-0.95+0.352026-08-09 → 2026-08-30
tmax00Znearest90d22 2 QC5.50 [4.53, 6.46]1.85-5.50 [-6.46, -4.53]5.95 [4.65, 6.78]0% [0%, 0%]5%9%-0.95+0.352026-08-09 → 2026-08-30
tmax00Znearest365d28 3 QC5.52 [4.60, 6.30]2.24-5.50 [-6.30, -4.58]6.20 [4.90, 7.14]4% [0%, 14%]7%14%-0.50+0.402026-01-09 → 2026-08-30
tmax00Znearestall28 3 QC5.52 [4.64, 6.35]2.24-5.50 [-6.30, -4.61]6.20 [5.11, 7.16]4% [0%, 15%]7%14%-0.50+0.402026-01-09 → 2026-08-30
tmax12Zbilinear30d22 2 QC4.92 [4.13, 5.79]1.98-4.92 [-5.79, -4.13]5.42 [4.46, 6.19]0% [0%, 0%]5%18%-0.75+0.302026-08-09 → 2026-08-30
tmax12Zbilinear90d22 2 QC4.92 [3.96, 5.84]1.98-4.92 [-5.84, -3.96]5.42 [4.26, 6.20]0% [0%, 0%]5%18%-0.75+0.302026-08-09 → 2026-08-30
tmax12Zbilinear365d27 3 QC4.95 [4.07, 5.81]2.35-4.84 [-5.57, -3.97]5.62 [4.59, 6.55]0% [0%, 0%]7%26%-0.32+0.382026-01-09 → 2026-08-30
tmax12Zbilinearall27 3 QC4.95 [4.16, 5.80]2.35-4.84 [-5.63, -3.98]5.62 [4.62, 6.54]0% [0%, 0%]7%26%-0.32+0.382026-01-09 → 2026-08-30
tmax12Znearest30d22 2 QC5.33 [4.65, 6.02]1.80-5.33 [-6.02, -4.65]5.74 [4.91, 6.41]0% [0%, 0%]0%9%-0.89+0.362026-08-09 → 2026-08-30
tmax12Znearest90d22 2 QC5.33 [4.55, 6.03]1.80-5.33 [-6.03, -4.55]5.74 [4.71, 6.37]0% [0%, 0%]0%9%-0.89+0.362026-08-09 → 2026-08-30
tmax12Znearest365d27 3 QC5.29 [4.52, 5.94]2.16-5.21 [-5.81, -4.42]5.86 [4.97, 6.62]0% [0%, 0%]4%15%-0.42+0.432026-01-09 → 2026-08-30
tmax12Znearestall27 3 QC5.29 [4.54, 5.96]2.16-5.21 [-5.86, -4.43]5.86 [4.98, 6.59]0% [0%, 0%]4%15%-0.42+0.432026-01-09 → 2026-08-30
tmin00Zbilinear30d22 2 QC1.63 [1.18, 2.22]1.61-0.38 [-1.40, 0.38]1.94 [1.46, 2.48]41% [10%, 67%]64%91%+0.51+0.522026-08-09 → 2026-08-30
tmin00Zbilinear90d22 2 QC1.63 [1.16, 2.28]1.61-0.38 [-1.77, 0.41]1.94 [1.43, 2.59]41% [0%, 67%]64%91%+0.51+0.522026-08-09 → 2026-08-30
tmin00Zbilinear365d28 2 QC2.61 [1.35, 4.83]2.44-1.63 [-4.46, 0.10]3.83 [1.66, 6.37]32% [0%, 58%]50%79%+0.52+0.572026-01-09 → 2026-08-30
tmin00Zbilinearall28 2 QC2.61 [1.34, 5.33]2.44-1.63 [-5.14, 0.17]3.83 [1.64, 6.61]32% [0%, 59%]50%79%+0.52+0.572026-01-09 → 2026-08-30
tmin00Znearest30d22 2 QC1.66 [1.30, 2.04]1.66+0.10 [-0.89, 0.87]1.92 [1.58, 2.27]27% [7%, 45%]64%91%+0.50+0.512026-08-09 → 2026-08-30
tmin00Znearest90d22 2 QC1.66 [1.29, 2.11]1.66+0.10 [-1.25, 0.95]1.92 [1.57, 2.34]27% [0%, 46%]64%91%+0.50+0.512026-08-09 → 2026-08-30
tmin00Znearest365d28 2 QC2.56 [1.43, 4.56]2.48-1.18 [-4.03, 0.55]3.66 [1.72, 6.01]21% [0%, 40%]50%79%+0.52+0.562026-01-09 → 2026-08-30
tmin00Znearestall28 2 QC2.56 [1.45, 5.03]2.48-1.18 [-4.80, 0.65]3.66 [1.74, 6.24]21% [0%, 42%]50%79%+0.52+0.562026-01-09 → 2026-08-30
tmin12Zbilinear30d22 2 QC1.64 [1.17, 2.12]1.62-0.49 [-1.33, 0.16]1.93 [1.47, 2.32]32% [7%, 59%]59%86%+0.51+0.522026-08-09 → 2026-08-30
tmin12Zbilinear90d22 2 QC1.64 [1.14, 2.28]1.62-0.49 [-1.61, 0.21]1.93 [1.43, 2.37]32% [0%, 60%]59%86%+0.51+0.522026-08-09 → 2026-08-30
tmin12Zbilinear365d27 2 QC2.11 [1.39, 3.29]1.98-1.18 [-2.94, -0.06]2.66 [1.71, 3.87]26% [0%, 50%]52%74%+0.62+0.662026-01-09 → 2026-08-30
tmin12Zbilinearall27 2 QC2.11 [1.32, 3.66]1.98-1.18 [-3.50, 0.01]2.66 [1.65, 4.18]26% [0%, 55%]52%74%+0.62+0.662026-01-09 → 2026-08-30
tmin12Znearest30d22 2 QC1.63 [1.28, 1.97]1.62+0.01 [-0.82, 0.67]1.84 [1.53, 2.11]23% [4%, 43%]68%91%+0.51+0.522026-08-09 → 2026-08-30
tmin12Znearest90d22 2 QC1.63 [1.24, 2.07]1.62+0.01 [-1.06, 0.73]1.84 [1.50, 2.17]23% [0%, 43%]68%91%+0.51+0.522026-08-09 → 2026-08-30
tmin12Znearest365d27 2 QC2.04 [1.41, 3.04]2.00-0.71 [-2.50, 0.44]2.50 [1.68, 3.60]19% [0%, 38%]59%81%+0.63+0.662026-01-09 → 2026-08-30
tmin12Znearestall27 2 QC2.04 [1.39, 3.33]2.00-0.71 [-3.06, 0.52]2.50 [1.64, 3.90]19% [0%, 40%]59%81%+0.63+0.662026-01-09 → 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 8, 00Z bilinear tmax+20+100-10-202026-08-09: -3.36 °F2026-08-10: -4 °F2026-08-11: -4.83 °F2026-08-12: -3.53 °F2026-08-13: -3.06 °F2026-08-14: -6.2 °F2026-08-15: -2.66 °F2026-08-16: -3.15 °F2026-08-17: -0.79 °F2026-08-18: -10.68 °F2026-08-19: -4.39 °F2026-08-20: -3.82 °F2026-08-21: -4.35 °F2026-08-22: -9.76 °F2026-08-23: -6.21 °F2026-08-24: -2.15 °F2026-08-25: -5.07 °F2026-08-26: -8.35 °F2026-08-27: -8.21 °F2026-08-28: -7.06 °F2026-08-29: -4.24 °F2026-08-30: -6.2 °F2026-08-092026-08-30°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 (22 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-6.20
2026-08-29-4.24
2026-08-28-7.06
2026-08-27-8.21
2026-08-26-8.35
2026-08-25-5.07
2026-08-24-2.15
2026-08-23-6.21
2026-08-22-9.76
2026-08-21-4.35
2026-08-20-3.82
2026-08-19-4.39
2026-08-18-10.68
2026-08-17-0.79
2026-08-16-3.15
2026-08-15-2.66
2026-08-14-6.20
2026-08-13-3.06
2026-08-12-3.53
2026-08-11-4.83
2026-08-10-4.00
2026-08-09-3.36

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMIA, lead day 8-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 1 day-12 to -10 °F: 1 day-10 to -8 °F: 3 days-8 to -6 °F: 6 days-6 to -4 °F: 6 days-4 to -2 °F: 9 days-2 to 0 °F: 1 day0 to 2 °F: 1 day2 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| 4.37 · P90 |e| 8.77 °F-200+20 °Fn = 28 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
-14.00-12.001
-12.00-10.001
-10.00-8.003
-8.00-6.006
-6.00-4.006
-4.00-2.009
-2.00+0.001
+0.00+2.001
+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
2026-08225.10 -5.10
2026-0165.50 -5.29

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-09 → 2026-08-30 28
GraphCast (IFS init)2026-01-09 → 2026-08-30 28
GraphCast (IFS init)2026-01-09 → 2026-08-30 27
GraphCast (IFS init)2026-01-09 → 2026-08-30 27

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 22 +0.89 [-0.46, 2.36] no
NCEP GFS 22 +2.54 [0.61, 4.80] yes
ECMWF IFS HRES 17 +1.05 [-1.05, 2.00] no
Persistence (baseline) 22 +2.28 [0.86, 3.56] yes

Cite this page

ShortCastCheck, KMIA graphcast_ifs lead 8, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMIA/model/graphcast_ifs/lead/8/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMIA (Miami Intl), model graphcast_ifs, lead day 8. 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/8/

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