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

Minneapolis-St Paul · GraphCast (IFS init) · lead day 7

KMSP · graphcast_ifs · lead 7

Target day = initialization date + 7 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
tmax00Zbilinear30d23 1 QC7.55 [5.77, 9.52]4.83-7.17 [-9.27, -5.27]9.03 [7.41, 10.52]17% [5%, 30%]30%30%-1.41-0.512026-08-08 → 2026-08-30
tmax00Zbilinear90d23 1 QC7.55 [5.68, 9.64]4.83-7.17 [-9.44, -5.16]9.03 [7.31, 10.56]17% [0%, 30%]30%30%-1.41-0.512026-08-08 → 2026-08-30
tmax00Zbilinear365d29 1 QC7.00 [5.06, 8.83]4.72-6.70 [-8.51, -4.79]8.52 [6.57, 9.98]17% [5%, 29%]31%34%-0.36+0.152026-01-08 → 2026-08-30
tmax00Zbilinearall29 1 QC7.00 [5.00, 8.73]4.72-6.70 [-8.43, -4.74]8.52 [6.59, 9.94]17% [6%, 30%]31%34%-0.36+0.152026-01-08 → 2026-08-30
tmax00Znearest30d23 1 QC7.57 [5.75, 9.58]4.84-7.21 [-9.34, -5.28]9.06 [7.43, 10.58]22% [6%, 38%]30%30%-1.42-0.512026-08-08 → 2026-08-30
tmax00Znearest90d23 1 QC7.57 [5.66, 9.72]4.84-7.21 [-9.53, -5.17]9.06 [7.33, 10.61]22% [0%, 37%]30%30%-1.42-0.512026-08-08 → 2026-08-30
tmax00Znearest365d29 1 QC7.02 [5.06, 8.88]4.71-6.73 [-8.58, -4.79]8.54 [6.57, 10.02]17% [0%, 32%]31%34%-0.37+0.152026-01-08 → 2026-08-30
tmax00Znearestall29 1 QC7.02 [5.03, 8.77]4.71-6.73 [-8.48, -4.76]8.54 [6.54, 10.05]17% [0%, 33%]31%34%-0.37+0.152026-01-08 → 2026-08-30
tmax12Zbilinear30d23 1 QC7.32 [6.03, 8.69]3.31-7.26 [-8.67, -5.96]8.41 [6.69, 9.80]4% [0%, 13%]9%22%-1.34-0.032026-08-08 → 2026-08-30
tmax12Zbilinear90d23 1 QC7.32 [5.96, 8.84]3.31-7.26 [-8.84, -5.91]8.41 [6.52, 9.93]4% [0%, 13%]9%22%-1.34-0.032026-08-08 → 2026-08-30
tmax12Zbilinear365d28 1 QC6.56 [4.43, 8.16]3.60-6.48 [-8.13, -4.32]7.87 [5.65, 9.28]14% [0%, 37%]18%29%-0.31+0.342026-01-08 → 2026-08-30
tmax12Zbilinearall28 1 QC6.56 [4.23, 8.11]3.60-6.48 [-8.06, -4.13]7.87 [5.57, 9.31]14% [0%, 42%]18%29%-0.31+0.342026-01-08 → 2026-08-30
tmax12Znearest30d23 1 QC7.35 [6.04, 8.72]3.30-7.28 [-8.70, -5.98]8.42 [6.72, 9.81]4% [0%, 13%]9%22%-1.35-0.032026-08-08 → 2026-08-30
tmax12Znearest90d23 1 QC7.35 [6.00, 8.89]3.30-7.28 [-8.89, -5.92]8.42 [6.59, 9.94]4% [0%, 13%]9%22%-1.35-0.032026-08-08 → 2026-08-30
tmax12Znearest365d28 1 QC6.57 [4.43, 8.18]3.63-6.48 [-8.16, -4.30]7.87 [5.63, 9.29]14% [0%, 37%]18%29%-0.31+0.332026-01-08 → 2026-08-30
tmax12Znearestall28 1 QC6.57 [4.24, 8.14]3.63-6.48 [-8.08, -4.10]7.87 [5.57, 9.33]14% [0%, 42%]18%29%-0.31+0.332026-01-08 → 2026-08-30
tmin00Zbilinear30d23 1 QC4.76 [3.00, 6.56]4.26-1.65 [-4.34, 1.25]5.70 [3.52, 7.50]9% [0%, 18%]22%30%+0.22+0.302026-08-08 → 2026-08-30
tmin00Zbilinear90d23 1 QC4.76 [2.77, 6.99]4.26-1.65 [-4.64, 1.59]5.70 [3.09, 8.02]9% [0%, 19%]22%30%+0.22+0.302026-08-08 → 2026-08-30
tmin00Zbilinear365d29 1 QC4.32 [2.63, 6.26]4.00-1.33 [-3.97, 1.02]5.28 [3.04, 7.22]10% [0%, 19%]28%38%+0.35+0.372026-01-08 → 2026-08-30
tmin00Zbilinearall29 1 QC4.32 [2.61, 6.02]4.00-1.33 [-3.86, 1.03]5.28 [3.01, 7.04]10% [0%, 20%]28%38%+0.35+0.372026-01-08 → 2026-08-30
tmin00Znearest30d23 1 QC4.75 [3.00, 6.58]4.22-1.77 [-4.45, 1.13]5.71 [3.50, 7.52]9% [0%, 18%]22%35%+0.22+0.312026-08-08 → 2026-08-30
tmin00Znearest90d23 1 QC4.75 [2.76, 7.04]4.22-1.77 [-4.76, 1.46]5.71 [3.13, 8.09]9% [0%, 19%]22%35%+0.22+0.312026-08-08 → 2026-08-30
tmin00Znearest365d29 1 QC4.34 [2.69, 6.29]3.99-1.45 [-4.08, 0.87]5.30 [3.09, 7.30]10% [0%, 19%]28%41%+0.35+0.372026-01-08 → 2026-08-30
tmin00Znearestall29 1 QC4.34 [2.66, 6.03]3.99-1.45 [-3.98, 0.89]5.30 [3.10, 7.09]10% [0%, 20%]28%41%+0.35+0.372026-01-08 → 2026-08-30
tmin12Zbilinear30d23 1 QC3.93 [3.06, 4.99]3.61-1.74 [-3.47, 0.27]4.96 [4.05, 5.79]17% [6%, 27%]35%48%+0.35+0.412026-08-08 → 2026-08-30
tmin12Zbilinear90d23 1 QC3.93 [3.00, 5.28]3.61-1.74 [-3.69, 0.32]4.96 [4.00, 6.00]17% [0%, 29%]35%48%+0.35+0.412026-08-08 → 2026-08-30
tmin12Zbilinear365d28 1 QC3.71 [2.84, 4.78]3.59-1.15 [-3.08, 1.00]4.68 [3.65, 5.59]18% [6%, 29%]36%54%+0.45+0.452026-01-08 → 2026-08-30
tmin12Zbilinearall28 1 QC3.71 [2.71, 4.75]3.59-1.15 [-3.08, 1.04]4.68 [3.53, 5.57]18% [7%, 27%]36%54%+0.45+0.452026-01-08 → 2026-08-30
tmin12Znearest30d23 1 QC4.00 [3.08, 5.13]3.60-1.89 [-3.64, 0.10]4.99 [4.06, 5.87]17% [6%, 27%]39%48%+0.34+0.412026-08-08 → 2026-08-30
tmin12Znearest90d23 1 QC4.00 [3.05, 5.44]3.60-1.89 [-3.83, 0.14]4.99 [3.99, 6.10]17% [0%, 29%]39%48%+0.34+0.412026-08-08 → 2026-08-30
tmin12Znearest365d28 1 QC3.77 [2.84, 4.90]3.60-1.30 [-3.22, 0.82]4.72 [3.66, 5.67]18% [6%, 29%]39%50%+0.44+0.452026-01-08 → 2026-08-30
tmin12Znearestall28 1 QC3.77 [2.75, 4.86]3.60-1.30 [-3.24, 0.86]4.72 [3.56, 5.66]18% [7%, 27%]39%50%+0.44+0.452026-01-08 → 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 KMSP, lead day 7, 00Z bilinear tmax+20+100-10-202026-08-08: +1.65 °F2026-08-09: -12.72 °F2026-08-10: -6.04 °F2026-08-11: -15.14 °F2026-08-12: -12.15 °F2026-08-13: -0.65 °F2026-08-14: -9.2 °F2026-08-15: -11.36 °F2026-08-16: -10.66 °F2026-08-17: -8.71 °F2026-08-18: +1.02 °F2026-08-19: -7.58 °F2026-08-20: +0.08 °F2026-08-21: -11.23 °F2026-08-22: +0.54 °F2026-08-23: -6.24 °F2026-08-24: -10.65 °F2026-08-25: -0.4 °F2026-08-26: +1.08 °F2026-08-27: -13.62 °F2026-08-28: -13.16 °F2026-08-29: -10.87 °F2026-08-30: -8.87 °F2026-08-082026-08-30°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 (23 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-8.87
2026-08-29-10.87
2026-08-28-13.16
2026-08-27-13.62
2026-08-26+1.08
2026-08-25-0.40
2026-08-24-10.65
2026-08-23-6.24
2026-08-22+0.54
2026-08-21-11.23
2026-08-20+0.08
2026-08-19-7.58
2026-08-18+1.02
2026-08-17-8.71
2026-08-16-10.66
2026-08-15-11.36
2026-08-14-9.20
2026-08-13-0.65
2026-08-12-12.15
2026-08-11-15.14
2026-08-10-6.04
2026-08-09-12.72
2026-08-08+1.65

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMSP, lead day 7-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 1 day-14 to -12 °F: 4 days-12 to -10 °F: 6 days-10 to -8 °F: 4 days-8 to -6 °F: 3 days-6 to -4 °F: 1 day-4 to -2 °F: 1 day-2 to 0 °F: 4 days0 to 2 °F: 5 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| 8.71 · P90 |e| 12.81 °F-200+20 °Fn = 29 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
-16.00-14.001
-14.00-12.004
-12.00-10.006
-10.00-8.004
-8.00-6.003
-6.00-4.001
-4.00-2.001
-2.00+0.004
+0.00+2.005
+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
2026-08237.55 -7.17
2026-0164.92 -4.92

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 28
GraphCast (IFS init)2026-01-08 → 2026-08-30 28

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 23 +0.90 [-0.88, 3.53] no
NCEP GFS 23 +2.40 [0.53, 5.33] yes
ECMWF IFS HRES 18 +3.46 [1.15, 6.01] yes
Persistence (baseline) 23 +4.42 [3.06, 6.08] yes

Cite this page

ShortCastCheck, KMSP graphcast_ifs lead 7, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMSP/model/graphcast_ifs/lead/7/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMSP (Minneapolis-St Paul), model graphcast_ifs, lead day 7. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KMSP/model/graphcast_ifs/lead/7/

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