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 0

KMSP · graphcast_ifs · lead 0

Target day = initialization date + 0 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
tmax00Zbilinear30d28 1 QC5.55 [4.94, 6.10]1.82-5.55 [-6.10, -4.94]5.97 [5.31, 6.55]4% [0%, 11%]4%21%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 1 QC5.55 [4.87, 6.21]1.82-5.55 [-6.21, -4.87]5.97 [5.23, 6.65]4% [0%, 11%]4%21%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 1 QC4.95 [3.60, 5.97]2.11-4.95 [-5.97, -3.60]5.54 [4.29, 6.43]9% [0%, 23%]9%29%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 1 QC4.95 [3.64, 5.95]2.11-4.95 [-5.95, -3.64]5.54 [4.39, 6.37]9% [0%, 21%]9%29%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 1 QC5.34 [4.66, 5.95]1.91-5.34 [-5.95, -4.66]5.82 [5.12, 6.43]4% [0%, 11%]14%21%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 1 QC5.34 [4.58, 6.11]1.91-5.34 [-6.11, -4.58]5.82 [5.00, 6.53]4% [0%, 11%]14%21%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 1 QC4.80 [3.53, 5.84]2.12-4.80 [-5.84, -3.53]5.41 [4.24, 6.29]9% [0%, 23%]18%29%——2026-01-01 → 2026-08-30
tmax00Znearestall34 1 QC4.80 [3.54, 5.79]2.12-4.80 [-5.79, -3.54]5.41 [4.22, 6.25]9% [0%, 21%]18%29%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 1 QC1.45 [1.10, 1.86]1.33-0.61 [-1.13, -0.05]1.88 [1.34, 2.38]43% [28%, 59%]71%93%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 1 QC1.45 [1.05, 1.88]1.33-0.61 [-1.22, -0.02]1.88 [1.20, 2.37]43% [28%, 61%]71%93%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 1 QC1.75 [1.14, 2.56]1.67-0.85 [-1.81, -0.08]2.39 [1.41, 3.41]41% [26%, 56%]65%88%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 1 QC1.75 [1.14, 2.55]1.67-0.85 [-1.77, -0.07]2.39 [1.38, 3.35]41% [26%, 59%]65%88%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 1 QC1.45 [1.09, 1.88]1.39-0.42 [-1.01, 0.17]1.94 [1.42, 2.42]50% [39%, 62%]68%93%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 1 QC1.45 [1.03, 1.92]1.39-0.42 [-1.09, 0.18]1.94 [1.31, 2.44]50% [35%, 65%]68%93%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 1 QC1.74 [1.12, 2.60]1.73-0.74 [-1.78, 0.09]2.45 [1.45, 3.48]47% [30%, 62%]62%88%——2026-01-01 → 2026-08-30
tmin00Znearestall34 1 QC1.74 [1.11, 2.56]1.73-0.74 [-1.73, 0.13]2.45 [1.45, 3.44]47% [32%, 62%]62%88%——2026-01-01 → 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 0, 00Z bilinear tmax+10+50-5-102026-08-01: -6.86 °F2026-08-02: -6.29 °F2026-08-03: -2.47 °F2026-08-04: -2.87 °F2026-08-05: -5.96 °F2026-08-06: -4.18 °F2026-08-07: -2.25 °F2026-08-08: -5.19 °F2026-08-09: -7.83 °F2026-08-10: -6.95 °F2026-08-11: -5.13 °F2026-08-12: -6.5 °F2026-08-13: -2.12 °F2026-08-14: -5.3 °F2026-08-15: -3.7 °F2026-08-16: -8.99 °F2026-08-17: -6.33 °F2026-08-18: -7.35 °F2026-08-19: -8.27 °F2026-08-20: -0.92 °F2026-08-21: -8.7 °F2026-08-22: -5.55 °F2026-08-23: -6.15 °F2026-08-24: -8.84 °F2026-08-25: -2.31 °F2026-08-28: -4.71 °F2026-08-29: -6.86 °F2026-08-30: -6.86 °F2026-08-012026-08-30°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 (30 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-6.86
2026-08-29-6.86
2026-08-28-4.71
2026-08-27—
2026-08-26—
2026-08-25-2.31
2026-08-24-8.84
2026-08-23-6.15
2026-08-22-5.55
2026-08-21-8.70
2026-08-20-0.92
2026-08-19-8.27
2026-08-18-7.35
2026-08-17-6.33
2026-08-16-8.99
2026-08-15-3.70
2026-08-14-5.30
2026-08-13-2.12
2026-08-12-6.50
2026-08-11-5.13
2026-08-10-6.95
2026-08-09-7.83
2026-08-08-5.19
2026-08-07-2.25
2026-08-06-4.18
2026-08-05-5.96
2026-08-04-2.87
2026-08-03-2.47
2026-08-02-6.29
2026-08-01-6.86

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMSP, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 4 days-8 to -7 °F: 2 days-7 to -6 °F: 8 days-6 to -5 °F: 5 days-5 to -4 °F: 3 days-4 to -3 °F: 2 days-3 to -2 °F: 7 days-2 to -1 °F: 0 days-1 to 0 °F: 3 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| 5.24 · P90 |e| 8.14 °F-100+10 °Fn = 34 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
-9.00-8.004
-8.00-7.002
-7.00-6.008
-6.00-5.005
-5.00-4.003
-4.00-3.002
-3.00-2.007
-2.00-1.000
-1.00+0.003
+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
2026-08285.55 -5.55
2026-0162.16 -2.16

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-01 → 2026-08-30 34
GraphCast (IFS init)2026-01-01 → 2026-08-30 34

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 28 +2.45 [2.11, 2.81] yes
Aurora (GFS init) 2 +0.49 [-2.07, 3.04] no
Aurora (IFS init) 3 +0.40 [-0.15, 1.14] no
FourCastNet v2 (GFS init) 2 +1.66 [0.03, 3.29] yes
FourCastNet v2 (IFS init) 3 +0.26 [-0.28, 1.34] no
NCEP GFS 28 +3.40 [2.55, 4.28] yes
GraphCast (GFS init) 2 -0.12 [-2.48, 2.25] no
ECMWF IFS HRES 22 +3.18 [2.58, 3.92] yes
Pangu-Weather (GFS init) 2 +2.44 [-0.30, 5.18] no
Pangu-Weather (IFS init) 3 +1.55 [0.97, 2.71] yes

Cite this page

ShortCastCheck, KMSP graphcast_ifs lead 0, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMSP/model/graphcast_ifs/lead/0/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMSP (Minneapolis-St Paul), model graphcast_ifs, lead day 0. 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/0/

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