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 2

KMSP · graphcast_ifs · lead 2

Target day = initialization date + 2 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
tmax00Zbilinear30d26 1 QC6.26 [5.05, 7.22]2.56-6.25 [-7.22, -5.04]6.97 [5.68, 7.83]8% [0%, 14%]12%15%-0.87+0.242026-08-03 → 2026-08-30
tmax00Zbilinear90d26 1 QC6.26 [4.75, 7.35]2.56-6.25 [-7.35, -4.75]6.97 [5.30, 7.96]8% [0%, 14%]12%15%-0.87+0.242026-08-03 → 2026-08-30
tmax00Zbilinear365d32 1 QC5.62 [3.98, 6.96]2.75-5.62 [-6.96, -3.97]6.47 [4.75, 7.64]9% [0%, 16%]19%25%-0.31+0.322026-01-03 → 2026-08-30
tmax00Zbilinearall32 1 QC5.62 [3.75, 6.98]2.75-5.62 [-6.97, -3.75]6.47 [4.35, 7.59]9% [0%, 16%]19%25%-0.31+0.322026-01-03 → 2026-08-30
tmax00Znearest30d26 1 QC6.27 [5.04, 7.25]2.61-6.26 [-7.25, -5.03]6.99 [5.67, 7.86]8% [0%, 14%]12%15%-0.87+0.232026-08-03 → 2026-08-30
tmax00Znearest90d26 1 QC6.27 [4.73, 7.41]2.61-6.26 [-7.41, -4.73]6.99 [5.31, 8.02]8% [0%, 14%]12%15%-0.87+0.232026-08-03 → 2026-08-30
tmax00Znearest365d32 1 QC5.67 [4.09, 7.00]2.76-5.67 [-7.00, -4.09]6.51 [4.86, 7.66]9% [0%, 16%]16%25%-0.32+0.322026-01-03 → 2026-08-30
tmax00Znearestall32 1 QC5.67 [3.81, 7.02]2.76-5.67 [-7.02, -3.81]6.51 [4.45, 7.63]9% [0%, 16%]16%25%-0.32+0.322026-01-03 → 2026-08-30
tmax12Zbilinear30d26 1 QC6.82 [5.55, 7.81]2.57-6.82 [-7.81, -5.55]7.44 [6.15, 8.31]0% [0%, 0%]8%15%-1.04+0.242026-08-03 → 2026-08-30
tmax12Zbilinear90d26 1 QC6.82 [5.30, 8.00]2.57-6.82 [-8.00, -5.30]7.44 [5.79, 8.45]0% [0%, 0%]8%15%-1.04+0.242026-08-03 → 2026-08-30
tmax12Zbilinear365d31 1 QC6.21 [4.38, 7.65]2.78-6.21 [-7.65, -4.38]6.98 [5.23, 8.18]3% [0%, 12%]13%23%-0.45+0.322026-01-03 → 2026-08-30
tmax12Zbilinearall31 1 QC6.21 [4.24, 7.60]2.78-6.21 [-7.60, -4.24]6.98 [4.89, 8.15]3% [0%, 12%]13%23%-0.45+0.322026-01-03 → 2026-08-30
tmax12Znearest30d26 1 QC6.85 [5.55, 7.87]2.64-6.85 [-7.87, -5.55]7.49 [6.16, 8.38]4% [0%, 11%]8%15%-1.05+0.222026-08-03 → 2026-08-30
tmax12Znearest90d26 1 QC6.85 [5.28, 8.17]2.64-6.85 [-8.17, -5.28]7.49 [5.80, 8.55]4% [0%, 11%]8%15%-1.05+0.222026-08-03 → 2026-08-30
tmax12Znearest365d31 1 QC6.25 [4.45, 7.71]2.82-6.25 [-7.71, -4.45]7.04 [5.34, 8.27]6% [0%, 15%]13%23%-0.46+0.312026-01-03 → 2026-08-30
tmax12Znearestall31 1 QC6.25 [4.28, 7.66]2.82-6.25 [-7.66, -4.28]7.04 [4.95, 8.26]6% [0%, 15%]13%23%-0.46+0.312026-01-03 → 2026-08-30
tmin00Zbilinear30d26 1 QC2.13 [1.64, 2.63]1.67-1.34 [-2.20, -0.47]2.49 [1.81, 3.05]23% [14%, 32%]54%77%+0.55+0.652026-08-03 → 2026-08-30
tmin00Zbilinear90d26 1 QC2.13 [1.50, 2.70]1.67-1.34 [-2.36, -0.43]2.49 [1.67, 3.10]23% [12%, 36%]54%77%+0.55+0.652026-08-03 → 2026-08-30
tmin00Zbilinear365d32 1 QC2.14 [1.55, 2.71]1.79-1.22 [-2.16, -0.40]2.56 [1.79, 3.13]28% [13%, 46%]53%75%+0.58+0.652026-01-03 → 2026-08-30
tmin00Zbilinearall32 1 QC2.14 [1.53, 2.67]1.79-1.22 [-2.15, -0.41]2.56 [1.79, 3.11]28% [15%, 45%]53%75%+0.58+0.652026-01-03 → 2026-08-30
tmin00Znearest30d26 1 QC2.21 [1.70, 2.77]1.68-1.46 [-2.36, -0.58]2.58 [1.86, 3.18]19% [12%, 28%]54%77%+0.54+0.652026-08-03 → 2026-08-30
tmin00Znearest90d26 1 QC2.21 [1.56, 2.82]1.68-1.46 [-2.53, -0.55]2.58 [1.74, 3.27]19% [10%, 33%]54%77%+0.54+0.652026-08-03 → 2026-08-30
tmin00Znearest365d32 1 QC2.24 [1.66, 2.85]1.79-1.36 [-2.35, -0.54]2.66 [1.86, 3.28]22% [10%, 35%]53%75%+0.56+0.652026-01-03 → 2026-08-30
tmin00Znearestall32 1 QC2.24 [1.65, 2.80]1.79-1.36 [-2.30, -0.57]2.66 [1.84, 3.23]22% [12%, 36%]53%75%+0.56+0.652026-01-03 → 2026-08-30
tmin12Zbilinear30d26 1 QC2.08 [1.52, 2.66]1.38-1.66 [-2.40, -1.00]2.46 [1.74, 3.03]19% [12%, 28%]50%81%+0.56+0.712026-08-03 → 2026-08-30
tmin12Zbilinear90d26 1 QC2.08 [1.44, 2.71]1.38-1.66 [-2.52, -0.95]2.46 [1.67, 3.10]19% [10%, 33%]50%81%+0.56+0.712026-08-03 → 2026-08-30
tmin12Zbilinear365d31 1 QC2.11 [1.57, 2.66]1.52-1.55 [-2.31, -0.91]2.50 [1.81, 3.06]19% [11%, 31%]48%77%+0.60+0.712026-01-03 → 2026-08-30
tmin12Zbilinearall31 1 QC2.11 [1.52, 2.66]1.52-1.55 [-2.32, -0.89]2.50 [1.75, 3.05]19% [11%, 30%]48%77%+0.60+0.712026-01-03 → 2026-08-30
tmin12Znearest30d26 1 QC2.16 [1.54, 2.79]1.41-1.79 [-2.55, -1.10]2.58 [1.80, 3.22]19% [5%, 33%]42%81%+0.55+0.712026-08-03 → 2026-08-30
tmin12Znearest90d26 1 QC2.16 [1.50, 2.86]1.41-1.79 [-2.69, -1.07]2.58 [1.73, 3.28]19% [4%, 33%]42%81%+0.55+0.712026-08-03 → 2026-08-30
tmin12Znearest365d31 1 QC2.20 [1.62, 2.82]1.56-1.68 [-2.48, -1.03]2.63 [1.87, 3.26]19% [7%, 32%]42%77%+0.58+0.702026-01-03 → 2026-08-30
tmin12Znearestall31 1 QC2.20 [1.58, 2.82]1.56-1.68 [-2.48, -1.01]2.63 [1.82, 3.23]19% [6%, 32%]42%77%+0.58+0.702026-01-03 → 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 2, 00Z bilinear tmax+20+100-10-202026-08-03: -3.66 °F2026-08-04: -2.98 °F2026-08-05: -6.39 °F2026-08-06: -3.9 °F2026-08-07: -1.31 °F2026-08-08: -5.21 °F2026-08-09: -7.14 °F2026-08-10: -9.63 °F2026-08-11: -7.97 °F2026-08-12: -8.21 °F2026-08-13: -0.12 °F2026-08-14: -4.36 °F2026-08-15: -4.56 °F2026-08-16: -8.45 °F2026-08-17: -8.5 °F2026-08-18: -10.97 °F2026-08-19: -8.5 °F2026-08-20: +0.06 °F2026-08-21: -6.68 °F2026-08-22: -6.36 °F2026-08-23: -8.06 °F2026-08-24: -11.67 °F2026-08-25: -3.2 °F2026-08-26: -8.59 °F2026-08-27: -9.3 °F2026-08-30: -6.95 °F2026-08-032026-08-30°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 (28 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-6.95
2026-08-29—
2026-08-28—
2026-08-27-9.30
2026-08-26-8.59
2026-08-25-3.20
2026-08-24-11.67
2026-08-23-8.06
2026-08-22-6.36
2026-08-21-6.68
2026-08-20+0.06
2026-08-19-8.50
2026-08-18-10.97
2026-08-17-8.50
2026-08-16-8.45
2026-08-15-4.56
2026-08-14-4.36
2026-08-13-0.12
2026-08-12-8.21
2026-08-11-7.97
2026-08-10-9.63
2026-08-09-7.14
2026-08-08-5.21
2026-08-07-1.31
2026-08-06-3.90
2026-08-05-6.39
2026-08-04-2.98
2026-08-03-3.66

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMSP, lead day 2-20 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: 2 days-10 to -8 °F: 8 days-8 to -6 °F: 7 days-6 to -4 °F: 4 days-4 to -2 °F: 5 days-2 to 0 °F: 5 days0 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| 6.27 · P90 |e| 9.23 °F-200+20 °Fn = 32 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
-12.00-10.002
-10.00-8.008
-8.00-6.007
-6.00-4.004
-4.00-2.005
-2.00+0.005
+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
2026-08266.26 -6.25
2026-0162.87 -2.87

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-03 → 2026-08-30 32
GraphCast (IFS init)2026-01-03 → 2026-08-30 32
GraphCast (IFS init)2026-01-03 → 2026-08-30 31
GraphCast (IFS init)2026-01-03 → 2026-08-30 31

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 26 +2.35 [1.60, 3.03] yes
Aurora (IFS init) 1 -0.84 [-0.84, -0.84] yes
FourCastNet v2 (IFS init) 1 +1.07 [1.07, 1.07] yes
NCEP GFS 26 +2.27 [-0.52, 4.24] no
ECMWF IFS HRES 20 +3.05 [1.56, 4.31] yes
Pangu-Weather (IFS init) 1 +2.84 [2.84, 2.84] yes
Persistence (baseline) 26 +2.91 [0.97, 4.83] yes

Cite this page

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

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

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