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 5

KMSP · graphcast_ifs · lead 5

Target day = initialization date + 5 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 QC6.73 [5.59, 7.70]2.63-6.57 [-7.70, -5.09]7.28 [6.21, 8.16]0% [0%, 0%]0%8%-0.63+0.362026-08-06 → 2026-08-30
tmax00Zbilinear90d25 1 QC6.73 [5.33, 7.92]2.63-6.57 [-7.92, -4.83]7.28 [6.00, 8.35]0% [0%, 0%]0%8%-0.63+0.362026-08-06 → 2026-08-30
tmax00Zbilinear365d31 1 QC6.22 [4.71, 7.49]2.71-5.87 [-7.34, -3.94]6.80 [5.16, 7.98]0% [0%, 0%]0%13%-0.10+0.532026-01-06 → 2026-08-30
tmax00Zbilinearall31 1 QC6.22 [4.68, 7.43]2.71-5.87 [-7.33, -3.63]6.80 [5.03, 7.92]0% [0%, 0%]0%13%-0.10+0.532026-01-06 → 2026-08-30
tmax00Znearest30d25 1 QC6.76 [5.60, 7.73]2.62-6.60 [-7.73, -5.11]7.31 [6.26, 8.19]0% [0%, 0%]0%8%-0.64+0.372026-08-06 → 2026-08-30
tmax00Znearest90d25 1 QC6.76 [5.38, 7.95]2.62-6.60 [-7.95, -4.86]7.31 [6.02, 8.37]0% [0%, 0%]0%8%-0.64+0.372026-08-06 → 2026-08-30
tmax00Znearest365d31 1 QC6.24 [4.71, 7.51]2.71-5.90 [-7.36, -3.94]6.82 [5.16, 7.99]0% [0%, 0%]0%13%-0.10+0.532026-01-06 → 2026-08-30
tmax00Znearestall31 1 QC6.24 [4.67, 7.47]2.71-5.90 [-7.36, -3.63]6.82 [5.01, 7.94]0% [0%, 0%]0%13%-0.10+0.532026-01-06 → 2026-08-30
tmax12Zbilinear30d25 1 QC6.47 [5.36, 7.15]2.64-6.43 [-7.12, -5.31]7.19 [6.21, 7.75]8% [0%, 19%]8%8%-0.57+0.362026-08-06 → 2026-08-30
tmax12Zbilinear90d25 1 QC6.47 [5.29, 7.14]2.64-6.43 [-7.12, -5.14]7.19 [6.21, 7.72]8% [0%, 21%]8%8%-0.57+0.362026-08-06 → 2026-08-30
tmax12Zbilinear365d30 1 QC5.96 [4.54, 6.91]2.66-5.91 [-6.88, -4.43]6.74 [5.22, 7.50]10% [0%, 20%]10%10%-0.04+0.542026-01-06 → 2026-08-30
tmax12Zbilinearall30 1 QC5.96 [4.11, 6.94]2.66-5.91 [-6.94, -3.97]6.74 [4.82, 7.55]10% [0%, 22%]10%10%-0.04+0.542026-01-06 → 2026-08-30
tmax12Znearest30d25 1 QC6.48 [5.35, 7.15]2.63-6.44 [-7.14, -5.30]7.20 [6.21, 7.75]8% [0%, 19%]8%12%-0.57+0.362026-08-06 → 2026-08-30
tmax12Znearest90d25 1 QC6.48 [5.29, 7.15]2.63-6.44 [-7.13, -5.15]7.20 [6.22, 7.74]8% [0%, 21%]8%12%-0.57+0.362026-08-06 → 2026-08-30
tmax12Znearest365d30 1 QC5.97 [4.53, 6.92]2.69-5.91 [-6.89, -4.42]6.75 [5.21, 7.52]10% [0%, 20%]10%13%-0.04+0.542026-01-06 → 2026-08-30
tmax12Znearestall30 1 QC5.97 [4.11, 6.96]2.69-5.91 [-6.94, -3.95]6.75 [4.78, 7.58]10% [0%, 22%]10%13%-0.04+0.542026-01-06 → 2026-08-30
tmin00Zbilinear30d25 1 QC2.84 [1.85, 3.89]2.42-1.69 [-3.08, -0.25]3.40 [2.18, 4.38]20% [4%, 37%]32%64%+0.40+0.492026-08-06 → 2026-08-30
tmin00Zbilinear90d25 1 QC2.84 [1.74, 3.95]2.42-1.69 [-3.22, -0.21]3.40 [2.01, 4.43]20% [0%, 40%]32%64%+0.40+0.492026-08-06 → 2026-08-30
tmin00Zbilinear365d31 1 QC2.96 [1.95, 3.93]2.62-1.34 [-2.89, 0.14]3.51 [2.38, 4.37]23% [6%, 41%]32%58%+0.54+0.602026-01-06 → 2026-08-30
tmin00Zbilinearall31 1 QC2.96 [1.97, 3.88]2.62-1.34 [-2.84, 0.21]3.51 [2.37, 4.35]23% [7%, 42%]32%58%+0.54+0.602026-01-06 → 2026-08-30
tmin00Znearest30d25 1 QC2.85 [1.80, 3.96]2.41-1.79 [-3.23, -0.33]3.45 [2.15, 4.49]20% [4%, 37%]40%60%+0.40+0.492026-08-06 → 2026-08-30
tmin00Znearest90d25 1 QC2.85 [1.69, 4.04]2.41-1.79 [-3.35, -0.31]3.45 [1.96, 4.59]20% [0%, 40%]40%60%+0.40+0.492026-08-06 → 2026-08-30
tmin00Znearest365d31 1 QC2.97 [1.93, 3.99]2.62-1.42 [-3.01, 0.08]3.56 [2.37, 4.49]19% [4%, 38%]39%55%+0.54+0.602026-01-06 → 2026-08-30
tmin00Znearestall31 1 QC2.97 [1.92, 3.93]2.62-1.42 [-2.95, 0.16]3.56 [2.34, 4.44]19% [5%, 38%]39%55%+0.54+0.602026-01-06 → 2026-08-30
tmin12Zbilinear30d25 1 QC2.41 [1.51, 3.38]1.98-1.49 [-2.90, 0.02]3.11 [2.00, 4.01]28% [8%, 48%]52%64%+0.49+0.582026-08-06 → 2026-08-30
tmin12Zbilinear90d25 1 QC2.41 [1.41, 3.47]1.98-1.49 [-3.06, 0.23]3.11 [1.76, 4.11]28% [0%, 50%]52%64%+0.49+0.582026-08-06 → 2026-08-30
tmin12Zbilinear365d30 1 QC2.56 [1.60, 3.59]2.26-1.09 [-2.73, 0.70]3.20 [2.04, 4.15]23% [0%, 43%]47%60%+0.59+0.652026-01-06 → 2026-08-30
tmin12Zbilinearall30 1 QC2.56 [1.67, 3.56]2.26-1.09 [-2.67, 0.79]3.20 [2.11, 4.10]23% [0%, 43%]47%60%+0.59+0.652026-01-06 → 2026-08-30
tmin12Znearest30d25 1 QC2.47 [1.52, 3.48]1.98-1.62 [-3.05, -0.11]3.17 [1.97, 4.12]32% [12%, 50%]52%68%+0.48+0.582026-08-06 → 2026-08-30
tmin12Znearest90d25 1 QC2.47 [1.43, 3.58]1.98-1.62 [-3.21, 0.09]3.17 [1.71, 4.23]32% [10%, 53%]52%68%+0.48+0.582026-08-06 → 2026-08-30
tmin12Znearest365d30 1 QC2.60 [1.61, 3.68]2.25-1.20 [-2.86, 0.62]3.23 [2.07, 4.24]27% [0%, 46%]47%63%+0.59+0.652026-01-06 → 2026-08-30
tmin12Znearestall30 1 QC2.60 [1.67, 3.65]2.25-1.20 [-2.81, 0.68]3.23 [2.09, 4.19]27% [0%, 47%]47%63%+0.59+0.652026-01-06 → 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 5, 00Z bilinear tmax+20+100-10-202026-08-06: -3.16 °F2026-08-07: +2.08 °F2026-08-08: -5.39 °F2026-08-09: -6.6 °F2026-08-10: -9.91 °F2026-08-11: -9.39 °F2026-08-12: -9.56 °F2026-08-13: -5.81 °F2026-08-14: -3.6 °F2026-08-15: -2.32 °F2026-08-16: -8 °F2026-08-17: -8.58 °F2026-08-18: -5.79 °F2026-08-19: -8.61 °F2026-08-20: -3.96 °F2026-08-21: -3.4 °F2026-08-22: -4.93 °F2026-08-23: -9.66 °F2026-08-24: -10.7 °F2026-08-25: -5.63 °F2026-08-26: -7.94 °F2026-08-27: -8.05 °F2026-08-28: -9.9 °F2026-08-29: -11.33 °F2026-08-30: -4 °F2026-08-062026-08-30°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 (25 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-4.00
2026-08-29-11.33
2026-08-28-9.90
2026-08-27-8.05
2026-08-26-7.94
2026-08-25-5.63
2026-08-24-10.70
2026-08-23-9.66
2026-08-22-4.93
2026-08-21-3.40
2026-08-20-3.96
2026-08-19-8.61
2026-08-18-5.79
2026-08-17-8.58
2026-08-16-8.00
2026-08-15-2.32
2026-08-14-3.60
2026-08-13-5.81
2026-08-12-9.56
2026-08-11-9.39
2026-08-10-9.91
2026-08-09-6.60
2026-08-08-5.39
2026-08-07+2.08
2026-08-06-3.16

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMSP, lead day 5-20 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: 9 days-8 to -6 °F: 2 days-6 to -4 °F: 9 days-4 to -2 °F: 7 days-2 to 0 °F: 0 days0 to 2 °F: 0 days2 to 4 °F: 2 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| 5.75 · P90 |e| 9.9 °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
-12.00-10.002
-10.00-8.009
-8.00-6.002
-6.00-4.009
-4.00-2.007
-2.00+0.000
+0.00+2.000
+2.00+4.002
+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
2026-08256.73 -6.57
2026-0164.09 -2.96

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 30
GraphCast (IFS init)2026-01-06 → 2026-08-30 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 +2.23 [0.96, 3.55] yes
NCEP GFS 25 +2.73 [-0.03, 5.21] no
ECMWF IFS HRES 18 +4.09 [1.94, 5.71] yes
Persistence (baseline) 25 +2.61 [0.25, 4.70] yes

Cite this page

ShortCastCheck, KMSP graphcast_ifs lead 5, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMSP/model/graphcast_ifs/lead/5/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMSP (Minneapolis-St Paul), model graphcast_ifs, lead day 5. 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/5/

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