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

Trenton Mercer · GraphCast (IFS init) · lead day 0

KTTN · 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 QC2.06 [1.15, 3.06]1.29-2.05 [-3.05, -1.13]2.61 [1.52, 3.56]29% [7%, 54%]61%75%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 1 QC2.06 [1.12, 3.22]1.29-2.05 [-3.22, -1.10]2.61 [1.43, 3.72]29% [6%, 58%]61%75%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 1 QC1.91 [1.08, 2.87]1.39-1.65 [-2.74, -0.51]2.45 [1.42, 3.37]35% [11%, 61%]62%79%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 1 QC1.91 [1.11, 2.96]1.39-1.65 [-2.85, -0.58]2.45 [1.44, 3.46]35% [10%, 60%]62%79%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 1 QC1.92 [1.09, 2.87]1.30-1.86 [-2.86, -0.93]2.47 [1.44, 3.41]32% [12%, 57%]61%82%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 1 QC1.92 [1.00, 3.04]1.30-1.86 [-3.04, -0.88]2.47 [1.35, 3.57]32% [8%, 60%]61%82%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 1 QC1.81 [1.07, 2.72]1.43-1.46 [-2.57, -0.34]2.35 [1.39, 3.24]38% [16%, 62%]62%85%——2026-01-01 → 2026-08-30
tmax00Znearestall34 1 QC1.81 [1.06, 2.80]1.43-1.46 [-2.66, -0.38]2.35 [1.39, 3.35]38% [14%, 61%]62%85%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 1 QC3.32 [2.89, 3.88]1.29+3.32 [2.89, 3.88]3.70 [3.20, 4.30]4% [0%, 10%]29%46%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 1 QC3.32 [2.86, 3.94]1.29+3.32 [2.86, 3.94]3.70 [3.17, 4.34]4% [0%, 10%]29%46%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 1 QC3.08 [2.42, 3.72]1.32+3.08 [2.42, 3.72]3.49 [2.79, 4.13]9% [0%, 23%]32%53%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 1 QC3.08 [2.51, 3.74]1.32+3.08 [2.51, 3.74]3.49 [2.87, 4.15]9% [0%, 21%]32%53%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 1 QC3.55 [3.14, 4.09]1.30+3.55 [3.14, 4.09]3.91 [3.41, 4.50]4% [0%, 10%]14%36%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 1 QC3.55 [3.11, 4.16]1.30+3.55 [3.11, 4.16]3.91 [3.39, 4.53]4% [0%, 10%]14%36%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 1 QC3.31 [2.66, 3.91]1.34+3.31 [2.66, 3.91]3.70 [3.00, 4.31]6% [0%, 14%]21%44%——2026-01-01 → 2026-08-30
tmin00Znearestall34 1 QC3.31 [2.77, 3.96]1.34+3.31 [2.77, 3.96]3.70 [3.08, 4.34]6% [0%, 14%]21%44%——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 KTTN, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -3.06 °F2026-08-02: -1.56 °F2026-08-03: -6.3 °F2026-08-04: -3.51 °F2026-08-05: -5.68 °F2026-08-06: -2.87 °F2026-08-07: -4.43 °F2026-08-08: -2.69 °F2026-08-09: -1.29 °F2026-08-10: -0.31 °F2026-08-11: -0.61 °F2026-08-12: -0.9 °F2026-08-13: -0.11 °F2026-08-14: +0.12 °F2026-08-15: -1.63 °F2026-08-16: +0.04 °F2026-08-17: -1.74 °F2026-08-18: -1.9 °F2026-08-19: -1.33 °F2026-08-20: -3.01 °F2026-08-21: -3.5 °F2026-08-22: -2.95 °F2026-08-23: -1.18 °F2026-08-24: -1.93 °F2026-08-25: -0.37 °F2026-08-28: -2.62 °F2026-08-29: -1.55 °F2026-08-30: -0.64 °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
2026-08-30-0.64
2026-08-29-1.55
2026-08-28-2.62
2026-08-27—
2026-08-26—
2026-08-25-0.37
2026-08-24-1.93
2026-08-23-1.18
2026-08-22-2.95
2026-08-21-3.50
2026-08-20-3.01
2026-08-19-1.33
2026-08-18-1.90
2026-08-17-1.74
2026-08-16+0.04
2026-08-15-1.63
2026-08-14+0.12
2026-08-13-0.11
2026-08-12-0.90
2026-08-11-0.61
2026-08-10-0.31
2026-08-09-1.29
2026-08-08-2.69
2026-08-07-4.43
2026-08-06-2.87
2026-08-05-5.68
2026-08-04-3.51
2026-08-03-6.30
2026-08-02-1.56
2026-08-01-3.06

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KTTN, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 0 days-7 to -6 °F: 1 day-6 to -5 °F: 1 day-5 to -4 °F: 1 day-4 to -3 °F: 4 days-3 to -2 °F: 5 days-2 to -1 °F: 9 days-1 to 0 °F: 7 days0 to 1 °F: 5 days1 to 2 °F: 0 days2 to 3 °F: 1 day3 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| 1.59 · P90 |e| 3.51 °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
-9.00-8.000
-8.00-7.000
-7.00-6.001
-6.00-5.001
-5.00-4.001
-4.00-3.004
-3.00-2.005
-2.00-1.009
-1.00+0.007
+0.00+1.005
+1.00+2.000
+2.00+3.001
+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
2026-08282.06 -2.05
2026-0161.16 +0.21

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 +0.26 [-0.97, 1.59] no
Aurora (GFS init) 2 -0.52 [-1.91, 0.86] no
Aurora (IFS init) 3 +0.64 [0.54, 0.79] yes
FourCastNet v2 (GFS init) 2 -1.51 [-3.13, 0.12] no
FourCastNet v2 (IFS init) 3 -1.12 [-2.47, -0.44] yes
NCEP GFS 28 +0.33 [-0.85, 1.41] no
GraphCast (GFS init) 2 -0.67 [-1.98, 0.64] no
ECMWF IFS HRES 22 -0.02 [-0.89, 1.02] no
Pangu-Weather (GFS init) 2 -1.16 [-1.29, -1.04] yes
Pangu-Weather (IFS init) 3 +0.01 [-1.27, 2.54] no

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KTTN (Trenton Mercer), 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/KTTN/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