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

Newark Liberty · GraphCast (IFS init) · lead day 0

KEWR · 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 QC4.57 [4.04, 5.14]1.24-4.57 [-5.14, -4.04]4.85 [4.21, 5.50]4% [0%, 11%]4%14%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 1 QC4.57 [3.93, 5.22]1.24-4.57 [-5.22, -3.93]4.85 [4.14, 5.65]4% [0%, 15%]4%14%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 1 QC4.35 [3.64, 4.96]1.22-4.35 [-4.96, -3.64]4.63 [3.83, 5.32]3% [0%, 12%]3%18%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 1 QC4.35 [3.72, 5.03]1.22-4.35 [-5.03, -3.72]4.63 [3.92, 5.40]3% [0%, 13%]3%18%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 1 QC4.07 [3.51, 4.57]1.18-4.07 [-4.57, -3.51]4.37 [3.71, 4.96]4% [0%, 11%]7%14%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 1 QC4.07 [3.44, 4.66]1.18-4.07 [-4.66, -3.44]4.37 [3.65, 5.08]4% [0%, 15%]7%14%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 1 QC3.97 [3.39, 4.50]1.11-3.97 [-4.50, -3.39]4.24 [3.58, 4.85]3% [0%, 12%]6%18%——2026-01-01 → 2026-08-30
tmax00Znearestall34 1 QC3.97 [3.40, 4.52]1.11-3.97 [-4.52, -3.40]4.24 [3.59, 4.91]3% [0%, 13%]6%18%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 1 QC1.97 [1.48, 2.46]1.72+1.19 [0.73, 1.61]2.41 [1.88, 2.86]32% [17%, 46%]61%68%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 1 QC1.97 [1.39, 2.49]1.72+1.19 [0.78, 1.61]2.41 [1.76, 2.91]32% [16%, 49%]61%68%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 1 QC1.77 [1.09, 2.28]1.73+0.84 [-0.00, 1.44]2.22 [1.35, 2.71]35% [21%, 52%]68%74%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 1 QC1.77 [1.17, 2.32]1.73+0.84 [0.12, 1.46]2.22 [1.52, 2.76]35% [20%, 50%]68%74%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 1 QC1.71 [1.23, 2.13]1.70+0.17 [-0.28, 0.60]2.10 [1.56, 2.51]39% [24%, 56%]64%86%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 1 QC1.71 [1.12, 2.17]1.70+0.17 [-0.32, 0.64]2.10 [1.40, 2.55]39% [23%, 61%]64%86%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 1 QC1.74 [1.25, 2.09]1.71-0.19 [-1.11, 0.45]2.06 [1.48, 2.43]32% [13%, 53%]65%88%——2026-01-01 → 2026-08-30
tmin00Znearestall34 1 QC1.74 [1.27, 2.11]1.71-0.19 [-0.95, 0.44]2.06 [1.56, 2.46]32% [14%, 54%]65%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 KEWR, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -5.1 °F2026-08-02: -2.81 °F2026-08-03: -9.18 °F2026-08-04: -3.64 °F2026-08-05: -6.8 °F2026-08-06: -6.01 °F2026-08-07: -6.86 °F2026-08-08: -3.83 °F2026-08-09: -3.74 °F2026-08-10: -2.62 °F2026-08-11: -3.78 °F2026-08-12: -4.8 °F2026-08-13: -3.45 °F2026-08-14: -2.86 °F2026-08-15: -4.8 °F2026-08-16: -3.19 °F2026-08-17: -3.61 °F2026-08-18: -5.59 °F2026-08-19: -4.98 °F2026-08-20: -4.12 °F2026-08-21: -6.17 °F2026-08-22: -4.09 °F2026-08-23: -4.13 °F2026-08-24: -4.95 °F2026-08-25: -4.37 °F2026-08-28: -6.95 °F2026-08-29: -4.7 °F2026-08-30: -0.97 °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
2026-08-30-0.97
2026-08-29-4.70
2026-08-28-6.95
2026-08-27—
2026-08-26—
2026-08-25-4.37
2026-08-24-4.95
2026-08-23-4.13
2026-08-22-4.09
2026-08-21-6.17
2026-08-20-4.12
2026-08-19-4.98
2026-08-18-5.59
2026-08-17-3.61
2026-08-16-3.19
2026-08-15-4.80
2026-08-14-2.86
2026-08-13-3.45
2026-08-12-4.80
2026-08-11-3.78
2026-08-10-2.62
2026-08-09-3.74
2026-08-08-3.83
2026-08-07-6.86
2026-08-06-6.01
2026-08-05-6.80
2026-08-04-3.64
2026-08-03-9.18
2026-08-02-2.81
2026-08-01-5.10

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KEWR, lead day 0-10 to -9 °F: 1 day-9 to -8 °F: 0 days-8 to -7 °F: 0 days-7 to -6 °F: 5 days-6 to -5 °F: 2 days-5 to -4 °F: 10 days-4 to -3 °F: 10 days-3 to -2 °F: 5 days-2 to -1 °F: 0 days-1 to 0 °F: 1 day0 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| 4.1 · P90 |e| 6.61 °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
-10.00-9.001
-9.00-8.000
-8.00-7.000
-7.00-6.005
-6.00-5.002
-5.00-4.0010
-4.00-3.0010
-3.00-2.005
-2.00-1.000
-1.00+0.001
+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
2026-08284.57 -4.57
2026-0163.31 -3.31

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 +1.92 [1.28, 2.42] yes
Aurora (GFS init) 2 -1.12 [-2.93, 0.68] no
Aurora (IFS init) 3 -0.01 [-0.20, 0.09] no
FourCastNet v2 (GFS init) 2 +0.05 [-3.14, 3.24] no
FourCastNet v2 (IFS init) 3 -0.55 [-2.38, 0.37] no
NCEP GFS 28 +2.81 [2.36, 3.17] yes
GraphCast (GFS init) 2 -1.04 [-2.41, 0.33] no
ECMWF IFS HRES 22 +0.69 [-0.02, 1.45] no
Pangu-Weather (GFS init) 2 +1.56 [-0.82, 3.94] no
Pangu-Weather (IFS init) 3 +1.74 [-1.17, 3.20] no

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KEWR (Newark Liberty), 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/KEWR/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