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 2

KEWR · 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 QC4.78 [3.87, 5.66]1.58-4.78 [-5.66, -3.87]5.22 [4.30, 6.08]4% [0%, 14%]8%19%-0.50+0.482026-08-03 → 2026-08-30
tmax00Zbilinear90d26 1 QC4.78 [3.59, 5.85]1.58-4.78 [-5.85, -3.59]5.22 [4.10, 6.21]4% [0%, 14%]8%19%-0.50+0.482026-08-03 → 2026-08-30
tmax00Zbilinear365d32 1 QC4.60 [3.61, 5.44]1.50-4.60 [-5.44, -3.61]5.01 [3.91, 5.84]3% [0%, 11%]6%22%-0.14+0.642026-01-03 → 2026-08-30
tmax00Zbilinearall32 1 QC4.60 [3.61, 5.46]1.50-4.60 [-5.46, -3.61]5.01 [3.95, 5.83]3% [0%, 12%]6%22%-0.14+0.642026-01-03 → 2026-08-30
tmax00Znearest30d26 1 QC4.32 [3.42, 5.14]1.61-4.29 [-5.13, -3.35]4.80 [3.92, 5.55]8% [0%, 17%]19%19%-0.35+0.472026-08-03 → 2026-08-30
tmax00Znearest90d26 1 QC4.32 [3.21, 5.27]1.61-4.29 [-5.26, -3.15]4.80 [3.70, 5.67]8% [0%, 17%]19%19%-0.35+0.472026-08-03 → 2026-08-30
tmax00Znearest365d32 1 QC4.22 [3.30, 4.96]1.45-4.19 [-4.94, -3.25]4.64 [3.66, 5.37]6% [0%, 14%]16%22%-0.05+0.652026-01-03 → 2026-08-30
tmax00Znearestall32 1 QC4.22 [3.31, 5.02]1.45-4.19 [-5.01, -3.25]4.64 [3.69, 5.41]6% [0%, 15%]16%22%-0.05+0.652026-01-03 → 2026-08-30
tmax12Zbilinear30d26 1 QC4.93 [4.09, 5.72]1.49-4.93 [-5.72, -4.09]5.26 [4.35, 6.03]0% [0%, 0%]4%23%-0.54+0.512026-08-03 → 2026-08-30
tmax12Zbilinear90d26 1 QC4.93 [3.87, 5.86]1.49-4.93 [-5.86, -3.87]5.26 [4.16, 6.15]0% [0%, 0%]4%23%-0.54+0.512026-08-03 → 2026-08-30
tmax12Zbilinear365d31 1 QC4.71 [3.67, 5.55]1.45-4.71 [-5.55, -3.67]5.04 [3.90, 5.84]0% [0%, 0%]6%23%-0.22+0.642026-01-03 → 2026-08-30
tmax12Zbilinearall31 1 QC4.71 [3.71, 5.59]1.45-4.71 [-5.59, -3.71]5.04 [4.01, 5.89]0% [0%, 0%]6%23%-0.22+0.642026-01-03 → 2026-08-30
tmax12Znearest30d26 1 QC4.43 [3.58, 5.21]1.55-4.43 [-5.21, -3.58]4.81 [3.93, 5.51]0% [0%, 0%]12%23%-0.39+0.492026-08-03 → 2026-08-30
tmax12Znearest90d26 1 QC4.43 [3.38, 5.32]1.55-4.43 [-5.32, -3.38]4.81 [3.75, 5.63]0% [0%, 0%]12%23%-0.39+0.492026-08-03 → 2026-08-30
tmax12Znearest365d31 1 QC4.31 [3.41, 5.06]1.44-4.31 [-5.06, -3.41]4.67 [3.67, 5.35]0% [0%, 0%]13%23%-0.11+0.642026-01-03 → 2026-08-30
tmax12Znearestall31 1 QC4.31 [3.39, 5.10]1.44-4.31 [-5.10, -3.39]4.67 [3.68, 5.40]0% [0%, 0%]13%23%-0.11+0.642026-01-03 → 2026-08-30
tmin00Zbilinear30d26 1 QC2.18 [1.53, 2.82]2.05+0.95 [0.25, 1.60]2.70 [2.01, 3.19]19% [4%, 39%]42%73%+0.49+0.512026-08-03 → 2026-08-30
tmin00Zbilinear90d26 1 QC2.18 [1.41, 2.96]2.05+0.95 [0.23, 1.65]2.70 [1.80, 3.28]19% [0%, 44%]42%73%+0.49+0.512026-08-03 → 2026-08-30
tmin00Zbilinear365d32 1 QC2.15 [1.41, 2.76]2.13+0.47 [-0.72, 1.41]2.71 [1.88, 3.25]22% [6%, 43%]47%75%+0.53+0.542026-01-03 → 2026-08-30
tmin00Zbilinearall32 1 QC2.15 [1.44, 2.81]2.13+0.47 [-0.66, 1.25]2.71 [1.85, 3.26]22% [5%, 43%]47%75%+0.53+0.542026-01-03 → 2026-08-30
tmin00Znearest30d26 1 QC2.13 [1.67, 2.77]2.14-0.10 [-0.84, 0.52]2.60 [2.09, 3.12]15% [0%, 30%]54%85%+0.50+0.492026-08-03 → 2026-08-30
tmin00Znearest90d26 1 QC2.13 [1.65, 2.92]2.14-0.10 [-0.99, 0.59]2.60 [1.99, 3.17]15% [0%, 32%]54%85%+0.50+0.492026-08-03 → 2026-08-30
tmin00Znearest365d32 1 QC2.29 [1.76, 2.90]2.25-0.59 [-1.88, 0.37]2.79 [2.18, 3.32]16% [0%, 30%]53%81%+0.50+0.512026-01-03 → 2026-08-30
tmin00Znearestall32 1 QC2.29 [1.78, 3.00]2.25-0.59 [-1.79, 0.26]2.79 [2.14, 3.39]16% [2%, 29%]53%81%+0.50+0.512026-01-03 → 2026-08-30
tmin12Zbilinear30d26 1 QC2.11 [1.47, 2.77]2.01+0.93 [0.32, 1.53]2.60 [1.97, 3.12]31% [12%, 54%]46%69%+0.50+0.522026-08-03 → 2026-08-30
tmin12Zbilinear90d26 1 QC2.11 [1.33, 2.96]2.01+0.93 [0.26, 1.58]2.60 [1.84, 3.23]31% [8%, 59%]46%69%+0.50+0.522026-08-03 → 2026-08-30
tmin12Zbilinear365d31 1 QC1.95 [1.10, 2.61]1.93+0.65 [-0.27, 1.38]2.45 [1.49, 3.01]35% [16%, 64%]52%74%+0.56+0.562026-01-03 → 2026-08-30
tmin12Zbilinearall31 1 QC1.95 [1.25, 2.69]1.93+0.65 [-0.16, 1.36]2.45 [1.64, 3.02]35% [14%, 63%]52%74%+0.56+0.562026-01-03 → 2026-08-30
tmin12Znearest30d26 1 QC2.15 [1.64, 2.77]2.14-0.06 [-0.75, 0.57]2.54 [2.05, 2.99]31% [5%, 58%]54%69%+0.49+0.492026-08-03 → 2026-08-30
tmin12Znearest90d26 1 QC2.15 [1.62, 2.96]2.14-0.06 [-0.85, 0.61]2.54 [1.97, 3.09]31% [0%, 58%]54%69%+0.49+0.492026-08-03 → 2026-08-30
tmin12Znearest365d31 1 QC2.14 [1.64, 2.72]2.07-0.39 [-1.56, 0.43]2.52 [2.04, 2.95]29% [6%, 54%]52%68%+0.52+0.532026-01-03 → 2026-08-30
tmin12Znearestall31 1 QC2.14 [1.65, 2.71]2.07-0.39 [-1.36, 0.38]2.52 [2.03, 2.93]29% [5%, 52%]52%68%+0.52+0.532026-01-03 → 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 KEWR, lead day 2, 00Z bilinear tmax+10+50-5-102026-08-03: -6.89 °F2026-08-04: -3.82 °F2026-08-05: -5.45 °F2026-08-06: -8.18 °F2026-08-07: -9.97 °F2026-08-08: -4.78 °F2026-08-09: -3.91 °F2026-08-10: -2.43 °F2026-08-11: -1.46 °F2026-08-12: -5.24 °F2026-08-13: -2.42 °F2026-08-14: -4.72 °F2026-08-15: -5.91 °F2026-08-16: -3.59 °F2026-08-17: -3.9 °F2026-08-18: -4.23 °F2026-08-19: -6.01 °F2026-08-20: -7.95 °F2026-08-21: -4.66 °F2026-08-22: -6.06 °F2026-08-23: -4.23 °F2026-08-24: -5.67 °F2026-08-25: -4.91 °F2026-08-26: -5.68 °F2026-08-27: -0.31 °F2026-08-30: -2.03 °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
2026-08-30-2.03
2026-08-29—
2026-08-28—
2026-08-27-0.31
2026-08-26-5.68
2026-08-25-4.91
2026-08-24-5.67
2026-08-23-4.23
2026-08-22-6.06
2026-08-21-4.66
2026-08-20-7.95
2026-08-19-6.01
2026-08-18-4.23
2026-08-17-3.90
2026-08-16-3.59
2026-08-15-5.91
2026-08-14-4.72
2026-08-13-2.42
2026-08-12-5.24
2026-08-11-1.46
2026-08-10-2.43
2026-08-09-3.91
2026-08-08-4.78
2026-08-07-9.97
2026-08-06-8.18
2026-08-05-5.45
2026-08-04-3.82
2026-08-03-6.89

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KEWR, lead day 2-10 to -9 °F: 1 day-9 to -8 °F: 1 day-8 to -7 °F: 1 day-7 to -6 °F: 3 days-6 to -5 °F: 6 days-5 to -4 °F: 8 days-4 to -3 °F: 5 days-3 to -2 °F: 5 days-2 to -1 °F: 1 day-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| 4.49 · P90 |e| 6.81 °F-100+10 °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
-10.00-9.001
-9.00-8.001
-8.00-7.001
-7.00-6.003
-6.00-5.006
-5.00-4.008
-4.00-3.005
-3.00-2.005
-2.00-1.001
-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
2026-08264.78 -4.78
2026-0163.82 -3.82

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 +1.64 [0.81, 2.22] yes
Aurora (IFS init) 1 +1.48 [1.48, 1.48] yes
FourCastNet v2 (IFS init) 1 -0.06 [-0.06, -0.06] yes
NCEP GFS 26 +1.77 [0.10, 2.88] yes
ECMWF IFS HRES 20 +0.65 [-0.77, 1.72] no
Pangu-Weather (IFS init) 1 +0.63 [0.63, 0.63] yes
Persistence (baseline) 26 +1.59 [0.53, 2.35] yes

Cite this page

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

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