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

New York Central Park · GraphCast (IFS init) · lead day 2

KNYC · 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 QC2.41 [1.83, 2.98]1.91-1.62 [-2.66, -0.53]2.93 [2.13, 3.61]19% [7%, 32%]46%73%+0.21+0.352026-08-03 → 2026-08-30
tmax00Zbilinear90d26 1 QC2.41 [1.75, 3.02]1.91-1.62 [-2.68, -0.25]2.93 [2.02, 3.61]19% [4%, 36%]46%73%+0.21+0.352026-08-03 → 2026-08-30
tmax00Zbilinear365d32 1 QC2.36 [1.77, 2.86]1.91-1.66 [-2.57, -0.66]2.96 [2.17, 3.55]22% [7%, 38%]50%75%+0.39+0.512026-01-03 → 2026-08-30
tmax00Zbilinearall32 1 QC2.36 [1.75, 2.85]1.91-1.66 [-2.56, -0.55]2.96 [2.15, 3.54]22% [8%, 37%]50%75%+0.39+0.512026-01-03 → 2026-08-30
tmax00Znearest30d26 1 QC2.39 [1.85, 2.93]1.94-1.52 [-2.57, -0.42]2.90 [2.10, 3.56]19% [7%, 32%]46%73%+0.21+0.342026-08-03 → 2026-08-30
tmax00Znearest90d26 1 QC2.39 [1.78, 3.01]1.94-1.52 [-2.58, -0.12]2.90 [1.99, 3.56]19% [4%, 36%]46%73%+0.21+0.342026-08-03 → 2026-08-30
tmax00Znearest365d32 1 QC2.33 [1.78, 2.82]1.94-1.55 [-2.45, -0.53]2.91 [2.14, 3.49]22% [7%, 38%]50%75%+0.40+0.512026-01-03 → 2026-08-30
tmax00Znearestall32 1 QC2.33 [1.76, 2.81]1.94-1.55 [-2.46, -0.44]2.91 [2.10, 3.48]22% [8%, 37%]50%75%+0.40+0.512026-01-03 → 2026-08-30
tmax12Zbilinear30d26 1 QC2.38 [1.63, 3.14]1.99-1.73 [-2.78, -0.67]2.93 [2.05, 3.66]27% [8%, 46%]42%69%+0.22+0.322026-08-03 → 2026-08-30
tmax12Zbilinear90d26 1 QC2.38 [1.50, 3.21]1.99-1.73 [-2.87, -0.50]2.93 [1.86, 3.70]27% [5%, 50%]42%69%+0.22+0.322026-08-03 → 2026-08-30
tmax12Zbilinear365d31 1 QC2.28 [1.49, 2.97]1.95-1.72 [-2.71, -0.68]2.91 [2.08, 3.59]32% [10%, 56%]48%71%+0.39+0.482026-01-03 → 2026-08-30
tmax12Zbilinearall31 1 QC2.28 [1.43, 3.00]1.95-1.72 [-2.68, -0.63]2.91 [1.96, 3.58]32% [10%, 56%]48%71%+0.39+0.482026-01-03 → 2026-08-30
tmax12Znearest30d26 1 QC2.37 [1.62, 3.11]2.04-1.64 [-2.70, -0.55]2.92 [2.04, 3.64]19% [4%, 36%]42%69%+0.22+0.312026-08-03 → 2026-08-30
tmax12Znearest90d26 1 QC2.37 [1.50, 3.21]2.04-1.64 [-2.78, -0.39]2.92 [1.83, 3.68]19% [0%, 40%]42%69%+0.22+0.312026-08-03 → 2026-08-30
tmax12Znearest365d31 1 QC2.26 [1.45, 2.96]2.00-1.62 [-2.62, -0.56]2.89 [2.03, 3.56]26% [7%, 48%]48%71%+0.39+0.472026-01-03 → 2026-08-30
tmax12Znearestall31 1 QC2.26 [1.40, 2.99]2.00-1.62 [-2.61, -0.53]2.89 [1.95, 3.55]26% [8%, 50%]48%71%+0.39+0.472026-01-03 → 2026-08-30
tmin00Zbilinear30d26 1 QC1.79 [1.11, 2.55]1.61+1.09 [0.39, 1.74]2.54 [1.58, 3.21]46% [22%, 67%]62%73%+0.46+0.502026-08-03 → 2026-08-30
tmin00Zbilinear90d26 1 QC1.79 [1.06, 2.75]1.61+1.09 [0.13, 1.80]2.54 [1.47, 3.27]46% [15%, 67%]62%73%+0.46+0.502026-08-03 → 2026-08-30
tmin00Zbilinear365d32 1 QC2.27 [1.37, 3.52]2.24+0.11 [-2.35, 1.52]3.15 [1.99, 4.26]41% [15%, 61%]56%66%+0.41+0.422026-01-03 → 2026-08-30
tmin00Zbilinearall32 1 QC2.27 [1.31, 3.56]2.24+0.11 [-2.18, 1.45]3.15 [1.86, 4.32]41% [13%, 61%]56%66%+0.41+0.422026-01-03 → 2026-08-30
tmin00Znearest30d26 1 QC1.90 [1.23, 2.61]1.59+1.28 [0.56, 1.92]2.59 [1.65, 3.28]38% [17%, 54%]62%73%+0.43+0.512026-08-03 → 2026-08-30
tmin00Znearest90d26 1 QC1.90 [1.20, 2.74]1.59+1.28 [0.33, 1.97]2.59 [1.51, 3.33]38% [12%, 54%]62%73%+0.43+0.512026-08-03 → 2026-08-30
tmin00Znearest365d32 1 QC2.33 [1.47, 3.44]2.21+0.29 [-2.17, 1.70]3.14 [2.03, 4.17]34% [13%, 50%]56%66%+0.39+0.432026-01-03 → 2026-08-30
tmin00Znearestall32 1 QC2.33 [1.42, 3.49]2.21+0.29 [-1.99, 1.63]3.14 [1.91, 4.22]34% [13%, 50%]56%66%+0.39+0.432026-01-03 → 2026-08-30
tmin12Zbilinear30d26 1 QC1.86 [1.15, 2.63]1.66+1.12 [0.54, 1.67]2.50 [1.62, 3.12]42% [17%, 62%]62%73%+0.44+0.482026-08-03 → 2026-08-30
tmin12Zbilinear90d26 1 QC1.86 [1.12, 2.85]1.66+1.12 [0.24, 1.70]2.50 [1.47, 3.22]42% [8%, 61%]62%73%+0.44+0.482026-08-03 → 2026-08-30
tmin12Zbilinear365d31 1 QC2.15 [1.32, 3.17]2.06+0.35 [-1.90, 1.51]2.91 [1.90, 3.92]39% [12%, 58%]58%68%+0.43+0.462026-01-03 → 2026-08-30
tmin12Zbilinearall31 1 QC2.15 [1.33, 3.24]2.06+0.35 [-1.64, 1.48]2.91 [1.81, 3.92]39% [12%, 58%]58%68%+0.43+0.462026-01-03 → 2026-08-30
tmin12Znearest30d26 1 QC1.95 [1.31, 2.66]1.64+1.30 [0.74, 1.85]2.56 [1.70, 3.16]38% [13%, 59%]62%81%+0.41+0.492026-08-03 → 2026-08-30
tmin12Znearest90d26 1 QC1.95 [1.26, 2.83]1.64+1.30 [0.44, 1.87]2.56 [1.54, 3.26]38% [8%, 58%]62%81%+0.41+0.492026-08-03 → 2026-08-30
tmin12Znearest365d31 1 QC2.20 [1.45, 3.12]2.04+0.54 [-1.71, 1.69]2.91 [1.94, 3.83]35% [10%, 56%]58%74%+0.41+0.462026-01-03 → 2026-08-30
tmin12Znearestall31 1 QC2.20 [1.42, 3.19]2.04+0.54 [-1.46, 1.65]2.91 [1.88, 3.80]35% [10%, 55%]58%74%+0.41+0.462026-01-03 → 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 KNYC, lead day 2, 00Z bilinear tmax+10+50-5-102026-08-03: +2.34 °F2026-08-04: -3.32 °F2026-08-05: -3.54 °F2026-08-06: -2.91 °F2026-08-07: -5.4 °F2026-08-08: -0.4 °F2026-08-09: +0.38 °F2026-08-10: +2.46 °F2026-08-11: +1.67 °F2026-08-12: -1.42 °F2026-08-13: +0.54 °F2026-08-14: -1.48 °F2026-08-15: -3.87 °F2026-08-16: -2.8 °F2026-08-17: -0.09 °F2026-08-18: -4.38 °F2026-08-19: -1.98 °F2026-08-20: -7.85 °F2026-08-21: -1.55 °F2026-08-22: -3.27 °F2026-08-23: -1.48 °F2026-08-24: -2.24 °F2026-08-25: -1.7 °F2026-08-26: -2.7 °F2026-08-27: +2.18 °F2026-08-30: +0.65 °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+0.65
2026-08-29—
2026-08-28—
2026-08-27+2.18
2026-08-26-2.70
2026-08-25-1.70
2026-08-24-2.24
2026-08-23-1.48
2026-08-22-3.27
2026-08-21-1.55
2026-08-20-7.85
2026-08-19-1.98
2026-08-18-4.38
2026-08-17-0.09
2026-08-16-2.80
2026-08-15-3.87
2026-08-14-1.48
2026-08-13+0.54
2026-08-12-1.42
2026-08-11+1.67
2026-08-10+2.46
2026-08-09+0.38
2026-08-08-0.40
2026-08-07-5.40
2026-08-06-2.91
2026-08-05-3.54
2026-08-04-3.32
2026-08-03+2.34

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KNYC, lead day 2-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 1 day-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: 7 days-1 to 0 °F: 4 days0 to 1 °F: 3 days1 to 2 °F: 2 days2 to 3 °F: 3 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| 2.08 · P90 |e| 4.33 °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
-10.00-9.000
-9.00-8.000
-8.00-7.001
-7.00-6.001
-6.00-5.001
-5.00-4.001
-4.00-3.004
-3.00-2.005
-2.00-1.007
-1.00+0.004
+0.00+1.003
+1.00+2.002
+2.00+3.003
+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
2026-08262.41 -1.62
2026-0162.16 -1.82

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 +0.54 [0.02, 1.00] yes
Aurora (IFS init) 1 -0.99 [-0.99, -0.99] yes
FourCastNet v2 (IFS init) 1 +0.14 [0.14, 0.14] yes
NCEP GFS 26 -1.60 [-3.24, -0.47] yes
ECMWF IFS HRES 20 +0.19 [-0.93, 0.92] no
Pangu-Weather (IFS init) 1 -3.18 [-3.18, -3.18] yes
Persistence (baseline) 26 -0.63 [-1.51, -0.08] yes

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KNYC (New York Central Park), 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/KNYC/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