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

Philadelphia Intl · GraphCast (IFS init) · lead day 2

KPHL · 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 QC4.84 [4.19, 5.53]1.77-4.84 [-5.53, -4.19]5.29 [4.72, 5.86]8% [0%, 17%]12%15%-0.66+0.392026-08-03 → 2026-08-30
tmax00Zbilinear90d26 1 QC4.84 [4.09, 5.56]1.77-4.84 [-5.56, -4.09]5.29 [4.67, 5.93]8% [0%, 17%]12%15%-0.66+0.392026-08-03 → 2026-08-30
tmax00Zbilinear365d32 1 QC4.54 [3.69, 5.32]1.80-4.52 [-5.32, -3.59]5.02 [4.17, 5.66]9% [0%, 19%]16%19%-0.15+0.562026-01-03 → 2026-08-30
tmax00Zbilinearall32 1 QC4.54 [3.78, 5.35]1.80-4.52 [-5.35, -3.71]5.02 [4.29, 5.69]9% [0%, 19%]16%19%-0.15+0.562026-01-03 → 2026-08-30
tmax00Znearest30d26 1 QC4.55 [3.90, 5.22]1.74-4.55 [-5.22, -3.90]5.01 [4.43, 5.57]8% [0%, 17%]12%23%-0.56+0.402026-08-03 → 2026-08-30
tmax00Znearest90d26 1 QC4.55 [3.77, 5.26]1.74-4.55 [-5.26, -3.77]5.01 [4.37, 5.62]8% [0%, 17%]12%23%-0.56+0.402026-08-03 → 2026-08-30
tmax00Znearest365d32 1 QC4.25 [3.40, 5.00]1.81-4.19 [-5.00, -3.22]4.74 [3.84, 5.38]9% [0%, 19%]16%28%-0.08+0.562026-01-03 → 2026-08-30
tmax00Znearestall32 1 QC4.25 [3.46, 5.05]1.81-4.19 [-5.05, -3.31]4.74 [3.98, 5.39]9% [0%, 19%]16%28%-0.08+0.562026-01-03 → 2026-08-30
tmax12Zbilinear30d26 1 QC4.93 [4.23, 5.64]1.60-4.93 [-5.64, -4.23]5.28 [4.67, 5.85]0% [0%, 0%]12%15%-0.69+0.452026-08-03 → 2026-08-30
tmax12Zbilinear90d26 1 QC4.93 [4.09, 5.70]1.60-4.93 [-5.70, -4.09]5.28 [4.59, 5.87]0% [0%, 0%]12%15%-0.69+0.452026-08-03 → 2026-08-30
tmax12Zbilinear365d31 1 QC4.56 [3.50, 5.41]1.69-4.56 [-5.41, -3.50]4.99 [4.00, 5.68]3% [0%, 12%]16%19%-0.22+0.562026-01-03 → 2026-08-30
tmax12Zbilinearall31 1 QC4.56 [3.62, 5.45]1.69-4.56 [-5.45, -3.62]4.99 [4.16, 5.67]3% [0%, 12%]16%19%-0.22+0.562026-01-03 → 2026-08-30
tmax12Znearest30d26 1 QC4.64 [3.93, 5.32]1.60-4.64 [-5.32, -3.93]5.01 [4.42, 5.53]8% [0%, 20%]12%23%-0.59+0.452026-08-03 → 2026-08-30
tmax12Znearest90d26 1 QC4.64 [3.81, 5.37]1.60-4.64 [-5.37, -3.81]5.01 [4.34, 5.55]8% [0%, 21%]12%23%-0.59+0.452026-08-03 → 2026-08-30
tmax12Znearest365d31 1 QC4.26 [3.15, 5.10]1.74-4.23 [-5.09, -3.04]4.70 [3.69, 5.37]10% [0%, 22%]16%29%-0.14+0.552026-01-03 → 2026-08-30
tmax12Znearestall31 1 QC4.26 [3.31, 5.13]1.74-4.23 [-5.13, -3.23]4.70 [3.85, 5.36]10% [0%, 22%]16%29%-0.14+0.552026-01-03 → 2026-08-30
tmin00Zbilinear30d26 1 QC2.03 [1.57, 2.42]2.03-0.67 [-1.30, 0.02]2.46 [1.96, 2.84]19% [4%, 35%]58%73%+0.41+0.422026-08-03 → 2026-08-30
tmin00Zbilinear90d26 1 QC2.03 [1.40, 2.43]2.03-0.67 [-1.32, 0.11]2.46 [1.73, 2.88]19% [3%, 36%]58%73%+0.41+0.422026-08-03 → 2026-08-30
tmin00Zbilinear365d32 1 QC1.87 [1.24, 2.29]1.86-0.61 [-1.23, 0.10]2.29 [1.45, 2.73]19% [5%, 33%]66%78%+0.49+0.492026-01-03 → 2026-08-30
tmin00Zbilinearall32 1 QC1.87 [1.30, 2.32]1.86-0.61 [-1.19, 0.02]2.29 [1.56, 2.75]19% [6%, 32%]66%78%+0.49+0.492026-01-03 → 2026-08-30
tmin00Znearest30d26 1 QC2.02 [1.61, 2.35]2.04-0.27 [-0.88, 0.40]2.37 [1.93, 2.69]23% [7%, 41%]58%73%+0.42+0.422026-08-03 → 2026-08-30
tmin00Znearest90d26 1 QC2.02 [1.54, 2.36]2.04-0.27 [-0.89, 0.49]2.37 [1.77, 2.71]23% [5%, 42%]58%73%+0.42+0.422026-08-03 → 2026-08-30
tmin00Znearest365d32 1 QC1.84 [1.24, 2.23]1.84-0.23 [-0.83, 0.42]2.19 [1.42, 2.60]25% [9%, 40%]66%78%+0.50+0.502026-01-03 → 2026-08-30
tmin00Znearestall32 1 QC1.84 [1.32, 2.24]1.84-0.23 [-0.78, 0.36]2.19 [1.56, 2.61]25% [9%, 40%]66%78%+0.50+0.502026-01-03 → 2026-08-30
tmin12Zbilinear30d26 1 QC2.02 [1.57, 2.42]2.06-0.70 [-1.30, -0.06]2.42 [1.96, 2.78]27% [8%, 46%]50%77%+0.42+0.422026-08-03 → 2026-08-30
tmin12Zbilinear90d26 1 QC2.02 [1.44, 2.42]2.06-0.70 [-1.29, -0.03]2.42 [1.80, 2.78]27% [9%, 52%]50%77%+0.42+0.422026-08-03 → 2026-08-30
tmin12Zbilinear365d31 1 QC1.86 [1.23, 2.27]1.83-0.71 [-1.23, -0.14]2.26 [1.47, 2.64]32% [13%, 57%]58%81%+0.49+0.502026-01-03 → 2026-08-30
tmin12Zbilinearall31 1 QC1.86 [1.27, 2.29]1.83-0.71 [-1.22, -0.20]2.26 [1.54, 2.67]32% [13%, 58%]58%81%+0.49+0.502026-01-03 → 2026-08-30
tmin12Znearest30d26 1 QC2.02 [1.65, 2.36]2.05-0.33 [-0.93, 0.29]2.30 [1.88, 2.62]19% [11%, 29%]58%81%+0.42+0.422026-08-03 → 2026-08-30
tmin12Znearest90d26 1 QC2.02 [1.53, 2.39]2.05-0.33 [-0.89, 0.32]2.30 [1.75, 2.65]19% [9%, 32%]58%81%+0.42+0.422026-08-03 → 2026-08-30
tmin12Znearest365d31 1 QC1.85 [1.23, 2.22]1.84-0.36 [-0.87, 0.21]2.15 [1.40, 2.51]26% [12%, 45%]65%84%+0.50+0.502026-01-03 → 2026-08-30
tmin12Znearestall31 1 QC1.85 [1.29, 2.24]1.84-0.36 [-0.83, 0.14]2.15 [1.52, 2.54]26% [13%, 44%]65%84%+0.50+0.502026-01-03 → 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 KPHL, lead day 2, 00Z bilinear tmax+10+50-5-102026-08-03: -5.98 °F2026-08-04: -8.46 °F2026-08-05: -3.86 °F2026-08-06: -6.27 °F2026-08-07: -6.92 °F2026-08-08: -3.1 °F2026-08-09: -3.97 °F2026-08-10: -1.83 °F2026-08-11: -0.46 °F2026-08-12: -6.63 °F2026-08-13: -3.25 °F2026-08-14: -8.21 °F2026-08-15: -8.06 °F2026-08-16: -2.79 °F2026-08-17: -4.85 °F2026-08-18: -3.84 °F2026-08-19: -6.36 °F2026-08-20: -5.63 °F2026-08-21: -5.6 °F2026-08-22: -4.64 °F2026-08-23: -3.61 °F2026-08-24: -6.29 °F2026-08-25: -3.05 °F2026-08-26: -6.68 °F2026-08-27: -0.7 °F2026-08-30: -4.91 °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
2026-08-30-4.91
2026-08-29—
2026-08-28—
2026-08-27-0.70
2026-08-26-6.68
2026-08-25-3.05
2026-08-24-6.29
2026-08-23-3.61
2026-08-22-4.64
2026-08-21-5.60
2026-08-20-5.63
2026-08-19-6.36
2026-08-18-3.84
2026-08-17-4.85
2026-08-16-2.79
2026-08-15-8.06
2026-08-14-8.21
2026-08-13-3.25
2026-08-12-6.63
2026-08-11-0.46
2026-08-10-1.83
2026-08-09-3.97
2026-08-08-3.10
2026-08-07-6.92
2026-08-06-6.27
2026-08-05-3.86
2026-08-04-8.46
2026-08-03-5.98

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KPHL, lead day 2-10 to -9 °F: 0 days-9 to -8 °F: 3 days-8 to -7 °F: 0 days-7 to -6 °F: 6 days-6 to -5 °F: 4 days-5 to -4 °F: 4 days-4 to -3 °F: 9 days-3 to -2 °F: 1 day-2 to -1 °F: 2 days-1 to 0 °F: 2 days0 to 1 °F: 1 day1 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| 4.45 · P90 |e| 6.89 °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
-9.00-8.003
-8.00-7.000
-7.00-6.006
-6.00-5.004
-5.00-4.004
-4.00-3.009
-3.00-2.001
-2.00-1.002
-1.00+0.002
+0.00+1.001
+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
2026-08264.84 -4.84
2026-0163.24 -3.11

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 +2.00 [1.51, 2.46] yes
Aurora (IFS init) 1 +0.42 [0.42, 0.42] yes
FourCastNet v2 (IFS init) 1 +0.32 [0.32, 0.32] yes
NCEP GFS 26 +1.18 [0.24, 2.08] yes
ECMWF IFS HRES 20 +1.10 [-0.53, 1.85] no
Pangu-Weather (IFS init) 1 +1.67 [1.67, 1.67] yes
Persistence (baseline) 26 +1.92 [1.06, 3.07] yes

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KPHL (Philadelphia Intl), 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/KPHL/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