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 5

KPHL · graphcast_ifs · lead 5

Target day = initialization date + 5 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
tmax00Zbilinear30d25 1 QC4.49 [3.76, 5.39]2.17-4.27 [-5.36, -3.31]5.17 [4.20, 6.24]8% [0%, 17%]20%28%-0.22+0.392026-08-06 → 2026-08-30
tmax00Zbilinear90d25 1 QC4.49 [3.68, 5.35]2.17-4.27 [-5.34, -3.20]5.17 [4.12, 6.20]8% [0%, 19%]20%28%-0.22+0.392026-08-06 → 2026-08-30
tmax00Zbilinear365d31 1 QC4.60 [3.75, 5.47]2.98-3.31 [-5.01, -0.78]5.43 [4.20, 6.43]6% [0%, 15%]23%29%+0.20+0.482026-01-06 → 2026-08-30
tmax00Zbilinearall31 1 QC4.60 [3.78, 5.46]2.98-3.31 [-4.96, -0.71]5.43 [4.22, 6.43]6% [0%, 15%]23%29%+0.20+0.482026-01-06 → 2026-08-30
tmax00Znearest30d25 1 QC4.23 [3.52, 5.11]2.16-3.98 [-5.10, -2.98]4.93 [3.98, 6.00]12% [4%, 20%]20%28%-0.15+0.402026-08-06 → 2026-08-30
tmax00Znearest90d25 1 QC4.23 [3.46, 5.11]2.16-3.98 [-5.07, -2.89]4.93 [3.89, 5.95]12% [3%, 21%]20%28%-0.15+0.402026-08-06 → 2026-08-30
tmax00Znearest365d31 1 QC4.40 [3.55, 5.24]3.00-2.99 [-4.72, -0.36]5.25 [4.00, 6.30]10% [0%, 18%]23%29%+0.23+0.472026-01-06 → 2026-08-30
tmax00Znearestall31 1 QC4.40 [3.57, 5.27]3.00-2.99 [-4.68, -0.30]5.25 [3.99, 6.33]10% [0%, 16%]23%29%+0.23+0.472026-01-06 → 2026-08-30
tmax12Zbilinear30d25 1 QC4.82 [3.70, 5.81]2.77-4.75 [-5.78, -3.58]5.80 [4.60, 6.65]16% [4%, 30%]24%36%-0.31+0.232026-08-06 → 2026-08-30
tmax12Zbilinear90d25 1 QC4.82 [3.64, 5.82]2.77-4.75 [-5.80, -3.46]5.80 [4.63, 6.64]16% [0%, 31%]24%36%-0.31+0.232026-08-06 → 2026-08-30
tmax12Zbilinear365d30 1 QC4.66 [3.65, 5.55]3.13-4.05 [-5.49, -2.08]5.56 [4.29, 6.45]17% [5%, 30%]23%33%+0.20+0.462026-01-06 → 2026-08-30
tmax12Zbilinearall30 1 QC4.66 [3.53, 5.56]3.13-4.05 [-5.44, -1.96]5.56 [4.26, 6.50]17% [4%, 29%]23%33%+0.20+0.462026-01-06 → 2026-08-30
tmax12Znearest30d25 1 QC4.62 [3.57, 5.55]2.72-4.47 [-5.50, -3.28]5.55 [4.40, 6.39]12% [4%, 21%]28%36%-0.26+0.242026-08-06 → 2026-08-30
tmax12Znearest90d25 1 QC4.62 [3.54, 5.55]2.72-4.47 [-5.53, -3.16]5.55 [4.43, 6.37]12% [0%, 22%]28%36%-0.26+0.242026-08-06 → 2026-08-30
tmax12Znearest365d30 1 QC4.48 [3.50, 5.30]3.10-3.76 [-5.22, -1.75]5.35 [4.16, 6.20]13% [5%, 24%]27%33%+0.23+0.472026-01-06 → 2026-08-30
tmax12Znearestall30 1 QC4.48 [3.47, 5.35]3.10-3.76 [-5.19, -1.64]5.35 [4.10, 6.22]13% [4%, 23%]27%33%+0.23+0.472026-01-06 → 2026-08-30
tmin00Zbilinear30d25 1 QC2.77 [1.81, 3.67]2.75-0.44 [-2.02, 0.93]3.22 [2.22, 3.99]16% [4%, 30%]40%52%+0.08+0.132026-08-06 → 2026-08-30
tmin00Zbilinear90d25 1 QC2.77 [1.65, 3.68]2.75-0.44 [-2.01, 1.21]3.22 [2.02, 4.01]16% [3%, 31%]40%52%+0.08+0.132026-08-06 → 2026-08-30
tmin00Zbilinear365d31 1 QC2.83 [1.82, 3.74]2.81+0.21 [-1.53, 2.35]3.69 [2.28, 4.89]26% [9%, 50%]45%55%+0.34+0.352026-01-06 → 2026-08-30
tmin00Zbilinearall31 1 QC2.83 [1.82, 3.72]2.81+0.21 [-1.63, 2.36]3.69 [2.26, 4.91]26% [9%, 52%]45%55%+0.34+0.352026-01-06 → 2026-08-30
tmin00Znearest30d25 1 QC2.71 [1.85, 3.48]2.71-0.03 [-1.55, 1.32]3.14 [2.26, 3.75]20% [7%, 41%]44%56%+0.10+0.142026-08-06 → 2026-08-30
tmin00Znearest90d25 1 QC2.71 [1.66, 3.51]2.71-0.03 [-1.54, 1.57]3.14 [2.10, 3.79]20% [5%, 50%]44%56%+0.10+0.142026-08-06 → 2026-08-30
tmin00Znearest365d31 1 QC2.83 [1.82, 3.70]2.78+0.60 [-1.10, 2.67]3.69 [2.33, 5.02]26% [8%, 50%]48%58%+0.34+0.362026-01-06 → 2026-08-30
tmin00Znearestall31 1 QC2.83 [1.87, 3.71]2.78+0.60 [-1.19, 2.69]3.69 [2.37, 5.03]26% [8%, 52%]48%58%+0.34+0.362026-01-06 → 2026-08-30
tmin12Zbilinear30d25 1 QC2.59 [1.56, 3.55]2.56-0.54 [-2.06, 0.80]3.23 [1.93, 4.12]28% [14%, 47%]44%68%+0.14+0.192026-08-06 → 2026-08-30
tmin12Zbilinear90d25 1 QC2.59 [1.44, 3.67]2.56-0.54 [-2.02, 0.96]3.23 [1.81, 4.26]28% [14%, 54%]44%68%+0.14+0.192026-08-06 → 2026-08-30
tmin12Zbilinear365d30 1 QC2.85 [1.62, 4.07]2.86+0.19 [-1.64, 2.51]3.76 [2.01, 5.13]30% [14%, 50%]43%67%+0.34+0.342026-01-06 → 2026-08-30
tmin12Zbilinearall30 1 QC2.85 [1.64, 4.02]2.86+0.19 [-1.76, 2.71]3.76 [2.04, 5.17]30% [15%, 53%]43%67%+0.34+0.342026-01-06 → 2026-08-30
tmin12Znearest30d25 1 QC2.56 [1.60, 3.48]2.56-0.13 [-1.63, 1.19]3.18 [1.86, 4.07]16% [4%, 30%]56%68%+0.15+0.192026-08-06 → 2026-08-30
tmin12Znearest90d25 1 QC2.56 [1.50, 3.64]2.56-0.13 [-1.60, 1.35]3.18 [1.75, 4.22]16% [4%, 33%]56%68%+0.15+0.192026-08-06 → 2026-08-30
tmin12Znearest365d30 1 QC2.86 [1.65, 4.15]2.86+0.55 [-1.23, 2.83]3.77 [1.96, 5.23]20% [6%, 38%]53%67%+0.33+0.342026-01-06 → 2026-08-30
tmin12Znearestall30 1 QC2.86 [1.70, 4.10]2.86+0.55 [-1.37, 2.92]3.77 [2.04, 5.32]20% [7%, 39%]53%67%+0.33+0.342026-01-06 → 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 5, 00Z bilinear tmax+20+100-10-202026-08-06: -4.07 °F2026-08-07: -5.49 °F2026-08-08: -5.12 °F2026-08-09: -5.9 °F2026-08-10: -2 °F2026-08-11: -0.73 °F2026-08-12: -9.79 °F2026-08-13: -3.25 °F2026-08-14: -5.02 °F2026-08-15: -11.8 °F2026-08-16: -3.8 °F2026-08-17: -4.1 °F2026-08-18: -5.65 °F2026-08-19: -7.17 °F2026-08-20: -1.05 °F2026-08-21: -5.09 °F2026-08-22: +1.98 °F2026-08-23: -1.93 °F2026-08-24: -5.05 °F2026-08-25: -3.86 °F2026-08-26: -6.29 °F2026-08-27: -2.92 °F2026-08-28: -5.98 °F2026-08-29: +0.73 °F2026-08-30: -3.35 °F2026-08-062026-08-30°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 (25 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-3.35
2026-08-29+0.73
2026-08-28-5.98
2026-08-27-2.92
2026-08-26-6.29
2026-08-25-3.86
2026-08-24-5.05
2026-08-23-1.93
2026-08-22+1.98
2026-08-21-5.09
2026-08-20-1.05
2026-08-19-7.17
2026-08-18-5.65
2026-08-17-4.10
2026-08-16-3.80
2026-08-15-11.80
2026-08-14-5.02
2026-08-13-3.25
2026-08-12-9.79
2026-08-11-0.73
2026-08-10-2.00
2026-08-09-5.90
2026-08-08-5.12
2026-08-07-5.49
2026-08-06-4.07

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KPHL, lead day 5-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 0 days-12 to -10 °F: 1 day-10 to -8 °F: 1 day-8 to -6 °F: 3 days-6 to -4 °F: 10 days-4 to -2 °F: 7 days-2 to 0 °F: 4 days0 to 2 °F: 3 days2 to 4 °F: 1 day4 to 6 °F: 0 days6 to 8 °F: 0 days8 to 10 °F: 0 days10 to 12 °F: 0 days12 to 14 °F: 1 day14 to 16 °F: 0 days16 to 18 °F: 0 days18 to 20 °F: 0 daysdashed: P50 |e| 4.07 · P90 |e| 7.86 °F-200+20 °Fn = 31 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
-20.00-18.000
-18.00-16.000
-16.00-14.000
-14.00-12.000
-12.00-10.001
-10.00-8.001
-8.00-6.003
-6.00-4.0010
-4.00-2.007
-2.00+0.004
+0.00+2.003
+2.00+4.001
+4.00+6.000
+6.00+8.000
+8.00+10.000
+10.00+12.000
+12.00+14.001
+14.00+16.000
+16.00+18.000
+18.00+20.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08254.49 -4.27
2026-0165.10 +0.67

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 30
GraphCast (IFS init)2026-01-06 → 2026-08-30 30

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 25 +1.53 [0.69, 2.34] yes
NCEP GFS 25 +0.16 [-1.02, 1.76] no
ECMWF IFS HRES 18 +0.13 [-1.00, 0.71] no
Persistence (baseline) 25 +0.81 [-0.67, 2.29] no

Cite this page

ShortCastCheck, KPHL graphcast_ifs lead 5, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KPHL/model/graphcast_ifs/lead/5/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KPHL (Philadelphia Intl), model graphcast_ifs, lead day 5. 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/5/

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