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 7

KPHL · graphcast_ifs · lead 7

Target day = initialization date + 7 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
tmax00Zbilinear30d23 1 QC4.53 [3.69, 5.47]2.55-4.22 [-5.07, -3.46]5.30 [4.34, 6.28]9% [0%, 17%]22%35%-0.18+0.302026-08-08 → 2026-08-30
tmax00Zbilinear90d23 1 QC4.53 [3.63, 5.46]2.55-4.22 [-5.10, -3.37]5.30 [4.29, 6.29]9% [0%, 16%]22%35%-0.18+0.302026-08-08 → 2026-08-30
tmax00Zbilinear365d29 1 QC4.69 [3.77, 5.71]3.13-3.68 [-4.92, -2.10]5.53 [4.46, 6.55]7% [0%, 14%]21%38%+0.19+0.472026-01-08 → 2026-08-30
tmax00Zbilinearall29 1 QC4.69 [3.82, 5.66]3.13-3.68 [-4.95, -2.06]5.53 [4.49, 6.51]7% [0%, 14%]21%38%+0.19+0.472026-01-08 → 2026-08-30
tmax00Znearest30d23 1 QC4.31 [3.47, 5.27]2.53-3.94 [-4.78, -3.17]5.06 [4.07, 6.09]9% [0%, 19%]26%35%-0.13+0.312026-08-08 → 2026-08-30
tmax00Znearest90d23 1 QC4.31 [3.40, 5.27]2.53-3.94 [-4.85, -3.06]5.06 [3.98, 6.12]9% [0%, 20%]26%35%-0.13+0.312026-08-08 → 2026-08-30
tmax00Znearest365d29 1 QC4.47 [3.56, 5.51]3.12-3.36 [-4.64, -1.71]5.33 [4.20, 6.39]7% [0%, 16%]28%38%+0.23+0.472026-01-08 → 2026-08-30
tmax00Znearestall29 1 QC4.47 [3.56, 5.45]3.12-3.36 [-4.65, -1.64]5.33 [4.22, 6.34]7% [0%, 16%]28%38%+0.23+0.472026-01-08 → 2026-08-30
tmax12Zbilinear30d23 1 QC4.46 [3.69, 5.50]2.35-4.23 [-5.40, -3.29]5.10 [4.32, 5.97]4% [0%, 12%]22%30%-0.16+0.362026-08-08 → 2026-08-30
tmax12Zbilinear90d23 1 QC4.46 [3.62, 5.69]2.35-4.23 [-5.61, -3.25]5.10 [4.19, 6.05]4% [0%, 12%]22%30%-0.16+0.362026-08-08 → 2026-08-30
tmax12Zbilinear365d28 1 QC4.53 [3.77, 5.56]2.87-3.30 [-5.10, -0.79]5.17 [4.34, 6.01]4% [0%, 11%]21%29%+0.22+0.512026-01-08 → 2026-08-30
tmax12Zbilinearall28 1 QC4.53 [3.78, 5.45]2.87-3.30 [-5.05, -0.76]5.17 [4.39, 5.95]4% [0%, 10%]21%29%+0.22+0.512026-01-08 → 2026-08-30
tmax12Znearest30d23 1 QC4.27 [3.55, 5.26]2.35-3.96 [-5.15, -3.00]4.88 [4.11, 5.73]4% [0%, 12%]26%35%-0.12+0.362026-08-08 → 2026-08-30
tmax12Znearest90d23 1 QC4.27 [3.46, 5.44]2.35-3.96 [-5.39, -2.95]4.88 [3.97, 5.81]4% [0%, 12%]26%35%-0.12+0.362026-08-08 → 2026-08-30
tmax12Znearest365d28 1 QC4.37 [3.62, 5.40]2.92-3.02 [-4.86, -0.46]5.02 [4.15, 5.91]4% [0%, 11%]25%32%+0.24+0.502026-01-08 → 2026-08-30
tmax12Znearestall28 1 QC4.37 [3.63, 5.27]2.92-3.02 [-4.77, -0.46]5.02 [4.19, 5.83]4% [0%, 10%]25%32%+0.24+0.502026-01-08 → 2026-08-30
tmin00Zbilinear30d23 1 QC3.18 [1.80, 4.42]3.14-0.34 [-1.72, 1.08]3.78 [2.09, 4.80]17% [0%, 33%]39%43%+0.21+0.152026-08-08 → 2026-08-30
tmin00Zbilinear90d23 1 QC3.18 [1.64, 4.46]3.14-0.34 [-1.83, 1.19]3.78 [1.87, 4.82]17% [0%, 35%]39%43%+0.21+0.152026-08-08 → 2026-08-30
tmin00Zbilinear365d29 1 QC3.46 [2.06, 4.74]3.43+0.67 [-1.21, 3.22]4.27 [2.48, 5.55]21% [6%, 36%]38%41%+0.32+0.322026-01-08 → 2026-08-30
tmin00Zbilinearall29 1 QC3.46 [2.08, 4.59]3.43+0.67 [-1.28, 3.45]4.27 [2.43, 5.43]21% [6%, 37%]38%41%+0.32+0.322026-01-08 → 2026-08-30
tmin00Znearest30d23 1 QC3.16 [1.86, 4.40]3.17+0.09 [-1.28, 1.51]3.79 [2.19, 4.73]17% [0%, 33%]39%43%+0.22+0.152026-08-08 → 2026-08-30
tmin00Znearest90d23 1 QC3.16 [1.69, 4.43]3.17+0.09 [-1.38, 1.64]3.79 [2.02, 4.84]17% [0%, 35%]39%43%+0.22+0.152026-08-08 → 2026-08-30
tmin00Znearest365d29 1 QC3.51 [2.10, 4.86]3.45+1.07 [-0.79, 3.59]4.36 [2.50, 5.74]17% [3%, 33%]38%41%+0.31+0.322026-01-08 → 2026-08-30
tmin00Znearestall29 1 QC3.51 [2.13, 4.73]3.45+1.07 [-0.84, 3.81]4.36 [2.60, 5.64]17% [3%, 33%]38%41%+0.31+0.322026-01-08 → 2026-08-30
tmin12Zbilinear30d23 1 QC2.42 [1.93, 2.91]2.41-0.17 [-1.62, 1.22]2.75 [2.22, 3.17]17% [3%, 33%]35%70%+0.40+0.352026-08-08 → 2026-08-30
tmin12Zbilinear90d23 1 QC2.42 [1.85, 2.94]2.41-0.17 [-1.64, 1.39]2.75 [2.10, 3.17]17% [0%, 34%]35%70%+0.40+0.352026-08-08 → 2026-08-30
tmin12Zbilinear365d28 1 QC2.94 [2.00, 4.25]2.90+0.58 [-1.36, 3.02]3.61 [2.33, 5.17]18% [4%, 33%]32%61%+0.43+0.432026-01-08 → 2026-08-30
tmin12Zbilinearall28 1 QC2.94 [2.02, 4.44]2.90+0.58 [-1.29, 3.21]3.61 [2.43, 5.38]18% [0%, 35%]32%61%+0.43+0.432026-01-08 → 2026-08-30
tmin12Znearest30d23 1 QC2.42 [1.97, 2.86]2.43+0.26 [-1.20, 1.63]2.77 [2.30, 3.11]13% [0%, 27%]48%74%+0.40+0.342026-08-08 → 2026-08-30
tmin12Znearest90d23 1 QC2.42 [1.88, 2.87]2.43+0.26 [-1.23, 1.86]2.77 [2.14, 3.11]13% [0%, 29%]48%74%+0.40+0.342026-08-08 → 2026-08-30
tmin12Znearest365d28 1 QC2.99 [2.09, 4.49]2.92+0.99 [-0.93, 3.38]3.72 [2.43, 5.43]14% [0%, 28%]43%64%+0.42+0.432026-01-08 → 2026-08-30
tmin12Znearestall28 1 QC2.99 [2.12, 4.65]2.92+0.99 [-0.87, 3.63]3.72 [2.52, 5.63]14% [0%, 29%]43%64%+0.42+0.432026-01-08 → 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 7, 00Z bilinear tmax+20+100-10-202026-08-08: -2.92 °F2026-08-09: -6.57 °F2026-08-10: -1.79 °F2026-08-11: -2.22 °F2026-08-12: -6.73 °F2026-08-13: -3.36 °F2026-08-14: -6.34 °F2026-08-15: -12.28 °F2026-08-16: +2.85 °F2026-08-17: -4.14 °F2026-08-18: -6.55 °F2026-08-19: -6.26 °F2026-08-20: +0.69 °F2026-08-21: -4.11 °F2026-08-22: -1.09 °F2026-08-23: -1.98 °F2026-08-24: -7.78 °F2026-08-25: -3.19 °F2026-08-26: -5.98 °F2026-08-27: -0.09 °F2026-08-28: -6.34 °F2026-08-29: -4.19 °F2026-08-30: -6.61 °F2026-08-082026-08-30°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 (23 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-6.61
2026-08-29-4.19
2026-08-28-6.34
2026-08-27-0.09
2026-08-26-5.98
2026-08-25-3.19
2026-08-24-7.78
2026-08-23-1.98
2026-08-22-1.09
2026-08-21-4.11
2026-08-20+0.69
2026-08-19-6.26
2026-08-18-6.55
2026-08-17-4.14
2026-08-16+2.85
2026-08-15-12.28
2026-08-14-6.34
2026-08-13-3.36
2026-08-12-6.73
2026-08-11-2.22
2026-08-10-1.79
2026-08-09-6.57
2026-08-08-2.92

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KPHL, lead day 7-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 1 day-12 to -10 °F: 1 day-10 to -8 °F: 0 days-8 to -6 °F: 9 days-6 to -4 °F: 4 days-4 to -2 °F: 5 days-2 to 0 °F: 5 days0 to 2 °F: 1 day2 to 4 °F: 2 days4 to 6 °F: 0 days6 to 8 °F: 0 days8 to 10 °F: 1 day10 to 12 °F: 0 days12 to 14 °F: 0 days14 to 16 °F: 0 days16 to 18 °F: 0 days18 to 20 °F: 0 daysdashed: P50 |e| 4.14 · P90 |e| 8.02 °F-200+20 °Fn = 29 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
-14.00-12.001
-12.00-10.001
-10.00-8.000
-8.00-6.009
-6.00-4.004
-4.00-2.005
-2.00+0.005
+0.00+2.001
+2.00+4.002
+4.00+6.000
+6.00+8.000
+8.00+10.001
+10.00+12.000
+12.00+14.000
+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
2026-08234.53 -4.22
2026-0165.30 -1.62

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 28
GraphCast (IFS init)2026-01-08 → 2026-08-30 28

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 23 +0.98 [0.13, 2.01] yes
NCEP GFS 23 +0.76 [-0.08, 2.00] no
ECMWF IFS HRES 18 +1.75 [-0.13, 2.96] no
Persistence (baseline) 23 +0.70 [-1.12, 2.52] no

Cite this page

ShortCastCheck, KPHL graphcast_ifs lead 7, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KPHL/model/graphcast_ifs/lead/7/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KPHL (Philadelphia Intl), model graphcast_ifs, lead day 7. 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/7/

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