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 · NCEP GFS · lead day 0

KPHL · gfs · lead 0

Target day = initialization date + 0 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, gfs-0p25, starting 2026-08-01 — 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
tmax00Zbilinear30d30 1 QC2.80 [2.06, 3.62]1.93-2.26 [-3.39, -1.27]3.57 [2.53, 4.44]17% [7%, 27%]47%70%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d30 1 QC2.80 [1.95, 3.77]1.93-2.26 [-3.64, -1.11]3.57 [2.39, 4.58]17% [6%, 29%]47%70%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d30 1 QC2.80 [1.90, 3.80]1.93-2.26 [-3.66, -1.05]3.57 [2.25, 4.62]17% [5%, 30%]47%70%——2026-08-01 → 2026-08-30
tmax00Zbilinearall30 1 QC2.80 [1.90, 3.93]1.93-2.26 [-3.76, -1.01]3.57 [2.23, 4.66]17% [4%, 29%]47%70%——2026-08-01 → 2026-08-30
tmax00Znearest30d30 1 QC3.07 [2.27, 3.91]2.21-2.42 [-3.61, -1.34]3.92 [2.83, 4.88]13% [3%, 23%]43%63%——2026-08-01 → 2026-08-30
tmax00Znearest90d30 1 QC3.07 [2.19, 4.07]2.21-2.42 [-3.89, -1.18]3.92 [2.70, 4.98]13% [0%, 24%]43%63%——2026-08-01 → 2026-08-30
tmax00Znearest365d30 1 QC3.07 [2.16, 4.19]2.21-2.42 [-3.93, -1.07]3.92 [2.60, 5.14]13% [0%, 23%]43%63%——2026-08-01 → 2026-08-30
tmax00Znearestall30 1 QC3.07 [2.12, 4.20]2.21-2.42 [-4.04, -1.05]3.92 [2.62, 5.07]13% [0%, 24%]43%63%——2026-08-01 → 2026-08-30
tmin00Zbilinear30d30 1 QC1.45 [1.22, 1.71]1.41+0.24 [-0.33, 0.79]1.69 [1.41, 1.95]37% [23%, 53%]77%93%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d30 1 QC1.45 [1.18, 1.70]1.41+0.24 [-0.37, 1.00]1.69 [1.37, 1.93]37% [21%, 54%]77%93%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d30 1 QC1.45 [1.19, 1.75]1.41+0.24 [-0.42, 1.08]1.69 [1.38, 1.97]37% [19%, 53%]77%93%——2026-08-01 → 2026-08-30
tmin00Zbilinearall30 1 QC1.45 [1.16, 1.71]1.41+0.24 [-0.39, 0.99]1.69 [1.33, 1.95]37% [21%, 53%]77%93%——2026-08-01 → 2026-08-30
tmin00Znearest30d30 1 QC1.49 [1.14, 1.87]1.45-0.54 [-1.19, 0.10]1.86 [1.48, 2.21]43% [27%, 60%]67%90%——2026-08-01 → 2026-08-30
tmin00Znearest90d30 1 QC1.49 [1.07, 1.89]1.45-0.54 [-1.29, 0.26]1.86 [1.42, 2.23]43% [24%, 67%]67%90%——2026-08-01 → 2026-08-30
tmin00Znearest365d30 1 QC1.49 [1.02, 1.91]1.45-0.54 [-1.31, 0.27]1.86 [1.33, 2.25]43% [23%, 67%]67%90%——2026-08-01 → 2026-08-30
tmin00Znearestall30 1 QC1.49 [1.07, 1.91]1.45-0.54 [-1.30, 0.28]1.86 [1.39, 2.24]43% [22%, 64%]67%90%——2026-08-01 → 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 NCEP GFS at KPHL, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -3.18 °F2026-08-02: -6.91 °F2026-08-03: -6.74 °F2026-08-04: -2.24 °F2026-08-05: -3.43 °F2026-08-06: -1.27 °F2026-08-07: -0.42 °F2026-08-08: +1.69 °F2026-08-09: -1.15 °F2026-08-10: -1.83 °F2026-08-11: -10 °F2026-08-12: -1.91 °F2026-08-13: -2.31 °F2026-08-14: -6.33 °F2026-08-15: -2.28 °F2026-08-16: -4.54 °F2026-08-17: -2.64 °F2026-08-18: -0.51 °F2026-08-19: -2.31 °F2026-08-20: +2.72 °F2026-08-21: -1.54 °F2026-08-22: -1.7 °F2026-08-23: -0.5 °F2026-08-24: -1.94 °F2026-08-25: -0.51 °F2026-08-26: -2.52 °F2026-08-27: +3.69 °F2026-08-28: -1.15 °F2026-08-29: -0.83 °F2026-08-30: -5.07 °F2026-08-012026-08-30°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 (30 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-5.07
2026-08-29-0.83
2026-08-28-1.15
2026-08-27+3.69
2026-08-26-2.52
2026-08-25-0.51
2026-08-24-1.94
2026-08-23-0.50
2026-08-22-1.70
2026-08-21-1.54
2026-08-20+2.72
2026-08-19-2.31
2026-08-18-0.51
2026-08-17-2.64
2026-08-16-4.54
2026-08-15-2.28
2026-08-14-6.33
2026-08-13-2.31
2026-08-12-1.91
2026-08-11-10.00
2026-08-10-1.83
2026-08-09-1.15
2026-08-08+1.69
2026-08-07-0.42
2026-08-06-1.27
2026-08-05-3.43
2026-08-04-2.24
2026-08-03-6.74
2026-08-02-6.91
2026-08-01-3.18

Error distribution

Distribution of the signed daily error in °F for NCEP GFS at KPHL, lead day 0-10 to -9 °F: 1 day-9 to -8 °F: 0 days-8 to -7 °F: 0 days-7 to -6 °F: 3 days-6 to -5 °F: 1 day-5 to -4 °F: 1 day-4 to -3 °F: 2 days-3 to -2 °F: 6 days-2 to -1 °F: 8 days-1 to 0 °F: 5 days0 to 1 °F: 0 days1 to 2 °F: 1 day2 to 3 °F: 1 day3 to 4 °F: 1 day4 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| 2.26 · P90 |e| 6.37 °F-100+10 °Fn = 30 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
-9.00-8.000
-8.00-7.000
-7.00-6.003
-6.00-5.001
-5.00-4.001
-4.00-3.002
-3.00-2.006
-2.00-1.008
-1.00+0.005
+0.00+1.000
+1.00+2.001
+2.00+3.001
+3.00+4.001
+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
2026-08302.80 -2.26

Availability

ModelScored period DaysCoverage
NCEP GFS2026-08-01 → 2026-08-30 30
NCEP GFS2026-08-01 → 2026-08-30 30

Compared with other models

Paired bootstrap on the days both models cover (90d, 00Z, bilinear, daily maximum). Negative means NCEP GFS has the lower MAE. "Distinguishable" means the 95 % interval excludes zero.

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 30 +0.48 [-0.47, 1.53] no
Aurora (GFS init) 2 -2.05 [-3.19, -0.92] yes
Aurora (IFS init) 3 -2.29 [-4.55, 1.00] no
FourCastNet v2 (GFS init) 2 -0.99 [-1.55, -0.42] yes
FourCastNet v2 (IFS init) 3 -2.76 [-5.67, -1.31] yes
GraphCast (GFS init) 2 -2.30 [-3.76, -0.85] yes
GraphCast (IFS init) 28 -1.98 [-2.95, -1.05] yes
ECMWF IFS HRES 24 -0.07 [-1.22, 1.33] no
Pangu-Weather (GFS init) 2 +0.86 [-0.03, 1.76] no
Pangu-Weather (IFS init) 3 -0.34 [-3.05, 2.13] no

Cite this page

ShortCastCheck, KPHL gfs lead 0, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KPHL/model/gfs/lead/0/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KPHL (Philadelphia Intl), model gfs, lead day 0. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KPHL/model/gfs/lead/0/

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