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 0

KPHL · graphcast_ifs · 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, 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
tmax00Zbilinear30d28 1 QC4.76 [4.27, 5.43]1.14-4.76 [-5.43, -4.27]4.98 [4.50, 5.61]0% [0%, 0%]4%11%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 1 QC4.76 [4.27, 5.50]1.14-4.76 [-5.50, -4.27]4.98 [4.49, 5.73]0% [0%, 0%]4%11%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 1 QC4.36 [3.47, 5.11]1.36-4.32 [-5.11, -3.35]4.66 [3.78, 5.35]3% [0%, 11%]6%21%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 1 QC4.36 [3.53, 5.19]1.36-4.32 [-5.17, -3.43]4.66 [3.89, 5.43]3% [0%, 11%]6%21%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 1 QC4.49 [4.03, 5.12]1.17-4.49 [-5.12, -4.03]4.73 [4.29, 5.30]4% [0%, 10%]7%11%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 1 QC4.49 [4.04, 5.19]1.17-4.49 [-5.19, -4.04]4.73 [4.29, 5.41]4% [0%, 11%]7%11%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 1 QC4.09 [3.20, 4.82]1.43-4.02 [-4.81, -2.99]4.41 [3.52, 5.07]3% [0%, 9%]15%24%——2026-01-01 → 2026-08-30
tmax00Znearestall34 1 QC4.09 [3.29, 4.90]1.43-4.02 [-4.88, -3.05]4.41 [3.65, 5.12]3% [0%, 9%]15%24%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 1 QC1.58 [1.31, 1.87]1.36-0.81 [-1.25, -0.38]1.96 [1.64, 2.26]39% [27%, 52%]68%89%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 1 QC1.58 [1.30, 1.87]1.36-0.81 [-1.37, -0.34]1.96 [1.58, 2.24]39% [27%, 52%]68%89%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 1 QC1.51 [1.24, 1.80]1.25-0.88 [-1.33, -0.42]1.89 [1.56, 2.17]41% [30%, 53%]71%88%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 1 QC1.51 [1.22, 1.79]1.25-0.88 [-1.34, -0.48]1.89 [1.55, 2.17]41% [29%, 54%]71%88%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 1 QC1.51 [1.15, 1.87]1.39-0.51 [-0.95, -0.10]1.89 [1.51, 2.21]43% [29%, 57%]64%89%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 1 QC1.51 [1.14, 1.87]1.39-0.51 [-1.10, -0.05]1.89 [1.47, 2.19]43% [28%, 56%]64%89%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 1 QC1.39 [1.03, 1.76]1.27-0.57 [-1.05, -0.15]1.79 [1.36, 2.12]44% [32%, 56%]68%91%——2026-01-01 → 2026-08-30
tmin00Znearestall34 1 QC1.39 [1.02, 1.76]1.27-0.57 [-1.04, -0.18]1.79 [1.36, 2.10]44% [30%, 57%]68%91%——2026-01-01 → 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 GraphCast (IFS init) at KPHL, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -4.65 °F2026-08-02: -5.21 °F2026-08-03: -7.76 °F2026-08-04: -7.61 °F2026-08-05: -5.87 °F2026-08-06: -4.43 °F2026-08-07: -5.41 °F2026-08-08: -3.36 °F2026-08-09: -4.25 °F2026-08-10: -1.18 °F2026-08-11: -4.53 °F2026-08-12: -6.09 °F2026-08-13: -3.81 °F2026-08-14: -4.84 °F2026-08-15: -5.98 °F2026-08-16: -2.05 °F2026-08-17: -5.03 °F2026-08-18: -3.7 °F2026-08-19: -5.51 °F2026-08-20: -4.12 °F2026-08-21: -6 °F2026-08-22: -3.34 °F2026-08-23: -3.79 °F2026-08-24: -5.61 °F2026-08-25: -2.67 °F2026-08-28: -6.38 °F2026-08-29: -5.13 °F2026-08-30: -4.84 °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
2026-08-30-4.84
2026-08-29-5.13
2026-08-28-6.38
2026-08-27—
2026-08-26—
2026-08-25-2.67
2026-08-24-5.61
2026-08-23-3.79
2026-08-22-3.34
2026-08-21-6.00
2026-08-20-4.12
2026-08-19-5.51
2026-08-18-3.70
2026-08-17-5.03
2026-08-16-2.05
2026-08-15-5.98
2026-08-14-4.84
2026-08-13-3.81
2026-08-12-6.09
2026-08-11-4.53
2026-08-10-1.18
2026-08-09-4.25
2026-08-08-3.36
2026-08-07-5.41
2026-08-06-4.43
2026-08-05-5.87
2026-08-04-7.61
2026-08-03-7.76
2026-08-02-5.21
2026-08-01-4.65

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KPHL, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 2 days-7 to -6 °F: 3 days-6 to -5 °F: 8 days-5 to -4 °F: 8 days-4 to -3 °F: 6 days-3 to -2 °F: 5 days-2 to -1 °F: 1 day-1 to 0 °F: 0 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| 4.48 · P90 |e| 6.07 °F-100+10 °Fn = 34 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
-8.00-7.002
-7.00-6.003
-6.00-5.008
-5.00-4.008
-4.00-3.006
-3.00-2.005
-2.00-1.001
-1.00+0.000
+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
2026-08284.76 -4.76
2026-0162.51 -2.31

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-01 → 2026-08-30 34
GraphCast (IFS init)2026-01-01 → 2026-08-30 34

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 28 +2.36 [1.98, 2.64] yes
Aurora (GFS init) 2 -0.02 [-1.15, 1.11] no
Aurora (IFS init) 3 +0.81 [0.69, 0.87] yes
FourCastNet v2 (GFS init) 2 +1.04 [-1.79, 3.87] no
FourCastNet v2 (IFS init) 3 +0.34 [-1.56, 1.28] no
NCEP GFS 28 +1.98 [1.05, 2.95] yes
GraphCast (GFS init) 2 -0.27 [-1.09, 0.54] no
ECMWF IFS HRES 22 +1.58 [1.08, 2.13] yes
Pangu-Weather (GFS init) 2 +2.89 [1.52, 4.27] yes
Pangu-Weather (IFS init) 3 +2.76 [1.90, 3.19] yes

Cite this page

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

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