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

Phoenix Sky Harbor · GraphCast (GFS init) · lead day 0

KPHX · graphcast_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, GRAP_v100, starting 2026-08-28 — 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
tmax00Zbilinear30d2 1 QC14.89 [14.50, 15.27]0.39-14.89 [-15.27, -14.50]14.89 [14.50, 15.27]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Zbilinear90d2 1 QC14.89 [14.50, 15.27]0.39-14.89 [-15.27, -14.50]14.89 [14.50, 15.27]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Zbilinear365d2 1 QC14.89 [14.50, 15.27]0.39-14.89 [-15.27, -14.50]14.89 [14.50, 15.27]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Zbilinearall2 1 QC14.89 [14.50, 15.27]0.39-14.89 [-15.27, -14.50]14.89 [14.50, 15.27]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Znearest30d2 1 QC14.99 [14.51, 15.48]0.49-14.99 [-15.48, -14.51]15.00 [14.51, 15.48]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Znearest90d2 1 QC14.99 [14.51, 15.48]0.49-14.99 [-15.48, -14.51]15.00 [14.51, 15.48]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Znearest365d2 1 QC14.99 [14.51, 15.48]0.49-14.99 [-15.48, -14.51]15.00 [14.51, 15.48]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmax00Znearestall2 1 QC14.99 [14.51, 15.48]0.49-14.99 [-15.48, -14.51]15.00 [14.51, 15.48]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Zbilinear30d2 1 QC6.39 [4.32, 8.45]2.07-6.39 [-8.45, -4.32]6.71 [4.32, 8.45]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Zbilinear90d2 1 QC6.39 [4.32, 8.45]2.07-6.39 [-8.45, -4.32]6.71 [4.32, 8.45]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Zbilinear365d2 1 QC6.39 [4.32, 8.45]2.07-6.39 [-8.45, -4.32]6.71 [4.32, 8.45]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Zbilinearall2 1 QC6.39 [4.32, 8.45]2.07-6.39 [-8.45, -4.32]6.71 [4.32, 8.45]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Znearest30d2 1 QC6.47 [4.35, 8.59]2.12-6.47 [-8.59, -4.35]6.81 [4.35, 8.59]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Znearest90d2 1 QC6.47 [4.35, 8.59]2.12-6.47 [-8.59, -4.35]6.81 [4.35, 8.59]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Znearest365d2 1 QC6.47 [4.35, 8.59]2.12-6.47 [-8.59, -4.35]6.81 [4.35, 8.59]0% [0%, 0%]0%0%——2026-08-29 → 2026-08-30
tmin00Znearestall2 1 QC6.47 [4.35, 8.59]2.12-6.47 [-8.59, -4.35]6.81 [4.35, 8.59]0% [0%, 0%]0%0%——2026-08-29 → 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 (GFS init) at KPHX, lead day 0, 00Z bilinear tmax+20+100-10-202026-08-29: -14.5 °F2026-08-30: -15.27 °F2026-08-292026-08-30°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 (2 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-15.27
2026-08-29-14.50

Error distribution

Distribution of the signed daily error in °F for GraphCast (GFS init) at KPHX, lead day 0-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 2 days-14 to -12 °F: 0 days-12 to -10 °F: 0 days-10 to -8 °F: 0 days-8 to -6 °F: 0 days-6 to -4 °F: 0 days-4 to -2 °F: 0 days-2 to 0 °F: 0 days0 to 2 °F: 0 days2 to 4 °F: 0 days4 to 6 °F: 0 days6 to 8 °F: 0 days8 to 10 °F: 0 days10 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| 14.89 · P90 |e| 15.2 °F-200+20 °Fn = 2 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
-16.00-14.002
-14.00-12.000
-12.00-10.000
-10.00-8.000
-8.00-6.000
-6.00-4.000
-4.00-2.000
-2.00+0.000
+0.00+2.000
+2.00+4.000
+4.00+6.000
+6.00+8.000
+8.00+10.000
+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
2026-08214.89 -14.89

Availability

ModelScored period DaysCoverage
GraphCast (GFS init)2026-08-29 → 2026-08-30 2
GraphCast (GFS init)2026-08-29 → 2026-08-30 2

Compared with other models

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

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 2 +5.13 [3.00, 7.26] yes
Aurora (GFS init) 2 +1.45 [0.32, 2.58] yes
Aurora (IFS init) 2 +7.90 [7.67, 8.13] yes
FourCastNet v2 (GFS init) 2 +4.32 [3.72, 4.92] yes
FourCastNet v2 (IFS init) 2 +5.64 [4.97, 6.30] yes
NCEP GFS 2 +7.13 [4.48, 9.78] yes
GraphCast (IFS init) 2 +7.64 [6.70, 8.58] yes
ECMWF IFS HRES 2 +6.92 [3.41, 10.42] yes
Pangu-Weather (GFS init) 2 +6.13 [5.80, 6.46] yes
Pangu-Weather (IFS init) 2 +7.89 [7.13, 8.64] yes

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KPHX (Phoenix Sky Harbor), model graphcast_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/KPHX/model/graphcast_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