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 (IFS init) · lead day 0

KPHX · 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 QC2.89 [2.08, 3.74]1.76-2.66 [-3.63, -1.76]3.50 [2.53, 4.36]14% [3%, 27%]39%68%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 1 QC2.89 [1.93, 4.05]1.76-2.66 [-3.98, -1.58]3.50 [2.28, 4.76]14% [0%, 27%]39%68%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 1 QC3.25 [2.27, 4.58]1.85-3.06 [-4.56, -1.97]3.85 [2.67, 5.07]12% [0%, 23%]32%56%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 1 QC3.25 [2.22, 4.42]1.85-3.06 [-4.34, -1.92]3.85 [2.72, 5.02]12% [0%, 23%]32%56%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 1 QC2.80 [1.99, 3.67]1.74-2.57 [-3.53, -1.69]3.42 [2.43, 4.29]14% [3%, 27%]39%68%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 1 QC2.80 [1.86, 3.97]1.74-2.57 [-3.88, -1.51]3.42 [2.21, 4.68]14% [0%, 27%]39%68%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 1 QC3.18 [2.17, 4.56]1.87-2.99 [-4.48, -1.86]3.80 [2.59, 5.06]12% [0%, 23%]32%56%——2026-01-01 → 2026-08-30
tmax00Znearestall34 1 QC3.18 [2.15, 4.36]1.87-2.99 [-4.28, -1.85]3.80 [2.63, 5.01]12% [0%, 23%]32%56%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 1 QC2.82 [2.03, 3.87]2.76-0.31 [-1.39, 0.87]3.55 [2.40, 4.80]18% [10%, 26%]43%57%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 1 QC2.82 [1.91, 3.86]2.76-0.31 [-1.49, 0.93]3.55 [2.26, 4.78]18% [10%, 28%]43%57%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 1 QC2.53 [1.57, 3.55]2.44-0.40 [-1.33, 0.62]3.28 [1.98, 4.43]24% [12%, 40%]50%65%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 1 QC2.53 [1.54, 3.52]2.44-0.40 [-1.41, 0.66]3.28 [1.90, 4.49]24% [13%, 38%]50%65%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 1 QC2.84 [2.03, 3.88]2.79-0.26 [-1.40, 0.98]3.59 [2.45, 4.85]18% [10%, 26%]46%54%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 1 QC2.84 [1.91, 3.88]2.79-0.26 [-1.51, 1.03]3.59 [2.28, 4.84]18% [10%, 28%]46%54%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 1 QC2.54 [1.58, 3.57]2.47-0.35 [-1.34, 0.71]3.31 [1.99, 4.49]24% [12%, 40%]53%62%——2026-01-01 → 2026-08-30
tmin00Znearestall34 1 QC2.54 [1.55, 3.52]2.47-0.35 [-1.45, 0.76]3.31 [1.92, 4.53]24% [13%, 38%]53%62%——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 KPHX, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -1.99 °F2026-08-02: -1.76 °F2026-08-03: -4.56 °F2026-08-04: +1.05 °F2026-08-05: +0.42 °F2026-08-06: -0.79 °F2026-08-07: -2.21 °F2026-08-08: -2.91 °F2026-08-09: -1.37 °F2026-08-10: -1.29 °F2026-08-11: +0.71 °F2026-08-12: -5.11 °F2026-08-13: -6.09 °F2026-08-14: -5.58 °F2026-08-15: -3.54 °F2026-08-16: -2.68 °F2026-08-17: -2.51 °F2026-08-18: -4.22 °F2026-08-19: -2.26 °F2026-08-20: +1.05 °F2026-08-21: -1.53 °F2026-08-22: -2.99 °F2026-08-23: -4.07 °F2026-08-24: -0 °F2026-08-25: -2.9 °F2026-08-28: -2.72 °F2026-08-29: -5.92 °F2026-08-30: -8.58 °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
2026-08-30-8.58
2026-08-29-5.92
2026-08-28-2.72
2026-08-27—
2026-08-26—
2026-08-25-2.90
2026-08-24-0.00
2026-08-23-4.07
2026-08-22-2.99
2026-08-21-1.53
2026-08-20+1.05
2026-08-19-2.26
2026-08-18-4.22
2026-08-17-2.51
2026-08-16-2.68
2026-08-15-3.54
2026-08-14-5.58
2026-08-13-6.09
2026-08-12-5.11
2026-08-11+0.71
2026-08-10-1.29
2026-08-09-1.37
2026-08-08-2.91
2026-08-07-2.21
2026-08-06-0.79
2026-08-05+0.42
2026-08-04+1.05
2026-08-03-4.56
2026-08-02-1.76
2026-08-01-1.99

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KPHX, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 1 day-8 to -7 °F: 1 day-7 to -6 °F: 2 days-6 to -5 °F: 3 days-5 to -4 °F: 5 days-4 to -3 °F: 3 days-3 to -2 °F: 8 days-2 to -1 °F: 5 days-1 to 0 °F: 2 days0 to 1 °F: 2 days1 to 2 °F: 2 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| 2.91 · P90 |e| 6.04 °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
-9.00-8.001
-8.00-7.001
-7.00-6.002
-6.00-5.003
-5.00-4.005
-4.00-3.003
-3.00-2.008
-2.00-1.005
-1.00+0.002
+0.00+1.002
+1.00+2.002
+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
2026-08282.89 -2.66
2026-0164.94 -4.94

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 -0.26 [-1.19, 0.39] no
Aurora (GFS init) 2 -6.19 [-8.26, -4.12] yes
Aurora (IFS init) 3 +0.03 [-0.45, 0.98] no
FourCastNet v2 (GFS init) 2 -3.32 [-4.86, -1.78] yes
FourCastNet v2 (IFS init) 3 -2.21 [-2.61, -1.73] yes
NCEP GFS 28 -0.02 [-0.73, 0.83] no
GraphCast (GFS init) 2 -7.64 [-8.58, -6.70] yes
ECMWF IFS HRES 23 +0.90 [0.14, 1.44] yes
Pangu-Weather (GFS init) 2 -1.51 [-2.78, -0.23] yes
Pangu-Weather (IFS init) 3 -0.36 [-1.58, 1.95] no

Cite this page

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

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