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

Los Angeles Intl · GraphCast (IFS init) · lead day 0

KLAX · 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, 2 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 2 QC3.31 [2.91, 3.75]1.05-3.31 [-3.75, -2.91]3.57 [3.06, 4.04]4% [0%, 11%]14%50%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 2 QC3.31 [2.82, 3.97]1.05-3.31 [-3.97, -2.82]3.57 [2.92, 4.25]4% [0%, 11%]14%50%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 2 QC3.21 [2.79, 3.81]1.01-3.21 [-3.81, -2.79]3.46 [2.93, 4.08]3% [0%, 9%]18%50%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 2 QC3.21 [2.78, 3.78]1.01-3.21 [-3.78, -2.78]3.46 [2.92, 4.07]3% [0%, 10%]18%50%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 2 QC6.65 [5.98, 7.42]1.30-6.65 [-7.42, -5.98]6.83 [6.06, 7.61]0% [0%, 0%]0%0%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 2 QC6.65 [5.89, 7.72]1.30-6.65 [-7.72, -5.89]6.83 [5.94, 7.87]0% [0%, 0%]0%0%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 2 QC6.06 [4.76, 7.25]1.52-6.06 [-7.25, -4.76]6.37 [5.13, 7.47]0% [0%, 0%]6%6%——2026-01-01 → 2026-08-30
tmax00Znearestall34 2 QC6.06 [4.82, 7.29]1.52-6.06 [-7.29, -4.82]6.37 [5.19, 7.47]0% [0%, 0%]6%6%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 2 QC0.70 [0.59, 0.84]0.72-0.20 [-0.55, 0.22]0.83 [0.71, 0.96]64% [50%, 78%]100%100%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 2 QC0.70 [0.53, 0.84]0.72-0.20 [-0.53, 0.29]0.83 [0.68, 0.96]64% [50%, 82%]100%100%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 2 QC0.90 [0.62, 1.35]0.88+0.16 [-0.41, 1.03]1.19 [0.75, 1.77]59% [43%, 74%]94%94%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 2 QC0.90 [0.61, 1.30]0.88+0.16 [-0.42, 1.01]1.19 [0.74, 1.72]59% [43%, 75%]94%94%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 2 QC0.75 [0.63, 0.91]0.74+0.22 [-0.17, 0.67]0.94 [0.75, 1.14]79% [64%, 90%]93%100%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 2 QC0.75 [0.58, 0.97]0.74+0.22 [-0.16, 0.77]0.94 [0.70, 1.20]79% [65%, 91%]93%100%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 2 QC1.01 [0.64, 1.60]0.96+0.56 [-0.05, 1.41]1.39 [0.80, 2.12]68% [42%, 85%]88%94%——2026-01-01 → 2026-08-30
tmin00Znearestall34 2 QC1.01 [0.64, 1.60]0.96+0.56 [-0.03, 1.44]1.39 [0.79, 2.07]68% [41%, 86%]88%94%——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 KLAX, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -2.87 °F2026-08-02: -2.52 °F2026-08-03: -3.07 °F2026-08-04: -4.12 °F2026-08-05: -2.98 °F2026-08-06: -1.68 °F2026-08-07: -2.51 °F2026-08-08: -1.72 °F2026-08-09: -3.62 °F2026-08-10: -2.16 °F2026-08-11: -3.35 °F2026-08-12: -2.99 °F2026-08-13: -3.42 °F2026-08-14: -4.37 °F2026-08-15: -2.98 °F2026-08-16: -2.42 °F2026-08-17: -2.11 °F2026-08-18: -6.28 °F2026-08-19: -5.23 °F2026-08-20: -0.67 °F2026-08-21: -4.12 °F2026-08-22: -1.97 °F2026-08-23: -4.87 °F2026-08-24: -2.51 °F2026-08-25: -3.87 °F2026-08-28: -3.08 °F2026-08-29: -6.41 °F2026-08-30: -4.8 °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
2026-08-30-4.80
2026-08-29-6.41
2026-08-28-3.08
2026-08-27—
2026-08-26—
2026-08-25-3.87
2026-08-24-2.51
2026-08-23-4.87
2026-08-22-1.97
2026-08-21-4.12
2026-08-20-0.67
2026-08-19-5.23
2026-08-18-6.28
2026-08-17-2.11
2026-08-16-2.42
2026-08-15-2.98
2026-08-14-4.37
2026-08-13-3.42
2026-08-12-2.99
2026-08-11-3.35
2026-08-10-2.16
2026-08-09-3.62
2026-08-08-1.72
2026-08-07-2.51
2026-08-06-1.68
2026-08-05-2.98
2026-08-04-4.12
2026-08-03-3.07
2026-08-02-2.52
2026-08-01-2.87

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KLAX, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 0 days-7 to -6 °F: 2 days-6 to -5 °F: 1 day-5 to -4 °F: 6 days-4 to -3 °F: 8 days-3 to -2 °F: 11 days-2 to -1 °F: 5 days-1 to 0 °F: 1 day0 to 1 °F: 0 days1 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| 3.01 · P90 |e| 4.85 °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
-8.00-7.000
-7.00-6.002
-6.00-5.001
-5.00-4.006
-4.00-3.008
-3.00-2.0011
-2.00-1.005
-1.00+0.001
+0.00+1.000
+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
2026-08283.31 -3.31
2026-0162.73 -2.73

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 +1.88 [1.34, 2.48] yes
Aurora (GFS init) 2 -0.30 [-1.13, 0.54] no
Aurora (IFS init) 3 +0.14 [-0.38, 0.39] no
FourCastNet v2 (GFS init) 2 +2.28 [0.91, 3.66] yes
FourCastNet v2 (IFS init) 3 +2.04 [0.93, 2.59] yes
NCEP GFS 28 +1.45 [0.86, 2.29] yes
GraphCast (GFS init) 2 +0.04 [-0.40, 0.47] no
ECMWF IFS HRES 23 +1.14 [0.78, 1.72] yes
Pangu-Weather (GFS init) 2 +3.14 [2.01, 4.26] yes
Pangu-Weather (IFS init) 3 +2.63 [1.75, 3.08] yes

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KLAX (Los Angeles 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/KLAX/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