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

Atlanta Hartsfield · ECMWF IFS HRES · lead day 0

KATL · ifs_hres · 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, ifs-gpid161, 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
tmax00Zbilinear30d24 1 QC2.64 [1.93, 3.24]1.55-2.38 [-2.99, -1.58]3.14 [2.26, 3.70]12% [0%, 30%]46%62%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d24 1 QC2.64 [1.98, 3.39]1.55-2.38 [-3.24, -1.18]3.14 [2.29, 3.77]12% [0%, 29%]46%62%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d24 1 QC2.64 [1.91, 3.36]1.55-2.38 [-3.23, -1.09]3.14 [2.23, 3.77]12% [0%, 33%]46%62%——2026-08-01 → 2026-08-30
tmax00Zbilinearall24 1 QC2.64 [1.88, 3.29]1.55-2.38 [-3.12, -0.92]3.14 [2.14, 3.74]12% [0%, 31%]46%62%——2026-08-01 → 2026-08-30
tmax00Znearest30d24 1 QC2.47 [1.81, 2.98]1.50-2.20 [-2.74, -1.41]2.99 [2.18, 3.55]21% [4%, 41%]46%75%——2026-08-01 → 2026-08-30
tmax00Znearest90d24 1 QC2.47 [1.83, 3.05]1.50-2.20 [-2.92, -1.02]2.99 [2.21, 3.57]21% [0%, 41%]46%75%——2026-08-01 → 2026-08-30
tmax00Znearest365d24 1 QC2.47 [1.78, 3.05]1.50-2.20 [-2.90, -0.95]2.99 [2.15, 3.58]21% [0%, 44%]46%75%——2026-08-01 → 2026-08-30
tmax00Znearestall24 1 QC2.47 [1.73, 2.99]1.50-2.20 [-2.88, -0.73]2.99 [2.06, 3.57]21% [0%, 42%]46%75%——2026-08-01 → 2026-08-30
tmin00Zbilinear30d24 1 QC1.28 [0.90, 1.65]1.24-0.36 [-1.25, 0.37]1.65 [1.23, 1.98]58% [41%, 77%]75%88%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d24 1 QC1.28 [0.82, 1.73]1.24-0.36 [-1.47, 0.43]1.65 [1.09, 2.04]58% [37%, 82%]75%88%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d24 1 QC1.28 [0.83, 1.83]1.24-0.36 [-1.54, 0.50]1.65 [1.07, 2.11]58% [36%, 86%]75%88%——2026-08-01 → 2026-08-30
tmin00Zbilinearall24 1 QC1.28 [0.82, 1.83]1.24-0.36 [-1.56, 0.39]1.65 [1.06, 2.09]58% [36%, 88%]75%88%——2026-08-01 → 2026-08-30
tmin00Znearest30d24 1 QC1.24 [0.89, 1.58]1.20-0.35 [-1.15, 0.37]1.59 [1.19, 1.90]54% [39%, 71%]75%88%——2026-08-01 → 2026-08-30
tmin00Znearest90d24 1 QC1.24 [0.79, 1.63]1.20-0.35 [-1.35, 0.39]1.59 [1.01, 1.94]54% [36%, 80%]75%88%——2026-08-01 → 2026-08-30
tmin00Znearest365d24 1 QC1.24 [0.80, 1.73]1.20-0.35 [-1.42, 0.45]1.59 [1.03, 1.98]54% [33%, 79%]75%88%——2026-08-01 → 2026-08-30
tmin00Znearestall24 1 QC1.24 [0.77, 1.70]1.20-0.35 [-1.40, 0.37]1.59 [0.99, 1.98]54% [35%, 79%]75%88%——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 ECMWF IFS HRES at KATL, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: +2.6 °F2026-08-08: -3.23 °F2026-08-09: -2.53 °F2026-08-10: -3.27 °F2026-08-11: -1.34 °F2026-08-12: -3.23 °F2026-08-13: -1.88 °F2026-08-14: -0.81 °F2026-08-15: -2.29 °F2026-08-16: -1.19 °F2026-08-17: -0.54 °F2026-08-18: +0.56 °F2026-08-19: -1.5 °F2026-08-20: -7.03 °F2026-08-21: -1.58 °F2026-08-22: -5.63 °F2026-08-23: -3.04 °F2026-08-24: -1.42 °F2026-08-25: -1.24 °F2026-08-26: -1.69 °F2026-08-27: -5.7 °F2026-08-28: -5 °F2026-08-29: -2.95 °F2026-08-30: -3.2 °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
2026-08-30-3.20
2026-08-29-2.95
2026-08-28-5.00
2026-08-27-5.70
2026-08-26-1.69
2026-08-25-1.24
2026-08-24-1.42
2026-08-23-3.04
2026-08-22-5.63
2026-08-21-1.58
2026-08-20-7.03
2026-08-19-1.50
2026-08-18+0.56
2026-08-17-0.54
2026-08-16-1.19
2026-08-15-2.29
2026-08-14-0.81
2026-08-13-1.88
2026-08-12-3.23
2026-08-11-1.34
2026-08-10-3.27
2026-08-09-2.53
2026-08-08-3.23
2026-08-07—
2026-08-06—
2026-08-05—
2026-08-04—
2026-08-03—
2026-08-02—
2026-08-01+2.60

Error distribution

Distribution of the signed daily error in °F for ECMWF IFS HRES at KATL, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 1 day-7 to -6 °F: 0 days-6 to -5 °F: 3 days-5 to -4 °F: 0 days-4 to -3 °F: 5 days-3 to -2 °F: 3 days-2 to -1 °F: 8 days-1 to 0 °F: 2 days0 to 1 °F: 1 day1 to 2 °F: 0 days2 to 3 °F: 1 day3 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| 2.41 · P90 |e| 5.44 °F-100+10 °Fn = 24 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
-8.00-7.001
-7.00-6.000
-6.00-5.003
-5.00-4.000
-4.00-3.005
-3.00-2.003
-2.00-1.008
-1.00+0.002
+0.00+1.001
+1.00+2.000
+2.00+3.001
+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
2026-08242.64 -2.38

Availability

ModelScored period DaysCoverage
ECMWF IFS HRES2026-08-01 → 2026-08-30 24
ECMWF IFS HRES2026-08-01 → 2026-08-30 24

Compared with other models

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

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 24 -1.50 [-1.81, -0.97] yes
Aurora (GFS init) 2 -3.51 [-3.75, -3.28] yes
Aurora (IFS init) 3 -2.97 [-3.15, -2.71] yes
FourCastNet v2 (GFS init) 2 -0.60 [-1.01, -0.19] yes
FourCastNet v2 (IFS init) 3 -0.72 [-2.24, 0.50] no
NCEP GFS 24 +1.60 [0.95, 2.29] yes
GraphCast (GFS init) 2 -1.96 [-2.27, -1.65] yes
GraphCast (IFS init) 22 -3.13 [-4.20, -2.07] yes
Pangu-Weather (GFS init) 2 +0.58 [0.50, 0.65] yes
Pangu-Weather (IFS init) 3 +0.44 [-0.13, 0.85] no

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KATL (Atlanta Hartsfield), model ifs_hres, 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/KATL/model/ifs_hres/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