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 · GraphCast (IFS init) · lead day 2

KATL · graphcast_ifs · lead 2

Target day = initialization date + 2 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
tmax00Zbilinear30d26 1 QC5.81 [5.02, 6.53]1.33-5.81 [-6.53, -5.02]6.12 [5.31, 6.85]4% [0%, 11%]4%4%-1.32+0.472026-08-03 → 2026-08-30
tmax00Zbilinear90d26 1 QC5.81 [5.00, 6.67]1.33-5.81 [-6.67, -5.00]6.12 [5.24, 7.01]4% [0%, 11%]4%4%-1.32+0.472026-08-03 → 2026-08-30
tmax00Zbilinear365d32 1 QC5.86 [5.06, 6.57]1.60-5.86 [-6.57, -5.06]6.29 [5.42, 7.08]3% [0%, 9%]6%6%-0.95+0.462026-01-03 → 2026-08-30
tmax00Zbilinearall32 1 QC5.86 [5.07, 6.58]1.60-5.86 [-6.58, -5.07]6.29 [5.39, 7.06]3% [0%, 10%]6%6%-0.95+0.462026-01-03 → 2026-08-30
tmax00Znearest30d26 1 QC5.83 [4.99, 6.58]1.39-5.83 [-6.58, -4.99]6.15 [5.31, 6.88]0% [0%, 0%]4%4%-1.33+0.452026-08-03 → 2026-08-30
tmax00Znearest90d26 1 QC5.83 [4.96, 6.71]1.39-5.83 [-6.71, -4.96]6.15 [5.20, 7.04]0% [0%, 0%]4%4%-1.33+0.452026-08-03 → 2026-08-30
tmax00Znearest365d32 1 QC5.92 [5.05, 6.66]1.67-5.92 [-6.66, -5.05]6.36 [5.43, 7.19]0% [0%, 0%]6%6%-0.97+0.442026-01-03 → 2026-08-30
tmax00Znearestall32 1 QC5.92 [5.10, 6.65]1.67-5.92 [-6.65, -5.10]6.36 [5.46, 7.17]0% [0%, 0%]6%6%-0.97+0.442026-01-03 → 2026-08-30
tmax12Zbilinear30d26 1 QC6.40 [5.75, 7.07]1.24-6.40 [-7.07, -5.75]6.62 [5.93, 7.29]0% [0%, 0%]0%0%-1.56+0.502026-08-03 → 2026-08-30
tmax12Zbilinear90d26 1 QC6.40 [5.69, 7.26]1.24-6.40 [-7.26, -5.69]6.62 [5.88, 7.45]0% [0%, 0%]0%0%-1.56+0.502026-08-03 → 2026-08-30
tmax12Zbilinear365d31 1 QC6.44 [5.68, 7.18]1.69-6.44 [-7.18, -5.68]6.85 [5.97, 7.82]0% [0%, 0%]3%3%-1.27+0.402026-01-03 → 2026-08-30
tmax12Zbilinearall31 1 QC6.44 [5.69, 7.17]1.69-6.44 [-7.17, -5.69]6.85 [5.96, 7.67]0% [0%, 0%]3%3%-1.27+0.402026-01-03 → 2026-08-30
tmax12Znearest30d26 1 QC6.37 [5.73, 7.05]1.22-6.37 [-7.05, -5.73]6.58 [5.90, 7.24]0% [0%, 0%]0%0%-1.55+0.512026-08-03 → 2026-08-30
tmax12Znearest90d26 1 QC6.37 [5.70, 7.18]1.22-6.37 [-7.18, -5.70]6.58 [5.86, 7.38]0% [0%, 0%]0%0%-1.55+0.512026-08-03 → 2026-08-30
tmax12Znearest365d31 1 QC6.45 [5.72, 7.22]1.67-6.45 [-7.22, -5.72]6.86 [5.95, 7.88]0% [0%, 0%]3%3%-1.27+0.412026-01-03 → 2026-08-30
tmax12Znearestall31 1 QC6.45 [5.72, 7.17]1.67-6.45 [-7.17, -5.72]6.86 [5.92, 7.71]0% [0%, 0%]3%3%-1.27+0.412026-01-03 → 2026-08-30
tmin00Zbilinear30d26 1 QC1.92 [1.08, 2.63]1.88-0.57 [-1.83, 0.34]2.50 [1.47, 3.26]31% [7%, 60%]58%81%+0.11+0.142026-08-03 → 2026-08-30
tmin00Zbilinear90d26 1 QC1.92 [0.89, 2.64]1.88-0.57 [-1.87, 0.31]2.50 [1.09, 3.25]31% [4%, 67%]58%81%+0.11+0.142026-08-03 → 2026-08-30
tmin00Zbilinear365d32 1 QC1.93 [1.23, 2.60]1.90-0.59 [-1.68, 0.26]2.46 [1.63, 3.16]31% [8%, 57%]53%84%+0.34+0.372026-01-03 → 2026-08-30
tmin00Zbilinearall32 1 QC1.93 [1.24, 2.57]1.90-0.59 [-1.74, 0.23]2.46 [1.62, 3.15]31% [8%, 57%]53%84%+0.34+0.372026-01-03 → 2026-08-30
tmin00Znearest30d26 1 QC1.98 [1.10, 2.76]1.90-0.67 [-1.99, 0.29]2.60 [1.49, 3.46]35% [10%, 67%]58%81%+0.08+0.132026-08-03 → 2026-08-30
tmin00Znearest90d26 1 QC1.98 [0.88, 2.74]1.90-0.67 [-2.02, 0.26]2.60 [1.08, 3.44]35% [12%, 75%]58%81%+0.08+0.132026-08-03 → 2026-08-30
tmin00Znearest365d32 1 QC2.00 [1.26, 2.74]1.93-0.72 [-1.90, 0.17]2.56 [1.68, 3.34]34% [12%, 61%]53%78%+0.32+0.362026-01-03 → 2026-08-30
tmin00Znearestall32 1 QC2.00 [1.29, 2.71]1.93-0.72 [-1.93, 0.17]2.56 [1.71, 3.34]34% [14%, 61%]53%78%+0.32+0.362026-01-03 → 2026-08-30
tmin12Zbilinear30d26 1 QC1.72 [1.00, 2.36]1.56-0.78 [-1.91, -0.02]2.29 [1.29, 3.15]38% [20%, 63%]69%88%+0.20+0.292026-08-03 → 2026-08-30
tmin12Zbilinear90d26 1 QC1.72 [0.82, 2.37]1.56-0.78 [-1.95, -0.02]2.29 [1.03, 3.11]38% [19%, 69%]69%88%+0.20+0.292026-08-03 → 2026-08-30
tmin12Zbilinear365d31 1 QC1.68 [1.05, 2.30]1.57-0.70 [-1.71, 0.01]2.24 [1.43, 2.99]42% [22%, 65%]68%87%+0.41+0.462026-01-03 → 2026-08-30
tmin12Zbilinearall31 1 QC1.68 [0.99, 2.27]1.57-0.70 [-1.74, -0.01]2.24 [1.35, 3.03]42% [22%, 69%]68%87%+0.41+0.462026-01-03 → 2026-08-30
tmin12Znearest30d26 1 QC1.82 [1.06, 2.50]1.61-0.89 [-2.09, -0.08]2.41 [1.34, 3.34]38% [20%, 63%]65%85%+0.16+0.272026-08-03 → 2026-08-30
tmin12Znearest90d26 1 QC1.82 [0.88, 2.50]1.61-0.89 [-2.13, -0.08]2.41 [1.09, 3.31]38% [19%, 69%]65%85%+0.16+0.272026-08-03 → 2026-08-30
tmin12Znearest365d31 1 QC1.77 [1.11, 2.46]1.62-0.83 [-1.89, -0.08]2.37 [1.51, 3.20]42% [22%, 65%]68%84%+0.38+0.452026-01-03 → 2026-08-30
tmin12Znearestall31 1 QC1.77 [1.05, 2.41]1.62-0.83 [-1.95, -0.10]2.37 [1.42, 3.23]42% [22%, 69%]68%84%+0.38+0.452026-01-03 → 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 KATL, lead day 2, 00Z bilinear tmax+20+100-10-202026-08-03: -5.46 °F2026-08-04: -3.69 °F2026-08-05: -8.2 °F2026-08-06: -5.46 °F2026-08-07: -4.41 °F2026-08-08: -7.84 °F2026-08-09: -5.84 °F2026-08-10: -10.1 °F2026-08-11: -5.7 °F2026-08-12: -5.98 °F2026-08-13: -7.42 °F2026-08-14: -5.86 °F2026-08-15: -5.43 °F2026-08-16: -4.71 °F2026-08-17: -5.41 °F2026-08-18: -0.84 °F2026-08-19: -4.41 °F2026-08-20: -6.19 °F2026-08-21: -4.56 °F2026-08-22: -5.23 °F2026-08-23: -5.87 °F2026-08-24: -5.45 °F2026-08-25: -6.64 °F2026-08-26: -3.27 °F2026-08-27: -10.57 °F2026-08-30: -6.44 °F2026-08-032026-08-30°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 (28 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-6.44
2026-08-29—
2026-08-28—
2026-08-27-10.57
2026-08-26-3.27
2026-08-25-6.64
2026-08-24-5.45
2026-08-23-5.87
2026-08-22-5.23
2026-08-21-4.56
2026-08-20-6.19
2026-08-19-4.41
2026-08-18-0.84
2026-08-17-5.41
2026-08-16-4.71
2026-08-15-5.43
2026-08-14-5.86
2026-08-13-7.42
2026-08-12-5.98
2026-08-11-5.70
2026-08-10-10.10
2026-08-09-5.84
2026-08-08-7.84
2026-08-07-4.41
2026-08-06-5.46
2026-08-05-8.20
2026-08-04-3.69
2026-08-03-5.46

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KATL, lead day 2-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 0 days-12 to -10 °F: 3 days-10 to -8 °F: 2 days-8 to -6 °F: 5 days-6 to -4 °F: 17 days-4 to -2 °F: 3 days-2 to 0 °F: 2 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| 5.7 · P90 |e| 9.34 °F-200+20 °Fn = 32 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
-16.00-14.000
-14.00-12.000
-12.00-10.003
-10.00-8.002
-8.00-6.005
-6.00-4.0017
-4.00-2.003
-2.00+0.002
+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
2026-08265.81 -5.81
2026-0166.08 -6.08

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-03 → 2026-08-30 32
GraphCast (IFS init)2026-01-03 → 2026-08-30 32
GraphCast (IFS init)2026-01-03 → 2026-08-30 31
GraphCast (IFS init)2026-01-03 → 2026-08-30 31

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 26 +1.45 [0.47, 2.18] yes
Aurora (IFS init) 1 -1.07 [-1.07, -1.07] yes
FourCastNet v2 (IFS init) 1 +1.81 [1.81, 1.81] yes
NCEP GFS 26 +2.70 [1.07, 4.04] yes
ECMWF IFS HRES 20 +3.11 [2.06, 4.22] yes
Pangu-Weather (IFS init) 1 +1.73 [1.73, 1.73] yes
Persistence (baseline) 26 +3.31 [2.26, 4.41] yes

Cite this page

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

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KATL (Atlanta Hartsfield), model graphcast_ifs, lead day 2. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KATL/model/graphcast_ifs/lead/2/

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