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 6

KATL · graphcast_ifs · lead 6

Target day = initialization date + 6 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
tmax00Zbilinear30d24 1 QC5.35 [3.76, 6.98]2.48-5.09 [-6.93, -3.26]6.16 [4.30, 7.74]0% [0%, 0%]17%25%-0.73+0.212026-08-07 → 2026-08-30
tmax00Zbilinear90d24 1 QC5.35 [3.56, 7.17]2.48-5.09 [-7.17, -3.02]6.16 [4.01, 7.98]0% [0%, 0%]17%25%-0.73+0.212026-08-07 → 2026-08-30
tmax00Zbilinear365d30 1 QC5.99 [4.20, 7.87]2.68-5.78 [-7.87, -3.74]6.77 [4.83, 8.44]0% [0%, 0%]13%20%-0.39+0.382026-01-07 → 2026-08-30
tmax00Zbilinearall30 1 QC5.99 [4.28, 8.06]2.68-5.78 [-8.06, -3.86]6.77 [4.92, 8.52]0% [0%, 0%]13%20%-0.39+0.382026-01-07 → 2026-08-30
tmax00Znearest30d24 1 QC5.31 [3.73, 6.88]2.45-5.05 [-6.86, -3.27]6.12 [4.27, 7.68]0% [0%, 0%]12%25%-0.72+0.222026-08-07 → 2026-08-30
tmax00Znearest90d24 1 QC5.31 [3.54, 7.06]2.45-5.05 [-7.05, -3.03]6.12 [3.96, 7.85]0% [0%, 0%]12%25%-0.72+0.222026-08-07 → 2026-08-30
tmax00Znearest365d30 1 QC5.99 [4.22, 7.94]2.71-5.78 [-7.94, -3.76]6.79 [4.85, 8.50]0% [0%, 0%]10%20%-0.39+0.372026-01-07 → 2026-08-30
tmax00Znearestall30 1 QC5.99 [4.29, 8.11]2.71-5.78 [-8.11, -3.86]6.79 [4.91, 8.60]0% [0%, 0%]10%20%-0.39+0.372026-01-07 → 2026-08-30
tmax12Zbilinear30d24 1 QC5.36 [4.08, 6.75]2.04-5.15 [-6.71, -3.62]5.90 [4.51, 7.28]4% [0%, 12%]4%12%-0.74+0.352026-08-07 → 2026-08-30
tmax12Zbilinear90d24 1 QC5.36 [3.91, 6.96]2.04-5.15 [-6.96, -3.42]5.90 [4.31, 7.58]4% [0%, 12%]4%12%-0.74+0.352026-08-07 → 2026-08-30
tmax12Zbilinear365d29 1 QC5.98 [4.41, 7.97]2.29-5.81 [-7.97, -3.95]6.56 [4.86, 8.33]3% [0%, 10%]3%10%-0.44+0.442026-01-07 → 2026-08-30
tmax12Zbilinearall29 1 QC5.98 [4.50, 8.10]2.29-5.81 [-8.10, -4.15]6.56 [4.97, 8.50]3% [0%, 10%]3%10%-0.44+0.442026-01-07 → 2026-08-30
tmax12Znearest30d24 1 QC5.35 [4.09, 6.67]2.01-5.14 [-6.65, -3.63]5.87 [4.51, 7.23]4% [0%, 12%]4%12%-0.73+0.362026-08-07 → 2026-08-30
tmax12Znearest90d24 1 QC5.35 [3.96, 6.89]2.01-5.14 [-6.89, -3.45]5.87 [4.33, 7.51]4% [0%, 12%]4%12%-0.73+0.362026-08-07 → 2026-08-30
tmax12Znearest365d29 1 QC6.00 [4.45, 7.99]2.33-5.83 [-7.99, -4.00]6.58 [4.90, 8.39]3% [0%, 10%]3%10%-0.45+0.432026-01-07 → 2026-08-30
tmax12Znearestall29 1 QC6.00 [4.50, 8.17]2.33-5.83 [-8.17, -4.18]6.58 [4.99, 8.59]3% [0%, 10%]3%10%-0.45+0.432026-01-07 → 2026-08-30
tmin00Zbilinear30d24 1 QC2.03 [1.29, 2.46]2.06+0.22 [-1.02, 1.29]2.42 [1.77, 2.80]33% [17%, 65%]50%75%+0.15+0.142026-08-07 → 2026-08-30
tmin00Zbilinear90d24 1 QC2.03 [1.04, 2.42]2.06+0.22 [-1.24, 1.37]2.42 [1.55, 2.75]33% [14%, 75%]50%75%+0.15+0.142026-08-07 → 2026-08-30
tmin00Zbilinear365d30 1 QC2.68 [1.69, 4.09]2.66-0.29 [-2.06, 1.20]3.45 [2.15, 4.94]30% [14%, 55%]43%63%+0.42+0.432026-01-07 → 2026-08-30
tmin00Zbilinearall30 1 QC2.68 [1.72, 4.17]2.66-0.29 [-2.19, 1.05]3.45 [2.18, 5.12]30% [14%, 53%]43%63%+0.42+0.432026-01-07 → 2026-08-30
tmin00Znearest30d24 1 QC2.07 [1.33, 2.48]2.08+0.12 [-1.08, 1.19]2.45 [1.80, 2.81]33% [17%, 65%]50%79%+0.13+0.132026-08-07 → 2026-08-30
tmin00Znearest90d24 1 QC2.07 [1.08, 2.44]2.08+0.12 [-1.30, 1.26]2.45 [1.58, 2.76]33% [14%, 75%]50%79%+0.13+0.132026-08-07 → 2026-08-30
tmin00Znearest365d30 1 QC2.73 [1.73, 4.16]2.69-0.42 [-2.25, 1.06]3.50 [2.18, 5.06]30% [14%, 55%]43%67%+0.41+0.422026-01-07 → 2026-08-30
tmin00Znearestall30 1 QC2.73 [1.77, 4.25]2.69-0.42 [-2.37, 0.94]3.50 [2.21, 5.20]30% [14%, 53%]43%67%+0.41+0.422026-01-07 → 2026-08-30
tmin12Zbilinear30d24 1 QC1.87 [1.17, 2.49]1.87-0.04 [-1.31, 1.09]2.53 [1.71, 3.03]42% [20%, 67%]50%88%+0.21+0.222026-08-07 → 2026-08-30
tmin12Zbilinear90d24 1 QC1.87 [0.97, 2.46]1.87-0.04 [-1.50, 1.10]2.53 [1.34, 2.98]42% [21%, 75%]50%88%+0.21+0.222026-08-07 → 2026-08-30
tmin12Zbilinear365d29 1 QC2.45 [1.48, 3.80]2.40-0.63 [-2.56, 0.93]3.31 [2.07, 4.63]34% [13%, 60%]45%76%+0.40+0.432026-01-07 → 2026-08-30
tmin12Zbilinearall29 1 QC2.45 [1.44, 4.07]2.40-0.63 [-2.79, 0.82]3.31 [2.16, 4.85]34% [10%, 60%]45%76%+0.40+0.432026-01-07 → 2026-08-30
tmin12Znearest30d24 1 QC1.90 [1.17, 2.48]1.88-0.15 [-1.45, 1.00]2.58 [1.75, 3.07]38% [17%, 62%]50%88%+0.20+0.212026-08-07 → 2026-08-30
tmin12Znearest90d24 1 QC1.90 [0.96, 2.48]1.88-0.15 [-1.61, 1.01]2.58 [1.36, 3.02]38% [17%, 67%]50%88%+0.20+0.212026-08-07 → 2026-08-30
tmin12Znearest365d29 1 QC2.51 [1.48, 3.93]2.44-0.77 [-2.73, 0.82]3.40 [2.16, 4.80]31% [11%, 53%]41%76%+0.39+0.422026-01-07 → 2026-08-30
tmin12Znearestall29 1 QC2.51 [1.47, 4.19]2.44-0.77 [-2.97, 0.72]3.40 [2.22, 5.01]31% [9%, 56%]41%76%+0.39+0.422026-01-07 → 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 6, 00Z bilinear tmax+20+100-10-202026-08-07: -4.73 °F2026-08-08: -9.3 °F2026-08-09: -5.32 °F2026-08-10: -6.72 °F2026-08-11: -4.7 °F2026-08-12: -6.95 °F2026-08-13: -8.95 °F2026-08-14: -6.56 °F2026-08-15: -5.11 °F2026-08-16: -2.38 °F2026-08-17: -4.4 °F2026-08-18: +1.99 °F2026-08-19: -1.97 °F2026-08-20: -3.3 °F2026-08-21: -2.04 °F2026-08-22: -4.59 °F2026-08-23: +1.13 °F2026-08-24: -4.73 °F2026-08-25: -5.3 °F2026-08-26: -1.56 °F2026-08-27: -13.86 °F2026-08-28: -11.11 °F2026-08-29: -6.49 °F2026-08-30: -5.21 °F2026-08-072026-08-30°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 (24 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-5.21
2026-08-29-6.49
2026-08-28-11.11
2026-08-27-13.86
2026-08-26-1.56
2026-08-25-5.30
2026-08-24-4.73
2026-08-23+1.13
2026-08-22-4.59
2026-08-21-2.04
2026-08-20-3.30
2026-08-19-1.97
2026-08-18+1.99
2026-08-17-4.40
2026-08-16-2.38
2026-08-15-5.11
2026-08-14-6.56
2026-08-13-8.95
2026-08-12-6.95
2026-08-11-4.70
2026-08-10-6.72
2026-08-09-5.32
2026-08-08-9.30
2026-08-07-4.73

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KATL, lead day 6-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 1 day-12 to -10 °F: 3 days-10 to -8 °F: 3 days-8 to -6 °F: 6 days-6 to -4 °F: 10 days-4 to -2 °F: 3 days-2 to 0 °F: 2 days0 to 2 °F: 2 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| 5.31 · P90 |e| 10.81 °F-200+20 °Fn = 30 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
-14.00-12.001
-12.00-10.003
-10.00-8.003
-8.00-6.006
-6.00-4.0010
-4.00-2.003
-2.00+0.002
+0.00+2.002
+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
2026-08245.35 -5.09
2026-0168.53 -8.53

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-07 → 2026-08-30 30
GraphCast (IFS init)2026-01-07 → 2026-08-30 30
GraphCast (IFS init)2026-01-07 → 2026-08-30 29
GraphCast (IFS init)2026-01-07 → 2026-08-30 29

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 24 +0.46 [-0.52, 2.08] no
NCEP GFS 24 +0.44 [-1.24, 2.70] no
ECMWF IFS HRES 19 +1.90 [0.26, 4.07] yes
Persistence (baseline) 24 +2.27 [1.12, 4.20] yes

Cite this page

ShortCastCheck, KATL graphcast_ifs lead 6, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KATL/model/graphcast_ifs/lead/6/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KATL (Atlanta Hartsfield), model graphcast_ifs, lead day 6. 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/6/

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