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 8

KATL · graphcast_ifs · lead 8

Target day = initialization date + 8 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
tmax00Zbilinear30d22 1 QC5.65 [3.62, 7.84]2.86-5.36 [-7.84, -2.97]6.52 [4.57, 8.32]18% [0%, 33%]18%23%-0.35+0.322026-08-09 → 2026-08-30
tmax00Zbilinear90d22 1 QC5.65 [3.45, 8.12]2.86-5.36 [-8.12, -2.68]6.52 [4.28, 8.62]18% [0%, 35%]18%23%-0.35+0.322026-08-09 → 2026-08-30
tmax00Zbilinear365d28 1 QC6.29 [4.30, 8.50]3.27-5.66 [-7.78, -3.33]7.10 [5.20, 8.87]14% [0%, 29%]14%18%-0.30+0.332026-01-09 → 2026-08-30
tmax00Zbilinearall28 1 QC6.29 [4.22, 8.58]3.27-5.66 [-7.94, -3.31]7.10 [5.20, 8.90]14% [0%, 30%]14%18%-0.30+0.332026-01-09 → 2026-08-30
tmax00Znearest30d22 1 QC5.69 [3.69, 7.84]2.84-5.39 [-7.84, -2.98]6.54 [4.60, 8.32]18% [0%, 33%]18%23%-0.36+0.322026-08-09 → 2026-08-30
tmax00Znearest90d22 1 QC5.69 [3.54, 8.13]2.84-5.39 [-8.13, -2.72]6.54 [4.38, 8.64]18% [0%, 35%]18%23%-0.36+0.322026-08-09 → 2026-08-30
tmax00Znearest365d28 1 QC6.35 [4.37, 8.56]3.28-5.72 [-7.82, -3.38]7.16 [5.29, 8.96]14% [0%, 29%]14%18%-0.32+0.322026-01-09 → 2026-08-30
tmax00Znearestall28 1 QC6.35 [4.28, 8.65]3.28-5.72 [-7.97, -3.35]7.16 [5.28, 9.03]14% [0%, 30%]14%18%-0.32+0.322026-01-09 → 2026-08-30
tmax12Zbilinear30d22 1 QC5.05 [3.35, 7.08]2.68-4.76 [-7.08, -2.70]5.84 [4.16, 7.57]9% [0%, 24%]18%23%-0.21+0.362026-08-09 → 2026-08-30
tmax12Zbilinear90d22 1 QC5.05 [3.24, 7.65]2.68-4.76 [-7.65, -2.63]5.84 [4.07, 8.07]9% [0%, 25%]18%23%-0.21+0.362026-08-09 → 2026-08-30
tmax12Zbilinear365d27 1 QC5.82 [3.82, 8.48]2.98-5.58 [-8.48, -3.28]6.68 [4.63, 8.82]7% [0%, 21%]15%19%-0.26+0.362026-01-09 → 2026-08-30
tmax12Zbilinearall27 1 QC5.82 [3.84, 8.83]2.98-5.58 [-8.83, -3.26]6.68 [4.67, 9.28]7% [0%, 20%]15%19%-0.26+0.362026-01-09 → 2026-08-30
tmax12Znearest30d22 1 QC5.01 [3.39, 6.95]2.68-4.71 [-6.95, -2.71]5.78 [4.19, 7.44]9% [0%, 24%]18%23%-0.20+0.362026-08-09 → 2026-08-30
tmax12Znearest90d22 1 QC5.01 [3.30, 7.52]2.68-4.71 [-7.52, -2.65]5.78 [4.12, 7.91]9% [0%, 25%]18%23%-0.20+0.362026-08-09 → 2026-08-30
tmax12Znearest365d27 1 QC5.82 [3.86, 8.51]3.01-5.58 [-8.51, -3.33]6.70 [4.66, 8.89]7% [0%, 21%]15%19%-0.26+0.352026-01-09 → 2026-08-30
tmax12Znearestall27 1 QC5.82 [3.91, 8.89]3.01-5.58 [-8.89, -3.29]6.70 [4.70, 9.40]7% [0%, 20%]15%19%-0.26+0.352026-01-09 → 2026-08-30
tmin00Zbilinear30d22 1 QC2.55 [1.83, 3.23]2.56+0.20 [-1.47, 1.84]2.96 [2.20, 3.51]18% [5%, 35%]41%64%-0.00+0.002026-08-09 → 2026-08-30
tmin00Zbilinear90d22 1 QC2.55 [1.66, 3.27]2.56+0.20 [-1.80, 2.04]2.96 [1.94, 3.56]18% [4%, 38%]41%64%-0.00+0.002026-08-09 → 2026-08-30
tmin00Zbilinear365d28 1 QC3.09 [2.07, 4.56]3.09-0.36 [-2.40, 1.49]4.13 [2.50, 6.35]18% [6%, 31%]36%57%+0.37+0.372026-01-09 → 2026-08-30
tmin00Zbilinearall28 1 QC3.09 [2.15, 4.60]3.09-0.36 [-2.68, 1.38]4.13 [2.56, 6.47]18% [5%, 30%]36%57%+0.37+0.372026-01-09 → 2026-08-30
tmin00Znearest30d22 1 QC2.55 [1.79, 3.20]2.55+0.09 [-1.60, 1.71]2.97 [2.23, 3.48]18% [5%, 35%]41%59%-0.00+0.012026-08-09 → 2026-08-30
tmin00Znearest90d22 1 QC2.55 [1.64, 3.23]2.55+0.09 [-1.89, 1.93]2.97 [1.96, 3.49]18% [4%, 38%]41%59%-0.00+0.012026-08-09 → 2026-08-30
tmin00Znearest365d28 1 QC3.08 [2.06, 4.53]3.08-0.49 [-2.56, 1.36]4.13 [2.54, 6.33]18% [6%, 31%]36%54%+0.37+0.372026-01-09 → 2026-08-30
tmin00Znearestall28 1 QC3.08 [2.14, 4.59]3.08-0.49 [-2.84, 1.24]4.13 [2.60, 6.45]18% [5%, 30%]36%54%+0.37+0.372026-01-09 → 2026-08-30
tmin12Zbilinear30d22 1 QC2.75 [2.13, 3.23]2.75+0.40 [-1.46, 2.00]3.14 [2.41, 3.71]9% [0%, 33%]41%68%-0.08-0.072026-08-09 → 2026-08-30
tmin12Zbilinear90d22 1 QC2.75 [2.01, 3.23]2.75+0.40 [-1.83, 2.15]3.14 [2.31, 3.66]9% [0%, 41%]41%68%-0.08-0.072026-08-09 → 2026-08-30
tmin12Zbilinear365d27 1 QC2.88 [2.16, 3.55]2.89+0.24 [-1.49, 1.82]3.50 [2.55, 4.49]11% [0%, 31%]44%67%+0.40+0.402026-01-09 → 2026-08-30
tmin12Zbilinearall27 1 QC2.88 [2.24, 3.48]2.89+0.24 [-1.55, 1.72]3.50 [2.55, 4.40]11% [0%, 33%]44%67%+0.40+0.402026-01-09 → 2026-08-30
tmin12Znearest30d22 1 QC2.78 [2.18, 3.22]2.78+0.29 [-1.67, 1.93]3.18 [2.51, 3.70]9% [0%, 33%]32%73%-0.09-0.082026-08-09 → 2026-08-30
tmin12Znearest90d22 1 QC2.78 [2.04, 3.20]2.78+0.29 [-2.02, 2.06]3.18 [2.42, 3.64]9% [0%, 41%]32%73%-0.09-0.082026-08-09 → 2026-08-30
tmin12Znearest365d27 1 QC2.92 [2.22, 3.57]2.93+0.10 [-1.69, 1.75]3.55 [2.66, 4.53]15% [0%, 44%]37%70%+0.39+0.392026-01-09 → 2026-08-30
tmin12Znearestall27 1 QC2.92 [2.29, 3.53]2.93+0.10 [-1.73, 1.64]3.55 [2.68, 4.46]15% [0%, 44%]37%70%+0.39+0.392026-01-09 → 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 8, 00Z bilinear tmax+20+100-10-202026-08-09: -6.14 °F2026-08-10: -9.66 °F2026-08-11: -6.67 °F2026-08-12: -8.02 °F2026-08-13: -8.18 °F2026-08-14: -6 °F2026-08-15: -6.67 °F2026-08-16: -6.39 °F2026-08-17: -5.02 °F2026-08-18: -0.35 °F2026-08-19: -3.19 °F2026-08-20: +0.46 °F2026-08-21: +2.72 °F2026-08-22: -0.32 °F2026-08-23: -5.32 °F2026-08-24: -4.86 °F2026-08-25: -5.54 °F2026-08-26: -0.98 °F2026-08-27: -11.37 °F2026-08-28: -10.66 °F2026-08-29: -5.45 °F2026-08-30: -10.41 °F2026-08-092026-08-30°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 (22 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-10.41
2026-08-29-5.45
2026-08-28-10.66
2026-08-27-11.37
2026-08-26-0.98
2026-08-25-5.54
2026-08-24-4.86
2026-08-23-5.32
2026-08-22-0.32
2026-08-21+2.72
2026-08-20+0.46
2026-08-19-3.19
2026-08-18-0.35
2026-08-17-5.02
2026-08-16-6.39
2026-08-15-6.67
2026-08-14-6.00
2026-08-13-8.18
2026-08-12-8.02
2026-08-11-6.67
2026-08-10-9.66
2026-08-09-6.14

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KATL, lead day 8-20 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: 4 days-10 to -8 °F: 3 days-8 to -6 °F: 8 days-6 to -4 °F: 5 days-4 to -2 °F: 1 day-2 to 0 °F: 3 days0 to 2 °F: 1 day2 to 4 °F: 1 day4 to 6 °F: 1 day6 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| 6.26 · P90 |e| 10.49 °F-200+20 °Fn = 28 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
-12.00-10.004
-10.00-8.003
-8.00-6.008
-6.00-4.005
-4.00-2.001
-2.00+0.003
+0.00+2.001
+2.00+4.001
+4.00+6.001
+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
2026-08225.65 -5.36
2026-0168.62 -6.72

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-09 → 2026-08-30 28
GraphCast (IFS init)2026-01-09 → 2026-08-30 28
GraphCast (IFS init)2026-01-09 → 2026-08-30 27
GraphCast (IFS init)2026-01-09 → 2026-08-30 27

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 22 +0.12 [-1.40, 1.38] no
NCEP GFS 22 +1.94 [-1.02, 5.82] no
ECMWF IFS HRES 17 +2.43 [0.56, 5.38] yes
Persistence (baseline) 22 +1.47 [0.17, 3.58] yes

Cite this page

ShortCastCheck, KATL graphcast_ifs lead 8, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KATL/model/graphcast_ifs/lead/8/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KATL (Atlanta Hartsfield), model graphcast_ifs, lead day 8. 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/8/

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