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 7

KATL · graphcast_ifs · lead 7

Target day = initialization date + 7 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
tmax00Zbilinear30d23 1 QC5.42 [3.51, 7.51]2.79-5.20 [-7.51, -3.08]6.23 [4.37, 7.93]9% [0%, 21%]17%26%-0.40+0.262026-08-08 → 2026-08-30
tmax00Zbilinear90d23 1 QC5.42 [3.35, 7.93]2.79-5.20 [-7.93, -2.87]6.23 [4.26, 8.28]9% [0%, 24%]17%26%-0.40+0.262026-08-08 → 2026-08-30
tmax00Zbilinear365d29 1 QC6.35 [4.19, 9.02]3.02-6.18 [-9.02, -3.88]7.20 [5.18, 9.32]7% [0%, 19%]14%21%-0.31+0.362026-01-08 → 2026-08-30
tmax00Zbilinearall29 1 QC6.35 [4.14, 9.39]3.02-6.18 [-9.39, -3.72]7.20 [5.07, 9.62]7% [0%, 18%]14%21%-0.31+0.362026-01-08 → 2026-08-30
tmax00Znearest30d23 1 QC5.41 [3.56, 7.43]2.77-5.19 [-7.43, -3.12]6.20 [4.42, 7.83]9% [0%, 21%]17%26%-0.40+0.262026-08-08 → 2026-08-30
tmax00Znearest90d23 1 QC5.41 [3.39, 7.79]2.77-5.19 [-7.79, -2.86]6.20 [4.28, 8.12]9% [0%, 24%]17%26%-0.40+0.262026-08-08 → 2026-08-30
tmax00Znearest365d29 1 QC6.36 [4.22, 9.07]3.00-6.19 [-9.07, -3.89]7.22 [5.22, 9.35]7% [0%, 19%]14%21%-0.32+0.362026-01-08 → 2026-08-30
tmax00Znearestall29 1 QC6.36 [4.16, 9.37]3.00-6.19 [-9.37, -3.80]7.22 [5.11, 9.64]7% [0%, 18%]14%21%-0.32+0.362026-01-08 → 2026-08-30
tmax12Zbilinear30d23 1 QC5.51 [3.85, 7.41]2.47-5.32 [-7.40, -3.41]6.19 [4.54, 7.77]4% [0%, 12%]13%26%-0.42+0.342026-08-08 → 2026-08-30
tmax12Zbilinear90d23 1 QC5.51 [3.67, 7.83]2.47-5.32 [-7.83, -3.21]6.19 [4.33, 8.09]4% [0%, 12%]13%26%-0.42+0.342026-08-08 → 2026-08-30
tmax12Zbilinear365d28 1 QC6.01 [4.15, 8.15]2.50-5.86 [-8.15, -3.78]6.66 [4.85, 8.39]4% [0%, 10%]11%21%-0.20+0.492026-01-08 → 2026-08-30
tmax12Zbilinearall28 1 QC6.01 [4.25, 8.21]2.50-5.86 [-8.21, -3.90]6.66 [4.95, 8.37]4% [0%, 10%]11%21%-0.20+0.492026-01-08 → 2026-08-30
tmax12Znearest30d23 1 QC5.50 [3.86, 7.32]2.43-5.29 [-7.32, -3.39]6.15 [4.54, 7.68]4% [0%, 12%]4%30%-0.42+0.352026-08-08 → 2026-08-30
tmax12Znearest90d23 1 QC5.50 [3.72, 7.73]2.43-5.29 [-7.73, -3.22]6.15 [4.35, 7.97]4% [0%, 12%]4%30%-0.42+0.352026-08-08 → 2026-08-30
tmax12Znearest365d28 1 QC6.01 [4.22, 8.16]2.48-5.85 [-8.16, -3.81]6.64 [4.83, 8.39]4% [0%, 10%]4%25%-0.20+0.492026-01-08 → 2026-08-30
tmax12Znearestall28 1 QC6.01 [4.28, 8.17]2.48-5.85 [-8.17, -3.91]6.64 [4.95, 8.41]4% [0%, 10%]4%25%-0.20+0.492026-01-08 → 2026-08-30
tmin00Zbilinear30d23 1 QC2.55 [1.74, 3.04]2.56+0.25 [-1.00, 1.46]3.20 [2.31, 3.68]30% [12%, 57%]48%57%-0.03-0.032026-08-08 → 2026-08-30
tmin00Zbilinear90d23 1 QC2.55 [1.51, 2.98]2.56+0.25 [-1.21, 1.65]3.20 [1.94, 3.65]30% [13%, 67%]48%57%-0.03-0.032026-08-08 → 2026-08-30
tmin00Zbilinear365d29 1 QC3.08 [1.95, 4.57]3.07-0.49 [-2.87, 1.16]4.42 [2.64, 6.68]31% [12%, 56%]48%55%+0.35+0.352026-01-08 → 2026-08-30
tmin00Zbilinearall29 1 QC3.08 [2.05, 4.63]3.07-0.49 [-2.99, 1.09]4.42 [2.78, 6.88]31% [14%, 56%]48%55%+0.35+0.352026-01-08 → 2026-08-30
tmin00Znearest30d23 1 QC2.55 [1.76, 3.04]2.56+0.15 [-1.10, 1.33]3.20 [2.30, 3.70]30% [12%, 57%]48%57%-0.03-0.032026-08-08 → 2026-08-30
tmin00Znearest90d23 1 QC2.55 [1.53, 2.97]2.56+0.15 [-1.29, 1.50]3.20 [1.97, 3.65]30% [13%, 67%]48%57%-0.03-0.032026-08-08 → 2026-08-30
tmin00Znearest365d29 1 QC3.10 [1.94, 4.64]3.06-0.63 [-3.06, 1.01]4.47 [2.64, 6.81]31% [12%, 56%]48%55%+0.34+0.352026-01-08 → 2026-08-30
tmin00Znearestall29 1 QC3.10 [2.05, 4.68]3.06-0.63 [-3.20, 0.97]4.47 [2.76, 7.01]31% [14%, 56%]48%55%+0.34+0.352026-01-08 → 2026-08-30
tmin12Zbilinear30d23 1 QC2.49 [1.72, 2.95]2.45-0.19 [-1.71, 1.11]2.89 [2.19, 3.28]17% [0%, 50%]48%61%-0.00+0.012026-08-08 → 2026-08-30
tmin12Zbilinear90d23 1 QC2.49 [1.56, 2.90]2.45-0.19 [-1.92, 1.20]2.89 [2.06, 3.26]17% [0%, 62%]48%61%-0.00+0.012026-08-08 → 2026-08-30
tmin12Zbilinear365d28 1 QC2.77 [1.99, 3.54]2.70-0.27 [-1.77, 0.99]3.38 [2.45, 4.41]18% [0%, 44%]43%61%+0.37+0.392026-01-08 → 2026-08-30
tmin12Zbilinearall28 1 QC2.77 [2.04, 3.59]2.70-0.27 [-1.81, 0.99]3.38 [2.46, 4.48]18% [0%, 44%]43%61%+0.37+0.392026-01-08 → 2026-08-30
tmin12Znearest30d23 1 QC2.54 [1.81, 2.97]2.48-0.29 [-1.84, 1.02]2.94 [2.23, 3.31]13% [0%, 39%]43%57%-0.03+0.002026-08-08 → 2026-08-30
tmin12Znearest90d23 1 QC2.54 [1.64, 2.93]2.48-0.29 [-2.03, 1.08]2.94 [2.11, 3.29]13% [0%, 50%]43%57%-0.03+0.002026-08-08 → 2026-08-30
tmin12Znearest365d28 1 QC2.83 [2.06, 3.61]2.72-0.39 [-1.93, 0.90]3.41 [2.53, 4.39]14% [0%, 39%]39%57%+0.36+0.382026-01-08 → 2026-08-30
tmin12Znearestall28 1 QC2.83 [2.12, 3.66]2.72-0.39 [-1.94, 0.88]3.41 [2.55, 4.47]14% [0%, 38%]39%57%+0.36+0.382026-01-08 → 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 7, 00Z bilinear tmax+20+100-10-202026-08-08: -7.61 °F2026-08-09: -11.64 °F2026-08-10: -6.36 °F2026-08-11: -4.41 °F2026-08-12: -4.6 °F2026-08-13: -7.4 °F2026-08-14: -9.53 °F2026-08-15: -6.36 °F2026-08-16: -3.9 °F2026-08-17: -4.11 °F2026-08-18: +1.72 °F2026-08-19: +0.27 °F2026-08-20: -2.22 °F2026-08-21: +0.47 °F2026-08-22: -4.36 °F2026-08-23: -1.52 °F2026-08-24: -4.99 °F2026-08-25: -6.68 °F2026-08-26: -7.81 °F2026-08-27: -2.66 °F2026-08-28: -10.93 °F2026-08-29: -7.35 °F2026-08-30: -7.63 °F2026-08-082026-08-30°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 (23 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-7.63
2026-08-29-7.35
2026-08-28-10.93
2026-08-27-2.66
2026-08-26-7.81
2026-08-25-6.68
2026-08-24-4.99
2026-08-23-1.52
2026-08-22-4.36
2026-08-21+0.47
2026-08-20-2.22
2026-08-19+0.27
2026-08-18+1.72
2026-08-17-4.11
2026-08-16-3.90
2026-08-15-6.36
2026-08-14-9.53
2026-08-13-7.40
2026-08-12-4.60
2026-08-11-4.41
2026-08-10-6.36
2026-08-09-11.64
2026-08-08-7.61

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KATL, lead day 7-20 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: 5 days-10 to -8 °F: 2 days-8 to -6 °F: 9 days-6 to -4 °F: 5 days-4 to -2 °F: 3 days-2 to 0 °F: 1 day0 to 2 °F: 3 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| 6.65 · P90 |e| 11 °F-200+20 °Fn = 29 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
-12.00-10.005
-10.00-8.002
-8.00-6.009
-6.00-4.005
-4.00-2.003
-2.00+0.001
+0.00+2.003
+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
2026-08235.42 -5.20
2026-0169.91 -9.91

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 28
GraphCast (IFS init)2026-01-08 → 2026-08-30 28

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 23 +0.53 [-0.43, 1.89] no
NCEP GFS 23 +2.44 [0.08, 5.91] yes
ECMWF IFS HRES 18 +1.53 [0.21, 6.36] yes
Persistence (baseline) 23 +1.55 [0.40, 3.54] yes

Cite this page

ShortCastCheck, KATL graphcast_ifs lead 7, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KATL/model/graphcast_ifs/lead/7/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KATL (Atlanta Hartsfield), model graphcast_ifs, lead day 7. 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/7/

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