Atlanta Hartsfield · NCEP GFS · lead day 4
KATL · gfs · lead 4
Target day = initialization date + 4 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, gfs-0p25, 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.
| Variable | Init | Method | Window | n | MAE °F | MAE debiased | Bias °F | RMSE °F | ±1 °F | ±2 °F | ±3 °F | Skill | Skill, debiased | Period |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| tmax | 00Z | bilinear | 30d | 26 1 QC | 3.26 [2.20, 4.53] | 3.37 | -0.72 [-2.89, 1.21] | 4.71 [2.78, 6.44] | 23% [15%, 31%] | 46% | 62% | -0.18 | -0.23 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | bilinear | 90d | 26 1 QC | 3.26 [2.12, 4.61] | 3.37 | -0.72 [-3.06, 1.21] | 4.71 [2.62, 6.50] | 23% [14%, 32%] | 46% | 62% | -0.18 | -0.23 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | bilinear | 365d | 26 1 QC | 3.26 [2.10, 4.80] | 3.37 | -0.72 [-3.49, 1.44] | 4.71 [2.61, 6.85] | 23% [14%, 33%] | 46% | 62% | -0.18 | -0.23 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | bilinear | all | 26 1 QC | 3.26 [2.12, 4.73] | 3.37 | -0.72 [-3.30, 1.33] | 4.71 [2.59, 6.77] | 23% [14%, 33%] | 46% | 62% | -0.18 | -0.23 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | nearest | 30d | 26 1 QC | 3.02 [2.11, 4.03] | 2.97 | +0.23 [-2.02, 2.01] | 4.27 [2.48, 5.78] | 19% [7%, 29%] | 50% | 69% | -0.09 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | nearest | 90d | 26 1 QC | 3.02 [2.03, 4.03] | 2.97 | +0.23 [-2.10, 2.11] | 4.27 [2.37, 5.81] | 19% [7%, 30%] | 50% | 69% | -0.09 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | nearest | 365d | 26 1 QC | 3.02 [2.03, 4.23] | 2.97 | +0.23 [-2.57, 2.22] | 4.27 [2.38, 6.10] | 19% [6%, 29%] | 50% | 69% | -0.09 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmax | 00Z | nearest | all | 26 1 QC | 3.02 [2.01, 4.16] | 2.97 | +0.23 [-2.33, 2.10] | 4.27 [2.36, 6.05] | 19% [4%, 30%] | 50% | 69% | -0.09 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | bilinear | 30d | 26 1 QC | 3.03 [2.05, 4.02] | 2.99 | +0.11 [-2.42, 2.18] | 3.98 [2.59, 5.02] | 19% [0%, 36%] | 38% | 69% | -0.09 | -0.09 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | bilinear | 90d | 26 1 QC | 3.03 [1.85, 4.01] | 2.99 | +0.11 [-2.82, 2.27] | 3.98 [2.18, 5.03] | 19% [0%, 38%] | 38% | 69% | -0.09 | -0.09 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | bilinear | 365d | 26 1 QC | 3.03 [1.80, 4.13] | 2.99 | +0.11 [-3.21, 2.36] | 3.98 [2.16, 5.19] | 19% [0%, 38%] | 38% | 69% | -0.09 | -0.09 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | bilinear | all | 26 1 QC | 3.03 [1.82, 4.10] | 2.99 | +0.11 [-2.93, 2.35] | 3.98 [2.15, 5.14] | 19% [0%, 36%] | 38% | 69% | -0.09 | -0.09 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | nearest | 30d | 26 1 QC | 3.18 [2.46, 3.88] | 2.87 | +0.76 [-1.74, 2.82] | 3.91 [2.91, 4.68] | 19% [8%, 28%] | 35% | 54% | -0.15 | -0.05 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | nearest | 90d | 26 1 QC | 3.18 [2.36, 3.90] | 2.87 | +0.76 [-2.16, 2.89] | 3.91 [2.70, 4.70] | 19% [7%, 29%] | 35% | 54% | -0.15 | -0.05 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | nearest | 365d | 26 1 QC | 3.18 [2.33, 3.92] | 2.87 | +0.76 [-2.82, 2.98] | 3.91 [2.67, 4.75] | 19% [7%, 29%] | 35% | 54% | -0.15 | -0.05 | 2026-08-05 → 2026-08-30 |
| tmax | 12Z | nearest | all | 26 1 QC | 3.18 [2.35, 3.90] | 2.87 | +0.76 [-2.25, 2.99] | 3.91 [2.64, 4.71] | 19% [0%, 29%] | 35% | 54% | -0.15 | -0.05 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | bilinear | 30d | 26 1 QC | 2.56 [1.77, 3.39] | 2.19 | +1.60 [0.60, 2.77] | 3.30 [2.07, 4.33] | 27% [16%, 39%] | 46% | 65% | -0.19 | -0.01 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | bilinear | 90d | 26 1 QC | 2.56 [1.72, 3.41] | 2.19 | +1.60 [0.40, 2.79] | 3.30 [2.01, 4.32] | 27% [14%, 41%] | 46% | 65% | -0.19 | -0.01 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | bilinear | 365d | 26 1 QC | 2.56 [1.67, 3.52] | 2.19 | +1.60 [0.29, 2.99] | 3.30 [1.98, 4.47] | 27% [14%, 43%] | 46% | 65% | -0.19 | -0.01 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | bilinear | all | 26 1 QC | 2.56 [1.71, 3.42] | 2.19 | +1.60 [0.33, 2.82] | 3.30 [1.98, 4.41] | 27% [13%, 40%] | 46% | 65% | -0.19 | -0.01 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | nearest | 30d | 26 1 QC | 3.34 [2.43, 4.33] | 2.23 | +2.59 [1.53, 3.95] | 4.06 [2.68, 5.24] | 8% [0%, 16%] | 27% | 58% | -0.55 | -0.03 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | nearest | 90d | 26 1 QC | 3.34 [2.31, 4.37] | 2.23 | +2.59 [1.32, 4.01] | 4.06 [2.54, 5.24] | 8% [0%, 15%] | 27% | 58% | -0.55 | -0.03 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | nearest | 365d | 26 1 QC | 3.34 [2.23, 4.49] | 2.23 | +2.59 [1.18, 4.20] | 4.06 [2.43, 5.38] | 8% [0%, 15%] | 27% | 58% | -0.55 | -0.03 | 2026-08-05 → 2026-08-30 |
| tmin | 00Z | nearest | all | 26 1 QC | 3.34 [2.31, 4.37] | 2.23 | +2.59 [1.20, 4.03] | 4.06 [2.52, 5.27] | 8% [0%, 15%] | 27% | 58% | -0.55 | -0.03 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | bilinear | 30d | 26 1 QC | 2.66 [1.63, 3.63] | 2.35 | +1.92 [0.21, 3.38] | 3.43 [2.09, 4.45] | 31% [18%, 47%] | 42% | 62% | -0.24 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | bilinear | 90d | 26 1 QC | 2.66 [1.51, 3.62] | 2.35 | +1.92 [-0.22, 3.35] | 3.43 [1.96, 4.41] | 31% [18%, 50%] | 42% | 62% | -0.24 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | bilinear | 365d | 26 1 QC | 2.66 [1.45, 3.78] | 2.35 | +1.92 [-0.58, 3.58] | 3.43 [1.92, 4.52] | 31% [16%, 50%] | 42% | 62% | -0.24 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | bilinear | all | 26 1 QC | 2.66 [1.41, 3.69] | 2.35 | +1.92 [-0.44, 3.50] | 3.43 [1.88, 4.48] | 31% [16%, 50%] | 42% | 62% | -0.24 | -0.08 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | nearest | 30d | 26 1 QC | 3.49 [1.96, 4.90] | 2.53 | +3.17 [1.21, 4.87] | 4.40 [2.60, 5.67] | 27% [7%, 48%] | 42% | 46% | -0.62 | -0.16 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | nearest | 90d | 26 1 QC | 3.49 [1.64, 4.85] | 2.53 | +3.17 [0.66, 4.84] | 4.40 [2.07, 5.63] | 27% [7%, 53%] | 42% | 46% | -0.62 | -0.16 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | nearest | 365d | 26 1 QC | 3.49 [1.38, 5.09] | 2.53 | +3.17 [0.19, 5.09] | 4.40 [1.84, 5.80] | 27% [0%, 57%] | 42% | 46% | -0.62 | -0.16 | 2026-08-05 → 2026-08-30 |
| tmin | 12Z | nearest | all | 26 1 QC | 3.49 [1.56, 5.02] | 2.53 | +3.17 [0.57, 5.02] | 4.40 [1.85, 5.77] | 27% [0%, 54%] | 42% | 46% | -0.62 | -0.16 | 2026-08-05 → 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
The same series as a table (26 days, newest first) · download CSV
| Climatological day | Error °F |
|---|---|
| 2026-08-30 | +0.51 |
| 2026-08-29 | -6.52 |
| 2026-08-28 | -1.77 |
| 2026-08-27 | +1.64 |
| 2026-08-26 | -8.36 |
| 2026-08-25 | -15.73 |
| 2026-08-24 | +2.78 |
| 2026-08-23 | -0.84 |
| 2026-08-22 | -3.17 |
| 2026-08-21 | -0.72 |
| 2026-08-20 | +1.71 |
| 2026-08-19 | +3.80 |
| 2026-08-18 | +8.80 |
| 2026-08-17 | +1.10 |
| 2026-08-16 | +0.30 |
| 2026-08-15 | -5.76 |
| 2026-08-14 | +3.61 |
| 2026-08-13 | -2.45 |
| 2026-08-12 | +1.01 |
| 2026-08-11 | +1.51 |
| 2026-08-10 | +0.33 |
| 2026-08-09 | -2.38 |
| 2026-08-08 | +0.10 |
| 2026-08-07 | -4.09 |
| 2026-08-06 | +3.24 |
| 2026-08-05 | +2.59 |
Error distribution
Bin counts as a table
| From °F | To °F | Days |
|---|---|---|
| -20.00 | -18.00 | 0 |
| -18.00 | -16.00 | 0 |
| -16.00 | -14.00 | 1 |
| -14.00 | -12.00 | 0 |
| -12.00 | -10.00 | 0 |
| -10.00 | -8.00 | 1 |
| -8.00 | -6.00 | 1 |
| -6.00 | -4.00 | 2 |
| -4.00 | -2.00 | 3 |
| -2.00 | +0.00 | 3 |
| +0.00 | +2.00 | 9 |
| +2.00 | +4.00 | 5 |
| +4.00 | +6.00 | 0 |
| +6.00 | +8.00 | 0 |
| +8.00 | +10.00 | 1 |
| +10.00 | +12.00 | 0 |
| +12.00 | +14.00 | 0 |
| +14.00 | +16.00 | 0 |
| +16.00 | +18.00 | 0 |
| +18.00 | +20.00 | 0 |
Month by month
00Z bilinear daily maximum, newest first.
| Month | n | MAE °F | Bias °F |
|---|---|---|---|
| 2026-08 | 26 | 3.26 | -0.72 |
Availability
| Model | Scored period | Days | Coverage |
|---|---|---|---|
| NCEP GFS | 2026-08-05 → 2026-08-30 | 26 | |
| NCEP GFS | 2026-08-05 → 2026-08-30 | 26 | |
| NCEP GFS | 2026-08-05 → 2026-08-30 | 26 | |
| NCEP GFS | 2026-08-05 → 2026-08-30 | 26 |
Compared with other models
Paired bootstrap on the days both models cover (90d, 00Z, bilinear, daily maximum). Negative means NCEP GFS has the lower MAE. "Distinguishable" means the 95 % interval excludes zero.
| Other model | Common days | MAE difference °F | Distinguishable |
|---|---|---|---|
| ECMWF AIFS Single | 26 | -0.95 [-2.76, 0.53] | no |
| GraphCast (IFS init) | 25 | -1.76 [-3.87, 0.21] | no |
| ECMWF IFS HRES | 20 | -0.55 [-3.28, 1.15] | no |
| Persistence (baseline) | 26 | +0.49 [-1.17, 1.97] | no |
Cite this page
ShortCastCheck, KATL gfs lead 4, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KATL/model/gfs/lead/4/
LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KATL (Atlanta Hartsfield), model gfs, lead day 4. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KATL/model/gfs/lead/4/
Accessed2026-08-31 (data through 2026-08-30; page built 2026-08-31T03:52:06+00:00)