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 · NCEP GFS · lead day 2

KATL · gfs · 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, 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.

VariableInitMethod Windown MAE °F MAE debiased Bias °F RMSE °F ±1 °F±2 °F±3 °F Skill Skill, debiased Period
tmax00Zbilinear30d28 1 QC3.13 [2.35, 3.99]3.13-0.78 [-2.90, 1.29]3.97 [3.11, 4.76]21% [4%, 37%]39%57%-0.31-0.312026-08-03 → 2026-08-30
tmax00Zbilinear90d28 1 QC3.13 [2.24, 4.10]3.13-0.78 [-3.33, 0.96]3.97 [3.01, 4.88]21% [0%, 38%]39%57%-0.31-0.312026-08-03 → 2026-08-30
tmax00Zbilinear365d28 1 QC3.13 [2.24, 4.20]3.13-0.78 [-3.73, 1.44]3.97 [2.96, 4.92]21% [0%, 39%]39%57%-0.31-0.312026-08-03 → 2026-08-30
tmax00Zbilinearall28 1 QC3.13 [2.20, 4.27]3.13-0.78 [-3.36, 1.20]3.97 [2.99, 4.94]21% [0%, 40%]39%57%-0.31-0.312026-08-03 → 2026-08-30
tmax00Znearest30d28 1 QC3.19 [2.47, 4.13]3.19-0.48 [-2.76, 1.69]4.11 [3.15, 5.16]29% [17%, 41%]39%54%-0.33-0.332026-08-03 → 2026-08-30
tmax00Znearest90d28 1 QC3.19 [2.34, 4.21]3.19-0.48 [-3.26, 1.43]4.11 [2.98, 5.25]29% [16%, 41%]39%54%-0.33-0.332026-08-03 → 2026-08-30
tmax00Znearest365d28 1 QC3.19 [2.33, 4.39]3.19-0.48 [-3.67, 1.84]4.11 [2.92, 5.39]29% [14%, 42%]39%54%-0.33-0.332026-08-03 → 2026-08-30
tmax00Znearestall28 1 QC3.19 [2.31, 4.42]3.19-0.48 [-3.25, 1.64]4.11 [2.88, 5.35]29% [15%, 43%]39%54%-0.33-0.332026-08-03 → 2026-08-30
tmax12Zbilinear30d28 1 QC3.16 [2.22, 4.26]3.13-0.92 [-2.68, 1.10]4.07 [2.68, 5.31]21% [4%, 36%]43%61%-0.32-0.312026-08-03 → 2026-08-30
tmax12Zbilinear90d28 1 QC3.16 [2.07, 4.41]3.13-0.92 [-3.23, 0.87]4.07 [2.46, 5.44]21% [0%, 37%]43%61%-0.32-0.312026-08-03 → 2026-08-30
tmax12Zbilinear365d28 1 QC3.16 [2.04, 4.40]3.13-0.92 [-3.16, 1.07]4.07 [2.42, 5.51]21% [0%, 36%]43%61%-0.32-0.312026-08-03 → 2026-08-30
tmax12Zbilinearall28 1 QC3.16 [2.04, 4.28]3.13-0.92 [-3.05, 1.01]4.07 [2.44, 5.36]21% [0%, 38%]43%61%-0.32-0.312026-08-03 → 2026-08-30
tmax12Znearest30d28 1 QC2.95 [2.18, 3.74]2.97-0.39 [-2.02, 1.39]3.87 [2.72, 4.81]21% [10%, 32%]50%57%-0.23-0.242026-08-03 → 2026-08-30
tmax12Znearest90d28 1 QC2.95 [2.09, 3.77]2.97-0.39 [-2.48, 1.32]3.87 [2.49, 4.95]21% [8%, 33%]50%57%-0.23-0.242026-08-03 → 2026-08-30
tmax12Znearest365d28 1 QC2.95 [2.04, 3.85]2.97-0.39 [-2.31, 1.53]3.87 [2.41, 4.98]21% [8%, 33%]50%57%-0.23-0.242026-08-03 → 2026-08-30
tmax12Znearestall28 1 QC2.95 [2.04, 3.77]2.97-0.39 [-2.43, 1.42]3.87 [2.46, 4.91]21% [6%, 33%]50%57%-0.23-0.242026-08-03 → 2026-08-30
tmin00Zbilinear30d28 1 QC2.07 [1.23, 2.76]1.93+1.29 [0.37, 2.21]2.66 [1.77, 3.30]36% [14%, 59%]50%75%+0.02+0.092026-08-03 → 2026-08-30
tmin00Zbilinear90d28 1 QC2.07 [1.01, 2.72]1.93+1.29 [0.10, 2.23]2.66 [1.46, 3.25]36% [14%, 62%]50%75%+0.02+0.092026-08-03 → 2026-08-30
tmin00Zbilinear365d28 1 QC2.07 [1.05, 2.83]1.93+1.29 [-0.01, 2.32]2.66 [1.51, 3.35]36% [11%, 60%]50%75%+0.02+0.092026-08-03 → 2026-08-30
tmin00Zbilinearall28 1 QC2.07 [1.01, 2.72]1.93+1.29 [-0.03, 2.21]2.66 [1.33, 3.24]36% [9%, 62%]50%75%+0.02+0.092026-08-03 → 2026-08-30
tmin00Znearest30d28 1 QC2.88 [1.92, 3.81]1.97+2.65 [1.60, 3.68]3.55 [2.46, 4.36]11% [0%, 24%]46%61%-0.37+0.072026-08-03 → 2026-08-30
tmin00Znearest90d28 1 QC2.88 [1.72, 3.79]1.97+2.65 [1.28, 3.68]3.55 [2.14, 4.33]11% [0%, 27%]46%61%-0.37+0.072026-08-03 → 2026-08-30
tmin00Znearest365d28 1 QC2.88 [1.64, 3.91]1.97+2.65 [1.19, 3.81]3.55 [2.18, 4.43]11% [0%, 29%]46%61%-0.37+0.072026-08-03 → 2026-08-30
tmin00Znearestall28 1 QC2.88 [1.74, 3.81]1.97+2.65 [1.20, 3.68]3.55 [2.12, 4.35]11% [0%, 27%]46%61%-0.37+0.072026-08-03 → 2026-08-30
tmin12Zbilinear30d28 1 QC1.91 [1.14, 2.74]1.84+0.92 [-0.28, 2.10]2.43 [1.40, 3.24]39% [21%, 62%]61%79%+0.09+0.142026-08-03 → 2026-08-30
tmin12Zbilinear90d28 1 QC1.91 [1.01, 2.68]1.84+0.92 [-0.56, 2.10]2.43 [1.22, 3.17]39% [17%, 67%]61%79%+0.09+0.142026-08-03 → 2026-08-30
tmin12Zbilinear365d28 1 QC1.91 [1.05, 2.87]1.84+0.92 [-0.75, 2.20]2.43 [1.22, 3.31]39% [17%, 65%]61%79%+0.09+0.142026-08-03 → 2026-08-30
tmin12Zbilinearall28 1 QC1.91 [1.04, 2.76]1.84+0.92 [-0.57, 2.10]2.43 [1.20, 3.25]39% [17%, 66%]61%79%+0.09+0.142026-08-03 → 2026-08-30
tmin12Znearest30d28 1 QC2.41 [1.21, 3.66]1.83+2.23 [0.98, 3.47]3.16 [1.60, 4.32]39% [20%, 61%]50%68%-0.14+0.142026-08-03 → 2026-08-30
tmin12Znearest90d28 1 QC2.41 [0.96, 3.55]1.83+2.23 [0.75, 3.44]3.16 [1.27, 4.20]39% [18%, 68%]50%68%-0.14+0.142026-08-03 → 2026-08-30
tmin12Znearest365d28 1 QC2.41 [0.92, 3.79]1.83+2.23 [0.68, 3.58]3.16 [1.19, 4.40]39% [18%, 71%]50%68%-0.14+0.142026-08-03 → 2026-08-30
tmin12Znearestall28 1 QC2.41 [0.86, 3.59]1.83+2.23 [0.72, 3.45]3.16 [1.16, 4.27]39% [18%, 67%]50%68%-0.14+0.142026-08-03 → 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 NCEP GFS at KATL, lead day 2, 00Z bilinear tmax+10+50-5-102026-08-03: -1.05 °F2026-08-04: -4.12 °F2026-08-05: -5.23 °F2026-08-06: +2.48 °F2026-08-07: +5.18 °F2026-08-08: +0.33 °F2026-08-09: -0.67 °F2026-08-10: -5.23 °F2026-08-11: +0.13 °F2026-08-12: +1.15 °F2026-08-13: +0.89 °F2026-08-14: +2.21 °F2026-08-15: +2.05 °F2026-08-16: +2.95 °F2026-08-17: +3.91 °F2026-08-18: +7.15 °F2026-08-19: +1.1 °F2026-08-20: -8.24 °F2026-08-21: +3.27 °F2026-08-22: -1.42 °F2026-08-23: -0.41 °F2026-08-24: -0.5 °F2026-08-25: -2.83 °F2026-08-26: -3.36 °F2026-08-27: -8.46 °F2026-08-28: -5.78 °F2026-08-29: -1.13 °F2026-08-30: -6.33 °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
2026-08-30-6.33
2026-08-29-1.13
2026-08-28-5.78
2026-08-27-8.46
2026-08-26-3.36
2026-08-25-2.83
2026-08-24-0.50
2026-08-23-0.41
2026-08-22-1.42
2026-08-21+3.27
2026-08-20-8.24
2026-08-19+1.10
2026-08-18+7.15
2026-08-17+3.91
2026-08-16+2.95
2026-08-15+2.05
2026-08-14+2.21
2026-08-13+0.89
2026-08-12+1.15
2026-08-11+0.13
2026-08-10-5.23
2026-08-09-0.67
2026-08-08+0.33
2026-08-07+5.18
2026-08-06+2.48
2026-08-05-5.23
2026-08-04-4.12
2026-08-03-1.05

Error distribution

Distribution of the signed daily error in °F for NCEP GFS at KATL, lead day 2-10 to -9 °F: 0 days-9 to -8 °F: 2 days-8 to -7 °F: 0 days-7 to -6 °F: 1 day-6 to -5 °F: 3 days-5 to -4 °F: 1 day-4 to -3 °F: 1 day-3 to -2 °F: 1 day-2 to -1 °F: 3 days-1 to 0 °F: 3 days0 to 1 °F: 3 days1 to 2 °F: 2 days2 to 3 °F: 4 days3 to 4 °F: 2 days4 to 5 °F: 0 days5 to 6 °F: 1 day6 to 7 °F: 0 days7 to 8 °F: 1 day8 to 9 °F: 0 days9 to 10 °F: 0 daysdashed: P50 |e| 2.65 · P90 |e| 6.57 °F-100+10 °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
-10.00-9.000
-9.00-8.002
-8.00-7.000
-7.00-6.001
-6.00-5.003
-5.00-4.001
-4.00-3.001
-3.00-2.001
-2.00-1.003
-1.00+0.003
+0.00+1.003
+1.00+2.002
+2.00+3.004
+3.00+4.002
+4.00+5.000
+5.00+6.001
+6.00+7.000
+7.00+8.001
+8.00+9.000
+9.00+10.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08283.13 -0.78

Availability

ModelScored period DaysCoverage
NCEP GFS2026-08-03 → 2026-08-30 28
NCEP GFS2026-08-03 → 2026-08-30 28
NCEP GFS2026-08-03 → 2026-08-30 28
NCEP GFS2026-08-03 → 2026-08-30 28

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 modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 28 -1.25 [-2.16, -0.24] yes
Aurora (IFS init) 1 -1.19 [-1.19, -1.19] yes
FourCastNet v2 (IFS init) 1 +1.69 [1.69, 1.69] yes
GraphCast (IFS init) 26 -2.70 [-4.04, -1.07] yes
ECMWF IFS HRES 22 +0.35 [-2.20, 1.57] no
Pangu-Weather (IFS init) 1 +1.62 [1.62, 1.62] yes
Persistence (baseline) 28 +0.73 [-0.16, 1.77] no

Cite this page

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

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