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

Boston Logan · GraphCast (IFS init) · lead day 0

KBOS · graphcast_ifs · lead 0

Target day = initialization date + 0 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
tmax00Zbilinear30d27 1 QC2.89 [2.48, 3.31]1.94-2.44 [-2.94, -1.88]3.43 [3.10, 3.76]22% [7%, 37%]30%56%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d27 1 QC2.89 [2.48, 3.31]1.94-2.44 [-3.04, -1.90]3.43 [3.10, 3.77]22% [6%, 39%]30%56%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d33 1 QC2.82 [2.45, 3.22]1.72-2.46 [-2.91, -1.96]3.31 [2.91, 3.63]18% [4%, 33%]27%55%——2026-01-01 → 2026-08-30
tmax00Zbilinearall33 1 QC2.82 [2.48, 3.23]1.72-2.46 [-2.95, -1.96]3.31 [2.94, 3.66]18% [3%, 33%]27%55%——2026-01-01 → 2026-08-30
tmax00Znearest30d27 1 QC2.84 [2.44, 3.29]1.96-2.36 [-2.85, -1.80]3.37 [3.04, 3.71]22% [7%, 37%]33%48%——2026-08-01 → 2026-08-30
tmax00Znearest90d27 1 QC2.84 [2.45, 3.32]1.96-2.36 [-2.97, -1.82]3.37 [3.03, 3.69]22% [6%, 39%]33%48%——2026-08-01 → 2026-08-30
tmax00Znearest365d33 1 QC2.73 [2.31, 3.16]1.74-2.33 [-2.80, -1.83]3.21 [2.74, 3.56]21% [8%, 36%]33%55%——2026-01-01 → 2026-08-30
tmax00Znearestall33 1 QC2.73 [2.34, 3.21]1.74-2.33 [-2.85, -1.84]3.21 [2.80, 3.56]21% [8%, 35%]33%55%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d27 1 QC1.33 [1.07, 1.61]1.33-0.01 [-0.63, 0.60]1.68 [1.33, 2.02]52% [36%, 66%]78%93%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d27 1 QC1.33 [1.09, 1.68]1.33-0.01 [-0.84, 0.61]1.68 [1.36, 2.08]52% [35%, 67%]78%93%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d33 1 QC1.42 [1.16, 1.77]1.36-0.28 [-1.26, 0.45]1.80 [1.45, 2.16]45% [26%, 62%]76%88%——2026-01-01 → 2026-08-30
tmin00Zbilinearall33 1 QC1.42 [1.14, 1.77]1.36-0.28 [-1.12, 0.42]1.80 [1.41, 2.17]45% [29%, 63%]76%88%——2026-01-01 → 2026-08-30
tmin00Znearest30d27 1 QC1.32 [1.03, 1.64]1.38+0.45 [-0.18, 1.07]1.82 [1.38, 2.21]52% [40%, 63%]74%89%——2026-08-01 → 2026-08-30
tmin00Znearest90d27 1 QC1.32 [1.01, 1.65]1.38+0.45 [-0.37, 1.05]1.82 [1.38, 2.24]52% [40%, 63%]74%89%——2026-08-01 → 2026-08-30
tmin00Znearest365d33 1 QC1.38 [1.10, 1.68]1.42+0.21 [-0.73, 0.92]1.86 [1.52, 2.19]52% [40%, 62%]70%88%——2026-01-01 → 2026-08-30
tmin00Znearestall33 1 QC1.38 [1.08, 1.68]1.42+0.21 [-0.63, 0.90]1.86 [1.48, 2.20]52% [40%, 61%]70%88%——2026-01-01 → 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 KBOS, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: -2.83 °F2026-08-02: -3.06 °F2026-08-03: -2.37 °F2026-08-04: -4.7 °F2026-08-05: -2.83 °F2026-08-06: +2.36 °F2026-08-07: -5.67 °F2026-08-08: -4.46 °F2026-08-09: -1.68 °F2026-08-10: -0.13 °F2026-08-11: -2.49 °F2026-08-12: -1.21 °F2026-08-13: -5.93 °F2026-08-14: -4.62 °F2026-08-15: +0.16 °F2026-08-16: -3.69 °F2026-08-17: -2.92 °F2026-08-18: +0.13 °F2026-08-19: -6.31 °F2026-08-20: -0.9 °F2026-08-21: -0.17 °F2026-08-22: -2.11 °F2026-08-23: +3.35 °F2026-08-24: -5.51 °F2026-08-25: -3.53 °F2026-08-29: -0.49 °F2026-08-30: -4.27 °F2026-08-012026-08-30°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 (30 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-4.27
2026-08-29-0.49
2026-08-28—
2026-08-27—
2026-08-26—
2026-08-25-3.53
2026-08-24-5.51
2026-08-23+3.35
2026-08-22-2.11
2026-08-21-0.17
2026-08-20-0.90
2026-08-19-6.31
2026-08-18+0.13
2026-08-17-2.92
2026-08-16-3.69
2026-08-15+0.16
2026-08-14-4.62
2026-08-13-5.93
2026-08-12-1.21
2026-08-11-2.49
2026-08-10-0.13
2026-08-09-1.68
2026-08-08-4.46
2026-08-07-5.67
2026-08-06+2.36
2026-08-05-2.83
2026-08-04-4.70
2026-08-03-2.37
2026-08-02-3.06
2026-08-01-2.83

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KBOS, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 0 days-8 to -7 °F: 0 days-7 to -6 °F: 1 day-6 to -5 °F: 3 days-5 to -4 °F: 4 days-4 to -3 °F: 6 days-3 to -2 °F: 8 days-2 to -1 °F: 3 days-1 to 0 °F: 4 days0 to 1 °F: 2 days1 to 2 °F: 0 days2 to 3 °F: 1 day3 to 4 °F: 1 day4 to 5 °F: 0 days5 to 6 °F: 0 days6 to 7 °F: 0 days7 to 8 °F: 0 days8 to 9 °F: 0 days9 to 10 °F: 0 daysdashed: P50 |e| 2.83 · P90 |e| 5.35 °F-100+10 °Fn = 33 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
-9.00-8.000
-8.00-7.000
-7.00-6.001
-6.00-5.003
-5.00-4.004
-4.00-3.006
-3.00-2.008
-2.00-1.003
-1.00+0.004
+0.00+1.002
+1.00+2.000
+2.00+3.001
+3.00+4.001
+4.00+5.000
+5.00+6.000
+6.00+7.000
+7.00+8.000
+8.00+9.000
+9.00+10.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08272.89 -2.44
2026-0162.55 -2.55

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-01 → 2026-08-30 33
GraphCast (IFS init)2026-01-01 → 2026-08-30 33

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 27 -0.40 [-0.60, -0.16] yes
Aurora (GFS init) 2 -1.56 [-1.62, -1.49] yes
Aurora (IFS init) 2 -0.46 [-0.76, -0.16] yes
FourCastNet v2 (GFS init) 2 -2.39 [-4.77, -0.01] yes
FourCastNet v2 (IFS init) 2 -1.93 [-3.76, -0.09] yes
NCEP GFS 27 +0.80 [0.23, 1.49] yes
GraphCast (GFS init) 2 -2.00 [-2.53, -1.47] yes
ECMWF IFS HRES 21 +0.60 [0.27, 1.07] yes
Pangu-Weather (GFS init) 2 -0.27 [-0.96, 0.43] no
Pangu-Weather (IFS init) 2 -0.17 [-1.53, 1.19] no

Cite this page

ShortCastCheck, KBOS graphcast_ifs lead 0, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KBOS/model/graphcast_ifs/lead/0/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KBOS (Boston Logan), model graphcast_ifs, lead day 0. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KBOS/model/graphcast_ifs/lead/0/

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