GFS vs ECMWF Demand-Hub Outlook: 2026-05-29 00z

Executive Summary

The 15-day outlook shows a material disagreement between models, with the GFS forecasting significantly higher weather-driven demand than the ECMWF. The GFS projects both more cooling demand (103.3 weighted CDDs vs. 92.8 for ECMWF) and more heating demand (18.8 weighted HDDs vs. 11.1 for ECMWF), driven by extreme warmth in the South and periods of cooler weather in the Midwest and East. Confidence in the forecast is low, reflected in a convergence score of just 0.54, though this is a slight improvement over the prior cycle.

Hub-Level Weather

The national forecast is shaped by major disagreements in the highest-weighted demand centers:

• South / South Central (Houston & Atlanta): The largest divergence is in the South, where the GFS is substantially hotter. Over the 15-day period, GFS projects 288.5 CDDs for Houston and 174.6 for Atlanta, while the ECMWF is far more moderate at 209.5 and 83.6 CDDs, respectively. This is the primary driver of the GFS's stronger national cooling demand signal.
• Midwest (Chicago): The models show opposing patterns. The ECMWF is consistently warm (89.5 cumulative CDDs), while the GFS is much cooler, forecasting several days of heating demand (15.8 cumulative HDDs) and only 49.4 CDDs.
• East (New York City & Boston): The GFS shows a more volatile pattern, with more heating demand early in the period and more cooling demand late compared to the ECMWF. For NYC, the GFS forecasts 22.4 HDDs vs the ECMWF's 10.1. In Boston, the ECMWF is colder, forecasting 80.6 HDDs to the GFS's 65.1.

Demand Implications (Natural Gas)

The competing forecasts imply different natural gas demand trajectories for early June.

• Bullish GFS: The GFS solution suggests robust national demand. Strong heat across the South (Houston, Atlanta) would drive significant power burn for air conditioning, while intermittent cool shots in the Midwest and Northeast would support lingering residential/commercial heating demand. The 00z GFS run trended significantly warmer, adding +25.4 weighted CDDs versus its prior run, strengthening this bullish case.
• Moderate ECMWF: The ECMWF paints a more bearish picture for gas demand. Its forecast for substantially cooler temperatures in the South implies much weaker power sector gas consumption. The ECMWF solution was largely stable compared to its prior run.

Given the season, the divergence in southern cooling demand is the most critical factor for the natural gas balance.

Model Agreement & Caveats

Confidence in this forecast should be low.

• Convergence: The model agreement score is 0.535, indicating significant forecast divergence. While this is a minor improvement from the last cycle (+0.11), the low absolute value highlights the uncertainty.
• Model Instability: The GFS showed a large shift warmer in this cycle (delta_weighted_cumulative_cdd_gfs: +25.37), suggesting it may be unstable and could revert in subsequent runs. The ECMWF was comparatively stable.
• Source Data: This analysis is based on deterministic runs from Open-Meteo. It does not incorporate ensemble means, which can provide a more stable and reliable signal. The forecast, especially beyond day 7-10, is subject to major revisions.