North America recorded global horizontal irradiance (GHI) 0.8% above normal in October, but that continental figure concealed sharply different conditions for solar assets. According to an analysis using the Solcast API, the Northeast and Great Lakes recorded substantial irradiance surpluses, while the Southwest and northern Mexico experienced widespread deficits during an unusually wet month. The result was a divided resource picture rather than a uniform shift, with regional anomalies far larger than the continental average.
Northeast and Great Lakes: a broad surplus
The Northeast and Great Lakes averaged GHI 14% above normal, and 95.5% of the region recorded above-average irradiance. The breadth of the surplus is notable: it was not confined to isolated sites but covered nearly the entire region in the analysis. The strongest anomalies were concentrated around Upstate New York and parts of Michigan, where GHI was approximately 20-25% above normal. Higher atmospheric pressure, lower humidity and below-normal precipitation across much of the region reduced the frequency of cloudy periods, allowing more solar radiation to reach the surface.
Two city records underlined the scale of the anomaly. Buffalo recorded its strongest October GHI in the 2007-2026 series, while Cleveland's average daily profile also set a high for the city in the Solcast record. Those records are consistent with the broader regional pattern rather than standing apart from it. The surplus was particularly relevant in a region with significant volumes of distributed solar installed in states including New York, New Jersey and Massachusetts.
Southwest: moisture, cloud and deficits
Conditions reversed farther southwest. A strong low-pressure trough approaching the West Coast drew Pacific moisture into the Four Corners region where Arizona, Colorado, New Mexico and Utah meet. Moisture associated with Tropical Storms Polo and Trudy also contributed to several days of heavy rainfall during October. The San Luis Valley received more than four times its usual October rainfall in four days, and the San Juan River at Pagosa Springs reached its second-highest crest since 1911. These precipitation signals point to an atmosphere that was both wetter and cloudier than normal across the region.
Elevated atmospheric moisture extended across much of the Southwest and northern Mexico, increasing cloud cover and coinciding with a broad irradiance deficit. GHI across the region averaged 9% below the October norm. Phoenix recorded GHI approximately 10% below its 2007-2025 October average, making October 2026 the second-lowest October for GHI in the Solcast record. The city figure is a local marker within a wider regional shortfall, not an isolated exception.
The below-normal conditions covered the solar-intensive corridor from Southern California into Arizona, exposing a substantial concentration of utility-scale solar assets to the same regional deficit. That geographic reach matters. A localized cloud anomaly would have limited relevance, but the analysis describes a broad area of below-normal irradiance across a corridor with significant utility-scale capacity. At the same time, the Northeast surplus covered states with substantial distributed solar, giving the month a distinct asset-type dimension as well as a geographic one.
Why the continental average misleads
The continental average is therefore a weak guide to asset-level resource conditions. A 0.8% national anomaly says little about a Northeast rooftop system operating under a regional surplus or a utility-scale plant in Arizona operating under a regional deficit. The 14% and 9% regional deviations were much larger than the continental mean, and their opposite signs help explain why the North American total stayed close to normal. In that sense, the October data illustrate how offsetting regional anomalies can produce a bland headline number. The headline average was therefore close to normal not because conditions were uniform but because surpluses and deficits pulled in opposite directions.
The figures describe irradiance, not measured generation. They do not quantify output, revenue or grid effects, and they should be read as October resource conditions rather than a complete performance assessment. Even so, the contrast is material because GHI is the solar resource reaching the surface, and the regional deviations were large enough to distinguish the month across the two areas. The article does not provide November guidance or asset-level error estimates. A monthly irradiance anomaly can inform resource assessment, but it is only one input among many that determine plant performance.
October thus presented two different resource regimes. The Northeast and Great Lakes had conditions that reduced cloud frequency, while the Southwest had conditions that increased cloud cover. The first favored solar radiation at the surface; the second worked against it. The article does not claim that these conditions determined plant output, but it does show that the resource side of the equation diverged sharply.
Data resolution and scope
Solcast produced the figures by tracking clouds and aerosols at 1-2 km resolution globally, using satellite data and proprietary AI and machine-learning algorithms. That data drives irradiance models, enabling the company to calculate irradiance at high resolution, with typical bias below 2%, as well as cloud-tracking forecasts. Solcast says the data are used by more than 350 companies managing over 350 GW of solar assets globally. These details describe the resolution and reach of the data; they do not turn a monthly irradiance analysis into a generation forecast.
For solar operators, the lesson is to look through the continental mean. In October, the same month produced a resource surplus for distributed solar in the Northeast and a resource deficit for utility-scale solar across the Southwest.
