Climavision opens radar and forecasting platform to AI data centers

Climavision is making its private radar network and high-resolution weather forecasts available to AI data centers, utilities and grid operators as power demand and weather risk rise across the U.S. The move comes as PJM pushes large electricity users to secure their own power by June 2027.

Why it matters: - AI data centers are increasing power demand at the same time the grid is becoming more exposed to weather-driven supply swings. - Climavision says decision-grade weather intelligence is becoming necessary for facilities that cannot afford downtime. - PJM's proposal would require data centers and other large electricity users to secure their own power by June 2027.

What happened: - Climavision opened its private radar network and high-resolution forecasting platform to AI data centers, utilities and grid operators serving them. - The company is targeting rising grid stress tied to AI growth and extreme weather. - Climavision already provides weather intelligence to government agencies, utilities and energy traders in rural areas that the federal radar network does not fully cover. - The company said the offering is aimed at helping large energy users and grid operators manage weather risk at the point of decision.

The details: - Climavision operates more than 30 radars across 15 states. - The radar network is designed to cover weather blind spots where federal radar coverage is thin. - The system tracks the lowest few thousand feet of the atmosphere, which can determine how much wind and solar power is actually produced. - The forecasting platform combines Climavision radar observations with public datasets, commercial sources and customer sensors. - The platform produces high-resolution forecasts for specific wind farms, substations or campuses rather than regional averages. - Climavision said its forecasting can help operators plan renewable supply with more precision. - Better solar forecasts can cut spare capacity held against forecast error by 30%, according to the company. - Climavision's HailCast, TornadoCast and LightningCast products can warn of hail, tornadoes and lightning up to 30 minutes ahead. - The company said those warnings can help protect turbines, inverters, substations and transmission lines. - Climavision's radar footprint spans Kentucky, North Carolina, Pennsylvania and Tennessee in PJM territory, along with states in the South and Midwest where data center growth is accelerating. - The company said its network already operates in Louisiana, Georgia and Mississippi, where major data center investments are underway.

Between the lines: - PJM has twice this year asked the U.S. Department of Energy for emergency authority to curtail data centers rather than cut power to homes and businesses. - FERC has directed all six U.S. grid operators to address how they connect large loads. - Under PJM's proposal, data centers could avoid curtailment by bringing qualifying capacity through new generation, storage, demand response, capacity uprates or fuel-switching. - Each of those compliance paths depends on weather-sensitive resources. - That makes weather forecasting a grid-planning issue, not just a protection tool. - Chris Goode, Climavision's co-founder and CEO, said AI has made the grid more sensitive to weather because the new supply being built is weather-driven and demand never pauses. - Goode said 21st century compute loads cannot be managed with 20th century weather infrastructure. - CenterPoint Energy selected Climavision in 2025 to integrate weather intelligence into grid operations after Hurricane Beryl damage in 2024. - Enverus uses Climavision forecasts for energy trading, and the two companies published an April analysis on weather signals before January's natural gas price spike.

What's next: - Data centers and grid operators can request coverage details for their service territories at more information. - Climavision appears to be positioning its weather network as infrastructure for both power procurement and grid resilience as the 2027 deadline approaches.

The bottom line: - Climavision is betting that AI-era electricity planning will depend as much on weather intelligence as on power supply itself.