Study says rural solar can feed AI data centers; Fresno County fits
A Next 10 and University of Pennsylvania study says data centers near Central Valley solar could use wasted power, a setup that aligns with the White House’s AI push.
Study says rural solar can feed AI data centers; Fresno County fits
Key Takeaways
- A Next 10 and University of Pennsylvania study says rural, solar-adjacent data centers could use curtailed power in California.
- E&E News frames the study as a possible assist to President Trump’s build-out of AI data centers while keeping bills down.
- Transmission congestion strands some Central Valley solar, making Fresno and Kings counties candidates for this approach.
- The California Energy Commission says data centers now draw about 1,000 MW, roughly 2% of CAISO peak demand.
What the new study argues
Twenty-three percent. That’s how much solar curtailment grew in California from 2023 to 2024, according to a study from nonprofit Next 10 led by University of Pennsylvania researcher Benjamin Lee. The report models a 20‑megawatt data center located near solar plants and finds bigger gains when paired with a 10‑megawatt battery. The idea is simple enough: put big computing where the stranded midday electrons already are.
E&E News connected that proposal to President Donald Trump’s stated goal of adding AI data centers while promising lower utility bills. California’s grid has struggled to move all of its midday solar to the cities that need it, and the study suggests rural siting could soak up power that now goes unused. That lands squarely on the Central Valley’s Westside.
Why it matters in the Valley
The Central Valley produces a lot of that midday solar. Transmission bottlenecks mean some power generated near Huron never makes it over the hills. Westlands Solar Park, spread across western Fresno and Kings counties, is the kind of area the study describes: utility‑scale solar along major lines like Gates. If companies put server halls near those interties and added on‑site batteries, they could run more hours on cheap local supply and cut curtailment that now frustrates generators.
For Fresno County readers, this is the local piece. Data center jobs are limited in number but tend to be steady and high wage, and projects bring construction work plus tax agreements the board will scrutinize. The California Energy Commission’s planning forecast pegs current data center load at about 1,000 megawatts statewide, roughly 2 percent of peak demand, and staff expect growth. Cooling water, noise, backup generation, and traffic during buildout are the questions the supervisors will press.
The valley heat shimmered across I‑5 by midafternoon.
What officials and companies are weighing
The White House has promoted a ratepayer pledge from utilities and tech firms that’s meant to keep household bills from rising as data centers scale up. Industry groups say new facilities will add their own generation or storage so they don’t pile onto stressed grids. Critics point out that transmission upgrades still cost money, and that on‑site power at a server farm doesn’t always run when the grid needs it most, so the siting details matter.
E&E’s piece notes California’s advantage, despite national policy swings. Solar growth here is real, grid constraints are specific, and rural parcels exist where environmental reviews are known quantities. For Fresno and Kings counties, that means proposals will likely target land already mapped for energy and within reach of gates and substations. The boardrooms will want firm answers on water supply, diesel backup limits, and how much local hiring gets written into agreements, because they should.
Rows of panels throwing a hard glare outside Huron told the story long before the study did.
Central Valley AI is produced by the CVAI Newsdesk team and developed by Kaweah Tech, a regional firm that builds, deploys, and integrates AI solutions for businesses across California's Central Valley.
Source
https://www.eenews.net/articles/how-solar-power-could-aid-trumps-ai-agenda/
