This is my favorite climate change chart. Japanese monks, aristocrats, and emperors kept meticulous records of cherry blossom festivals for 1,200 years and accidentally built the world's longest climate dataset.

Tomorrow is my last day at EPRI after 13 years as a full-time employee (closer to 20 if you count my internship in 2007). This place has been my second home. I'm proud of what we accomplished together, and I'll continue to be a champion, collaborator, and advocate for EPRI's mission.

Texas is deploying battery storage so fast that it's almost keeping up with the other 49 states... by itself!

Wild figures in two decades of EIA retrospective data. Solar getting it wrong gets all the airplay. But the consistent underestimation of natural gas exports and the speed of coal's collapse are just as striking, even at 3-5 year horizons.

We spent a lot of time building scenarios for offshore wind futures. Still didn't capture "administration pays ~$1B for a French company to abandon its US leases and promise never to build offshore wind here again." New tail risk unlocked.

One thing jumps out on this map: The U.S. is overwhelmingly a natural gas country for home heating today. But look at electricity in the Southeast. For the state outlines, the darker they are, the more heat pumps. The electrification of heating is a Southern story first but is spreading.

$1.96 trillion in clean energy investment in 2025. That's nearly a tripling in seven years. And yet growth is slowing, especially manufacturing investment. The headline number masks a more complicated story underneath.

My article @climateconnections.bsky.social today examines how, despite their polar-opposite politics, California and Texas became leaders in solar, wind, and battery deployment. The most important thing you need to know is that I came up with the headline – I'm very proud of it 😅 🧵 follows [1/8]:

Data centers: 4-5% of U.S. electricity today. By 2030: 9-17%. Those projections vary wildly by state. We built a dashboard so you can see your state's numbers and download the data. Check out the full state-by-state picture here: powering-intelligence.epri.com/dashboard/

New LBNL data: Connecting a clean energy project to the grid now costs an average of $194/kW in non-ISO territories, and that's only for projects that actually finished the process. Active and withdrawn projects pay far more. More here: eta-publications.lbl.gov/sites/defaul...

It's happening!

Here's something that might make policy wonks bristle. Demand subsidies for solar had the weakest effect on innovation and output in China. Production and R&D subsidies drove the biggest gains.

In 2007, Congress created the Hurricane Forecast Improvement Program after Katrina, Rita, and Wilma. Budget over its first decade: ~$250 million total. Value of the forecast improvements it produced: $2 billion per hurricane in reduced damages. The program paid for itself in a single storm.

Texas and California still dominate the discourse, but the bigger story in this map is diffusion. Solar and storage are showing up almost everywhere, no longer a Sun Belt niche. The energy transition is becoming geographically messier, and that's a big deal.

This chart is wild. Virginia is the world's largest data center market. It currently has DCs consuming >25% of its electricity. By 2030, it could have 39-57%, according to our latest Powering Intelligence analysis. No other state is remotely close.

My new commentary in Science recounts the history of stabilization wedges and features the updates to the wedges by Johnson and Staffell. Read the commentary: www.science.org/doi/10.1126/... Read the Wedge 2026: www.science.org/doi/10.1126/...

Wild stat: the "home" county often isn't the one breathing it. Authors estimate that within ~6 hours, ~99% of coal emissions leave the source county.

Modeling Monday 🧵 Data centers are the new load forecasting problem So we updated EPRI's Powering Intelligence methodology to move beyond trend extrapolations and "chip shipment math" and into something planners can actually use: a state-level project pipeline to grid load mapping

Now that our updated data center electricity analysis is out, a few things I see people often get wrong about DC load growth:

New from EPRI: Our updated Powering Intelligence analysis of U.S. data center electricity demand. Data centers could consume 9-17% of U.S. electricity by 2030, up from 4-5% today. Our projections are ~60% higher than last year's. Here's why. 🧵

According to MethaneSAT measurements, the Powder River Basin loses 7.5% of gas production to the atmosphere. More results from MethaneSAT (one year in orbit before failure) available here: www.methanesat.org/project-upda...

EPRI's updated Powering Intelligence data center scenarios drop tomorrow. Here's where our projections stood in 2024. How much higher do you think the new numbers will be?

In 2025, the U.S. added 53 GW of new electricity capacity, the most since 2002. In 2026, it's on track for 86 GW, mostly solar/storage. To put that in perspective, the U.S. is expected to add more capacity in the next two years than it did in an entire decade of the 2000s combined.

Are you a fan of decarbonization scenarios, oblique Samuel Beckett puns, and didgeridoos? If that's you, today is your (improbably) lucky day! My new piece in Nature Climate Change touches on all three in summarizing the important new paper from Hamish Beath and coauthors.

Adaptation is millions of decentralized decisions: where people live/work, what they buy, what they build, what they insure. Policy that improves incentives/info can move those margins at scale. Great new Science article from my colleague Delavane Diaz and others: www.science.org/doi/10.1126/...

Renewables and retail power prices: the "they must be making bills skyrocket" story doesn't show up in the data. New paper by Alec Truax finds ~0.3% lower retail prices for each +1p.p. renewables share (U.S., 2004-2023) with causal ID.

Hot modeling take

China now consumes a third of global electricity, more than twice that of the United States.

Transmission is the "Swiss Army knife" investment: it can lower capex, lower opex, support resource adequacy, connect wind/solar... and still clears benefit-cost >1.5 in most studies.* *If you can permit it.

Modelers when someone calls scenarios "predictions"

Wuthering Heights premieres tomorrow. Fun fact: The Brontë sisters may have died young because their drinking water was literally contaminated by an overfull graveyard next door. #GothicAF

Final EIA-860 is out, and 2025 really did it: 54 GW of new U.S. electric capacity, 96% clean. Solar again carries the offense, and we built more storage in 2025 than the cumulative total through 2023. More and more and more additions... but also record low retirements.

The EV debate in one picture: Even the smallest, most efficient gas car has a higher life-cycle footprint than the largest BEV. Size matters. Power matters. Powertrain matters more. This figure comes from this great paper: www.nature.com/articles/s44...

"The grid is decarbonizing" is not the same as "everyone benefits." New EPRI analysis connects nodal power system modeling with community vulnerability data to track distributional air quality impacts. The who is the point.

We just released an EPRI report on how market drivers and state policy could alter U.S. energy investments through 2050. The headline isn't subtle: planning is now dominated by load growth and build-rate constraints.

Big EPRI study out today on "Energy System Implications of Market Drivers and State Policy" Short version: load growth, state policy, federal incentives, and exports are colliding. These collisions alter build rates, fuel deliverability, and customer bills.

🚨Our new paper led by Dr. @zialyle.bsky.social looked at how climate change poses risks to US drinking water utilities & if their bonds disclose these risks to investors. Utilities serving 67 million people have high risks, but 36% of their bonds don't mention climate: www.nature.com/articles/s43...

New paper on US iron and steel decarbonization pathways from our crew based on the global NetZeroSteel model. Lead by the indomitable fgnxl.bksy.social. NetZeroIndustry.org Free access here to March 22 kwnsfk27.r.eu-west-1.awstrack.me/L0/https:%2F...

Weird. That didn't feel like anything more than a 1.6993571859512.

Can load growth ever lower electricity prices? EPRI's new Win-Win Watts paper: sometimes yes; it depends.

Clean energy cost projections that ignore actual financing terms are wishful thinking. Cool new paper finally gives empirical, national-level WACC for 10 key technologies across 176 countries. www.nature.com/articles/s41...

Two lines tell the story of U.S. coal jobs 📉 U.S. coal production: ~884M tons (1985) → ~513M (2024) 📉 Coal mining employment: ~174k (1985) → ~43k (2024) Automation meant jobs fell faster than production. For scale, Tesla has ~120k U.S. employees... ~3x the mining workforce.

Texas power buildout in one picture: wind (green) laid the foundation, solar (yellow) is the new wave, and energy storage (purple) is clustering around load. This is the grid transition, mapped.

This is an impressive data visualization.

Every circle on this map is a real project with steel (or lithium) in the ground or on the way. Zoom in on the Central Valley and tell me again that nothing ever gets built in California.

New Rhodium data show U.S. economy-wide emissions rose in 2025. To hit net-zero by 2050, the pace of decarbonization would need to roughly triple. We're now closer to 2050 than the first season of Survivor.

Let's check in on clean energy deployment in California and Texas. Texas surpassed California as the leading utility-scale solar state in 2025 with roughly 12.9 GW added last year alone. CA still has more total solar when you include distributed capacity. And lots of batteries everywhere.

Your great-great-grandparents commuted about an hour a day. So did Roman merchants. So do you. No matter how fast we make transportation, people just live farther from work regardless of income. It's called Marchetti's constant, basically Jevons paradox for your morning commute.

Transmission discourse is usually: "Build more lines = cheaper + cleaner." This great new paper by @knittelmit.bsky.social and others is spicier: The big tradeoff is cost vs. extreme weather reliability, where interconnection is an insurance policy.

Carbon isn’t the only thing that matters for carbon capture. New modeling from my colleagues shows amine solvent emissions can create nitrosamines/nitramines and even affect PM2.5. But acid wash can dramatically cut these levels. www.sciencedirect.com/science/arti...