Decarbonizing travel: shared self-driving fleets and extended-range EVs
Two ways to take carbon out of car travel: share the cars, or put a bigger battery in them. What does each one buy?
Every number here comes from published SEAR Lab papers, listed at the bottom with the table or figure each one comes from. Where a paper has an error, the demo shows the corrected value and says so. Cite the papers, not this page.
A. Austin, 2015–2050: what if shared self-driving cars carry most of the driving?
Pick one of the ten futures the paper modeled. Compare the fleet, the power plants and the CO₂ with the private-cars-only future under the same carbon policy.
Show the numbers as a table
Scenario 3 of Table 2. The paper shows its results as charts; the year-by-year values here come from the ten scenarios re-solved in 2026 with the paper's public model, which reproduces the electricity-mix figures the paper states (natural gas 80% of generation in 2035 without a tax, 79.6% here; peaking at 56% in 2021 with it, 55.6% here). Build years and storage schedules are one optimal solution among several; totals are not.
B. The United States: how much driving could an extended-range EV do on battery?
An extended-range EV always drives on its electric motor; a small gasoline generator recharges the battery when it runs out. Slide the electric range and see what share of 2023 U.S. driving it covers, how much battery that takes, and how much CO₂ it saves.
Share of U.S. miles on electricity, against the battery it takes to convert every car
CO₂ saved each year against an all-gasoline fleet
Show the numbers as a table
- The 50-mile battery is 13.7 kWh in the abstract and 13.9 kWh in Table 6. 50 miles at 3.6 miles per kWh is 13.9 kWh, which the demo uses.
- CO₂ saved at 150 miles appears as 679 (introduction), 679.7 (Table 5) and 680 Mt (conclusions). The demo uses Table 5.
- Table 6's fleet battery capacities are about 10% lower than battery size times the 286.1 million vehicles in Table 4 (13.9 kWh × 286.1 million = 4.0 TWh, printed 3.6; the all-electric car's 83.3 kWh gives 23.8 TWh, printed 21.5). They correspond to about 258 million vehicles. The demo shows the printed capacities; the comparison between ranges is unaffected.
- Table 4 lists fuel economy as 18/26.4/36 mpg and grid intensity as 125/348/714 g CO₂/kWh under “Worst/Average/Best”, but Table 5 and the text use them the other way round (the worst case has efficient gasoline cars and a dirtier grid: 36 mpg, 714 g/kWh). The demo follows Table 5.
What this shows
Both studies ask how much carbon comes out of car travel when the cars change. In Austin, the shared fleet's biggest effect is on numbers: a car that is shared is driven all day, so the city needs far fewer of them (about 430,000 instead of 1.08 million in 2050 in the central case). Because each shared car covers so many miles, cheap-to-run battery-electric cars pay off quickly, and the shared fleet electrifies years ahead of private cars. Even if every private mile it replaces becomes two shared miles, total cost and cumulative CO₂ stay below the private-cars-only future.
The fleet also changes the power system around it. Letting fleet cars charge in the middle of the day, when solar power is plentiful, is worth 3.6% of total system cost without a carbon tax and 3.5% with one, and it lets more of the fleet go electric. Try switching charging to “Night only” and watch the gasoline and hybrid share of the fleet grow back. The carbon tax mostly changes the power plants: without it natural gas and solar share the grid in 2050; with it, wind replaces the gas.
The national EREV study asks the same question for one car at a time. Most U.S. trips are short, so the first 50 miles of battery cover nearly three-quarters of all miles driven. Each further 25 miles needs about the same extra battery but adds fewer electric miles: from 50 to 100 miles adds 9.8 percentage points, from 100 to 150 only 3.7. Switch the assumptions to “Worst” and the CO₂ saving almost disappears: with very efficient gasoline cars (36 mpg) and a dirtier grid (714 g of CO₂ per kWh), an electric mile is barely cleaner than a gasoline one. The Austin model's answer to that is the other half of this page: a grid that decarbonizes as the cars electrify.
The lab's follow-on study, which adds high-speed rail to shared fleets for the Texas Triangle, is under revision; its results will appear here when it is published.
Behind this demo
Sources
- Jones, E.C., & Leibowicz, B.D. (2019). Contributions of shared autonomous vehicles to climate change mitigation. Transportation Research Part D: Transport and Environment, 72, 279–298. doi:10.1016/j.trd.2019.05.005. Panel A: scenario definitions from Table 2 and Sections 4.1–4.3; stated results from Sections 5.1, 5.2 and 5.4 (pp. 287, 289, 293). Year-by-year values from the paper's public code and re-solved results, sear-labs/sav-osemosys-trd-2019 (results/clean/annual.csv and objective.csv; archived as doi:10.5281/zenodo.22715620).
- Patil, H.V., Kumbhar, A.A., & Jones, E.C., Jr. (2025). Contributions of extended-range electric vehicles (EREVs) to electrified miles, emissions and transportation cost reduction. Energies, 18(24), 6448. doi:10.3390/en18246448 (open access, CC BY 4.0). Panel B: Table 4 (assumptions), Table 5 (electric miles, CO₂ saved) and Table 6 (electric share, battery size, fleet battery capacity).