Revealed by a study done by a team from Boston University and the University of California, Berkeley
The study, revealed in Energy Policy, reveals how a team from Boston University and the University of California, Berkeley came to some interesting predictions for electricity production and usage by 2050. By utilizing an expanded Kaya identity framework, the team modeled vehicle stock, energy intensity, and vehicle miles traveled, progressively considering the effects of each of the three revolutions - electrification, autonomy, and sharing. The results project that electricity usage from light-duty vehicle (LDV) transport is likely to rise to the 570–1140 TWh range - 13–26%, respectively - of total electricity demand by the year 2050.
Furthermore, the researchers project that the decrease in LDV greenhouse gas emissions could go all the way up to 80% - depending on the pace at which the electric sector decarbonizes. Decarbonization means that newer, cleaner technologies are employed in the production of electricity - those that result in lowered emissions of greenhouse gases or a complete removal of those in the production process. In turn, a bigger an emphasis needs to be put on the renewable energy sector to improve the rates at which the decarbonization happens.
When a conventional Kaya identity is used when forecasting transport energy and emissions forecasting, it considers vehicle miles traveled (VMT) and the average energy density in kWh/mile to calculate the total energy usage.
This approach is useful when models predicting aggregate total travel are stable enough to perform well over long forecast periods and fleetwide average energy intensity can also be projected with confidence. Unfortunately, few of the conditions that make this aggregate approach useful hold today. Traditional forecasts of aggregate VMT began losing accuracy following the Great Recession of 2008, well before the sharing and autonomy disruptions had much of an effect. Autonomy is expected to greatly disrupt these forecasts, possibly along with new preferences for walkable urbanism, ridesharing, and other changes.
There is no silver bullet to address these difficulties, but we gain a little tractability with a conceptual framework based on an expanded identity of the following form.
Σi = Φl,t
where the stock in year t of EVs of a motorized vehicle type i is denoted by ĸi,t, vi,t is the average miles traveled by that vehicle type in year t, and ei,t is the average electricity use of the vehicle type i per mile traveled during year t, which we refer to as electric intensity (EI).
… The uncertainties and potential errors in this approach remain large, but at least they are disaggregated within a more flexible and transparent framework. For example, this framework allows us to treat electric non-autonomous and autonomous cars and light trucks all separately, adjusting use intensity for vehicle type as well as allowing the composition of the fleet to migrate from one type to another.—Fox-Penner et al.
Even though the electric and autonomous passenger vehicles will represent a large and important new driver of electricity demand, the energy sector should be able to satisfy the increased demand - as based on the team's modeling. Furthermore, the team suggests that the cornerstone of transport decarbonization policy in the near term should remain within the rapid and complete transport electrification throughout all sectors - both private and public.
The only caveat - according to the result of the modeling - might be the increase in driving mileage, urban and suburban sprawl, alongside increased traffic congestion. This can be curbed by the long-term policy and incentivization of potential energy efficiency improvements through both better system management and the improvements set forth by the accident-free vehicle fleet.
Source: Green Car Congress
Resources: Peter Fox-Penner, Will Gorman, Jennifer Hatch (2018) “Long-term US transportation electricity use considering the effect of autonomous-vehicles: Estimates & policy observations,” Energy Policy Volume 122, Pages 203-213 doi: 10.1016/j.enpol.2018.07.033.