Additionally, an EV performing the same trip in August but charging under the Workplace Morning charging profile was found to emit nearly 14% more CO2 per kilometer when using hourly grid emissions profiles instead of annual averages and nearly 10% more instead of monthly averages. This variance is even more pronounced under the marginal scenarios, though not always with the same directionality. Assuming the Hourly Marginal Mix tends to reduce the per-kilometer CO2 emissions of an EV charged with the Workplace Morning profile in the summer months, making that charging profile the most attractive in terms of environmental benefit in some cases. Under Hourly Marginal, Resource X assumptions, the CO2 emitted per kilometer for an EV can vary as much as 58% depending on the time it is charged on a given day. Importantly, almost all simulated EV scenarios realized reduced CO2 emissions per kilometer compared to ICEVs. However, the magnitude of these reductions varies substantially under different emissions assumptions, charging profiles, and seasons. The extreme hourly and seasonal variations in effective emissions rates of EVs found in this study indicate that reliance upon annual or monthly average emissions rates for the modeling of EV environmental benefits is inadequate. Table 3-2 depicts the wide variation in emissions rates from simulation to simulation relative to the ICEV baseline.Average emissions rates at lower resolutions obscure vital information that could otherwise be used to optimize environmental benefits as well as inform policy. Effective communication of hourly or higher resolution of grid CO2 intensity would help the consumer make an informed choice on when to charge their EV to maximize environmental benefits. With the maturation of “smart-charging,” enabling communication between the grid or utility and smart charger units would allow the smart charger to control the rate of charge to minimize effective EV emissions, cannabis drying subject to user-configured constraints involving required time of use, desired battery capacity, and cost.

As expected, the HEV reduced CO2 emissions compared to an ICEV, but it also performed consistently on par or better than many EV scenarios. When annual or monthly average emissions rates are assumed, HEVs already perform better on a per-kilometer basis than EVs beyond a certain distance threshold. These de-facto superiorities of HEVs become less pronounced at certain times when hourly emissions rates are assumed. Failing to understand and incorporate higher-resolution evolutions in grid emissions intensities can lead decision makers to ill-informed conclusions that could be sub-optimal for reducing environmental externalities. It is worth noting that both EVs and grid generating resources are evolving dynamically, placing renewed emphasis on studies that consider environmental impacts during this transition period . There is likely some threshold of EV penetration that will trigger a realignment in marginal emissions trends. As electrical power demand increases at peak charging times when consumers are incentivized to charge their vehicles, marginal resources in addition to those observed in this study, will eventually be required to supply sufficient power. Often, due to the inherent need for dispatch ability, marginal resources are fossil-based or non-renewable in nature. Thus, if additional marginal resources need to be brought online, it could alter grid emissions profiles and lead to shifting environmentally optimal charging periods. Understanding the scaling behaviors of marginal power demand for growing rates of EV adoption will be critical for decision-makers to stay one step ahead of lagging realignments, anticipate them, and communicate optimal charging periods to consumers as well as intelligent infrastructure. An important benefit of the methodology employed in this study is that it can be easily adapted to model additional use cases, charging profiles, emissions profiles, and additional pollutants. To prove the feasibility of such adaptations, SO2 and NOx grid emissions were simulated for the same charging profiles as CO2 for the Suburban Errands trip in August and October.

For these simulations, there were assumed to be no SO2 emissions for the ICEV and HEV baselines. NOx emissions for the ICEV and HEV baselines were calculated using a conversion factor of 0.000167 , as informed by the MoVES model. Figure 3-4 and Figure 3-5 depict pollutants that are greater per kilometer in EVs than ICEVs. SO2 and NOx are examples of pollutants emitted in the production of electrical energy at fossil power plants that are essentially absent from or significantly reduced in vehicular tailpipe emissions. While these additional pollutants should be acknowledged, it is critical to understand the spatial confines of their dispersion. While tailpipe, mobile-source emissions are present anywhere motor vehicles travel, emissions from electricity generation are localized in smaller areas immediately surrounding power plant facilities. Given that power plants are typically located in more rural areas, the per-unit damages from pollutants emitted by electricity generation can be much less. However, there are environmental justice issues inherent in these trade-offs that need to be addressed and explored further.Because of the potential opportunity of vehicle electrification to help decarbonize the transportation sector and improve air quality, the technical findings of this research could have some significant policy implications. To optimize the potential benefits, much greater attention will be required to the incremental difference in EV emissions relative to baseline ICEVs and HEVs. As some of the findings suggest, the individual vehicle and fleet wide improvements may be much lower than expected by some studies as scale-up occurs. However, the findings also provide some suggested means of ensuring that environmental and social improvements can be realized, and at the scales needed. Because this research begins to quantify technical parameters related to both the magnitude and the range of possible emissions impacts as compared to multiple baselines , the study’s findings can be useful for education and awareness by all EV users.

They also have clear implications on policy and public investment, including the urgent need for managed and coordinated charging, and greater attention to resource planning, in terms of generation resources, dispatch decision making, infrastructure funding, and the long-run environmental benefits and impacts for EVs across a range of use cases and time horizons.To investigate the true variability associated with CO2 and other transportation-related vehicle emissions, this study has developed a simulation framework that explores multiple parameters concurrently. The goal has not been to determine with high precision a given case as much as it is to develop a broad comparison among major inputs and factors. In this way, we explore electric vehicles as compared to a baseline case . We explore several driving cycles and charging profiles that represent typical approaches both for residential and workplace charging at various times of the day. And then, we develop various methods for estimating CO2 and other vehicle emissions. As noted, studies that have addressed this previously have often utilized annualized averages to simplify the analysis. In our research review of other tools and dashboards , we confirmed that a very basic algorithm is utilized . We acknowledge such traditional approaches provide a kind of first-order, initial estimation that can be useful to some audiences in some contexts. However, it is imperative to recognize and explain the limitations of accepted approaches, and the risk of relying too heavily on average emissions estimates, as they are highly subject to change in the future, and to variability during the present . In short, new tools and methodologies are needed that can estimate the impact of taking various assumptions for how the grid will meet marginal demands in the near, intermediate and long terms. This transition period from a few million EVs to 100 million EVs will take some time, and environmental impacts will need to be more fully understood. As EV adoption increases and the grid is expanded to meet new demands for electrification, such transition tools and methods can be increasingly valuable to researchers, planners, policymakers, drying cannabis and infrastructure decision-makers. As such, our present work provides much needed additional insight and may be useful to inform 2nd order factors and more complex and integrated guidance. Going a step further by exploring limitations and pursuing additional rough orders of magnitude could have tremendous value for the transportation research community. It would also facilitate a more direct and apples-apples comparison of EVs to other technologies there are substantial shortcomings as penetration rates grow. We conclude with a brief recap. It is clear that at certain very low very modest levels of EV deployment, something like an average assessment of the weighted mix of resources may not be illogical or even inaccurate. It is beyond the scope of this study to determine exactly at what penetration rates things change, but it can be stated that at significant increases in EV charging, in particular at certain hours of the day and months and seasons of the year, the assumption of weighted mixes breaks down. Some preliminary findings from the use cases investigated in the present study reveal that conditions most at risk of yielding higher than desired CO2 emissions rates include afternoon charging at the workplace and early evening charging at the residence. Conversely, it seems that residential overnight charging may, at present, be one of the lowest impact scenarios for EV charging.On the behavior side, it’s clear that managing charging events throughout the 24 hours of the day should merit greater attention. It appears that residential charging may be environmentally preferred compared to workplace charging under certain conditions. This may be an important near-term way to mitigate the unintended effects of higher marginal emissions impacts. This, however, is a simplified observation, since it assumes adequate all-electric range and adequate access to residential EV charging. Neither of these assumptions is necessarily sound at higher penetration rates or given certain economic barriers and social inequities. Furthermore, charging management and behavior alone will likely be inadequate as EV shares grow to levels that push up against current resource capacities and are not yet envisioned fully and accommodated for utility resource plans in the 3-to-10-year horizon. Future study is anticipated to further inform decision-making around such scenarios, including the ability to convert the predominantly historical approach dispatch to a predictive forecasting approach where a 2030 scenario is developed that can better simulate future resources, both fossil, and non-fossil, and how they will be deployed to meet a growing load to support electric transportation. While the use case results are of interest in their own right, it should be noted they are sensitive to the source data for the selected region and approaches taken to integrate electric charging behaviors. While the particular insights may not apply to other regions, it is important to note that the research term also intended to conduct a regional use case as a validation for the methodology. The methodology has been constructed and described in sufficient detail that it can be combined with other data sets and regional attributes for the purpose of adapting it as a decision support tools, beyond the particular region selected herein. Thus, the methodology is shown to be scalable and more broadly applied to other time horizons and regions, and leverage other data sets.This research study focuses on a few specific vehicle types around light-duty vehicle uses in order to develop a comparative framework with a manageable technical scope. While additional use cases and extensions are suggested as areas of future work, the frameworks and simulation-based comparisons are extremely generalizable and extendable because of the organizational approach. A lot of attention to detail has been paid to the development of physics-based vehicle models, including consideration of architectures, power train, overall accessory loads, and sensitivity to drive cycle and external ambient temperatures. Similar attention to detail has been paid to developing practical and representative EV charging profiles, reasonable mapping of standard drive cycles to real-world trips and travel behavior, and high-fidelity analyses of existing grid dispatch methods based on real-world data. A primary contribution of this effort is therefore the integration of each of the individual subsystems and independent data sources, including hourly grid characteristics, toward a novel understanding of the complex impacts of vehicle electrification. In short, the team has successfully achieved its chief aim of laying the groundwork for a more complete understanding of these results at scale. Regarding EV charging behavior, we have considered data from multiple concurrent sources which provides insight into when people are most likely to be charging their EVs today.