| Input | Assumption |
| Technologies evaluated | PV, battery storage |
| Model objective | Minimize life cycle costs (electricity purchased from the grid, as well as cost of purchase, installation, and O&M for a renewable system) |
| Ownership model | Third-party ownership |
| Analysis period | 25 years (standard analysis period and conservative life estimate) |
| Inflation rate (nominal) | 2.5% per 2022 NREL Annual Technology Baseline |
| Discount rate (nominal) | Host: 5.64% Developer: 5.64% |
| Electricity cost escalation rate (nominal) | 1.9%/year per U.S. Energy Information Administration for U.S. commercial electricity, 25-year analysis period |
| Tax rates | Host: 26% Developer: 26% |
| Net metering | Excluded from this analysis |
Third-party ownership is modeled in REopt as described in Figure 22. With third-party financing, the developer bears the capital and operational costs for any installed DERs and is assumed to monetize incentives such as investment tax credit (ITC) and modified accelerated cost recovery schedule (MACRS). The installed system provides utility cost savings to the system host (vertiport). In return, the system host pays an annualized “energy services” payment to the system developer.
Figure 20. Third-party ownership model as implemented in REopt.
## 2.3.4 Technology Considerations
This analysis considers the following DERs to serve existing site loads plus eVTOL charging infrastructure at sites in the scope of the analysis. Technical details for each technology are described in Anderson et al. (2021).
## Solar PV
REopt considers south-facing standard rooftop PV with a 10° tilt angle and fixed axis. Each site’s corresponding coordinates were used to determine the solar resource from NREL’s National Solar Resource Database, which drives the energy production of a PV system. Solar PV is AC-coupled with the loads, which means all DC power output of PV gets converted to AC and takes a 4% efficiency hit in the conversion process. PV inverter clipping was modeled with a DC-AC ratio of 1.2. REopt default capital cost and operational cost values were used for PV. Due to third-party ownership of DERs, the PV system was modeled with a 30% ITC, 5-year MACRS, and 80% bonus depreciation, per updates from U.S. Congress (2022). These incentives are meant to come together to reduce the federal tax liability of the system developer. These assumptions and relevant data sources are described in Table 3. Per the latest FAA guidelines, airports are required to measure the impact of hazardous glint and glare from solar projects developed at airports on air traffic control operations and pilots (Federal Aviation Administration, DOT, 2021). This analysis does not consider the impact of glint and glare of any identified solar PV systems on airport operations.
Table 3. Solar PV Assumptions in REopt
| Input | Assumption |
| System type | Roof mount, fixed axis, standard module |
| Technology resource | Typical meteorological year (TMY) weather data from the National Solar Resource Database a |
| Inverter efficiency | 96% |
| Installed capacity density | 10 DC watts/ft² (O.01 kW/ft2); PV capacity that fits in roof top area |
| Tilt | Roof mount, 10° |
| Azimuth | 180°(south-facing) |
| DC-to-AC size ratio | 1.2 |
| System capital cost | $1,592/kW per 2022 NREL Annual Technology Baseline |
| O&M cost | $17/kW/year per NREL Annual Technology Baseline |
| Incentives b | 30%ITC |
| MACRS: 5-year depreciation with 80% bonus MACRS |
a https://pvwatts.nrel.gov/version\_6.php, https://nsrdb.nrel.gov/data-sets/tmy, https://sam.nrel.gov/weather-data.html
b REopt assumes all cost savings from incentives are passed through developer to system host.
## BESS
A lithium-ion BESS with 89.9% round-trip AC-AC efficiency was considered for the analysis. A battery’s performance can degrade if its state of charge drops too low (<10%) or if it is kept too high (>90%), making some battery capacity unusable. Therefore, REopt modeled the battery to hold at least 20% minimum state of charge to mathematically allow utilization of only 80% battery capacity. BESS capital cost was \$388/kWh, and cost of associated power electronics to charge and discharge the battery was \$775/kW. Because the performance of a battery degrades over its lifetime, REopt schedules a battery replacement in the 10th year of the analysis period where energy storage capacity costs \$220/kWh and power electronics cost \$440/kW. The replacement BESS is anticipated to last the remaining 15 years of the analysis period. The BESS is allowed to charge from the grid as well as any on-site PV. Additionally, BESS was modeled as eligible for 30% ITC and a 7-year MACRS depreciation schedule with no bonus depreciation, per updates from U.S. Congress (2022).
Table 4. BESS Assumptions in REopt
| Input | Assumption |
| Battery type | Lithium-ion |
| AC-AC round-trip efficiency | 89.9% (97.5% internal, 96% inverter, 96% rectifier) |
| Minimum state of charge | 20% (battery charge is managed to stay above this minimum) |
| Capital costs | $775/kW + $388/kWh based on Wood Mackenzie U.S. Energy Storage Monitor |
| Replacement costs (year 10) a | $440/kW + $220/kWh based on Wood Mackenzie U.S. Energy Storage Monitor |
| Allow the utility grid and any on-site PV Yes to charge the battery? | |
| Incentives b | 30% ITC |
| MACRS: 7-year depreciation with 0% bonus MACRS |
a Default one-time battery replacement occurs during year 10 of the analysis. REopt can also model replacement and augmentation battery replacement strategies.
b REopt assumes all cost savings from incentives are passed through developer to system host.
## 2.3.5 REopt Analysis Process and Scenarios
The process of analyzing site data and transforming them to usable REopt inputs is summarized in Figure 23.
Figure 21. Steps taken as part of the REopt analysis for sites.
First, electricity costs and consumption data were collected from provided electric bills for analysis to determine the electric utilities and rate schedule servicing each site. Because hourly site operational load profile was unavailable for all three sites, DOE’s CRB load profiles were used to simulate the operational load profiles. Next, the following scenarios were executed for these sites, which are also summarized in Table 5:
BAU (Scenario 1): REopt was run using the simulated load profiles and electric tariff2 in an attempt to replicate the billed electricity consumption and costs in REopt. The purpose of this step is to validate the simulated load profile and electric tariff inputs and ensure no components are missing. This validation step is done by finding the percent difference between annual
electric bill costs per billing data and REopt results and reiterating over input data if the percent difference is higher than a certain threshold.
BAU With EVSE Charging (BAU+CH) (Scenario 2): This scenario considers the impact of EVSE load additions on utility costs. Results from this scenario quantify the implications of adding EVSE loads to site operational loads without any new on-site DER capacity.
Optimal (Scenario 3): This scenario considers the costs and benefits of including solar PV and BESS in conjunction with additional load of EVSE.
Optimal\_Restricted (Scenario 4): This is similar to the Optimal scenario (Scenario 3) but restricts the on-site solar PV size to rooftop area or existing solar PV system sizes.
Resilience: Results from Optimal and Optimal\_Restricted scenarios were used for resilience assessments for these sites. For HHI, the existing on-site generator was also included in the analysis to reflect the impact of on-site dispatchable DERs on resilience per assumptions described in Table 5.
Table 5. Summary of Scenarios Evaluated in REopt for Each Site
| # | Scenario | Purpose | Considers Existing Load | Considers EVSE Loads | Considers Cost-Optimal PV + BESS |
| １BAU | | Site continues normal operations without any changes to its electric loads | | | |
| 2BAU+CH | Site adds eVTOL charging loads to its operational load profile and continues operations without any new DERs | √ | | |
| 3 | Optimal | Site adds eVTOL charging loads to its operational load profile and continues operations with an option of adding on-site solar PV, BESS | | 」 | |
| | 4 Optimal_restricted Site adds eVTOL charging loads to its operational load profile and continues operations with an option of adding on-site solar PV, BESS constrained by rooftop/land area restrictions | √ | | |
| 5 | Resilience runs | Assessing the resilience of cost- optimal systems to withstand a 4- hour outage | + | | |
## 2.3.6 Site Electric Load and Electric Tariff Summary
This section summarizes the electric tariffs (demand and energy charges) along with parameters used to synthesize the electric load profiles at sites. Due to the absence of interval data (or utilityprovided site-specific electric load data), a blend of DOE’s CRB load profiles was used to simulate site operational loads after reviewing services offered at sites via websites and Google Maps.3 This section also provides a breakdown of the proportion of various load profiles used for each site.
Teterboro Signature Flight Support Building (TEB)
• Electric utilities: Public Service Electric and Gas Company (PSE&G) and Talen Energy Marketing LLC
o Electricity generation service is provided by Talen. Because no rate plan information was available, average energy rates (\$/kWh) for each month were calculated and used for in inputs to REopt (Figure 24).
o Electricity distribution service is provided under PSE&G’s Large Power and Lighting – Secondary (LPLS) rate schedule (Table 6).
No interval data were available for TEB, so the load profile was simulated using the following “blend” of DOE CRB load profiles: 40% medium office and 10% warehouse. Additionally, 50% of the site’s load was modeled as flat load. Monthly consumption in kilowatt-hours extracted from billing data was used to scale up load profile appropriately.
Table 6. PSE&G LPLS Rate Schedule Breakdown
| Electric Tariff Component | t Charge | Applies To |
| Fixed monthly charge | $4.95 | Per month |
| Monthly demand charge | $3.9802/kW | Peak kW measured per month |
| Summer-only demand charge | $9.7113/kW | Peak kW measured in summer months |
| Distribution kWh charges | $0.007789/kWh October-May $0.003052/kWh June-September | Applies to all energy consumed on- site, but varies by month of year |
| Societal benefits charges | $0.009766/kWh | Applies to all energy consumed on- site |
| BGS energy charges | Real-time hourly varying energy charges | PJM Load Weighted Average Residual Metered Load Aggregate Locational Marginal Prices for the Public Service Transmission Zone |
| BGS energy charge adjustments | Adjusted for losses | Adjusted for hourly losses and adjusted to remove the mean hourly PJM marginal losses of 0.7164% |
| Ancillary service charges | $0.006/kWh | Applies to energy rates in all hours. This value also includes PJM's administrative charges and should be adjusted to hourly losses and mean hourly PJM losses of 0.7164%. |
| Monthly BGS capacity and transmission charges | $21.1552/kW summer/winter coincident peak | Meant for PSE&G to recover costs of operating and maintaining electricity generation services to service its area/zone |
Figure 24. Plot of hourly energy charges used as HHI’s energy rate.
Two additional sites (HRHC and AtlantiCare), which were included earlier in the report, were not considered for a REopt analysis. HRHC’s existing power consumption is higher than the anticipated peak EVSE charging loads, which means grid impact would be minimal. AtlantiCare Hospital was not considered for REopt modeling because no EVSE chargers can be installed onsite due to it being a hospital.
## 2.4 Estimating Greenhouse Gas Emissions
The methodology for estimating GHG emissions resulting from total operational electricity consumption at the facilities included in this study adheres to the accounting framework provided in The GHG Protocol for Project Accounting (Greenhouse Gas Protocol 2005) and the supplemental guidelines provided in the Guidelines for Quantifying GHG Reductions from Grid-Connected Electricity Projects (Greenhouse Gas Protocol 2007). Key elements of this accounting framework include:
• Defining the boundary of the assessment
• Identifying a baseline scenario and estimating baseline emissions
• Identifying project scenarios and estimating the emissions for each.
The three primary GHGs included in this assessment are carbon dioxide (CO2), methane (CH4), and nitrous oxide $\left( \mathrm { N } _ { 2 } \mathrm { O } \right)$ , which are converted into a common reporting unit called “carbon dioxide equivalent” $\left( \mathrm { C O } _ { 2 } \mathrm { e } \right)$ using each gas’s global warming potential. As such, results will be reported in units of $\mathrm { C O } _ { 2 } \mathrm { e }$
## 2.4.1 Defining the Boundary of the GHG Assessment
The evaluation of GHG emissions for this study is limited to the facility electricity consumption at sites identified in Section 2.1.1 and selected for eVTOL assessment (ACY, TEB, and HHI). GHG emissions are estimated using electricity data provided by these sites and modeled for project scenarios along with location-specific emissions factors for grid-sourced electricity. To align with other model outputs and available emissions factor data, the temporal span of this GHG emissions analysis is 2024 through 2050, with calculations biennially from 2024–2030 and every 5 years from 2030–2050.
## 2.4.2 Analysis Scenarios
For continuity with other analysis completed by NREL, the outputs of the REopt models are used as inputs for the GHG emissions calculations. The scenarios described in Table 5 are also carried forward to the GHG emissions analysis.
## 2.4.2.1 Baseline Scenario
Quantifying a projection of the estimated change in GHG emissions from projects under consideration at these sites involves comparing the emissions modeled in project scenarios with a baseline scenario of what GHG emissions would be generated in the absence of project scenarios over the same time period (i.e., “business as usual”). As such, one of the most critical elements of the GHG emissions assessment is deriving a reasonable and accurate estimate of baseline emissions.
For this assessment, the baseline emissions scenario relies on three primary factors:
1. Historical metered electricity consumption at each site.
These data are used to determine the base year for the assessment at each site: ACY (2021), TEB (2021), and HHI (2021/2022).
2. Assumed load changes at each site over the 10-year assessment period (independent of load changes modeled in project scenarios).
FAA has advised NREL to assume zero organic load growth/decline over the assessment period.
3. Projected changes in the emissions intensity of the regional power grid surrounding the sites.
Annual emissions factors for grid electricity in New Jersey from the NREL 2022 Cambium data are used to estimate future emissions associated with electricity consumed from the power grid at the assessed FAA sites. These emissions factors are annual average values that represent the CO2e emissions resulting from power generation in a given geographic region during a given year.
## 2.4.2.2 Project Scenarios
This assessment evaluates projected GHG emissions resulting from three project scenarios in comparison with the baseline scenario:
Increased electricity demand for eVTOL aircraft charging loads at the identified sites, in alignment with outputs from BAU+CH from the REopt model.
Demand for grid electricity after the installation of cost-optimal on-site renewable energy generation system(s) that supply the electricity needs at each site, with the assumed increase in electrical charging demand in the BAU+CH scenario.
Demand for grid electricity after the installation of on-site renewable energy generation system(s) that meet the cost-optimal and site roof area constraints in the REopt model Optimal\_Restricted scenario (if applicable).
## 2.4.3 GHG Emissions Accounting Methods
Two different approaches to GHG emissions accounting are used in this analysis to express the way the project scenarios are estimated to influence GHG emissions: (1) attributional accounting and (2) consequential accounting. Both methods are widely accepted and useful in their own way, but it is important to note that they are not meant to be combined or compared.
## 2.4.3.1 Attributional Accounting
Attributional accounting of GHG emissions is used to assign ownership or responsibility to a given organization or entity for the emissions that are caused by their activities. This method is applied in the development of organizational GHG emissions inventories and can be used to express the emissions “footprint” associated with a certain activity.
To calculate emissions using this method, activity data (i.e., kilowatt-hours of electricity consumption) are gathered for a given time period and typically multiplied by an emissions factor associated with that activity (i.e., kilograms of CO2e per kilowatt-hour of grid electricity). To most accurately attribute the emissions to the entity being evaluated, the activity data must be actual measured data or reliable projections/forecasts of expected measured data in the future. In other words, the activity data are built “from the ground up” to account for the total level of activity carried out by the entity during the assessment period.
In this analysis, attributional accounting is used to estimate the GHG emissions that would be caused by electricity consumption at each of the included FAA sites from 2024 through 2050. These emissions would be attributable to (i.e., “owned by”) each site based on the amount of electricity consumed and the source(s) that supply that electricity. The calculation is simply:
## ???????? ???? ???????????????????????????? = ???????????????????????????????????????????? ???? ???? ???? ???? ???? ???? ∗ ???? ???????????????????????????? ???? ???? ????
The sources of electricity in this analysis include the regional power grid and on-site solar PV. While the emissions rate of solar PV is zero, the emissions rate of the regional power grid is dependent upon the mix of power plants that supply electricity to New Jersey and the surrounding area during a given period. For the baseline year of this analysis, the U.S. Environmental Protection Agency (EPA) Emissions & Generation Resource Integrated Database (eGRID) 2020 total output emissions factor for New Jersey was used in conjunction with the actual metered electricity consumption at each site. For the future years, the projected grid electricity emissions rates rely on various assumptions about the potential technology and policy development in the future. The NREL 2022 Cambium data (Gagnon, Cowiestoll, and Schwarz 2023) provide modeled forecasts of regional grid emissions rates under several scenarios. For this assessment’s calculation of GHG emissions using the attributional accounting approach, the annual average emissions rate for New Jersey under the mid-case scenario was used.
## 2.4.3.2 Consequential Accounting
Contrary to how attributional accounting of GHG emissions assigns ownership of/responsibility for emissions to a given entity and builds a “footprint” out of actual activity data, the consequential accounting method is used to evaluate the emissions that would be caused or avoided by a particular change in activity without attributing the emissions to any one entity. This method is often used to evaluate hypothetical scenarios in comparison with a counterfactual scenario in which the decision/action being evaluated does not occur.
The calculation of emissions uses the same general formula as the one used in attributional accounting; however, the input data are different in the consequential accounting method. The activity data is the change in activity and the emissions rate is specific to the change in activity. For this assessment of GHG emissions related to electricity consumption, the marginal change in electricity consumption in each scenario must be multiplied by a specific marginal emissions rate.
Because the project scenarios that are assessed involve changes to the FAA sites’ loads, marginal emissions rates are applied to the incremental change in electricity consumption to calculate the estimated change in GHG emissions compared to the baseline scenarios. Marginal emissions factors differ from annual average emissions factors in that they represent the emissions generated by power plants that are forced to operate in response to marginal changes in demand on the power grid. These power plants are typically less economical, less efficient, and more polluting than other power plants and are therefore dispatched by power grid operators only
when needed. These emissions rates are also referred to as “short-run marginal emissions rates” (SRMERs) because the grid is responding to marginal load changes it has not planned for.
Alternatively, long-run marginal emissions rates (LRMERs) can be used to analyze the emissions resulting from sustained long-term changes in electricity demand (e.g., projects that increase site power loads, energy efficiency projects, on-site power generation) that result in structural changes to the power grid. In other terms, the LRMER represents a marginal emissions factor for a future in which the grid has built new capacity (or some other structural intervention) in response to the long-term change in demand.
The NREL Cambium data sets contain modeled SRMERs and LRMERs for a range of possible futures of the U.S. electricity sector through 2050. This analysis utilizes mid-case scenarios for both SRMERs and LRMERs to estimate future emissions changes in the project scenarios. In alignment with recommended uses of SRMERs and LRMERs by NREL Cambium authors, this analysis for FAA assumes SRMERs for 2024–2028 and LRMERs for 2030–2050.
## 2.5 Hazards and Risk Analysis
## 2.5.1 Electric Vehicle Supply Equipment Hazards
EVSE infrastructure faces natural, human, and technological hazards—with significant overlap between each in many scenarios. Natural hazards derive from climate- and weather-induced events or seismic activity. Human hazards may be unintentional errors or intentional attacks to critical infrastructure. Technological hazards are most easily understood when delineating the effects of human or natural hazards between the electrical distribution system and the energy storage system.
Natural hazards vary between locations due to topology and geographic location. Unlike climaterelated hazards, seismic risks must be addressed through the National Electrical Code for any relevant equipment or structures. Warm and humid coastal regions are more prone to flooding, hurricanes, and tropical storms, unlike other regions that experience yearly snowfall at higher altitudes, and also can vary from regions with high tornado incidences or seismic activity. Coastal regions must also consider saltwater corrosion, water table location for any underground structures, and high-strength winds on aboveground electrical infrastructure. Regions with high tornado activity, often not the same regions with high tropical storm activity, must also consider infrastructure reinforcements to survive high winds. Regions with extreme temperatures must comply with the manufacturer-suggested operating conditions to achieve optimal performance. Severely cold conditions may decrease battery discharge rates and in turn alter vertiport operations or aircraft scheduling. Snow removal, either with heavy equipment or MgCl for deicing, also poses collision, corrosion, and fire risks if any unintentional operator errors occur.
Human hazards must be mitigated in ways that minimize harm to the electrical infrastructure as well as the operating personnel. As in the case of snow clearing or de-icing in colder climates, a human hazard can be operator error of heavy equipment when removing snow, eVTOL operations, or the potential electrocution or arc flash associated with the damaged electrical equipment. In severe cases, fires and explosions can occur when electrical equipment is damaged. Thus, human hazards exist alone or in tandem with natural hazards that need to be alleviated.
When identifying technological hazards, most risk is inherently present with the equipment in place and can be heightened as an effect of natural or human hazards. The transformers, conductors, connectors, and other equipment associated with the distribution system and charging equipment all have electrical hazards—shock, arc flash, electrocution, and fire—that are present all the time. Faults or damage to the system can be induced by an outside agent (natural or human hazards) or degradation over time. In both situations, the risks and their severity must be mitigated to avoid cascading system damage and/or harm to the operating personnel. A battery storage system poses electrical, thermal, and chemical hazards of its own. Battery management systems can monitor the risk of overcharging/discharging, thermal runaway, overheating, and electrode destabilization. In the case that there are catastrophic events occurring in or around a BESS, many National Fire Protection Association regulations provide guidelines for handling and suppressing any fire or explosive hazards. Additional details are provided in the separate report including hazards associated with eVTOL charging infrastructure by Rane et al (2023).
## 2.5.2 Cybersecurity Evaluation
Associated with intentional human hazards, the risk of cyberattacks on critical electrical infrastructure is brought about by physical means or by remotely deploying malicious malware that results in hindered or terminated operations. These types of attacks can affect not only the charging infrastructure at the site, but also the aircraft and local grid if the control system or charging connectors are compromised. Thus, supply chain and network access management are key focal points to provide holistic charging infrastructure protection. For example, the EVSE communications network at the site between charging infrastructure and central controllers is often interconnected through Open Charge Point Protocol, and if not implemented securely, various attack angles are left vulnerable (Markel and Sanghvi 2022). With significant overlap with vehicle charging systems, many of the lessons learned and best practices should continue to be implemented in the aviation industry to achieve seamless, integral, and functional cybersecurity.
## 2.6 Estimating Economic and Job Impacts
For this modeling work, the required charging infrastructure for eVTOL aircraft in New Jersey and New York is expected to create local jobs in multiple sectors, as well as stimulate the local economy of both states. To estimate the economic impacts of these ground infrastructure projects, we use an input–output (IO) model, one of the most commonly used and straightforward methods for estimating economywide impacts due to a change in regional demand (e.g., a new construction project). IO models are composed of several equations reflecting each sector’s production function (i.e., their production “recipes”) that together show how these sectors interact in a region. This is accomplished by modeling the structure of an economy as a network of sectors buying and selling to one another, local households, government, and external markets. Its results reflect the supply chain’s responses and the total macro-level impacts from changes in demand for goods or services in a region.
IO models capture “multiplier effects” in the economy that arise from sectoral interaction across different supply chains. A change in demand acts as a drop of water in a still pond creating waves (“ripples”) that are large in the beginning and fade over time. The first wave is the largest: It represents the change in the purchasing pattern of directly affected firms. To meet the demand for goods and services of this first wave, suppliers of these firms need to change their purchasing patterns, generating a secondary smaller ripple. Then, suppliers of these suppliers also need to change their purchasing patterns, and so on, generating a sequence of fading ripples in the economy. The sum of all ripple effects is the total economic impact of that initial change in demand.
A few examples of IO analysis in energy-related projects are Jeffers et al. (2022), who used the IO framework to estimate the economic effects of the deployment of zero-emission buses (battery electric and fuel cell) and related infrastructure in California; Navigant (2020), who estimated the impacts of a replacement project for Astoria Gas Turbine Power in New York state; and Lahr, Coughlin, and Felder (2010), who analyzed the statewide impacts of natural gas infrastructure projects and residential solar lease programs in New Jersey.
## 2.6.1 Approach and Data
Construction and operation costs from different charging infrastructure projects will be separated by state to estimate temporary and permanent economic effects. From the total amount of investments in a region, part of the goods and services are provided by companies outside the state (nonlocal purchases) and do not generate local impacts (Figure 27). Those are excluded from the analysis using state-specific regional purchase coefficients that determine the percentage of local purchases for each good/service in the model (these vary by year due to the evolving regional economic structure). The amount of local purchases is then used to introduce a demand shock in the model and to determine the total economic impact including jobs created in each state due to these investments.
Figure 25. Economic impact analysis overview.
Impacts can be classified as direct (immediate economic impact from the change in demand), indirect (from supply chain linkages), or induced (resulting from the spending of wages/salaries by workers) by region. The results include total temporary and permanent jobs created, changes in gross regional product (value added), and sectoral output:
• Jobs are defined as the sum of full-time and part-time workers (measured in full-timeequivalent jobs) employed at the place of business. All jobs supported by local companies are accounted for, including those of out-of-state commuters (who might spend part of their wages outside the state).
• Output represents the value of production and includes all sales and purchases of a particular sector.
Value added represents the wealth generated by an economic activity and includes compensation of employees (wages and benefits), profit-type income, property income, and taxes on production.
Gross regional product measures the monetary value of final goods and services produced in a region and is the sum of all sectors’ value added.
For this analysis we used IO data from the U.S. Environmentally-Extended Input-Output (USEEIO) project (EPA 2023a). USEEIO is a publicly available platform developed and maintained by the EPA that provides IO tables connected to a series of socioeconomic, environmental, and resource use metrics (EPA 2023b). Its stateior (EPA 2023a) extension provides data for each U.S. state at the U.S. Bureau of Economic Analysis Summary Level (71- sector disaggregation) for 2012–2020. From this data set, we used the 2019 multiregional IO table for New Jersey, which has two interconnected regions: New Jersey and the rest of the United States. Data for 2020 were not considered due to the transient effects of the COVID-19 pandemic on the economic structure of these states.
Infrastructure data for each site were obtained from the previous analyses and represent installed costs in purchasing prices (i.e., include transportation and wholesale/retail margins). Each itemized expense was broken down into direct costs (labor, materials, equipment rental, and subcontract), indirect costs (construction management, engineering, startup, and permitting), and contingency costs according to the assumptions used in Burns and McDonnell (2019) and shown in Table 10. Next, each cost was allocated to an economic sector in the model according to the Bureau of Economic Analysis summary-level disaggregation. Costs were deflated to 2012 prices, and purchasing values were separated into producer costs, transportation costs, and wholesale/retail margins using the 2012 Benchmark Input-Output Margins table (BEA 2018). Finally, the adjusted costs were deflated to 2019 prices to match the year of the USEEIO IO tables.
Table 10. Expense Allocation Assumptions
| Site Name | Provide Site Constraints | Annual Usage (MWh) | Annual Cost ($) | Utilitya | Rate | Summary of Provided Interval Data | % Difference in Billed vs. REopt BAU Year1Bil b |
| ACY Atlantic City, NJ (combine FIS, | Land: 0.0 sq. ft. Rooftop: 31,100 sq. ft. No existing | 5,588 MWh | $707,300 | Atlantic City Electric (ACE) [D] | Annual General Service - Secondary (AGS-S) | | 1.24% |
| terminal, and parking garage loads) | DERS No on-site generator | | | Constellation Energy [G] | Fixed Price Solution, details not provided | Interval data were not | |
| TEB 103 Lindbergh Dr, Teterboro, NJ07608 | Land: unlimited Rooftop: 0 sq ft Existing on-site PV: 640 kW- | 834 MWh (net) $121,800 | | PSE&G [D] | LPLS | provided for these sites. Load profiles were simulated | |
| DC No on-site generator | | | Talen Energy Marketing, LLC [G] | Not provided | using a combination (i.e., blend) of DOE CRB load | 1.7% |
| Land: 0.0 sq. ft. Rooftop: 42,800 sq. ft. No existing on- | 490 MWh | $56,500 | PSE&G [D, G] | General Light and Power Secondary (GLP) | profiles. PSE&G's Basic | -3.9% |
a [D] refers to electricity distribution utility and [G] refers to electricity generation utility.
b Year 1 bill comparison was done using electricity costs from the same time period as available electricity bills. highlights the implications of adding EVSE loads at each of these sites in terms of expected electric utility cost increases and changes in average measured (i.e., averaged over a specific time interval) monthly peak demand and annual energy consumption.
Both ACY and TEB see an approximate 20% increase in their Year 1 electricity bills, which can be largely attributed to increases in demand charges at each of these sites. At HHI, the magnitude of EVSE-related costs is considerably higher than BAU site loads, which results in an approximate 300% increase in Year 1 electricity bills. All sites also register a considerable increase in average measured monthly peak demand, with the largest percent increase noted at HHI. Because the EVSE loads are peaky in nature, they mainly influence the demand charges. Rise in marginal emissions due to charging is between 3.7% and 4.4% for all sites. Annual electricity consumption also increases about 5% for ACY and TEB, and 35% for HHI.
Table 23. A Comparison of BAU and BAU+CH Scenario Shows the Impact of EVSE and Charging Loads on Each Site’s Electric Utility Charges and Electricity Consumption Trends
| Site | Scenario | Optimal PV Size | Optimal BESS Size and Duration | NPV of Savings of Analysis Period | Annual Electricity Consumption | Average Monthly Peak Load | Overnight Capital Costs a | Annualized Payments to Third Party b |
| ACY | BAU | NA | NA | NA | 5,588 MWh | 1,086 kW | NA | NA |
| BAU+CH | NA | NA | NA | 5,835 MWh | 1,927 kW 688 kW | NA $2.37 | NA |
| Optimal | 2,091 kW- DC | 1,004 kW/ 1,224 kWh (~1.25 hours) | $2.89 million | 3,493 MWh | | million | $319,400 |
| Optimal_restricted | 311 kW-DC | 771 kW/264 kWh (~20 minutes) | $1.76 million | 5,450 MWh | 1,130 kW | $0.63 million | $88,200 |
| TEB BAU | NA | NA | NA | 1,662 MWh | 170 kW | NA | NA |
| BAU+CH | NA | NA | NA | 1,739 MWh | 995 kW | NA | NA |
| Optimal | 1,628 kW- DCc | 521kW/93 kWh(~10 minutes) | $0.925 million | 614MWh | 139 kW | $1.04 million | $138,800 |
| Optimal_restricted | 640kW-DCc | 410 kW/56 kWh(~10 minutes)i) | $0.109 million | 1,740 MWh | 587kW | $0.181 million | $26,900 |
| BAU | NA | NA | NA | 488.4 | 84 kW | NA | NA |
| HHI | BAU+CH | NA | NA | NA | 662.9 | 979 kW | NA | NA |
| Optimal | 0.0 kW-DC | 821 kW/388 kWh (~30 minutes) | $0.414 million | 686.1 | 166 kW | $787,200 | $62,400 |
| Optimal_restricted | 0.0 kW-DC | 821 kW/388 kWh(~30 | $0.414 milion | 686.1 | 166 kW | $787,200 | $62,400 |
| | minutes) | | | | | |
a Overnight capital costs for recommended cost-optimal systems after incentives.
b Intended to pay for developer’s capital related costs.
c Site has an estimated 640 kW of existing solar PV. The optimal REopt results recommend 988 kW of new PV capacity.
Figure 62 presents a stacked area plot displaying the dispatch for PV and BESS to meet the site load at ACY under the restricted rooftop area scenario. This figure highlights how BESS is discharging to meet the charging demand during the afternoon hours and charging during offpeak hours and reducing overall peak consumption from the grid. Note that the grid is supplying the BESS charging load during off-peak hours.
Figure 60. Example of solar PV and BESS dispatch to meet site loads at ACY under the restricted rooftop area scenario.
Figure 63, Figure 64, and Figure 65 present the changes in measured monthly peak demand for all three sites due to the addition of charging loads, as well as the addition of any REoptrecommended DERs. Observe that for all the sites, the BAU+CH scenario has higher monthly peak demand compared to other scenarios.
Figure 61. Monthly peak demand values at TEB for various scenarios.
As shown in Figure 63 for TEB, the monthly peak demand for the BAU+CH case increases 5 times compared to the case for BAU, while for cost-optimal restricted case, it decreases by 40% compared to the BAU+CH case.
Figure 62. Monthly peak demand values at ACY for various scenarios.
As shown in Figure 64 for ACY, the monthly peak demand for the BAU+CH case increases around 2 times compared to the case for BAU, while for cost-optimal restricted case, it decreases back to being close to the BAU case.
Figure 63. Monthly peak demand values at HHI for various scenarios.
As shown in Figure 65 for HHI, the monthly peak demand for the BAU+CH case increases approximately 10 times compared to the case for BAU, while for cost-optimal restricted case, it decreases by approximately 80% compared to the BAU+CH case.
Table 26, Table 27, Table 28, Table 29, and Table 30 present the detailed financial and electric tariff results for all three sites for both optimal and optimal restricted scenarios.
Table 26. Comparison of Financial Results for ACY’s BAU+CH and Optimal Scenarios