xvEPA
EPA/600/R-25/135 | May 2025 | www.epa.gov/research
United States
Environmental
Protection Agency
Documentation for application of
City-based Optimization Model for
Energy Technologies (COMET) to New
York City to support metropolitan-scale
air, climate, and energy planning
Office of Research and Development
Center for Environmental Measurement and Modeling
-------
vvEPA
United States
Environmental Protection
Agency
EPA/600/R-25/135
May 2025
Documentation for application of City-based
Optimization Model for Energy Technologies
(COMET) to New York City to support
metropolitan-scale air, climate, and energy
planning
By
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
CONTRIBUTORS
P. Ozge Kaplan, EPA ORD
Zane Carroll, EPA ORD
Marie Pied, ESMIA Consultants
Romain Chaffanjon, ESMIA Consultants (formerly)
Kathleen Vaillancourt, ESMIA Consultants
RECOMMENDED CITATION
Kaplan, P.O., Z. Carroll, M. Pied, R. Chaffanjon, K. Vaillancourt (2025). Documentation for
application of City-based Optimization Model for Energy Technologies (COMET) to New York
City to support metropolitan-scale air, climate, and energy planning. EPA/600/R-25/135. DOI:
10.23719/1532264
DATA AVAILABILITY
All the Appendices and the COMET vl5.0.9 and COMET vl6.0.1 model files can be downloaded
at DOI: 10.23719/1532264
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection Agency
policy and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation of use.
Development of COMET-NYC was supported by Abt Global and ESMIA consultants -
subcontractor to UNC Institute for The Environment- under Contract 68HERD21A0002 Task
Order #68HERH22F0164.
This work was conducted under the Agency's Quality Assurance (QA) program for
environmental information, with an approved Quality Assurance Project Plans for (Support for
Energy System Modeling Activities in EPA/ORD), J-AESMD-0033640-QP-1-0 and (Use of The
Integrated MARKAL-EFOM System (TIMES) for Energy Technology Scenario Analysis), J-AESMD-
0000246-QP-1-6.
ACKNOWLEDGEMENTS
This effort owes its success to the dedicated involvement and meticulous care, particularly in
managing data and assumptions, of EPA's partners in NYC government specifically Office of
Management and Budget's Environmental Sustainability & Resiliency Task Force.
-------
Table of Contents
Abstract 1
1 Introduction 2
2 Background on TIMES 4
2.1 Description 4
2.2 Data Requirements 7
3 COMET-NYC Structure 11
3.1 Units 12
3.2 System-wide Model Assumptions 12
3.3 Pollutant Coverage 13
3.4 Version control 13
4 Buildings Module 14
4.1 Calibration to base years 14
4.2 Residential Sector 16
4.2.1 Residential Sector Demand Projections 16
4.2.2 Residential Emissions Accounting 20
4.2.3 Residential Sector Constraints 21
4.3 Commercial Sector 22
4.3.1 Commercial Energy Demand Services 22
4.3.2 Commercial Technology Structure 24
4.3.3 Commercial Emissions Accounting 25
4.3.4 Commercial Sector Constraints 25
5 Transportation Sector 27
5.1 Light-Duty Vehicles 27
5.1.1 Light-Duty Vehicle Energy Demand Services 27
5.1.2 Technology Structure 28
5.1.3 Light-Duty Vehicle Emissions Accounting 29
5.1.4 Light-Duty Vehicle Constraints 29
5.2 Heavy-Duty Vehicles 30
5.2.1 Energy Demand Services 30
5.2.2 Technology Structure 31
5.2.3 Heavy Duty Vehicle Constraints 32
6 Electric Sector 32
-------
7 Reference Case 34
7.1 Electricity Sector 34
7.2 Building Sector 37
7.2.1 Residential 38
7.2.2 Commercial 41
7.3 Transportation 45
7.3.1 Fuel Consumption for the Transportation Sector by Region 47
7.4 System Wide Emissions 49
8 Final Remarks 59
9 References 60
Appendix A 62
Appendix B 62
Appendix C 62
Appendix D 62
APPENDIX E 62
APPENDIX F 62
-------
Abstract
The City-based Optimization Model for Energy Technologies (COMET-NYC) is an energy system
modeling tool developed by the U.S. Environmental Protection Agency's Office of Research and
Development to support long-term, metropolitan-scale air, climate, and energy planning for
New York City. Built on the internationally recognized TIMES modeling framework, COMET-NYC
identifies the least-cost mix of technologies and fuels required to meet projected energy
demands from 2010 to 2055 across NYC's buildings, transportation, and electricity sectors.
COMET-NYC uses a scenario-based optimization approach to simulate the deployment of
energy technologies under various assumptions, policies, and constraints. It incorporates local
data sources to estimate and calibrate energy consumption and emissions at the borough level.
It tracks both greenhouse gases (GHGs) and criteria air pollutants, supporting city-level climate
and air quality policy evaluation.
The model includes detailed modules for the residential, commercial, industrial, and
transportation sectors, accounting for current and future technology costs, fuel types, and
efficiency parameters. It uses linear programming to minimize system-wide costs while meeting
energy service demands and emissions targets. COMET-NYC supports both retrospective
analysis (e.g., calibration to 2010, 2015, and 2020) and future scenario exploration, such as
electrification strategies.
Two versions of the model are documented here: vl5.0.9, which underpinned emissions
reduction planning during the 2023-2024 NYC budgeting cycle, and vl6.0.1, which includes
updated buildings data and improved calibration.
COMET-NYC is a critical decision-support tool that enables policymakers to evaluate the costs,
benefits, and tradeoffs of various technology mixes and policy strategies across NYC's complex
urban energy system.
In addition to NYC specific COMET, US EPA developed an open-source version of the model
called "Generative COMET1" that can be applied to medium- to large-scale cities in the U.S.
1 DOI: 10.23719/1532263
1
-------
1 Introduction
Local, state and regional authorities are facing challenges caused by a changing climate,
urbanization, limited natural resources, and aging infrastructure. As of 2021, more than 130
million people in the U.S. are estimated to live in areas that exceed one or more National
Ambient Air Quality Standards (NAAQS) (EPA, 2024). Challenges may also come from the
increasing energy demands associated with population and economic growth. Increasing
temperatures introduce challenges as well, including greater space cooling requirements in
buildings, decreased efficiency of thermo-electric cooling at power plants, limits on the
discharge of cooling water into rivers & lakes, and the air quality impacts associated with
increased photochemical reaction rates.
Many are interested in pursuing environmental goals while stimulating economic growth.
Governments have begun to set emissions reduction targets to protect human health and the
environment. At the federal level, there are various standards and regulations, and at the
regional level, states have agreed to work together to reduce emissions through programs such
as the Regional Greenhouse Gas Initiative (RGGI), an electric sector cap-and-trade program,
Section 177 light-duty Zero-Emission Vehicle (ZEV) sales targets, and the multi-state medium-
and heavy-duty ZEV Initiative. Furthermore, at the state level, 23 states and the District of
Columbia (DC) have adopted GHG emission reduction targets, while 30 states and DC have
adopted either Renewable Portfolio Standards or Clean Energy Standards in the electric sector
(NCSL, 2021).
The energy system and its resulting emissions will be impacted by difficult-to-predict factors
such as technology development and adoption, climate, the availability of water and energy
resources, and current and future energy and environmental policy. Another complicating
factor is that the time frames typically associated with air quality management can be very
different. Air quality management can often involve time horizons of a decade or less, while
energy and infrastructure decisions and build out may stretch to longer time horizons. In this
complex landscape, planners need tools and information that will allow them to understand the
synergies and tradeoffs among air and energy objectives and to develop robust and cost-
effective management strategies. Given a limited number of resources, planners can benefit
from systematically evaluating multiple potential strategies for achieving economic and
environmental goals related to energy transition issues. Specifically, state and local decision
makers need to understand the environmental and health implications of energy supply and
use in their regions, as well as the extent to which energy resources and technologies may
contribute to achieving current and future environmental goals.
U.S. Environmental Protection Agency's Office of Research and Development (EPA/ORD) has
been developing and applying various energy system tools to evaluate the long-term economic
and environmental benefits of technology and infrastructure deployment strategies, to
understand the environmental and health implications of energy supply and use in their
regions, and to analyze which energy resources and technologies may contribute to achieving
current and future environmental goals.
COMET-City-based Optimization Model for Energy Technologies, developed by EPA/ORD, is an
application of TIMES energy-environment-economic optimization framework. In its New York
2
-------
City application (COMET-NYC), the model identifies the least-cost mix of technologies and fuels
needed to meet projected energy demands from 2010 to 2055. It accounts for market trends,
federal policies, and state actions, and evaluates their impacts on GHG emissions and air
pollution. COMET optimizes technology investments and fuel use across end-use sectors such
as buildings and transportation, considering constraints like emissions limits and electrification
standards. The model includes supply curves for primary energy carriers (e.g., oil, natural gas,
coal, hydrogen, and renewables), and deployment of energy conversion technologies (e.g.,
power plants, combined heat and power) based on capital costs, efficiency, and other
performance parameters.
COMET-NYC offers a detailed representation of the city's energy system, including electric
generating units (EGUs) dispatching power via the New York Independent System Operator. It
models energy flows from resource extraction/import to end-use across the five boroughs and
New York State. Using linear programming, it minimizes the system's net present cost while
satisfying energy demands and user-defined constraints. Outputs include technology pathways,
total system costs, emissions (GHG and CAP), and energy prices. The model also supports
scenario analysis to evaluate the effects of new technologies or policies. Designed for city-level
application, COMET helps assess technology portfolios to meet urban energy, climate, and
environmental goals.
This report provides an overview of the COMET, data sources, and calibration against actual
energy consumption data and discusses a reference case providing a future year energy
outlook. We provide assumptions for the versions 15.0.9 and 16.0.1.
3
-------
2 Background on TIMES
TIMES is an economic model generator for local, national, multi-regional, or global energy
systems, which provides a technology-rich basis for representing energy dynamics over a multi-
period time horizon. TIMES is maintained through the Energy Technology and Systems Analysis
Program (ETSAP) of the International Energy Agency (IEA). ETSAP currently has as contracting
parties 21 countries and one private sector sponsor. TIMES can assist in the design of least-cost
pathways for sustainable energy systems and is ideally suited for the preparation of Low-
Emissions Development Strategies (LEDS) and Intended Nationally Determined Contributions
(INDC) and Nationally Determined Contributions (NDC) roadmaps. It is usually applied to the
analysis of the entire energy sector but may also be applied to study single sectors such as the
electricity and district heat sector.
2.1 Description
TIMES consists of generic variables and equations constructed from the specification of sets
and parameter values depicting an energy system for each distinct region in a model. To
construct a TIMES model, a preprocessor first translates all data defined by the modeler into
special internal data structures representing the coefficients of the TIMES matrix applied to
each variable for each equation in which the variable may appear. This step is called Matrix
Generation. Once the model is solved (optimized) a Report Writer assembles the results of the
run for analysis by the modeler. The matrix generation, report writer, and control files are
written in GAMS (the General Algebraic Modelling System). GAMS is a powerful high-level
language specifically designed to facilitate the process of building large-scale optimization
models. GAMS accomplishes this by relying heavily on the concepts of sets, compound indexed
parameters, dynamic looping and conditional controls, variables and equations. Thus, there is a
very strong synergy between the philosophy of GAMS and the overall concept of the Reference
Energy System (RES) specification embodied in TIMES, making GAMS very well suited to the
TIMES paradigm. Furthermore, by nature of its underlying design philosophy, the GAMS code is
very similar to the mathematical description of the equations. Thus, the approach taken to
implement a TIMES model is to "convert" the input data by means of a (rather complex)
preprocessor while taking care of the necessary exceptions to properly construct the matrix
coefficients for the Linear Programming (LP) model. In addition, the GAMS platform integrates
seamlessly with a wide range of commercially available optimizers such as CPLEX and/or
XPRESS. To build, run, and analyze a TIMES model, several software tools have been developed
in the past or are currently under development, so that the modeler does not need to provide
the input information needed to build a TIMES model directly in GAMS. These tools are the
model interfaces VEDA2.0. EPA/ORD currently holds licenses to utilize VEDA2.0 to build TIMES
models. The TIMES model generator has extensive documentation and demo models to build
instances of TIMES models (Loulou et al 2016, Loulou et al 2016, Goldstein et al 2016, &
Goldstein et al 2016).
In TIMES, a complete scenario consists of four types of inputs: energy service demand curves,
primary resource supply curves, a policy setting, and the descriptions of a complete set of
technologies. The basis of a TIMES model is a network diagram called a Reference Energy
System (RES), which depicts an energy system from resource supply to end-use demand (Figure
4
-------
1) The RES constructs an energy system up from a list of technology types, energy carriers, and
user demands. The four technology types represented are resource, process, conversion, and
demand technologies as defined in detail below:
1) Resource technologies represent the extraction cost and availability of resources such as
coal, oil, and natural gas.
2) Conversion technologies represent the conversion of fuel inputs into electricity.
3) Process technologies represent other means of converting resources into end-use fuels
including refineries and coal-to-liquid processes.
4) Demand technologies represent the technologies that meet specific user demands, such as
vehicles, air conditioners, and water heaters.
These technologies feed into a final stage consisting of end-use demands for useful energy
services. End-use demands include items such as residential lighting, commercial air
conditioning, and automobile passenger miles traveled. Estimates of end-use energy service
demands (e.g., vehicle miles traveled; residential lighting, steam heat requirements in the
paper industry; etc.) are provided by the user for each region to drive the reference scenario. In
addition, the user provides estimates of the existing stocks of energy related equipment in all
sectors, and the characteristics of available future technologies, as well as present and future
sources of primary energy supply and their potentials.
Using these as inputs, the TIMES model aims to supply energy services at minimum global cost
(more accurately at minimum loss of total surplus) by simultaneously making decisions on
equipment investment and operation; primary energy supply; and energy trade for each region.
For example, if there is an increase in residential lighting energy service relative to the
reference scenario (perhaps due to a decline in the cost of residential lighting, or due to a
different assumption on GDP growth), either existing generation equipment must be used more
intensively or new - possibly more efficient - equipment must be installed. The choice by the
model of the generation equipment (type and fuel) is based on the analysis of the
characteristics of alternative generation technologies, on the economics of the energy supply,
and on environmental criteria.
5
-------
"5 t=
o
~ o
"5 =
O
Gas
extraction
Coal
extraction
, Oil
extraction
Oil
Import 'y
o
1 I
Gas
Furnace
Electric
Heater
Oil
Furnace
Figure 1 Illustrative Reference Energy System
TIMES is thus a vertically integrated model of the entire extended energy system. The scope of
the model extends beyond purely energy-oriented issues, to the representation of
environmental emissions, and perhaps materials, related to the energy system. In addition, the
model is suited to the analysis of energy-environmental policies, which may be represented
with accuracy thanks to the explicitness of the representation of technologies and fuels in all
sectors. In TIMES, the quantities and prices of the various commodities are in equilibrium, i.e.,
their prices and quantities in each "time period" are such that the suppliers produce exactly the
quantities demanded by the consumers. This equilibrium has the property that the total
economic surplus is maximized. It is useful to distinguish between a model's structure and a
particular instance of its implementation. A model's structure exemplifies its fundamental
approach for representing a problemit does not change from one implementation to the
next. Therefore, all TIMES models exploit an identical underlying structure.
Thus, the structure of a TIMES model is ultimately defined by variables and equations created
from the union of the underlying TIMES equations and the data input provided by the user. This
information collectively defines each TIMES regional model database, and therefore the
resulting mathematical representation of the RES for each region.
6
-------
2.2 Data Requirements
The user input sets contain the fundamental information regarding the structure and the
characteristics of the underlying energy system model. The user input sets can be grouped
according to the type of information related to them:
One dimensional sets defining the components of the energy system: regions,
commodities, processes;
Sets defining the Reference Energy System (RES) within each region;
Sets defining the inter-connections (trade) between regions;
Sets defining the time structure of the model: periods, time slices, time slice
hierarchy;
Sets defining various properties of processes or commodities.
The following is a list of the classifications of data needed to build instances of TIMES models,
and the most common data parameters for each classification. For the purposes of brevity,
TIMES documentation files include all the necessary information regarding input data needs to
build a basic TIMES model. Following are parameters needed to build a typical energy system
model using TIMES.
Energy Service Demands
o Demand projections for buildings and transportation sectors
o Season/time-of-day pattern of the demand
Energy Carrier Profiles
o Input energy
o Output energy
Costs
o Resource supply
o Investment in new capacity
Fixed and variable operations and maintenance (O&M)
o Fuel delivery
o "Hurdle" rates
Technology Profiles
o Resource supply steps and cumulative resource limits
o Existing installed capacity and limits on new investment
o Fuels in and out
o Efficiency and Availability
Environmental Indicators
o Unit emissions per resource
o Emission constraints/taxes per pollutant
o Unit emissions per resource
o Emission constraints/taxes by pollutant
System and other parameters
o Electric reserve margin
7
-------
o Season/time-of-day fractions describing the electrical load
o System-wide discount rate
Furthermore, the TIMES models include time periods for modeling horizon. TIMES is 'demand
driven' in that feasible solutions are obtained only if all the specified end-use demands for
energy services are satisfied for every time-period. Table 1 summarizes the parameters needed
to build a typical energy system model using TIMES.
TIMES also distinguishes between two types of units for characterizing energy system
technologies, activity, and capacity. Activity represents the use of a technology. Most
technology activity is measured in petajoules (PJ). Capacity represents the size (installed
capacity) of the technology stock and is measured according to the ability to provide for some
amount of activity per unit time. Accordingly, capacities for most technologies are measured in
petajoules per year (PJ/yr). Electricity generation technology capacities are measured in
gigawatts (GW), and transportation technology activities are measured in billions of miles per
year.
Table 1 Variable Types in the Model and Corresponding Data Requirements
Variable Type
Input Requirements
End-Use Energy Service
Demands
Projections for energy service demands for:
TRANSPORTATION: Light-duty vehicle demand (bn-vmt-yr), bus transportation demand
(bn-vmt-yr), heavy-duty short-haul truck transportation demand (bn-vmt-yr), passenger rail
transportation demand (pn-passs-miles), medium-duty truck transportation demand (bn-vmt-
yr),
RESIDENTIAL BUILDINGS: space cooling (PJ/yr), space heating (PJ/yr), water heating
(PJ/yr), lighting (billion lumens/yr), other electricity demand (PJ/yr), other natural gas
demand (PJ/yr),
COMMERCIAL BUILDINGS: space cooling (PJ/yr), space heating (PJ/yr), water heating
(PJ/yr), lighting (billion lumens/yr), other electricity demand (PJ/yr), other natural gas
demand (PJ/yr),
All demands include load shape for electric demand profiles.
Energy Carriers
Any kind of entity that is a form of
energy that is produced or
consumed in the energy system
(e.g., coal, refined oil, natural gas,
gasoline, electricity)
- Transmission efficiency
- Transmission capacity
- Investment cost
- Operation and maintenance cost
- Electricity Transmission and distribution cost
- Reserve margin for electricity
Resource Technologies
Technologies that characterize raw
fuels exported or imported into the
energy system
- Resource supply cost for each supply step
- Cumulative Resource limits for an energy carrier for each period
- Cumulative Resource limits for an energy carrier over the entire modeling horizon
- cost and capacity limits of Resource transportation
- cost of extraction and production of Resource
8
-------
Variable Type
Input Requirements
Process, Conversion, and
Demand Technologies
Any kind of technology that can
change the location, form, and/or
structure of the energy carriers
- New capacity investment cost
- Fixed operation and maintenance cost
- Variable operation and maintenance cost as a function of activity
- Fuel delivery charges
- Technical efficiency as a ratio between input and output
- Technology investment availability year
- Availability factor
- Capacity utilization factors
- Base year installed capacity
- Upper bound on new capacity investment (if exists)
- Upper bound on incremental new investment (growth rate)
- Upper bound on total capacity installed over the modeling horizon
- "Hurdle" rate for a technology
Emissions
- Emissions factor per unit of fuel consumed
- Emissions factor for per unit of activity
- Emissions factor for per unit of installed capacity
- Upper bound for emission for each period
- Emission constraints over the entire modeling horizon
- Emission constraints for any given sector
Scenario Framework
A scenario approach is appropriate to the assessment of long-term technological development
in the energy system. Extended research, policy, and assessment horizons make business-as-
usual extrapolations, conventionally used in shorter-term energy futures analyses,
inappropriate. The technology innovation process is inherently uncertain and unpredictable.
Over a period of decades, we simply cannot know which technologies will achieve fundamental
breakthroughs and which will not. Changes in economic structures, consumer preferences,
resource supplies, and other variables similarly lead to inherent unpredictability. With these
factors in consideration, COMET-NYC was constructed under the energy system optimization
modeling principles presented in DeCarolis et al. (2017). These principles include considerations
such as minimizing model bias, setting clear spatio-temporal boundaries and goals, maximizing
model and data transparency and quality assurance.
The scenario approach to assessing technology futures requires that the menu of technology
options being built into the models be appropriately connected to a set of driving forces to
produce informative and internally consistent scenarios. Driving forces are the key elements
that influence how the future turns out. Any scenario approach must identify the key driving
forces that are expected to have an impact on the issues under consideration. Scenarios are
then built from combinations of values or realizations of these driving forces. Major driving
forces for the energy system technology futures include:
Economic growth
Population growth
Changes in the structure of the economy, work, and recreation
Land use and transportation policy
Air pollution and environmental policy
Oil and natural gas supply
9
-------
Consumer attitudes
Rates and patterns of technological change
Future Technologies and Scenarios
For the technologies of interest in our scenario assessment, we collected estimates of
technology costs, performance, and availability. Because many of these technologies are still
under development, these data will be estimates of future cost, performance, and availability.
There is therefore considerable uncertainty about these parameters. Indeed, it is this
uncertainty that is the motivation for and source of our scenario assessment.
We are looking for a range of values that covers plausible future outcomes. Therefore, for each
parameter it is best to gather data from several different sources and to provide some
evaluation of the reliability of each source and the assumptions supporting each estimate.
Having this well-documented range of values will allow us to construct scenarios that explore
the range of possible futures. The researcher will determine, based on an analysis of available
data, what data parameters will be used initially to input the technology into the database.
10
-------
3 COMET-NYC Structure
COMET-NYC uses New York City's annual greenhouse gas inventory (GHGI) reports to estimate
energy consumption in residential, commercial, and industrial buildings and transportation. The
modeling time-period runs from 2010 until 2055 with 5-year time intervals for reporting. The
2010 (City of New York, 2011), 2015 NYC GHGI reports (City of New York, 2017) and 2020 NYC
GHGI reports are used to calibrate the model's results for 2010, 2015 and 2020.
New York City-specific data sources are used for most inputs within COMET-NYC. Vehicle miles
travelled (VMT) projections from New York Metropolitan Transport Council (NYMTC) are used
for transportation demand (2023). Population projections from NYMTC, combined with
statistics of residential and commercial real estate from PLUTO, are used to find the projected
demand for energy in residential and commercial buildings. Projections from NYSERDA are used
to adjust these energy demand forecasts to account for increased cooling load and decreased
heating load due to rising temperatures. The New York Independent System Operator (NYISO)
Gold Book is used to find projected electricity demand in the rest of New York State and to find
the locations of current electric generators in New York State (2023). National projections from
the ElA's Annual Energy Outlook (AEO) are used to find the projected cost and efficiency of new
energy technologies, as well as cost curves for generating energy from fossil or renewable
sources (EIA, 2016). The model was also calibrated to match historical New York City GHG
emissions from the GHG Inventory (City of New York, 2023).
Furthermore, the COMET-NYC includes data specific to NYC such as New York City Department
of Environmental Protection (NYCDEP), the New York State Energy Research & Development
Authority (NYSERDA), and a variety of other sources. Where local data is unavailable, the
COMET-NYC relies on a database created for U.S. energy system. For instance, the U.S.
Environmental Protection Agency (EPA) has been maintaining a model representing the U.S.
national energy system through nine census divisions (known as the EPAUS9rT model) for use
within the TIMES energy-economy-environment modeling framework (Lenox et al., 2013). Data
for the EPAUS9rT are derived primarily from the U.S. Energy Information Administration (EIA)'s
National Energy Modeling System (NEMS) model, and the results are calibrated every two years
to the corresponding Annual Energy Outlook (AEO) published by EIA.
The COMET-NYC consists of six regions including Brooklyn, Bronx, Manhattan, Staten Island,
Queens and New York State (to cover EGUs in the state). Each of the six-regions is structured in
a different RES diagram. Those diagrams are interconnected through technology links (i.e. fuel
trades). The naming conventions for each fuel type remain same from one region to another.
The naming identification of regions are presented in Figure 2. For instance, in the model, R1
represents the all the EGUs in New York State except the ones in the New York City. This region
is the source of electricity and transfers electricity to other regions via trade technologies.
In addition to the six regions, there is an outer region for model fuel supply (R0-"dummy"). This
is the supply region of the model that characterizes the fossil fuel sources located outside of the
(or trade) option, a transportation cost, capacity limits, and capacity extension cost (investment
cost) are defined.
11
-------
Bronx \'
R1
New York State
R2
Brooklyn
R3
Bronx
R4
Manhattan
R5
Statcn Island
R6
Ouecns
Figure 2 COMET-NYC regional coverage - New York State and Boroughs of New York City
3.1 Units
The cost data is given in 2005 million U.S. dollars. Energy carriers are given in terms of PJ. Most
end-use demands are given in terms of PJ with the following exceptions:
Commercial and Residential Lighting Demand: billion lumens per year (bn-lum-yr),
Light-duty Transportation: billion vehicle miles traveled (bn-vmt),
Medium-and Heavy-duty Transportation: billion vehicle miles traveled (bn-vmt),
Passenger rail: billion passenger miles (bn-pass-miles).
3.2 System-wide Model Assumptions
There are numerous assumptions that are used to compute the annual investment cost
such as annual discount rate, also referred to as "hurdle rate." It is applied as 3% and 4%
to the system-wide economy (that covers all 6 regions). This discount rate can be
adjusted for a specific technology if this technology requires a different rate.
The year is divided into 12 different time slices over the planning horizon as seen in
Table 2. The fraction of the year is specified in the database. These time slices were
derived from the ElA's National Energy Model System (NEMS) to appropriately
represent the seasons and intraday time slices (Goudarzi, 2007; Appendix A).
Grid transmission losses are characterized as "transmission efficiency." This value is
selected as 95% based on EIA state profile.
The reserve margin/capacity for electricity is 20%.
12
-------
Table 2 Time-Slice Fractions Used to Characterize Load- Duration Curves
Description
Time Fraction
Intermediate Day - AM
8.36%
Intermediate Night - PM
9.95%
Intermediate Night
13.93%
Intermediate Peak
1.19%
Summer Day - AM
9.75%
Summer Day - PM
10.54%
Summer Night
11.14%
Summer Peak
1.99%
Winter Day - AM
6.91 %
Winter Day - PM
9.87%
Wnter Night
15.19%
Wnter Peak
1.18%
SOURCE: Goudarzi (2007); Appendix B
3.3 Pollutant Coverage
The COMET-NYC includes emission factors for: carbon dioxide (CO2), nitrogen oxides (NOx),
particulate matter < 10 |am (PM10), particulate matter < 2.5 |am (PM2.5), sulfur dioxide (SO2),
volatile organic compounds (VOC), methane (CH4), carbon monoxide (CO), Organic Carbon (OC),
and Black Carbon (BC) for each region and sector across the whole energy system related to
fuel consumption. System-level carbon dioxide is reported in million tons (Mt) per year, where
all other emissions within sectors are reported in thousand tons (kton) per year.
3.4 Version control
This documentation provides details on the V15.0.9 of the COMET-NYC model. This version was
modified and updated to aid NYC Office of Management and Budget with their city budgeting
process during time frame 2023 through 2024. The COMET-NYC is used to generate various
emission reduction scenarios for the city in line with their climate goals.
Since more updates are conducted to the model, V16.0.1 includes some updates to the
buildings module. In the following sections, we will describe each update separately.
13
-------
4 Buildings Module
The building end-use energy demands are split into residential, commercial, and industrial
(facility level) buildings. A majority of the industrial sector is lumped in with the commercial for
V15.0.9. In V16.0.1, manufacturing and concentration will be broken out into in the industrial
sector. The level of end-use demand in each of the three sub-sectors is estimated using a
bottom-up approach based on the U.S. ElA's AEO (Specifically, the U.S. ElA's Commercial
Building Energy Consumption Survey (CBECS) and the U.S. ElA's Residential Energy
Consumption Survey (RECS)), the NYC Primary Land Use Tax Lot Output (PLUTO), official NYC
energy and emissions and other related official data.
4.1 Calibration to base years
COMET-NYC characterizes existing building stock through its end-use energy service demand
and includes suite of future technologies and retrofits to meet these demands. This sector is
built using the data collected under the NYC Benchmarking Law (LL84) along with Primary Land
Use Tax Lot Output (PLUTO) files. PLUTO dataset contains data on all buildings in NYC - where
each building has a unique Borough-Block-Lot (BBL) number. LL84 provides annual
measurements of energy and water consumption for some building types. The data set is
included in the public data repository associated with this report (Appendix B). In addition,
city's GHG inventory data provide various fuel consumption levels per each building type. All
data is matched (by BBL as well as reporting year) to allocate existing building stock to the
associated energy use for each building. We utilized the PLUTO data to build baseline
calibration framework. The GHG inventory data has fuel consumption by type for each building
type categorized as Residential (1-4 units and multifamily), Commercial and Institutional
(commercial, institutional, and streetlights) and Manufacturing and construction (industrial).
Each individual fuel consumption (e.g., natural gas consumption) per building type (e.g.,
residential buildings) is needed to calibrate end-use energy service demands. We obtained data
set from NYC Department of Health and Mental Hygiene that characterizes fuel allocations per
space heating, water heating, space cooling, lighting, conveyance, process loads, and
miscellaneous for calendar year 2014 for each building type. This data did not provide borough-
specific values. A close look at PLUTO 2010 shows us total building area by building type in each
borough (Figure 3). Thus, we used this information to allocate energy consumption values per
building type per end-use service demand into five boroughs. This data set is included in the
public data repository associated with this report (Appendix C).
14
-------
& 1/800
£ 1,600
c 1,400
U00 ¦ ¦
S 1,000
5> 800
1=1 I I I .
Brooklyn Bronx Manhattan Queens Staten Island
¦ lto4 ¦Commercial ¦ Industrial ¦ Institutional BMultifamily
Figure 3 Total building area in sq.ft by building type and region. Data sourced from PLUTO 2010.
"1-4" refers to 1-4 unit family and multi-family households.
Next is to characterize existing technology stock per end-use service demand. This activity
includes finding, for instance, capacity and efficiency of boilers, furnaces, and heat pumps for
space heating per each borough. This type of detailed technology data does not exist specific to
NYC, therefore we rely on ElA's Commercial and Residential Energy Consumption Surveys
(CBECS and RECS) and ElA's AEO to gather information specific to the NYC census division.
Technology capacity, costs, and efficiency data for Middle Atlantic Census Division from CBECS
and RECS is gathered for our calculations. Future technology representations are gathered from
EIA2.
2 https://www.eia.gov/analvsis/studies/buildings/equipcosts/pdf/appendix-a.pdf also provided in Appendix D.
15
-------
Figure 4 Data Sources for Buildings Sector
4.2 Residential Sector
4.2.1 Residential Sector Demand Projections
Residential sector energy service demand includes 1-4-unit family and multifamily households.
Total energy demand for the residential sector is classified under four main sections (space
heating, space cooling, water heating, lighting) and two aggregated fuel consumptions (other-
electricity and other-natural gas). The nomenclature and related units are given in Table 3.
Table 3 Residential End-use Service Demands
Demand
Units
Description
RSC
PJ/vr
Space Cooling
RSH
PJ/vr
Space Heating
RWH
PJ/yr
Water Heating
RLT
billion lumens/yr
Lighting
ROE
PJ/vr
Other - Electricity
ROG
PJ/yr
Other - Natural Gas
16
-------
The end-use service demand projections for buildings need to be determined as this is one of
the key model inputs. The projections depend on various drivers such as population projections
(Table 4), economic growth, number of people per household (Table 5), type of housing,
building envelope efficiency and additional need for cooling and heating (due to changes in
Heating Degree Day (HDD) and Cooling Degree Day (CDD) (Table 5)). Population average square
feet of space per household are the key drivers of the demand growth. We gathered borough
specific population projections from the city (see Table 4) (NYMTC, 2020). The raw data from
NYMTC is included in Appendix E. In addition, we gather HDD and CDD projections specific to
NYC (see Table 5) (NYSERDA, 2024).
Table 4 New York City Population Projections
Region
2010
2015
2020
2025
2030
2035
2040
2045
2050
Brooklyn
2,552,911
2,593,655
2,647,112
2,760,391
2,820,822
2,860,506
2,894,388
2,928,160
2,956,932
Bronx
1,385,108
1,423,160
1,454,816
1,515,667
1,548,245
1,573,786
1,595,881
1,616,845
1,633,550
Manhattan
1,585,873
1,636,537
1,668,548
1,698,050
1,735,482
1,754,534
1,768,412
1,781,885
1,791,292
Staten Island
468,730
477,525
484,897
491,202
495,047
498769
502,327
505,464
507,920
Queens
2,250,002
2,294,943
2,349,324
2,418,636
2,463,405
2,483716
2,500,457
2,517,076
2,528,763
SOURCE: NYMTC 2020
Table 5 Average number of persons per household in 20053
Region
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
2055
Brooklyn
2.6
2.6
2.7
2.7
2.7
2.7
2.6
2.6
2.6
2.6
2.6
Bronx
2.6 2.6 2.7 2.7 2.7 2.7 2.6 2.6 2.6 2.6 2.6
Manhattan
2.0
2.0
2.1
2.1
2.1
2.0
2.0
2.0
2.0
2.0
2.0
Queens
2.9 2.9 3.0 3.0 3.0 2.9 2.9 2.9 2.9 2.9 2.9
Staten Island
2.9
2.9
2.9
3.0
2.9
2.9
2.8
2.8
2.8
2.8
2.8
NYC
2.6 2.6 2.7 2.7 2.7 2.6 2.6 2.6 2.6 2.6 2.6
Table 6 Heating and Cooling Degree Days
Region
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
COOLING DEGREE DAYS, CDD
NYC
1112
1142
1503
1544
1635
1636
1637
1653
1801
1815
1938
HEATING DEGREE DAYS, HDD
NYC
4376
4376
4376
4376
3930
3933
3958
3938
3759
3741
3576
SOURCE: NYSERDA 2024
3 https://www.nvc.gov/assets/planning/download/pdf/data-maps/census/census2010/t sfl p5 nvc.pdf
The city level growth rate is applied to borough level 2005 baseline value. The data is for 2010, therefore we
assumed that this value was the same for prior years. Values came from NYC Planning Department.
17
-------
The total city-wide energy consumption per end-use service demand reported in 2015 is used
as benchmark for projections. Below we walk through how to determine Residential Space
Heating Demand for 2045 in Brooklyn (RSH2045, bk):
RSH 2045, bk - HousingHeatingCoefficient * AverageHouseSquareFootage bk * NumberOfHouseholds 2045,bk
* HDD_Projection 2045, nyc / Conversion Factor
Housing Heating Coefficient = AdjustedHeatingServiceDemand2o45, bk * Conversion Factor /
(AverageHouseSquareFootage2045 * NumberOfHouseholds2045, bk * HDD_Projection2045,NYc)
AdjustedHeatingServiceDemand2o45,bk = AdjustedHeatingServiceDemand2oo5 * (1-
RetrofitEnvelopGain2045) * (l-NewBuildEnvelopeGain2045) * (HDD2045/HDD2005) *
(NumberOfHouseholds2045/NumberOfHouseholds2005) *
(AverageHouseSquareFootage2045/AverageHouseSquareFootage2005) *
(SpaceHeatingCoefficientsK)
SpaceHeatingCoefficientsK = BuildingAdjustment/PopulationEffect/1.1
BuildingAdjustment = FuelConsumption2oio, BK/FuelConsumption2oio, nyc
PopulationEffect = NumberOfHouseholds2oio, BK/NumberOfHouseholds2oio, nyc
NumberOfHouseholds2045,BK = PopulationProjection2045, BK/AverageNumberOfPersonsPerHousehold2045, bk
AdjustedHeatingServiceDemand2oo5 = RSH2oos,nyc * NumberOfHouseholds2oo5, bk/
NumberOfHouseholds2oo5, nyc * SpaceHeatingCoefficientBK
RSH 2005,NYC - RSH 2010,nyc * AEOSpaceHeatingFinalEnergy2oo5 / AEOSpaceHeatingFinalEnergy2oio
RSH 2010,NYC - RSH ELC,2010,NYC + RSH NGA,2010,NYC + RSH DSL,2010,NYC + RSHsTM, 2010, NYC
RSH 2015,NYC - RSH 2010,NYC
RSH 2020,NYC - RSH 2015,nyc * AEOSpaceHeatingFinalEnergy2o2o / AEOSpaceHeatingFinalEnergy2ois
RSH 2025,NYC - RSH 2020,nyc * AEOSpaceHeatingFinalEnergy2o25 / AEOSpaceHeatingFinalEnergy202o
RSH 2050,NYC - RSH 2045,nyc * AEOSpaceHeatingFinalEnergy2o5o / AEOSpaceHeatingFinalEnergy2045
FuelConsumption2oio, bk = FuelConsumptionELc,2oio,BK + FuelConsumptionNGA,2oio,BK +
FuelConsumptionDsL,2oio,BK + FuelConsumptionSTM,2oio,BK
FuelConsumptionELc,2oio,BK = FuelConsumptionELc,2oio,BK, 1-4 + FuelConsumptionELc,2oio,BK, Muitifamiiy
FuelConsumptionELC,2oio,BK, 1-4 = (PLUTOBuildingAreaBK, ia / PLUTOBuildingAreaNYc, 1-4) *
(FuelConsumptionELC, 2014,1-4 / FuelConsumptionELC,2014, nyc) * FuelConsumptionELC, 2010, nyc *
(FuelConsumptionSH,ELc,2014,1-4 / FuelConsumptionELC,2014,1-4)
Similarly, the end-use service demands are calculated for all types per borough, and
summarized in Table 6.
18
-------
Table 7 Residential Sector Demand Projections (COMET_NYC V15.0.9)
Borough 2015
2020
2025
2030
2035
2040
2045
2050
Residential Cooling Demand (PJ)
Bronx
4.47
4.91
5.14
5.28
5.42
5.96
6.04
6.48
Brooklyn
8.15
8.94
9.36
9.61
9.86
10.80
10.94
11.73
Manhattan
6.66
7.29
7.46
7.65
7.83
8.54
8.62
9.20
Queens
6.50
7.15
7.39
7.56
7.71
8.41
8.48
9.04
Staten Island
1.39
1.51
1.54
1.56
1.59
1.73
1.74
1.86
Residential Heating Demand (PJ)
Bronx
23.25
21.66
22.67
23.40
23.70
22.80
22.85
21.93
Brooklyn
39.29
36.54
38.28
39.53
39.94
38.34
38.36
36.81
Manhattan
42.72
39.70
40.58
41.92
42.22
40.37
40.23
38.43
Queens
19.80
18.47
19.10
19.66
19.75
18.86
18.78
17.93
Staten Island
7.27
6.73
6.85
6.97
7.00
6.69
6.65
6.35
Residential Lighting Demand (Billion Lumens/Year)
Bronx
1.64
1.59
1.67
1.73
1.77
1.79
1.82
1.84
Brooklyn
2.81
2.72
2.87
2.96
3.02
3.06
3.10
3.13
Manhattan
2.69
2.60
2.68
2.77
2.82
2.84
2.86
2.88
Queens
1.42
1.38
1.43
1.48
1.50
1.51
1.52
1.53
Staten Island
0.54
0.52
0.53
0.54
0.55
0.56
0.56
0.56
Residential Miscellaneous Electric Demand (PJ)
Bronx
4.87
4.74
4.69
4.51
4.45
4.38
4.32
4.25
Brooklyn
9.56
9.29
9.21
8.86
8.71
8.57
8.43
8.30
Manhattan
6.79
6.59
6.37
6.13
6.01
5.89
5.78
5.66
Queens
4.91
4.79
4.68
4.49
4.39
4.30
4.21
4.12
Staten Island
2.45
2.37
2.28
2.16
2.11
2.07
2.02
1.98
Residential Miscellaneous Gas Demand (PJ)
Bronx
0.49
0.69
0.69
0.69
0.67
0.66
0.65
0.65
Brooklyn
0.91
0.90
0.90
0.89
0.87
0.85
0.85
0.85
Manhattan
1.41
1.40
1.36
1.34
1.31
1.27
1.27
1.27
Queens
0.18
0.18
0.18
0.17
0.17
0.16
0.16
0.16
Staten Island
0.09
0.09
0.09
0.09
0.09
0.08
0.08
0.08
Residential Water Heating Demand (PJ)
Bronx
7.67
7.94
8.29
8.43
8.36
8.26
8.12
7.98
Brooklyn
12.48
12.90
13.48
13.71
13.58
13.38
13.14
12.90
Manhattan
13.28
13.72
13.99
14.23
14.05
13.79
13.48
13.18
Queens
6.25
6.49
6.69
6.78
6.68
6.55
6.40
6.25
Staten Island
2.07
2.13
2.16
2.17
2.13
2.09
2.04
1.99
SOURCE: U.S. EPA, with ElA's Commercial and Residential Energy Consumption Surveys
19
-------
In COMET-NYC V15.0.9, PLUTO 2010 values are used as basis for the calibration, along with
2010 inventories. In COMET-NYC V16.1, we pulled historic data from the main PLUTO database,
and able to calibrate the model using PLUTO 2010, 2015 and 2020 data with corresponding
inventory data. Since the floorspace allocation differed from one year to another the historic
demand values are shifted slightly.
Next to populate the future technology portfolio that can meet the end-use service demands.
The technology and fuel combinations are given in Table 7. Future technology cost and
efficiency values for residential space heating, space cooling, water heating and lighting are
taken from ElA's Updated Buildings Sector Appliance and Equipment Costs and Efficiencies
(2023). All parameters related to residential sector technologies are provided exogenously into
the model.
Table 8 Residential Technology and Fuel Combinations
End-use Demand
Technology Type
Fuel
Space Heating
Radiant - Boiler System
Electric, Natural Gas, Distillate
Furnace
Natural Gas, Distillate, Kerosene
Space Cooling
Room AC
Electric
Central AC
Electric
Space Heating and Cooling (Simultaneous)
Air-Source Heat Pump
Electric
Ground-Source Heat Pump
Electric
Water Heating
Electric, Natural Gas, Distillate, Solar
Lighting
Incandescent
Electric
CFL
Electric
LED
Electric
Halogen
Electric
Linear Fluorescent
Electric
Reflector
Electric
4.2.2 Residential Emissions Accounting
COMET-NYC tracks fuel combustion related emissions as well as some process and leakage
emissions occurring along the energy system. For instance, CO2 emissions are tracked through
quantity of fuel combusted and verified for 2010, 2015 and 2020 using New York City's
Greenhouse Gas Inventory. Methane emissions are tracked throughout the system, with the
main contribution coming from oil and gas operations, which are beyond the geographical
scope of this analysis. Criteria air emission factors are derived from U.S. EPA's National
20
-------
Emissions Inventory (NEI) platform and AP-42 datasets (EPA November 2024, and EPA 2015).
These emission factors are tied to fuel and technology combinations.
4.2.3 Residential Sector Constraints
COMET-NYC utilizes constraints to mimic more realistic outputs in accordance with the existing
city policy implications. For instance, to model the city's plan to phase out petroleum-based
space heating options, an upper bound on diesel consumption is set for the 2015-2055 period
(Table 8). However, we also include a lower bound on electricity consumption on the space
heating to assure that the share of electricity-based space heating will not drop unrealistically
over the modeling period. Additionally, district heat was constrained to Manhattan (R4) to
mimic real world conditions. In addition to fuel share constraints, technology splits are included
to mimic AEO 2016 Residential Unit Consumption of Energy with respect to the equipment
classes.
Table 9 Residential Fuel Use and Technology Mix Constraints
End-use Service Demand
Fuel/Tech
At Least
At Most
Year
Diesel
17.4%
20.2%
2015
Diesel
12.3%
12.3%
2055
Diesel
0%
2055
Electric
0.9%
1.8%
2015
Electric
2.2%
2020
Electric
3.0%
100.00%
2055
Natural Gas
73.4%
2015
Natural Gas
73.4%
2020
Natural Gas
0%
2055
Residential Space Heating
Furnace
46.6%
2015
Furnace
34.0%
2055
Furnace- Diesel
15.0%
2015
Furnace- Diesel
9.2%
2055
Furnace- Electric
4.40%
2015
Furnace- Electric
6.60%
2055
Heat Pump
0.60%
2015
Heat Pump
0.50%
2055
Radiant
42.7%
2015
Radiant
31.2%
2055
Diesel
13.4%
29.9%
2015
Diesel
4.1%
18.3%
2020
Residential Water Heating
Natural Gas
81.60%
2015
Natural Gas
59.60%
2020
NG - Instantaneous
2.00%
2015
NG - Instantaneous
2.00%
2055
21
-------
End-use Service Demand
Fuel/Tech
At Least
At Most
Year
Solar
10.00%
2050
Electric
2.8%
3.5%
2015
Electric
2.0%
100.00%
2055
Residential Space Cooling
Central Heat Pump
2.3%
2015
Central Heat Pump
1.7%
2055
Central AC
38.1%
2015
Central AC
27.8%
2055
Central AC
59.4%
2015
Central AC
43.0%
2055
4.3 Commercial Sector
Commercial sector energy service demand covers Commercial, Institutional and Industrial
buildings. Total energy demand for the commercial sector is classified under four main sections
(space heating, space cooling, water heating, lighting) and two aggregated fuel consumptions
(other-electricity and other-natural gas).
4.3.1 Commercial Energy Demand Services
The commercial sector in COMET-NYC V15.0.9 is an aggregation of Commercial, Institutional
and Industrial Buildings. Industrial building emissions are defined in the industrial sector files,
however institutional and industrial buildings demands are lumped in and included in the
demand calculations for the commercial sector. These data inputs are sourced from the 2015
NYC GHG inventory under manufacturing and construction.
In the COMET-NYC V16.1, the industrial building demand is separated from the Commercial and
Institutional Buildings. The industrial building demand which is listed under Manufacturing and
Construction in the inventories are represented in the IND workbook.
The methodology and technology structure are similar the residential sector. Hence some
sections of the commercial sector section are curtailed. The commercial sector module includes
details of commercial sector energy demands and their corresponding end-used technologies.
The nomenclature and corresponding units for those end use energy demands are listed in
Table 10.
The main driver for commercial demand would be projection of total floor space. Conventional
methods rely on gathering the best estimate of the current floor space dedicated to the
different types of businesses and the current employees working at these businesses. This
approach leads to estimation of the average floor space needed per employee for different
types of commercial businesses. The Northwest Power and Conservation Council implemented
a methodology where they made estimates using employee trends (2021). In future work, it is
possible to implement a similar methodology to then look at the future number of employees
in the region and derive the estimated need for commercial floor space from the projected
employment trends. However, in our data search we could not allocate any future projections
for employment trends per business type. Therefore, in interim, we calculated total commercial
22
-------
floor area per capita using 2020 population data per borough and reported aggregate floor area
from PLUTO (Table 9).
Table 10 2020 Floor Area (Commercial + Institutional + Industrial)
Borough
2020 (sqft)
BK
402,983,294
BX
193,021,586
MN
789,411,121
SI
61,710,960
QN
333,675,891
Table 11 Commercial Demand
Demand
Units
Description
CSH
PJ/yr
Space Heating
CSC
PJ/yr
Space Cooling
CWH
PJ/yr
Water Heating
CLT
billion lumens/yr
Lighting
CME
PJ/yr
Misc-ELC
CMN
PJ/yr
Misc - NG
Final energy consumption in 2010 and 2015 are calibrated against reported actual final energy
consumption data provided by NYC Department of Health and Mental Hygiene. Demands are
then calculated similar to what is outlined for residential sector and presented in Table 11.
Table 12 Commercial Sector Demand Projections
Borough
2015
2020
2025
2030
2035
2040
2045
2050
Commercial Cooling Demand (PJ)
Brooklyn
26.29
26.88
28.45
29.05
29.05
29.10
31.35
31.21
Bronx
12.43
12.86
13.63
13.90
13.90
13.95
15.06
15.02
Manhattan
50.76
52.71
55.73
55.54
55.55
55.47
59.52
59.03
Staten Island
21.62
22.19
23.55
23.75
23.67
23.57
25.27
25.03
Queens
4.04
4.14
4.36
4.32
4.26
4.24
4.55
4.50
Commercial Heating Demand (PJ)
Brooklyn
24.70
22.17
22.64
22.78
22.47
21.21
20.86
19.65
Bronx
11.81
10.61
10.83
10.89
10.77
10.19
10.03
9.46
Manhattan
47.33
42.43
42.29
42.56
41.86
39.35
38.54
36.16
Staten Island
3.75
3.35
3.32
3.29
3.23
3.03
2.97
2.78
Queens
20.06
18.06
18.21
18.26
17.91
16.82
16.46
15.43
23
-------
Borough
2015
2020
2025
2030
2035
2040
2045
2050
Commercial Lighting Demand (Billion Lumens/Year)
Brooklyn
14.71
15.67
17.01
17.78
18.69
19.58
19.81
20.00
Bronx
7.04
7.51
8.14
8.51
8.96
9.41
9.53
9.63
Manhattan
28.85
30.69
32.52
34.00
35.62
37.18
37.46
37.66
Staten Island
2.26
2.40
2.53
2.61
2.72
2.84
2.86
2.87
Queens
12.15
12.97
13.91
14.49
15.14
15.78
15.89
15.96
Commercial Miscellaneous Electric Demand (PJ)
Brooklyn
14.58
16.25
18.31
20.11
22.06
24.20
26.56
29.16
Bronx
6.97
7.78
8.76
9.62
10.58
11.63
12.78
14.04
Manhattan
28.59
31.83
35.01
38.46
42.05
45.96
50.23
54.89
Staten Island
2.24
2.49
2.72
2.95
3.22
3.51
3.83
4.19
Queens
12.03
13.45
14.97
16.39
17.87
19.51
21.30
23.26
Commercial Miscellaneous Gas Demand (PJ)
Brooklyn
1.20
1.27
1.39
1.55
1.82
2.18
2.18
2.19
Bronx
0.58
0.61
0.67
0.74
0.87
1.05
1.05
1.05
Manhattan
2.36
2.49
2.66
2.97
3.48
4.13
4.13
4.12
Staten Island
0.19
0.19
0.21
0.23
0.27
0.32
0.31
0.31
Queens
0.99
1.05
1.14
1.27
1.48
1.75
1.75
1.75
Commercial Water Heating Demand (PJ)
Brooklyn
4.04
4.12
4.30
4.39
4.46
4.51
4.56
4.61
Bronx
1.93
1.98
2.06
2.10
2.14
2.17
2.20
2.22
Manhattan
7.92
8.08
8.22
8.40
8.49
8.56
8.63
8.67
Staten Island
0.62
0.63
0.64
0.64
0.65
0.65
0.66
0.66
Queens
3.34
3.41
3.52
3.58
3.61
3.63
3.66
3.68
SOURCE: U.S. EPA, with ElA's Commercial and Residential Energy Consumption Surveys
4.3.2 Commercial Technology Structure
Several demand technology and fuel combinations are included in the model (Table 12). Each of
these technology and fuel combinations have distinct technology attributes such as investment
cost, O&M cost, starting year, and fuel efficiency. We utilized AEO's Commercial Technology
Equipment Type Description File (EIA, 2023) and Commercial Building Energy Consumption
Surveys (CBECS) to determine values for various end-use sectors such as space heating, cooling
etc. Specifically, CBECS is utilized to allocate technology shares among end-use service demands
(2018).
Table 13 Commercial Technology and Fuel Combinations
24
-------
End-Use Demand
Technology Type
Fuel
Space Heating
Boiler
Electric, Natural Gas, Diesel
Furnace
Natural Gas, Diesel
Space Cooling
Centrifugal Chiller
Electric, Natural Gas
Reciprocating Chiller
Electric
Scroll Chiller
Electric
Screw Chiller
Electric
Rooftop AC
Electric, Natural Gas
Window/Wall AC
Electric
Central AC
Electric
Space Heating and Cooling (Simultaneous)
Air-Source Heat Pump
Electric
Ground-Source Heat Pump
Electric
Water Heating
Electric, Natural Gas, Diesel, Solar
Lighting
Incandescent
Electric
CFL
Electric
LED
Electric
Halogen
Electric
Linear Fluorescent
Electric
Metal Halide
Electric
4.3.3 Commercial Emissions Accounting
Emission accounting follow same procedure as residential sector.
4.3.4 Commercial Sector Constraints
Similarly to in the residential sector, COMET-NYC uses constraints to achieve realistic
adoption of certain commercial technologies.
Table 14 Commercial Fuel Use and Technology Mix Constraints
Fuel/Tech
At Least
At Most
Year
Electric
9.20%
2015
Electric
10.80%
2020
Electric
6.70%
21.60%
2055
Natural Gas
49.10%
2015
Commercial Space
Natural Gas
49.10%
2020
Heating
Natural Gas
0%
2055
Diesel
8.10%
2015
Diesel
16.2%
2055
Boiler
31.00%
2015
Boiler
0%
2055
25
-------
Furnace
47.80%
2015
Furnace
0%
2055
Commercial Space
Cooling
Electric
77.90%
2015
Electric
56.90%
2055
Natural Gas
17.10%
2015
Natural Gas
12.50%
2055
Rooftop
49.30%
2015
Rooftop
36.00%
2055
Central
14.50%
2015
Central
10.60%
2055
Window/Wall
10.80%
2015
Window/Wall
7.90%
2055
Ground-Source Heat
Pump
3.20%
2015
Ground-Source Heat
Pump
2.30%
2055
Air-Source Heat Pump
5.20%
2015
Air-Source Heat Pump
3.80%
2055
Electric
0.60%
3.50%
2015
Electric
0.40%
100%
2055
Natural Gas
76.30%
83.90%
2015
Commercial Water
Heating
Natural Gas
0%
61.30%
2055
Diesel
20.80%
2015
Diesel
15.20%
2055
Solar
1.00%
2055
SOURCE: U.S. EPA, with LL84 Datasetand ElA's Commercial and Residential Energy Consumption Surveys
26
-------
5 Transportation Sector
The transportation sector covers the vehicle technologies that are used to meet the passenger
and freight demand. Technologies are classified under two main technology sets namely light-
duty vehicles (LDV) and heavy-duty vehicles (HDV) - which includes medium duty vehicles.
LDV technologies include gasoline, diesel, compressed natural gas (CNG), hydrogen (H2), and
electric powered cars including plug-in, electric vehicle (EV), and hybrid, which meet passenger
demand measured in billion vehicle miles traveled per year (bn-vmt-yr). HDV technologies
include heavy-duty short haul trucks, buses, and electric passenger rail to account for NYC's
extensive public transit system.
5.1 Light-Duty Vehicles
Light-Duty Vehicle (LDV) demand account for personal vehicle miles travelled (VMT) for
passenger demand. Transportation Light-duty (TL) demand is represented not only by various
demand technologies (including different fuel type and efficiency levels) but also fuel
distribution networks for gasoline, diesel, electricity, etc. Mini-compact, compact, full size,
minivan, pick-up truck, small SUV, and large SUV are the main vehicle class sizes.
5.1.1 Light-Duty Vehicle Energy Demand Services
Light-duty vehicle demand for base year is calculated with respect to the total fuel consumption
provided in NYC greenhouse gas emission inventory report using base year average vehicle
efficiency, aggregate vehicle miles travelled. The historic NYC VMT values are based on NYMTC
outputs. These values cover trips originating in the city and ends in the city, plus trips ending in
the city and plus originating in the city. The vehicles passing through the city is not included.
Therefore, long-haul freight and interstate passenger transport are not included. VMT
projections are gathered from NYMTC'sTransportation Conformity Determination's regional
transportation forecast (2023) (Appendix 2A; Summer Values). Demand trajectories are
adjusted for each borough according to population forecasts. 2010 LDV fleet distribution for the
NYC is set as a constraint.
27
-------
Table 15 NYC's Historic Vehicle Miles Traveled and Future Projections for COMET
Year
COMET Inputs
VMT Data
Sources for the VMT
2010
19,657,033,169
19,657,033,169
NYC GHG Inventory - passenger cars
2015
19,662,888,322
19,662,888,322
NYC GHG Inventory - passenger cars
2016
20,244,327,807
NYC GHG Inventory - passenger cars
2017
20,439,338,696
NYC GHG Inventory - passenger cars
2018
21,080,754,312
NYC GHG Inventory - passenger cars
2019
21,523,925,338
NYC GHG Inventory - passenger cars
2020
20,141,922,090
16,593,133,288
NYC GHG Inventory - passenger cars
2021
20,188,205,867
NYC GHG Inventory - passenger cars
2022
21,323,591,647
NYC GHG Inventory - passenger cars
2023
2024
2025
21,581,580,205
21,581,580,205
NYMTC-CONFORMITY Appendix 2A (Summer)
2026
21,634,210,285
NYMTC-CONFORMITY Appendix 2A (Summer)
2030
22,083,405,748
Linear extrapolation between 2025 and 2035
2035
22,585,231,290
22,585,231,290
NYMTC-CONFORMITY Appendix 2A (Summer)
2040
23,015,341,268
Linear extrapolation between 2035 and 2045
2045
23,445,451,245
23,445,451,245
NYMTC-CONFORMITY Appendix 2A (Summer)
2050
23,367,934,005
23,367,934,005
NYMTC-CONFORMITY Appendix 2A (Summer)
Table 16 Light-Duty Vehicle Demand Projections per Borough
Region
(billion VMT)
2010
2015
2020
2025
2030
2035
2040
2045
2050
Brooklyn
4.35
4.53
4.71
7.98
8.18
8.37
8.55
8.72
8.48
Bronx
3.01
3.09
3.23
3.14
3.20
3.27
3.32
3.36
3.45
Manhattan
3.11
3.08
3.14
3.56
3.63
3.70
3.76
3.82
3.87
Staten
Island
2.03
2.08
2.14
2.18
2.24
2.29
2.33
2.37
2.41
Queens
7.16
6.88
6.92
4.71
4.84
4.96
5.07
5.17
5.16
TOTAL
19.66
19.66
20.14
21.58
22.08
22.59
23.02
23.45
23.37
SOURCE: U.S. EPA, with NYMTC
5.1.2 Technology Structure
The light-duty demand (TL) is met by eleven different engine types for seven car classes (Table
16).
28
-------
Table 17 Light-Duty Vehicle Fuel and Technology Combination
Mini-
Compact
Compact
Full-Size
Minivan
Pickup
Small
SUV
Large
SUV
Gasoline
Conventional
X
X
X
X
X
X
X
Hybrid
X
X
X
X
X
X
Plug-in hybrid
(20 miles per
charge)
X
X
X
X
X
X
Plug-in hybrid
(40 miles per
charge)
X
X
X
X
X
X
Diesel
Conventional
X
X
X
X
X
X
Hybrid
X
X
X
X
X
CNG
Conventional
X
X
X
X
Flex fuel
X
X
X
X
H2
Fuel cell
X
X
X
X
X
X
Electric
100-mile range
X
X
X
X
X
X
X
200-mile range
X
X
X
X
X
X
X
5.1.3 Light-Duty Vehicle Emissions Accounting
COMET-NYC assigns CO2 emission factors to each transportation fuel based on carbon content
of the fuel. The emissions are then calculated by means of the total consumption of the fuel
within the transportation technologies. For criteria air pollutants, emission factors are defined
on the technology itself to represent transportation related air regulations. Transportation
sector criteria pollutant emission factors for each vehicle type and fuel are gathered from the
U.S. EPA's Motor Vehicle Emission Simulator (MOVES2014b). MOVES creates emission factors
for on-road motor vehicles and gathers estimate of emissions from cars and trucks under a
wide range of user-defined conditions e.g., vehicle types, time periods, geographical areas,
pollutants, and vehicle operating characteristics. Emissions factors were obtained by
postprocessing MOVES simulations outputs with custom MySQL scripts. The ratio between
emissions and activity (distance traveled) were used to create activity-weighed emissions
factors. COMET-NYC includes county-level emissions factors simulated via MOVES using vehicle
in-place data per county obtained from NYSDEC. More information on how the improved
emissions factors were calculated using MOVES can be found in Appendix A.
5.1.4 Light-Duty Vehicle Constraints
The LDV sector includes seven car classes including Mini-compact, Compact, Full-size, Minivan,
Pick-up, Small and Large SUV. Car shares are based on regional car and truck sales (sales data by
class) for the Middle Atlantic region presented in AEO.
29
-------
5.2 Heavy-Duty Vehicles
Table 18 presents demand naming conventions for the heavy-duty vehicles (HDV) sector. The
HDV sector includes buses (TB), short haul heavy-duty trucks (THS), medium size, passenger rail
transport (TRP) and subway in the COMET-NYC database. The reported fuel consumption
inventory values from NYC cover trips originating in the city and ends in the city, plus trips
ending in the city and plus originating in the city. The vehicles passing through the city is not
included; therefore, long-haul freight is not included.
5.2.1 Energy Demand Services
Input data that are concerning heavy-duty technologies are collected from NYC 2010 fuel
consumption data and various fleet constraints based on Transportation Data Book and AEO
(2014).
"TH" end use energy demands are calculated with the assumption that calibration year existing
technology combinations in EPAUS9rT are also valid for NYC. NYC energy consumption value for
transportation sector is combined with the average efficiency of existing fleet to calculate the
TH demand, then the demand is extended according to the AEO demand projections.
All heavy-duty vehicles transportation demands are exogenous to the model. The demands are
projected using population and economic activity data. The inventory years are calibrated in the
model.
Table 18 Heavy-Duty Vehicle Demand Projection
Borough
2015
2020
2025
2030
2035
2040
2045
2050
Bus Transportation (bn-vmt)
Bronx
0.06
0.06
0.06
0.07
0.07
0.07
0.08
0.08
Brooklyn
0.11
0.11
0.12
0.12
0.13
0.13
0.14
0.14
Manhattan
0.07
0.07
0.07
0.07
0.08
0.08
0.08
0.09
Queens
0.09
0.1
0.1
0.1
0.11
0.11
0.12
0.12
Staten Island
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
Medium-Duty Trucks (bn-vmt)
Bronx
0.09
0.09
0.09
0.09
0.09
0.09
0.1
0.1
Brooklyn
0.16
0.16
0.16
0.16
0.17
0.17
0.17
0.18
Manhattan
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Queens
0.14
0.14
0.14
0.14
0.14
0.14
0.15
0.15
Staten Island
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
Passenger Rail (bn-pass-miles)
Bronx
1.66
1.73
1.85
1.96
2.07
2.21
2.34
2.48
Brooklyn
3
3.17
3.37
3.56
3.73
3.97
4.18
4.43
Manhattan
1.87
1.96
2.07
2.17
2.25
2.37
2.49
2.62
30
-------
Borough
2015
2020
2025
2030
2035
2040
2045
2050
Queens
2.66
2.79
2.94
3.07
3.19
3.38
3.59
3.79
Staten Island
0.53
0.58
0.61
0.64
0.67
0.7
0.76
0.81
Short-Haul Heavy-Duty (bn-vmt)
Bronx
0.11
0.12
0.12
0.13
0.14
0.15
0.16
0.16
Brooklyn
0.2
0.21
0.22
0.23
0.25
0.26
0.28
0.29
Manhattan
0.13
0.13
0.14
0.14
0.15
0.16
0.17
0.17
Queens
0.18
0.19
0.2
0.2
0.21
0.22
0.24
0.25
Staten Island
0.04
0.04
0.04
0.04
0.04
0.05
0.05
0.05
TRN.SHIP (bn-t-m)
Brooklyn
1.87
1.76
1.65
1.57
1.56
1.58
1.58
1.58
SOURCE: U.S. EPA, with NYMTC and AEO
Table 19 Heavy-Duty Transportation Demands
Name
Description
Units
Unit Description
TB
Bus
bn-vmt
billion vehicle milestraveled
TMS
Medium DutyTrucks
bn-vmt
billion vehicle milestraveled
THS
Short Haul Heavy Duty
Tru cks
bn-vmt
billion vehicle milestraveled
TRP
Passenger Rail
(includesSubway)
bn-pass-miles
billion passengermiles
5.2.2 Technology Structure
Table 19 represents the available engine and fuel type pairs in the COMET-NYC and
distinguishes them with respect to the efficiency improvements and different vintage years
with available fuel options. User-defined constraints are set for the calibration year to mimic
the real fuel investment data.
Table 20 Heavy-Duty Vehicle Demand Types, Fuel, and Technology Combinations
End-Use Demand
Fuel
Efficiency Improvements
Bus Demand
Diesel
Improved Eff, Adv. Tech, Adv. Hybrid, Conventional
Electric
Improved Eff, Conventional
CNG
Improved Eff, Adv. Tech, Adv. Hybrid, Conventional
Hydrogen fuel cell
Hybrid, Conventional
Medium- and Heavy-Duty Vehicles
- Short-Haul Demand
Diesel
Improved Eff, Adv. Tech, Adv. Hybrid, Conventional
CNG
Improved Eff, Adv. Tech, Adv. Hybrid, Conventional
Rail Passenger Demand -
Commuter
Diesel
Electric
31
-------
End-Use Demand
Fuel
Efficiency Improvements
Rail Passenger Demand -
Passenger Rail Subways &
Streetcars
Electric
5.2.3 Heavy Duty Vehicle Constraints
In HDV sector, the model has several constraints. CNG-powered buses are given a fixed amount
of investment for 2010 to represent existing stock of CNG bus fleet. Additionally, medium-duty
CNG and heavy-duty gasoline and CNG vehicles were given a fixed amount of investment cost
by defining an upper bound. The model has both commuter rail and subway to meet TRP
demand. To keep the balance to mimic the actual sector conditions the percent of total
demand that can be met by commuter rail is protected by lower bounds that belong to the
actual NYC transportation data for 2010 (City of New York, 2012). Diesel and electric buses are
protected by a lower bound in this same way.
6 Electric Sector
The COMET calculates annual emissions from electric generating facilities providing electricity
to New York City. NYC (NYISO Zone J) generates electricity, while the rest of the demand is met
through electricity from facilities in New York (NY) state and New Jersey (NJ) state.
The ELC workbook contains technology characterization for all electric generating units (EGU)
located in New York and New Jersey. In addition, imports from Canada and neighboring states
are represented. The EGUs in New York City are dual fuel generators using natural gas or oil.
The ELC sector also includes CHP capacity, and CHP for district heating. The CHP details are
taken from the U.S. Department of Energy (DOE) CHP database and U.S. ElA's Historical State
database for New York (2023). The transmissions and distribution network capacity for electric
trade linkages are included.
All generators in NY and NJ are grouped into 3 different regions based upon their location as
listed below.
In City (NYISO zone J which corresponds to New York City)
Zone ABCDEF + GHI (NYISO zones A through F; NYISO zones G, H and I)
PSEG (Portion of PJM in New Jersey) and Imports from Canada
32
-------
A - West
B - Genesee
C - Central
D ¦ North
E - Mohawk Valley
F - Capital
G - Hudson Valley
H - Millwood
I - Dunwoodie
J - New York City
K - Long Island
K
Figure 5 NYISO Load Zone Map | Source: NYISO 2023
Total generation for each of the three different regions are then calculated based upon
generation reported from input data sources (EIA 923, EIA 860, EPA CEMS and NYISO Gold
Book) and membership of generating facilities to each region. Model determines the electricity
demand for the New York City's boroughs through detailed technology representation in
buildings and transportation. The electricity demand for the rest of the state is taken
exogenously from NYISO GoldBook high demand scenario. Based on the total electric demand,
model acts as a capacity expansion model, and calculates future model years EGU capacity
based on capital, O&M and fuel costs. In addition, we incorporate constraints to mimic city's
access to upstate renewable sources in zones G,H, and I.
It is assumed that a portion of generation from each of those regions listed above goes to serve
NYC's load. All the electricity generated in the city is used in the city. Total emissions from each
generating facility in NY and NJ are calculated based upon total generation, fuel emission
coefficient and heat rate. COMET is not a dispatch model; therefore, we created rules on how
generation is allocated to the city. This information came from the NYC sources based on prior
electric dispatch studies. These assumptions are then used to calculate the emission intensity of
electricity. Following list includes main assumptions;
1. 100% of generation in Zone J is used to serve Zone J's load. For example, if NYC's
load is 50,000 GWh and Zone J's annual generation is 25,000 GWh, Zone J's
regional distribution factor is 50%.
2. Model includes imports from Canada and PJM grid specifically from PSEG. The
total aggregate electricity generation from PSEG plants are represented as
import flows, and we calculated an emission intensity for that flow based on the
generator heat rate and fuel consumption data.
3. The remainder of NYC's load (i.e. after subtracting zone J generation and PSEG
imports) is served by generation in ABDCEF and GHI. The model optimizes on the
least cost pathway for the capacity expansion. For calculating NYC specific
electricity intensity, we assume a 50/50 split between zones A-F and zones GHI.
We added an installed reserve margin of 20% based on information from NYSIO (2023).
E F
A B
33
-------
7 Reference Case
The Reference Case is defined as a business-as-usual case that contains all implemented federal
and state policies relevant to energy and environment starting in 2010. The methodology and
data sources described in the previous sections are utilized to generate a baseline reference
case for New York City such that the energy consumption values reported in the NYC GHG
Inventory is matching sector-by-sector energy consumption for 2015. The COMET-NYC model
outputs can be broken down by sector, technology, region, and several other ways. Figure 6
depicts the overall trends in fuel consumption for each sector and fuel type for the reference
case scenario. The following sections will further break down these trends.
800 -
700 -
600 -
~ 500 -
o
g 400 -
3
VI
C
S 300 -
~
-------
energy targets when state and local actions are implemented. This trend is further seen in
Figure 8 which depicits the projected CO2 emissions from the electiricty sector in the reference
case scenario. The refrence case shows these emissions staying relatively steady and not
dipping below 20 MT of CO2.
Fuel consumption mix by fuel type for scenario sl_nyc_ref 2010-2050
1200
1000
E?
| 800
Q.
E
I 600
o
u
~cD
£ 400
200
0
2010 2015 2020 2025 2030 2035 2040 2045 2050
"fear
ELCCOAL
ELCNGA
ELCWND
ELCSOL
ELCDSL
ELCHYD
Figure 7 Electricity sector fuel generation by fuel type for the reference scenario from 2010-2050
I 20
Figure 8 Electricity Sector Carbon dioxide Emissions by Fuel Type for the reference case scenario COMET-
NYC also allows users to break down changes in scenarios by region. In this case, the regions were
grouped by in the city (NYC) and outside of the city (NY State) (Figure 8). In scenarios where more energy
reduction policies or clean energy indicatives are enacted, NYC fuel consumption shifts from natural gas
to wind energy. In NY State, the energy mix has a larger breadth as there are more reliable energy
production technology that may be implemented outside of the city.
Electricity Sector CO2 emissions for scenario sl nyc ref over time
2010 2015 2020 2025 2030 2035 2040 2045 2050
Vfear
35
-------
Region: NYC, Scenario: sl_nyc_ref
Region: NY State, Scenario: sl_nyc_ref
2010 2015 2020 2025
2030
\fear
2035 2040 2045 2050
2010 2015 2020 2025
2030
"fear
2035 2040 2045 2050
Figure 9 Fuel consumption by fuel type and region from years 2010-2050for the reference case scenario
In addition to emission and energy consumption projections, COMET-NYC also provides
financial information such as investment cost. Figure 10 breaks down the reference case
scenario investment cost by fuel type and region. The investment cost for the projected shifts in
the electricity sector largely occur in the state, as this is where most electricity generation takes
place. Additionally, the reference case scenario optimization projects see large increases in
investment for renewable sources, such as wind and solar, and in natural gas over time.
Financial figures get further broken down in the Cost Implications section.
Investment Cost by Fuel Type for Region: NYC
2000
-------
7.2 Building Sector
COMET-NYC allows users to look at building sector results in several ways. Users may pull data
on end-use service demands by appliance/technology, by the specific building sector, by fuel
type, etc. Figure 11 displays an overall decrease in energy consumption for the building sector
from 2010 to 2050 for the reference case scenario. The overall energy consumption in the
building sector is projected to moderately decrease over time, however substantial decreases in
natural gas consumption in the sector is unlikely without federal, state, or local action. Similar
trends are seen amongst the different regions, seeing the greatest changes in consumption and
fuel type mix in R4, Manhattan (Figure 12).
Building Sector Energy Consumption for Scenario: sl nyc ref
Figure 11 Building sector energy consumption by fuel type from years 2010-2050for the reference case
scenario
Figure 12 Building sector radar chart depicting building sector energy consumption by fuel type and
region for the reference case scenario. Left Panel: year 2015. Right Panel: year 2050
37
-------
There has been a shift in number of HDDs and CDDs over the years. With the impact of
increasing global temperatures, the number days for cooling needs is projected to increase and
the number of days for heating needs is steady. These yields following trends in buildings. The
space heating demand for residential (Figure 17) and commercial buildings (Figure 23) is
projected to decrease 8 and 22 percent in 2050 compared to 2015, respectively. Contrary to
this, the space cooling demand for residential and commercial buildings is projected to increase
41 and 14 percent in 2050 compared to 2015, respectively. These shifts in demands yield a
decrease in total energy demand, resulting in need for less fuel in the buildings sector. On top
of this, the technology turnover rate, efficiency improvements in technologies, switching to
electric appliances yields further decrease in fuel and electricity consumption for buildings
sector. As a result, we observe reductions in CO2 emissions in the reference scenario.
End-Use Service Demand by Technology Category - Scenario: si nyc ref
Heating Lighting
Figure 13 End-use service demand by technology type for all sectors from years 2010-2055for the
reference case scenario
7.2.1 Residential
Residential sector technology is broken down into the following categories: lighting (RLT), space
cooling (RSC), space heating (RSH), water heating (RWH), electricity miscellaneous (ROE), and
natural gas miscellaneous (ROG). Most of the annual end-use demand from the residential
sector technologies is coming from space heating (Figure 14). Figures 15-18 demonstrate the
projected changes over time for technology types creating the end-use demand. For technology
categories lighting, water heating, and space cooling these categories experience the greatest
amount of electrification by 2050. Alternatively, likely due to its size, technology performance,
and cost in NYC, space heating is still projected to use natural gas and diesel technologies by
2050 to meet end-use demand.
38
-------
J?
<> ^
* ^
eft
G°°
v"
.c&
,*w
b
-------
End-Use Demand for Residential Lighting by Technology - slnycref
2010 2015 2020 2025 2030 2035 2040 2045 2050
>fear
Compact Fluorescent High-Pressure Sodium ¦¦ Incandescent LED Linear Fluorescent
Figure 16 Breakdown of end-use demand by lighting technology in the residential sector for years 2010-
2050for the reference case scenario
End-Use Demand for Residential Space Heating by Technology - slnycref
DSLCHP
Distillate Furnace
Electric Radiant
Natural Gas CHP
Distillate Boiler Radiant
Electric Heat Pump
Natural Gas Boiler Radiant
Natural Gas Furnace
Figure 17 Breakdown of end-use demand by space heating technology in the residential sector for years
2010-2050for the reference case scenario
40
-------
End-Use Demand for Residential Cooling by Technology - sl nyc ref
60 -
50 -
i 1 1 1 1 1 1 1 1
2010 2015 2020 2025 2030 2035 2040 2045 2050
>fear
Electric Central AC Electric Heat Pump Natural Gas Heat Pump Room AC
Figure 18 Breakdown of end-use demand by space cooling technology in the residential sector for years
2010-2050for the reference case scenario
7.2.2 Commercial
Residential sector technology is broken down into the following categories: lighting (CLT), space
cooling (CSC), space heating (CSH), water heating (CWH), electricity miscellaneous (CME), and
natural gas miscellaneous (CMN). In the base case scenario, the projected end-use demand for
every technology category increases over time except for space heating (Figure 19). Similarly to
the residential sector, electrification is most prevalent in the space cooling and water heating
technology categories in COMET-NYC optimization projections (Figure 20-23). Space heating
maintains the largest share of natural gas as a fuel type amongst the technology categories.
41
-------
Figure 19 End-use demand for the commercial sector in years 2015 and 2050from the reference case
scenario
End-Use Demand for Commercial Water Heating by Technology- sl nyc ref
2010 2015
2020 2025
i i
2030 2035
tear
i i i
2040 2045 2050
DSL ¦¦
Electric Heat Pump
Natural Gas HE
Natural Gas steam turbine
¦¦ DSL CHP ¦
Natural Gas CHP
Natural Gas Instant
Solar
¦¦ Electiric
Figure 20 Breakdown of end-use demand by water heating technology in the commercial sector for years
2010-2050for the reference case scenario
42
-------
End-Use Demand for Commercial Lighting by Technology - sl nyc ref
2010
Compact Fluorescent
Hallogen
Incandescent
LED
2045
Linear Fluorescent
2050
Figure 21 Breakdown of end-use demand by lighting technology in the commercial sector for years 2010-
2050for the reference case scenario
43
-------
End-Use Demand for Commercial Cooling by Technology - sl nyc ref
2010
2015
2020 2025
2030
Year
2035 2040
2045 2050
DSL Steam Turbine
ELC Air Source Heat Pump
ELC Air Source Heat Pump
111
ELC Central AC
ELC Centrifugal Chiller
ELC Ground Source Heat Pump
III
ELC Reciprocating Chiller
ELC Roof AC
ELC Window/Wall AC
Natural Gas Centrifugal Chiller
Natural Gas Roof AC
Figure 22 Breakdown of end-use demand by space cooling technology in the commercial sector for years
2010-2050for the reference case scenario
End-Use Demand for Commercial Space Heating by Technology - sl_nyc_ref
2010
2015 2020 2025
2030
"fear
2035 2040
2045 2050
DSL Boiler
ELC Air Source Heat Pump
ELC RSU Standard Small
Natural Gas Furnace
¦
DSL CHP
ELC Ground Source Heat Pump
Natural Gas Boiler
Natural Gas Steam Turbine
DSL Furnace
ELC RSU Standard Large
Natural Gas CHP
Figure 23 Breakdown of end-use demand by space heating technology in the commercial sector for years
2010-2050for the reference case scenario
44
-------
7.3 Transportation
In 2015, gasoline was the main fuel meeting the light-duty demand, whereas diesel is mainly
consumed by buses and heavy-duty short-haul vehicles (Figure 24-26). Although an increase in
CO2 emissions is expected with population growth, urbanization and economic development,
the implementation of national light-duty fuel efficiency standards and vehicle turnover to
more efficient technologies lead to reduced fuel consumption and therefore reductions in city-
wide emissions, and transportation CO2 emissions (Figure 27). In addition, penetration of
electric vehicles contributes to reduction in CO2 emissions. In the heavy-duty sector, the diesel
consumption is still prominent and grows steadily, however compared the light-duty sector,
their contribution is to overall CO2 emissions is low.
Figure 24 Light-Duty vehicle fuel consumption by fuel type from years 2010-2050for the reference case
scenario
45
-------
Fuel Consumption by Fuel Category in sl_nyc_ref
0 J 1 1
2010 2015
2020
2025 2030 2035
'fear
2040 2045
2050
¦
CNG - Heavy-Duty
DSL - Medium-Duty
DSL - Bus
GSL- Bus
¦
DSL - Heavy-Duty
¦
ELC - Medium-Duty
CNG - Bus
DSL - Rail
GSL - Heavy-Duty
¦¦
GSL - Medium-Duty
ELC - Bus
ELC - Rail
Figure 25 Fuel Consumption for non-light-duty vehicles 2010-2050for the reference case scenario
Transportation Sector Fuel Consumption by Technology and Fuel in sl nyc ref
250-
200
100
GSL - Light-Duty
CNG - Heavy-Duty
ELC - Medium-Duty
ELC - Bus
¦
ELC - Light-Duty
¦
DSL - Heavy-Duty
GSL - Medium-Duty
GSL-Bus
CNG - Light-Duty
GSL - Heavy-Duty
DSL - Bus
DSL - Rail
DSL - Light-Duty
LPG - Light-Duty
DSL - Medium-Duty
CNG - Bus
ELC - Rail
Figure 26 Reference case fuel consumption by fuel type and vehicle type for the transportation sector
2010-2050
46
-------
C02 Emissions by Transportation Category in sl_nyc_ref
Figure 27 Reference Case C02 emissions for the transportation sector by vehicle type 2010-2050
7.3.1 Fuel Consumption for the Transportation Sector by Region
Transportation sector light-duty vehicles fuel consumption is projected to decrease and shift
towards electrification from 2010 to 2050. Specifically, fuel consumption is projected to shift
from R6, Queens, to R2, Brooklyn, for light-duty vehicles in the reference case {Figure 28). On
the other hand, heavy-duty, medium-duty, and buses are not seeing the same geographical
shifts in fuel consumption and electrification (Figures 29-31). Rail is projected to increase in all
regions and to see a shift to full electrification (Figure 32).
R6 R6
Figure 28 Left Panel: Light-Duty Fuel Consumption (PJ) by region for Reference Case 2015, Right Panel:
Light-Duty Fuel Consumption (PJ) by region for Reference Case 2050
47
-------
Heavy-Duty Fuel Consumption (PJ) by Region (sl_nyc_ref, 2015)
R3
Heavy-Duty Fuel Consumption (PJ) by Region (sl_nyc_ref, 2050)
R3
Figure 29 Left Panel: Heavy-Duty Fuel Consumption by region for Reference Case 2015, Right Panel:
Heavy -Duty Fuel Consumption by region for Reference Case 2050
Figure 30 Left Panel: Medium-Duty Fuel Consumption by region for Reference Case 2015, Right Panel:
Medium-Duty Fuel Consumption by region for Reference Case 2050
Medium-Duty Fuel Consumption (PJ) by Region (sl_nyc_ref, 2050)
Medium-Duty Fuel Consumption (PJ) by Region (sl_nyc_ref, 2015)
48
-------
Bus Fuel Consumption (PJ) by Region (sl_nyc_ref, 2015)
R3
DSL
Figure 32 Left Panel: Rail Fuel Consumption by region for Reference Case 2015, Right Panel: Rail
Fuel Consumption by region for Reference Case 2050
7.4 System Wide Emissions
In the reference case, ail air pollution emissions are projected to reduce over time (Figures 33-
42). For the reference case scenario, the greatest relative reductions are seen for NOx and S02.
Figure 31 Left Panel: Bus Fuel Consumption by region for Reference Case 2015, Right Panel: Bus Fuel
Consumption by region for Reference Case 2050
Rail Fuel Consumption (PJ) by Region (slnycref, 2015)
Rail Fuel Consumption (PJ) by Region (sl nyc ref, 2050)
Bus Fuel Consumption (PJ) by Region (sl_nyc_ref, 2050)
R3
49
-------
System Wide Emissions by Pollutant Type for Scenario: sl_nyc_ref
CH4
CO
N20
NH3
NOX
PM10
PM25
S02
Figure 33 System wide air pollution emissions by pollutant type for years 2010-2050for the reference
case scenario
C02 Emissions by Sector for Scenario: sl_nyc_ref
O 40000
C02C
C02E
C02I
C02R
C02T
V V
tear
Figure 34 System wide C02 emissions by sector for the reference case scenario. Sectors: C02C
(commercial), C02E (Electricity), C02I (Industrial% C02R (Residential), and C02T (Transportation).
50
-------
CCh Emissions by Sector Region: R1
CO2 Emissions by Sector Region: R2
CO2 Emissions by Sector Region: R3
C02C
C02E
C02R
C02T
C02C
C02E
C02R
C02T
CO2 Emissions by Sector Region: R4
C02C
C02E
C02R
C02T
v' ^ & (jP t*'
^ rfy r5y rSP rSP rtf 0CT Jp
'V'V'V'V'V'V'V'V'V
\fear
CO2 Emissions by Sector Region: R5
C02C
C02E
C02R
C02T
V f f f f V t
\fear
CO2 Emissions by Sector Region: R6
C02C
C02E
C02R
C02T
<\y wj tv^
-------
NOX Emissions by Sector for Scenario: sl_nyc_ref
NOXC
¦¦
NOXE
NOXI
uu
NOXR
NOXT
tear
Figure 36 System wide NOX emissions by sector for the reference case scenario. Sectors: NOXC
(commercial), NOXE (Electricity), NOXI (Industrial), NOXR (Residential), and NOXT (Transportation).
52
-------
NOx Emissions by Sector, Region: R1 NO* Emissions by Sector, Region: R2 NO* Emissions by Sector, Region: R3
NOXC
¦¦
NOXE
NOXR
NOXT
NOXC
NOXE
NOXR
NOXT
NOx Emissions by Sector, Region: R4 NOx Emissions by Sector, Region: R5 NOx Emissions by Sector, Region: R6
Figure 37 System wide NOX emissions by sector and region for the reference case scenario. Sectors:
NOXC (commercial), NOXE (Electricityj, NOXI (Industrial), NOXR (Residential), and NOXT (Transportation).
53
-------
PM10 Emissions by Sector for Scenario: sl_nyc_ref
PM10C
PM10E
PM10I
PM10R
PM10T
4-
tear
Figure 38 System wide PM10 emissions by sector for the reference case scenario. Sectors: PM10C
(commercial), PM10E (Electricity), PM10I (Industrial), PM10R (Residential), and PM10T (Transportation).
54
-------
PMio Emissions by Sector, Region: R1
PMio Emissions by Sector, Region: R2
PMio Emissions by Sector, Region: R3
PMIOC
PMIOE
PM10R
PMIOT
PMIOC
PMIOE
PMIOR
PMIOT
PMio Emissions by Sector, Region: R4
PMIOC
PMIOE
PMIOR
PMIOT
0 1.0 -
1
PMio Emissions by Sector, Region: R5
PMio Emissions by Sector, Region: R6
PMIOC
PMIOE
PMIOR
PMIOT
PMIOC
PMIOE
PMIOR
PMIOT
it nr
tear
J) oP kJ
n? T? nP i?
tear
S& # tP
1? 1? 1?
tear
Figure 39 System wide PMio emissions by sector and region for the reference case scenario. Sectors:
PMIOC (commercial), PMIOE (Electricity), PM10I (Industrial), PMIOR (Residential), and PMIOT
(Transportation).
55
-------
S02 Emissions by Sector for Scenario: sl_nyc_ref
40-
¦£ 30-
20
10 -
0-
S02C
S02E
S02I
S02R
S02T
Figure 40 System wide S02 emissions by sector for the reference case scenario. Sectors: S02C
(commercial), S02E (Electricity), S02I (Industrial), S02R (Residential), and S02T (Transportation).
56
-------
SO; Emissions by Sector, Region: R1 SO2 Emissions by Sector, Region: R2 SO2 Emissions by Sector, Region: R3
SO2 Emissions by Sector, Region: R4 SO2 Emissions by Sector, Region: R5 SO2 Emissions by Sector, Region: R6
Figure 41 System wide S02 emissions by sector and region for the reference case scenario. Sectors: S02C
(commercial), S02E (Electricity), S02I (Industrial), S02R (Residential), and S02T (Transportation).
57
-------
Figure 42 System wide PM2.5 emissions by sector for the reference case scenario. Sectors: PM25C
(commercial) and PM25T (Transportation).
58
-------
8 Final Remarks
The COMET is developed to perform scenario analysis at the city and regional level. The
COMET-NYC is calibrated to current technology stock and fuel consumption values for New York
City. The model can be used to make long-term projections for energy consumption. The
COMET-NYC uses several official data sources to provide useful model outcomes. The data is
planned to be updated on a regular basis in accordance with the release of the document
updates including borough-based population forecasts, AEO forecasts, local GHG emission
reports, EIA state level electricity generation reports, etc.
COMET-NYC finds the cheapest technology and fuel combination portfolio that meets demands
in transportation and buildings. COMET-NYC calculates levelized cost of investing and operating
a technology to meet end-use demand using engineering economics principles. COMET-NYC
makes decisions using capital costs endogenously generated fuel price, electricity price, and
salvaging costs for older technologies. The objective function then incorporates annualized
costs using a global discount rate to calculate the net present value of all life-cycle costs of
investments. The costs which are incurred in all regions (i.e., resource supply region, New York
City boroughs, and rest of New York State), are included in the objective function. Hence, the
total system cost contains all energy sector related costs such as investment, operating and
maintenance costs of the technologies within New York City's whole energy system (including
electricity generation units in the city, transportation, and building sectors) and New York
State's power sector. In addition, cost of fuel delivery, extraction, refinery, and import from
outer regions are covered in the total system cost.
The COMET-NYC model serves as an example for other cities and communities who are
interested in leveraging the benefits of performing energy planning as population growth and
rising temperatures place increasing pressure on aging infrastructure. This type of modeling
framework could aid policy making process through generating technically robust and high-
fidelity technology evaluations, therefore leading to more efficient policy design mitigate of
emissions while properly identifying costs.
59
-------
9 References
DeCarolis, J., Daly, H., Dodds, P., Keppo, I., Li, F., McDowall, W., Pye, S., Strachan, N., Trutnevyte, E., Usher, W.,
Winning, M., Yeh, S., & M. Zeyringer. (2017). Formalizing best practice for energy system optimization
modeling. Applied Energy. 194, 184-198. https://doi.Org/10.1016/j.apenergy.2017.03.001
Energy Technology Systems Analysis Program (ESTAP). n.d.. TIMES, https://iea-etsap.org/index.php/etsap-
tools/model-generators/times
Goldstein, G., Kanudia, A., Lettila, A., Remme, U., & E. Wright. 2016. Documentation for the TIMES Model Part III.
https://github.com/etsap-
TIMES/riMES_Documentation/blob/master/Documentation_for_the_TIMES_model-Part-lll.docx
Goldstein, G., Gargiulo, M., & A. Kanudia. 2016. Documentation for the TIMES Model Part IV.
https://github.com/etsap-
TIMES/riMES_Documentation/blob/master/Documentation_for_the_TIMES_model-Part-IV.docx
Kaplan, P. O., & Isik, M. (2020) City-based Optimization Model for Energy Technologies: COMET-New York City
Documentation (EPA 600/R-19/124). US Environmental Protection Agency. Washington, DC.
Loulou, R., Goldstein, G., Kanudia, A., Lettila, A., Remme, U., Wright, E., Giannakidis, G., & Noble, K., 2016.
Documentation for the TIMES Model Part I. https://github.com/etsap-
TIMES/riMES_Documentation/blob/master/Documentation_for_the_TIMES_model-Part-l.docx
Loulou, R., Lettila, A., Kanudia, A., Remme, U., & G. Goldstein. 2016. Documentation for the TIMES Model Part II.
https://github.com/etsap-
TIMES/riMES_Documentation/blob/master/Documentation_for_the_TIMES_model-Part-ll.docx
National Confrence of State Legislators (NCSL). (2021). State Renewable Portfolio Standards and Goals.
https://www.ncsl.org/energy/state-renewable-portfolio-standards-and-goals
New York City Mayor's Office of Climate and Environmental Justice. (2023). NYC Greenhouse Gas Inventories.
https://climate.cityofnewyork.us/initiatives/nyc-greenhouse-gas-inventories/
New York State Energy Research and Development Authority (NYSERDA). (2019, July). 2020 NYStretch Energy
Code Commercial Cost Effectiveness Analysis. https://www.nyserda.ny.gOv/-/media/
Project/Nyserda/Files/Programs/Energy-Code-Training/commercial-cost-analysis-report.pdf
New York State Energy Research and Development Authority (NYSERDA). (2024). NY-
Sun. https://www.nyserda.ny.gov/AII-Programs/NY-Sun
The New York Independent System Operator (NYIS01). (2023). How the Installed Reserve Margin Supports
Reliability in New York. https://www.nyiso.com/-/how-the-installed-reserve-margin-supports-reliability-
in-new-york
The New York Independent System Operator (NYIS02). (2023, April). Load & Capacity Data: Gold Book.
https://www.nyiso.com/documents/20142/2226333/2023-Gold-Book-Public.pdf
New York State Energy Research and Development Authority (NYSERDA). (2019, July). Energy Savings and Cost-
Effectiveness Analysis of the 2020 NYStretch Energy Code Residential Provisions.
https://www.nyserda.ny.gov/-/media/Project/Nyserda/Files/Programs/Energy-Code-
Training/%20Residential-Cost-Analysis-Report.pdf
New York State Energy Research and Development Authority (NYSERDA). (n.d.). Con Edison Dashboard.
https://www.nyserda.ny.gov/AII-Programs/NY-Sun/Contractors/Dashboards-and-
incentives/%20%20ConEd-Dashboard
60
-------
New York State Energy Research and Development Authority (NYSERDA). (2024). Monthly Cooling and Heating
Degree Day Data. https://www.nyserda.ny.gov/About/Publications/Energy-Analysis-Reports-and-
Studies/Weather-Data/ Monthly-Cooling-and-Heating-Degree-Day-Data#NewYorkCity
New York Metropolitan Transportation Council (NYMTC). (2020, October 20). Socioeconomic and Demographic
Forecasts for Horizon Year 2055. https://www.nymtc.org/en-us/Data-and-Modeling/ Socioeconomic-
and-Demographic-SED-Forecasts/2055-Forecasts
New York Metropolitan Transportation Council (NYMTC). (2023, September 21). Transportation Conformity
Determination. https://www.nymtc.Org/Portals/0/Pdf/Conformity/Conformity%202023/
Adopted%20Documents/2023%20Transportation%20Conformity%20Determination%20Final%20
Adopted%2092123.pdf?ver=MOeeqooFnfO3koBgp6sOCg%3d%3d
Northwest Power and Conservation Council (NWPCC). (2021). "Forecast of Commercial Floor Space
Requirements", https://www.nwcouncil.org/2021powerplan_forecast-commercial-floor-space-
requirements/
Goudarzi, L (2007). Deliverable No. 1: Recommendation for establishing time slices in MARKAL modeling
framework. OnLocation, Inc. Energy System Consulting.
United States Energy Information Administration (EIA) (2012). Annual Energy Outlook, 2012.
https://www.eia.gov/outlooks/aeo/data/browser/#/?id=30-AE02012®ion=0-
0&cases=ref2012&start=2009&end=2035&f=A&sourcekey=0
United States Energy Information Administration (EIA). (2018). "Commercial Buildings Energy Consumption
Survey", https://www.eia.gov/consumption/commercial/reports.php
United States Energy Information Administration (EIA). (2023). "Historical State Data".
https://www.eia.gov/electricity/data/state/
United States Energy Information Administration (EIA). (2023). Technology Forecast Updates - Residential and
Commercial Building Technologies - Reference Case.
https://www.eia.gov/analysis/studies/buildings/equipcosts/pdf/appendix-a.pdf
United States Environmental Protection Agency. (2015). Source Category: Locomotives.
https://www.epa.gov/sites/default/files/2015-08/documents/locomotives.
pdf#:~:text=%C2%B7%20PM2.5%20emission%20factor%20PM2.5:%208.28%20g%2Fgallon%20
(yard,operations)%20PM2.5:%206.03%20g%2Fgallon%20(all%20line%20haul%20operations)
United States Environmental Protection Agency. (November 19, 2024). AP-42: Compilation of Air Emissions
Factors from Stationary Sources, https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-
compilation-air-emissions-factors-stationary-sources
United States Environmental Protection Agency. (2024). Summary Nonattainment Area Population Exposure
Report., https://www3.epa.gov/airquality/greenbook/popexp. html
United States Internal Revenue Service (U.S. IRS). (2025). Residential Clean Energy Credit.
https://www.irs.gov/credits-deductions/residential-clean-energy-
credit#:~:text=The%20Residential%20Clean%20Energy%20Credit,placed%20in%20service%20in%20203
4.
United States White House, The. (2023, September 21). Clean Energy Tax Provisions in the Inflation Reduction Act.
https://www.whitehouse.gov/cleanenergy/clean-energy-tax-provisions/
61
-------
Appendix A
Time slice documentation
Appendix B
PLUTO 2010 data
Appendix C
Original 2014 building end-use demand splits for NYC
Appendix D
EIA 2023 building technology data
APPENDIX E
2015 NYMTC SEDS population and employment forecasts
APPENDIX F
Documentation of Transportation Sector Emission Factors Updates
62
-------
vvEPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
Recycled/Recyclable Printed on paper that contains a minimum of
50% postconsumer fiber content processed chlorine free
------- |