Summary of Comments and Responses from State Technical Review of Inventory of U.S. Greenhouse Gas Emissions and Sinks by State: 1990-2019 March 2022


«>EPA

United States
Environmental Protection
AgtriCY

Summary of Comments and
Responses from State Technical

Review of

Inventory of U.S. Greenhouse Gas Emissions

and Sinks by State: 1990-2019

March 2022

Prepared by

U.S. Environmental Protection Agency
Climate Change Division
Office of Atmospheric Programs
1200 Pennsylvania Avenue, NW
Washington, DC 20004

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

TABLE OF CONTENTS

1.	Introduction	1

2.	Summary of Comments and EPA responses	2

2.1	General Questions	2

2.2	Energy	9

2.2.1	Combustion	12

2.2.2	Fugitive	21

2.3	Industrial Processes and Product Use (IPPU)	26

2.3.1	Minerals	28

2.3.2	Chemicals	32

2.3.3	Metals	36

2.3.4	Product Use	39

2.4	Agriculture	41

2.4.1	Livestock	43

2.4.2	Agricultural Soil Management	44

2.4.3	Other Charge Questions	45

2.5	Land Use, Land-Use Change, and Forestry	46

2.5.1	Forest Lands and Lands Converted to Forest Land	49

2.5.2	Agricultural Lands (Croplands and Grasslands)	49

2.5.3	Wetlands and Lands Converted to Wetlands	50

2.5.4	Settlements and Lands Converted to Settlements	51

2.5.5	Other Lands and Lands Converted to Other Lands	52

2.6	Waste	52

2.6.1	Solid Waste	54

2.6.2	Wastewater	58

2.7	Comments from Independent Organizations	63

3.	Individual State Comments	66

3.1	Alaska Department of Environmental Conservation (DEC)	66

3.2	Arizona Department of Environmental Quality Air Quality Division Technical Analysis
Unit	85

3.3	Iowa Department of Natural Resources	96

3.4	Maine Department of Environmental Protection	106

3.5	Minnesota Pollution Control Agency	108

3.6	North Dakota Department of Environmental Quality	110

3.7	Rhode Island Department of Environmental Management (DEM)	Ill

3.8	Texas CEQ	132

3.9	Virginia Department of Environmental Quality (DEQ)	151

4.	Independent Organization Responses	160

4.1 U.S. Climate Alliance on Behalf of Alliance Member States	160

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

List of Tables

Table 1: Potential Gaps in Sectors in Iowa	4

Table 2: Sectors Identified with Significant Differences between Iowa DNR and EPA	5

ii

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

1. Introduction

The U.S. Environmental Protection Agency EPA thanks all commenters for their interest and
feedback on this first annual Inventory of U.S. Greenhouse Gas Emissions and Sinks by State.
The U.S. Environmental Protection Agency (EPA), Climate Change Division (CCD) sought
comments from states to review the draft final report Methods Support Document: Inventory of
U.S. Greenhouse Gas Emissions and Sinks by State, developed by the EPA. This methods
support document is referred to throughout this review as the "State Methods document" or
"State MethodsThe review was by sector, lasted 40 days (i.e., implemented from September
17 thru November 1, 2022)1 and included charge questions to focus review on areas identified by
EPA as needing a more in-depth review of approaches to disaggregate the national Inventory,
including feedback on available state-level data. Feedback received after those dates is also
reflected in this report. EPA received comments from 9 States and one organization.

The goal of the technical review by state experts is to provide an objective review of the
approaches to developing state-level data consistent with the national Inventory to ensure that the
final state-level estimates, and document reflect sound technical information and analysis.
Conducting a basic expert peer review of all categories before completing the inventory in order
to identify potential problems and make corrections where possible is also consistent with IPCC
good practice as outlined in Volume 1, Chapter 6 of the 2006 IPCC Guidelines for National
Greenhouse Gas Inventories.

Concurrent with the review by state experts, EPA also implemented a letter peer review
consistent with EPA's Peer Review Handbook. The peer review involved 17 independent experts
identified for their sectoral expertise. Findings from the peer review, including responses to
findings are also available online here at https://www.epa.gov/ghgemissions/state-ghg-emissions-and-
removals.

The report organizes and summarizes comments received by sector and sub sector consistent with
the guiding questions EPA distributed during the review. EPA responses to comments are
included under each summary. Section 2 summarizes the comments received from all states by
Charge Question. It also includes a summary of the comments received from independent
organizations. Section 3 provides the individual comments received from states. Section 4
provides comments received from independent organizations.

1 Note, EPA did distribute a subset of state-level estimates in the Agriculture and LULUCF sectors for review later,
sharing data and documentation with state experts on January 18 and 19 for a 30-day review. EPA noted that
comments received after the 30 days would still be considered for the next annual publication of this data. EPA has
not received any comments from state experts on this data at the time of publication of this report.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2. S u in in a r y of Comments and EPA responses
2.1 General Questions

1. What are your overall impressions of the clarity of the methods described in this
report?

State reviewers from Maine's Department of Environmental Protection (DEP) listed the
following primary concerns regarding the results from the State Methods:

1) Data are inconsistent between various EPA GHG datasets;

2) Methods are inconsistent between the national Inventory, the disaggregated state GHG
inventories, and the State Inventory Tool (SIT); and

3) Stakeholder confusion and associated decreased confidence are likely to result from
comparison of EPA-developed state GHG inventories and SIT-based state-developed
GHG inventories.

They emphasized the importance of data consistency because it is critical for confidence in a
dataset and any strategies or policies that are based on that foundational data. Otherwise,
stakeholders will question which dataset is accurate. Specifically, the differences in results from
the State Methods and the SIT differ significantly. They noted that while the reasons for
variability in the State Methods and the national Inventory results are explained in the State
Methods report, but the differences between state GHGI data and SIT default data are not
defined. They commented that this is concerning since these are the two datasets likely to be
compared as many states currently rely on the SIT for development of state GHG inventories.

Maine's DEP continued, commenting that state stakeholders compare SIT default data with the
Maine-produced GHGI inventory, and the SIT is augmented when more accurate, state-specific
data are available. These changes are tracked and transparent in the revised SIT models, and in
contrast input differences and methodology of the State Methods and the SIT are unclear.
Reviewers from Maine's DEP noted that unlike in the State Methods, the SIT makes explicit
input data and data sources and is more transparent. Furthermore, the State Methods doesn't
explain where and how data are used when they are included. They recommended that the EPA
hold the release of the disaggregated state data until the SIT default data for each state better
matches the state data disaggregated from the national Inventory.

Reviewers from Rhode Island's Department of Environmental Management (DEM) found the
report had clear and thorough descriptions of methods. They suggested including references at
the end of each chapter. Reviewers from Alaska's Department of Environmental Conservation
(DEC) found that the methods were clearly laid out and established.

EPA Response: The EPA recognizes that a number of states rely on the SIT as a starting point
for preparing their state GHG inventory estimates. The SIT includes default activity data and
estimates that states can use as a starting point for compiling a state-level GHG inventory. The
default data included in SIT are largely consistent with the EPA '.s state-level inventory
because the data are based on methodologies, emissions factors, and other data from the
national Inventory. However, some differences exist between default data in SIT and the
GHGI by state estimates due to differences in methods, data, and level of completeness. As

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

additional state-level data and/or methodological approaches become available through the
national Inventory disaggregation, they will be used to supplement or improve the embedded
calculations and defaults in SIT as appropriate. SIT users will retain the ability to customize
the tool with their own data in lieu of using defaults.

The EPA has also provided a fact sheet with information cross walking the state-level GHG
estimates with the SIT methods. The fact sheet is available online here:
https://www.epa.gov/ghgemissions/state-ghg-emissions-and-removals.

Finally, EPA has included data appendices to accompany the State Methods report. The data
appendices do include input data to enhance transparency of methods and data.

2. What recommendations do you have to add to or improve the overall transparency,
completeness, consistency and accuracy of this report?

Reviewers from Alaska's DEC noted differences in Alaska specific values from national
averages in land use and land use change (LULUCF), decomposition rates in landfills,
population using waste treatment systems and electrical grid, agriculture, and many other sectors.
Responses to their detailed questions are listed below in the chapter specific questions.

Reviewers from Texas Commission on Environmental Quality (CEQ) suggested providing an
executive summary. Reviewers from Minnesota's Pollution Control Agency (PCA) commented
that there are significant differences between the boundaries of the U.S. state Inventory and the
inventory produced by the Minnesota Pollution Control Agency. This creates challenges for
explaining the different results, but they also recognized that the complexity of estimating
emissions for different purposes creates distinct but complementary results.

EPA Response: EPA thanks Texas CEQ for their suggestion and will consider including an
executive summary in future annual publications of the Inventory by State. EPA thanks
MPCA for their thoughts on explaining differences and acknowledging the different but
complementary purpose of this information. EPA will consider including comparative results
and an executive summary in future annual publications of the GHG Inventory by U.S. State.

Reviewers from Rhode Island's DEM suggested including a comparison of the methodology
used with the EPA's SIT since many states use the SIT to calculate state GHG emissions. They
also suggested including sources and estimates of any state level data used in emissions estimates
and including specific data sources and estimates, including the data used to estimate GHGs.

EPA Response: The EPA recognizes that a number of states rely on the SIT as a starting point
for to preparing their state GHG inventory estimates. The SIT includes default activity data
and estimates that states can use as a starting point for compiling a state-level GHG inventory.
The default data included in SIT are largely consistent with the EPA '.s state-level inventory
because the data are based on methodologies, emissions factors, and other data from the
national Inventory. However, some differences exist between default data in SIT and the
GHGI by state estimates due to differences in methods, data, and level of completeness. As
additional state-level data and/or methodological approaches become available through the
national Inventory disaggregation, they will be used to supplement or improve the embedded

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

calculations and defaults in SIT as appropriate. SIT users will retain the ability to customize
the tool with their own data in lieu of using defaults.

The EPA has also provided a fact sheet with information cross walking the state-level GHG
estimates with the SIT methods. This fact sheet is available online here:
https://www.epa.iiov/iihiiemissions/state-iihii-emissions-and-removals.

Reviewers from Arizona's Department of Environmental Quality (DEQ) Air Quality Division,
Technical Analysis Unit noted that it would be helpful to include more details on the calculation
of the emissions for each sector. They continued that control measures or existing best practices
aimed at reducing GHG emissions should be considered in the emission estimates where
possible. Any activity data that incorporates these measures should be specified in the report. For
example, for Electric Power sector, is it possible to consider the heat rate improvement actions?
Many power plants have applied that to reduce emissions but it seems that these improvements
are not considered in this report, according to reviewers from Arizona's DEQ.

EPA Response: The EPA has provided additional level of detail on how state-level emissions
were calculated and also further clarified Appendices with additional data on calculations. In
terms of the questions on if the report accounts for mitigation efforts, it depends on the source
and method. In particular with the electric power sector emissions by state are based on fuel
used so it would account for heat rate improvements. It is outside the scope of this report to
provide information on control measures more generally but there are other existing GHG
mitigation resources from EPA available (see for example:
https://www.epa.gov/statelocalenergy).

Reviewers from Iowa's Department of Natural Resources (DNR) compared Iowa DNR
emissions for 2015 through 2019 as published in the 2019 GHG Inventory Report & Technical
Support Document (TSD) to the EPA's state GHG Trends and Emissions and Sinks by Gas.

They mapped the sectors as closely as possible since the two sources have different
disaggregation. They cautioned that Iowa DNR's Inventory data refer to activity during the
previous calendar year, making it out of sync with the EPA's national Inventory and SIT because
they calculate emissions for two years before the current year. Furthermore, reviewers from Iowa
DNR recommended the EPA carefully consider the wording of its announcement of publicly
releasing the data and to include information on the webpage housing the data that helps to
clarify the data that the EPA is posting. Regarding the completeness, they identified possible

gaps in the Inventory (Table 1) for Iowa.

Table 1: Potential Gaps in Sectors in Iowa

Non-Energy Use of Fuels CO

Natural Gas Systems CO , N O

Glass Production CO

Carbon Dioxide Consumption CO

Urea Consumption for Non-Agricultural Purposes CO

Petrochemical Production CO

Carbide Production and Consumption CO

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

International Bunker Fuels

co2, ch4, n2o

Composting

ch4, n2o

Anaerobic Digestion at Biogas Facilities

ch4

N20 from Product Uses

n2o

Iowa's DNR also identified areas where there were significant discrepancies between DNR
EPA values for the same category. These sectors can be found in Table 2.

Table 2: Sectors Identified with Significant Differences between Iowa DNR and EPA

Sector

Pollutant(s)

Industrial Fossil Fuel Combustion

o
u

Cement Production

o
u

Lime Production

o
u

Other Process Uses of Carbonates

o
u

Incineration of Waste

co2, ch4, n2o

Ammonia Production

o
u

Liming

o
u

Urea Fertilization

o
u

Mobile Combustion

co2, n2o

Natural Gas Systems

o
u

Enteric Fermentation

o
u

Manure Management

co2, n2o

Wastewater T reatment

co2, n2o

Stationary Combustion

n2o

Electronics Industry

HFCs

Electrical Transmission and Distribution

SFs

LULUCF

o
u

EPA Response: The EPA appreciates the comments on the comparison of Iowa DNR's
inventory with that provided in the GHG Inventory by U.S. State. For many of the differences,
it is clear where there are differences in scope/methodology that account for differences. For
other categories, the EPA will work with Iowa to help identify the reason for discrepancies.
The EPA will add a caveat to the release of the data to indicate they are not considered official
state-level GHG data and provide references to state-level official reports.

3. Data availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are not
currently included, but could be incorporated into this analysis?

b. For national level datasets that are currently used, are you aware of other
comparable datasets of activity, emission factor, or emissions data that are available
at the state, county, or zip-code levels?

Reviewers from Minnesota's PC A commented that they hope that ongoing collaboration with
EPA will help MPCA improve the estimates for their GHG inventory. There were some areas

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

where MPCA does not have disaggregated data available. They noted that full transparency and
access to input data are important for their ability to compare methods and estimates and to
incorporate information into their inventory.

Reviewers from Arizona DEQ noted that providing facility-level data such as CO2 emissions
reported to state and Local Emissions Inventory System (SLEIS) from each facility would be
useful. They noted also that weights and measures might be able to help provide local data on
transportation fuel consumption. They referenced EPA's Facility level Information on
Greenhouse Gases Tool (FLIGHT) database which includes statewide CO2 emissions for the
year 2019 and was recently updated in September, 2020.2 They also referenced the EIA's energy
related emissions.3

EPA Response: The EPA appreciates the responses from Minnesota PCA and Arizona DEQ.
The EPA has provided some additional information and clarity on the use of other data
sources such as the EIA SEDS and GHGRP data. The EPA also notes that GHGRP data by
state are available through their State and Tribal Fact Sheets found here:
https://www.epa,sov/shsreportins/shsrp-state-and-tribal-fact-sheet.

4. Uncertainty. Currently uncertainty ranges are not included for the state level estimates.
Please provide feedback on what qualitative and quantitative information would be
useful. Timeseries Coverage. Currently state data covers 1990-2019 consistent with the
2021 National GHG Inventory, and inclusive of most known baseline periods for
climate policy. Subsequent publications of this data will also strive to maintain this
consistency with the National Inventory. As state-specific input datasets are not always
available over the entire timeseries, understanding which years may be more important
can help us to better prioritize our backcasting and methodological efforts across the
time series. EPA appreciates feedback on which, if any years should be prioritized for
future state-level estimates (e.g., 2000 and later, 2005 and later, 2010 and later, or the
full time series).

Reviewers from Virginia's Department of Environmental Quality (DEQ) compared
disaggregated values from the state draft methodology report with state emissions obtained from
SIT and/or GHGRP oriented methods and found them consistent within ±10% accuracy. They
noted that summary emissions trends are very similar, while emissions disaggregated on the
basis of the national Inventory are the least in recent years compared to values reckoned by the
state. They emphasized, "Differences in the accounting decisions do not necessarily indicate that
one of the estimates is accurate, or 'correct', but it will make comparability more difficult."

2 U.S. Environmental Protection Agency. (08/07/2021). Facility Level Information on Greenhouse gases Tool,"
Available at: httos ://ghgdata. epa. gov/ghgp/main. do

3 U.S. Energy Information Administration. (2020). "Energy-Related CO2 Emission Data Tables," Available at:
https://www.eia.gov/environment/emissions/state/

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Figure 1 shows the comparison of the gross emissions in Virginia between the State Methods and
scenarios related to the SIT.

EPA Response: The EPA appreciates the comments from Virginia's DEQ. The EPA notes that
several emission sources were missing from the first release of the state-level data, including
for example, some key Ag sources (e.g, Ag soils) and some key LULUCF categories, which
could be leading to differences. The EPA also recognizes that there are differences with the
SIT but that as additional state-level data and/or methodological approaches become available
through the national Inventory disaggregation, they will be used to supplement or improve the
embedded calculations and defaults in SIT as appropriate. SIT users will retain the ability to
customize the tool with their own data in lieu of using defaults.

Reviewers from Rhode Island's DEM found the timeseries of 1990 through 2019 to be consistent
with the Rhode Island GHG Emissions Inventory and had no additional feedback on any years
which should be prioritized.

EPA Response: The EPA thanks Rhode Island's DEMfor their feedback.

5. Key Category Analysis. EPA anticipates prioritizing methodological refinements for
more significant categories to make efficient use of available resources over time. EPA
appreciates feedback on which categories are more relevant for further refining for
your state.

a. Given that the emissions profile of some states will be different from the national
average, which categories that are more significant in terms of absolute emissions,
or have changing emission trends (e.g., increasing, variable)?

Reviewers from Rhode Island's DEM noted that in Rhode Island, transportation, electricity
consumption and residential heating are the largest sources of GHG emissions. They wondered
why EPA had electricity emissions estimated at 0.660 MMT CChe in 1990 since Rhode Island's
estimated electricity consumption in 1990 was 2.82 MMT CO2Q in 1990, which is a 327%
difference in estimates. They wondered if there is an idea of why these estimates vary so
significantly. Finally, they noted in every year, electricity emissions are higher in Rhode Island's
GHG Inventory when compared to EPA's state level inventory and emissions vary by 10-20%.

EPA Response: The EPA used the EIA SEDS data on fuel used to produce electricity in a
given state to determine emissions from electric power generation within a state, including for
RI. The SEDS data are based on power plant data reported by state and represent fuel use and
emissions associated with electric power produced in the state. For 1990, the SEDS data show
much lower fuel use for electricity production than in 1991 and later years, causing lower
emissions in that year. The electricity consumption data for RI (also from SEDS) show fairly
consistent electricity consumption for 1990 and later years, so if emissions are based on
consumption, the 1990 emissions would be higher than those based on production in RI.
Electric power sector emissions in other years are likely different for the same reason. The
EPA will follow up with RI to better understand differences in electric power sector emissions.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

b. The national Inventory includes a key category analysis (KCA) consistent with
2006 IPCC Guidelines. Would it be useful for states if a key category analysis
(KCA) was completed for each state?

Reviewers from Rhode Island's DEM found the KCA consistent with 2006 IPCC Guidelines.

EPA Response: The EPA thanks Rhode Island's DEMfor their feedback.

6. Data Presentation and Usability.

a. Are there other ways the state-level emissions data could be presented to facilitate
their use (e.g., in the EPA GHG Inventory Data Explorer available online at:
https://cfpub.epa.gov/ghgdata/inventoryexplorer/)?

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could
be helpful?

b. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

c. What data format would best facilitate the use of the state-level emissions data (e.g.,
.xlsx download, etc.)?

d. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

e. EPA plans to provide users additional information on where they can find official
state data, where it exists. Do you have suggestions on how we should direct users to
official state data (i.e., section in methods report including links to state data)?

Reviewers from Rhode Island's DEM recommended having data available in .xlsx, or .zip files.
Reviewers from Arizona's DEQ suggested .xlsx format and suggested that available text and .csv
downloads and an FTP site would be useful too. They suggested that, if possible, monthly data
disaggregation could be useful, especially for sectors with seasonable variability. Furthermore,
they suggested that browsing by state, year, fuel, sector, or facility level could be useful. Finally,
they suggested including a separate section containing all necessary links for each section and
disseminating information via workgroups EPA is involved with would also be helpful.

Reviewers from Rhode Island's DEM also suggested any data used to estimate GHG emissions
in Rhode Island (e.g., energy sales, population, er capita emissions factors used) be included as
additional datasets to increase the usability of state-level emissions data. Finally, they referenced
the Rhode Island Department of Environmental Management (RIDEM) GHG Emissions
Inventory.4 Reviewers from Alaska's DEC suggested including links to state websites and links
to updated NEI data to direct users to official state data.

EPA Response: The EPA appreciates the specific suggestions on additional datasets, data
disaggregation, and data formats to enhance usability of the state-level emission data. The
EPA will consider these suggestions with future publications of the state data. The EPA has
published the data in the GHG Data Explorer and it can be downloaded in .xlsx and.csv

4 Rhode Island Department of Environmental Management. (2021). "Greenhouse Gas Emissions Inventory,"
Available at: http://www.dem.ri.gov/programs/air/ghg-emissions-inventorv.php

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

formats. The GHG Data Explorer also allow users to export charts and graphs in .png and
jpeg formats. The EPA recognizes that there will be differences between the EPA's state-level
estimates and some inventory estimates developed independently by individual state
governments. The EPA will provide caveats that inventory data presented here should not be
viewed as official data of any state government. Additional information will also be provided
on official state data, where those data exist, including information on potential areas of
difference between the EPA's data and official state data. Learn more about GHG emissions
and the NEI here: https://www.epa.gov/air-emissions-inventories/are-ghgs-national-
emissions-inventory-complete.

2.2 Energy

1. What are your overall impressions of the clarity of this section?

Reviewers from Arizona's DEQ thought that the methodology and emissions estimate for Fossil
Fuel Combustion sectors were clear, especially the description of why some fuel use reported in
SEDS may be different from reporting in the national Inventory. They found the tables provided
in the appendices helped to explain adjustments made to different sectors. However, they noted
more bottom up data collection would be helpful. For the Non-Energy Uses of Fossil Fuels, they
thought the emissions calculation method was clearly laid out.

EPA Response: The EPA thanks the reviewers for their comments. The EPA has added some
more detail and clarification to the report and appendices to help clarify the data and methods
used

2. What recommendations do you have to add to or improve the overall transparency,
completeness, consistency, and accuracy of this chapter?

Reviewers from Rhode Island's DEM suggested discussing the transportation C02 emissions
from mobile combustion in one section. Additionally, they noted that RIDEM uses MOVES to
calculate transportation emissions and they calculate electricity consumption and not generation.
They agreed with the methodology for determining the carbon content of fuels.

Reviewers from Arizona's DEQ noted that Section 2.6 of the State Methods "describes that the
total amount of distillate fuel and motor gasoline used in Transportation sector was taken from
the national Inventory (based on FHWA data). The totals are said to be based on multiple factors
to determine transportation sector fuel use. A more specific description of the mentioned factors
would be useful. It would also be useful to know how EPA plans to investigate why the total fuel
use on the FHWA forms used to allocate transportation sector gasoline and diesel fuel use across
states doesn't match with the EIA statistics. More details regarding how FHWA determines fuel
consumption would be helpful as well. The transportation section provided information on the
carbon content of the fuel, among other details. Some discussion of carbon content in MSW
fossil components/emission factors would be helpful."

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: The EPA appreciates the comments. The EPA has added more detail to the
methodology report on calculation of both national- and state-level transportation sector
emissions and specifically the differences with the state-level and national-level data used The
EPA continues to allocate transportation sector emissions to states based on fuel sales data,
consistent with the IPCC Guidelines, but will continue to look at other methods including the
MOVES model as part of ongoing QA/QC. The EPA is updating the national-level waste
incineration calculations as part of ongoing planned improvements, and those updates will be
incorporated into the state-level estimates when finalized.

3. Data Availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are not
currently included, but could be incorporated into this analysis?

b. For national-level datasets that are currently used, are you aware of other
comparable datasets of activity, emission factor, or emissions data that are available
at the state, county, or zip-code level?

Reviewers from Arizona's DEQ suggested that for the Transportation sector, prioritizing years
after 2012 because this is when E15 fuel was first approved. They suggested that for the Electric
Power sector, emissions for 2010 and later should be prioritized for Arizona because there was a
significant increase followed by a decrease in emissions, which is not the same trend observed
for other states. They suggested for NEU of Fossil Fuels, emissions from 2008 and later should
be prioritized because NEI data began in 2008 and the two databases can complement each other.
Finally, for Stationary Combustion, they suggested prioritizing emissions from 2005 and later
because peak use of coal and natural gas for Station Combustion occurred in 2006 and began to
decrease after 2010.

EPA Response: The EPA appreciates the comments on time frames to consider for prioritizing
analysis.

Reviewers from Alaska's DEC suggested that the NEI for the state of Alaska could be included
for Fugitive Emissions and for Point and Non-point sources. They also suggested the data from
Alaska Oil and Gas Conservation Council (AOGCC) on flaring (over one hour) and fills in the
gaps on routine practices and flaring less than one hour. Finally, they noted for the Oil and Gas
Sector, the state contractor used the NEI to generate GHG emissions and they found this
representative.

EPA Response: The EPA thanks Alaska's DEC for their comment. We note that there is much
coordination between the EPA '.s NEI and GHG Inventory teams, and we often use the same
input data (e.g., activity data) to calculate emissions. While the NEI does quantify CH4
emissions for some sources in the NEI O&G Tool as an intermediate step to quantifying
VOCs, those estimates are not available for all sources. The national GHG Inventory uses
data reported to GHGRP to quantify emissions from flaring. Data from flaring in Alaska are
included in that dataset. For future annual publications of this report, the EPA will consider

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

conducting a comparison of the GHG Inventory state flaring values with other state data
sources (e.g., AOGCC) to identify potential improvements.

4. Uncertainty. Currently uncertainty ranges are not included for the state-level estimates.
Please provide feedback on what qualitative and quantitative information would be
useful. Time Series Coverage. Currently state data cover 1990 through 2019 consistent
with the 2021 National Inventory and are inclusive of most known baseline periods for
climate policy. Subsequent publications of these data will also strive to maintain this
consistency with the national Inventory. Because state-specific input datasets are not
always available over the entire time series, understanding which years may be more
important can help us better prioritize our backcasting and methodological efforts
across the time series. The EPA appreciates feedback on which, if any, years should be
prioritized for future state-level estimates (e.g., 2000 and later, 2005 and later, 2010 and
later, or the full time series).

Reviewers from Virginia's DEQ compared State Methods emissions with SIT and GHGRP
results and found similarity in trends irrespective of method for computing emissions and the
spread converges especially in recent years supporting near consistency. They noted, "however,
GHGRP report oriented data, the highest of all the assessments is slightly higher than
disaggregated values. It may be pointed out that total emissions reported by GHGRP between
2010 & 2019 account for above 90% of all emissions from power industry in the state."

EPA Response: The EPA appreciates the comment. The EPA is preparing a fact sheet
comparing SIT data with those presented as part of this analysis to help clarify where there
may be differences. The EPA uses the GHGRP data in multiple sectors to help allocate
national-level total emissions to states.

5. Key Categories. The EPA anticipates prioritizing methodological refinements for more
significant categories to make efficient use of available resources over time. The EPA
appreciates feedback on which categories are more relevant for further refining for the
sector you are reviewing.

Reviewers from Arizona's DEQ suggested a KCA for Electric Power sector because it has
changed significantly over time and decreased in past years. They recommended a KCA for each
state because the energy structures for power plants are different. Furthermore, they
recommended a KCA for Stationary Combustion for Arizona because there is a significant
difference between the national level and the state level emissions Inventory. A KCA of the
Mobile Source sector may be useful, they suggested. Finally, they suggested refining Natural
Gas systems in Arizona because it is the fourth largest GHG sector. Reviewers from Alaska's
DEC suggested refining assumptions and results regarding pipeline compressor stations,
pipeline, and tank emissions. Oil and gas rigs are different on the North Slope, they noted. The
infrastructure for the lower 48 states does not match what is in Alaska.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: EPA thanks the Arizona DEQfor their perspectives on categories more
relevant for refining methods to estimate emissions. As commenters suggested, for future
iterations of the GHG Inventory by U.S. State, the EPA is considering alternate state-level
activity data for allocating state emissions such as processing plant information for the
natural gas processing segment, and station data rather than pipelines for the natural gas
transmission segment. The EPA will also consider approaches that use additional GHGRP
data to further refine state-specific estimates, such as for tanks and other production
equipment.

6. Data Presentation and Usability.

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could be
helpful?

iii. What data format would best facilitate the use of the state-level emissions data
(e.g., .xlsx download)?

b. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

For the Transportation Sector, reviewers from Arizona's DEQ suggested further disaggregation
by vehicle age when looking at light-duty vehicle emissions and breaking vehicles up by Tier
type (Tier II or Tier III) to illustrate lower emission rates for newer vehicles. For the Electric
Power sector, they recommended finer disaggregation and providing facility-level data. Finally,
they suggested disaggregating data for the Stationary Combustion sector at the facility, county,
or non-attainment level.

EPA Response: The EPA provided data as part of the state review process in the level of
disaggregation that was available. With final release of the data, other disaggregation and
ways to look at the data will be available in Data Explorer to the extent the state-level data are
available. Data on state-level fuel use by vehicle type were not readily available; more detail
was added to the report on differences in national- and state-level transportation sector data
used

2.2.1 Combustion
2.2.1.1 Fossil Fuel Combustion

1. Some fuels have differences in consumption data between the aggregated state-level
totals and national totals. The current approach is to use data from the national
Inventory in those cases. Are there other approaches that could be taken? Do you know
of cases where others have dealt with the differences in the totals, and if so how?

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Alaska's DEC suggested calculating a percentage of emissions from NEI data of
large facilities of state emissions and updating the scaling factor every three years. They also
noted that SEDS data has been beneficial and accounts for small facilities that don't report
emissions through the permit program to complement NEI data. Additionally, they noted that
more engagement with state-level planners is needed to generate aggregate totals. This is because
Alaska has a unique relationship with both aviation and maritime transportation due to the state's
infrastructure arrangements. They noted that fuel consumption habits vary in the state compared
to utilization in the continental US. There is continued use of large amounts of light aircraft to
bring passengers and cargo to and from remote communities, some of which still burn leaded
AvGas and non-standard fuels. They also commented that Alaska receives a large amount of
international marine traffic via the Great Circle Route which needs to be considered for long-
term fuel consumption data. Many international vessels arrive in state waters using International
Maritime Organization (IMO)-designated low sulfur content fuel which generates other criteria
area pollutant (CAP) and GHG emissions due to fuel chemistry from prior bunker fuel. The
state's adjacent location to major developing trans-Arctic shipping routes could also change how
the state's fuel consumption habits appear on national-level inventories.

Reviewers from Alaska's DEC also strongly recommended that rather than relying on generic
nationwide data, EPA coordinate with the NEI team on aviation and maritime inventories. Both
have developed specific activity data which incorporates satellite location-drive emissions data
and actual engine activity data. They recommended this approach because it would be more
accurate and CAP emissions have already been generated by this data. The only missing
component at that stage would be the application of emissions factors for GHG emissions, they
noted. The same applies for state's aviation inventory which is already generated by landing and
takeoff data from NEI. This NEI dataset also contains activity and landing and takeoff data from
all categories of aircraft.

EPA Response: The EPA appreciates the comments on potential data sources for
transportation sector emissions. The EPA continues to use SEDS as the primary data source
for allocating combustion emissions to states. The EPA has compared the SEDS data with
NEI data for transportation and will continue to look at NEI data as an alternative data
source and to be used as a QA/QC comparison. Updates will be incorporated into future
reports as applicable.

For Point Source emissions, reviewers from Texas CEQ recommended developing electric utility
fuel consumption at the state or site level using EIA data. Regarding Area Source categories,
reviewers from Texas CEQ suggested adjusting state-level totals to match the national totals for
consistency is reasonable since the adjustments are mostly small (less than 5%). Reviewers from
Rhode Island's DEM identified no other approaches and used SIT default fuel consumption data
when developing Rhode Island's annual GHG Inventory.

EPA Response: The EPA appreciates the comments on data sources. The EPA relies on EIA
SEDS data, which are based on the same data sources used by EIA in their electric power
sector results. The EPA is planning on providing more detail on the comparison of this state-
level data with results of the SIT.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Arizona's DEQ commented that for most sectors, this approach seemed
appropriate and accurate. However, they noted that for some sectors such as Electric Power, a
bottom-up method to collect activity data or CEMS data from the facility would be more
accurate. They acknowledged that applying this method may not be easy for the Stationary
Combustion sector, but it can be considered an alternative way to collect data.

EPA Response: The EPA appreciates the comment. The EPA '.s approach to allocating
emissions to the state level is to rely on a similar approach to the national Inventory in terms
of calculating fuel use emissions based on fuel use multiplied by an emissions factor. The
EPA does not rely on CEMS data for the national Inventory, but it is a good QA/QC check of
national-level emissions. The EPA will investigate comparisons of power sector CEMS data
with the current approach of using SEDS data at the state level and incorporate into future
reports as applicable.

Regarding Industrial Fossil Fuel Combustion, reviewers from Iowa DNR estimated their Fossil
Fuel Combustion from the industrial sector to be significantly higher than EPA's state GHGI
results, varying from 29% to 56% from between 2015 and 2019. They suggested that Iowa
DNR's inventory includes emissions from this sector that may be double-counted in some IPPU
sectors of Cement Production, Lime Production, Iron & Steel Production, Ammonia Production,
etc. They noted that in the national Inventory, portions of fuel consumption data for several fuel
categories (coking coal, other coal, natural gas, residual fuel, and distillate fuel) were reallocated
to IPPU, as these portions were consumed as raw materials during the non-energy related
industrial processes. This is an area for the Iowa DNR to improve its inventory to be better
aligned with EPA, they commented.

EPA Response: The EPA appreciates the comment and agrees that some of the differences
between the Iowa DNR industrial sector fuel combustion emissions estimates and those
developed by the EPA could be due to adjustments the EPA makes. The EPA adjusts the
industrial sector fuel use combustion emissions to account for some fuel use and emissions
that are accountedfor specifically under the IPPU sector. The EPA will follow up with Iowa
DNR to better understand differences in emissions estimates.

Regarding transportation, reviewers from Iowa DNR calculated highway transportation
emissions using the SIT mobile combustion module. For CH4 and N2O, they used the actual total
annual VMT from Iowa Department of Transportation (DOT).5 Since Iowa DOT does not have
VMT data by model years, DNR allocated Iowa VMT using default national on-road distribution
by vehicle/fuel type. Furthermore, Iowa uses the annual vehicle mile accumulation, age
distribution, and control technology values from the most recent national Inventory. They
calculated CO2 emissions from highway vehicles and all emissions from non-highway vehicles
using fuel sales data from either SEDS, FHWA, or the SIT default value. This hybrid calculation
method results in Iowa emissions that are lower than the disaggregated Iowa values provided by

5 Iowa Department of Transportation. (2019). "VMT by County/System," Available at:
https://iowadot.gov/maps/msp/vmt/countwmtl9.pdf

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA (for CO2 5%, for CH4 2%, and for N2O 14% for 2019). Iowa DNR's calculations may have
a higher level of uncertainty because DNR did not use the true vehicle/fuel distribution for Iowa.
Reviewers from Iowa DNR had no specific concerns regarding the CO2 emissions estimation by
allocated VMT as it seems to be the best method available. They noted that it would be
interesting to see the difference between CO2 emissions based on VMT versus fuel sales,
however.

EPA Response: The EPA appreciates the comment and notes the approaches used by the EPA
and Iowa are very similar for calculating transportation sector emissions. The differences
could be due to a number of reasons, including Iowa DNR using their own specific VMT data
and the fact that the EPA uses national average default factors. The EPA will continue to
evaluate other approaches for allocating transportation sector emissions across state's VMT
and include comparisons in future reports as applicable.

2. Consistent with the IPCC Guidelines, we have adjusted fuel consumption totals in the
energy sector to account for consumption in the IPPU sector. In some cases, this step
could lead to a negative emission total for a state if the subtracted amount (as
determined from the assumed distribution) was greater than consumption data from
the State Energy Data System (SEDS). This outcome was corrected to zero if that was
the case, but are there other approaches for correcting for that difference?

Regarding Point Source emissions, reviewers from Texas CEQ noted in-house energy generation
occurs within some industrial sites (e.g., petrochemical production sites) and adjusting fuel
consumption in the energy sector may not be needed for those industrial sites. They noted EIA
collects consumption and generation data on all generators at sites greater than one megawatt on
the Annual Power Plant Operations Report.

EPA Response: The EPA appreciates the comment. The EPA relies on SEDS data for
allocating electric power sector emissions across states and notes that the SEDS data are
consistent with other EIA-reportedpower plant data. In terms of adjustments in the industrial
sector fuel use, the adjustments are only made to areas where fuel is used as part of the
process and not to generate power. Further information about the adjustments made were
added to the report.

Regarding Area Source emissions, reviewers from Texas CEQ noted that for some Industrial,
Commercial, and Institutional Combustion sources, they have seen an approach where remaining
negative emissions are re-allocated among the other non-zero states to fully account for the
adjustment. However, making the adjustment to zero without re-allocating the remaining
negative emissions is simpler and easier to understand.

EPA Response: The EPA appreciates the comment and will continue to evaluate methods of
making adjustments to account for the potential of negative emissions. Updates will be
included in future reports as applicable.

Reviewers from Rhode Island's DEM identified no other approaches. Reviewers from Arizona's
DEQ found the approach to be generally reasonable but noted whenever a negative value is

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

increased to 0, there should be a deduction somewhere else. They suggested that the total of the
negative values should be tabulated, converted to zero, and then all state totals should be scaled
down uniformly by the appropriate amount.

3. Consistent with the national Inventory, the default approach taken here was to allocate
transportation sector CO2 emissions based on Federal Highway Administration
(FHWA) fuel use/sales by state. For some states, this may not be accurate because fuel
sold in a state may be combusted in other states. Another option is to use vehicle-miles-
traveled (VMT) data by state but that approach does not factor in vehicle fuel economy.
Are there other alternative or complementary approaches to allocate transportation
fuel across states, including VMT data and other sources (e.g., NEI—based on county-
level fleet and activity data to generate a bottom-up inventory) that the EPA should
consider? If so, what data sources exist to help with that alternative approach? Would
it be helpful to present transportation sector emissions using multiple approaches in
future inventories?

Reviewers from Alaska's DEC noted that the existing methodology is appropriate since Alaska
doesn't have adjacent states. They also noted that the SEDS dataset is reliable and should be
used, as it is with the SIT. They also recommended looking at NEI data on mobile sources
combustion, especially on- and off-road maritime, and aviation emissions. They provided a final
caveat that FHWA data on registered vehicles are not complete because native populations do
not have to register vehicles that don't travel on highway systems. In the triennial NEI, Alaska
uses a 10% adjustment to compensate for these unregistered vehicles.

EPA Response: The EPA appreciates the comment on the potential for unregistered vehicles
and will evaluate the possibility of making adjustments similar to the NEI in future reports as
applicable.

Reviewers from Texas CEQ noted that, consistent with EPA requirements, on-road emissions
inventories are developed using emissions factors and corresponding activity levels. The same
method can be used to generate both CO2 and non-CCh emissions rates for on-road mobile
sources. Using the CO2 emissions rates from the national emission factor model combined with
the corresponding activity levels is the preferred method for producing emissions estimates
consistent with other pollutants.

EPA Response: The EPA appreciates the comments on data sources for developing state-level
transportation sector emissions. The EPA continues to use SEDS as the primary data source
for allocating combustion emissions to states. The EPA will continue to evaluate alternative
data sources, and updates will be incorporated into future reports as applicable.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Rhode Island's DEM noted that Rhode Island has rough transportation data
provided by the Rhode Island DOT. Additionally, they noted, "current transportation data have
significant limitations such as county level VMT data and speed distributions, but updates to
transportation data collection are being considered. Rhode Island is in the process of creating
Motor Vehicle Emissions Simulator (MOVES) inputs using in state data. For MOVES inputs
prior to 2017, Northeast States for Coordinated Air Use Management (NESCAUM) and other
contractors provided Rhode Island with the inputs required. No MOVES inputs have been
created for 2018-2020. The transportation data necessary for MOVES is in the process of being
created for 2020 ahead of the National Emissions Inventory (NEI) submission."

EPA Response: The EPA appreciates the comments on data sources for developing state-level
transportation sector emissions. Although the EPA is not adopting the NEI/MOVES approach
as the primary method for this annual effort, the EPA will continue comparisons with the NEI
and other approaches of allocating vehicle emissions.

Reviewers from Arizona's DEQ commented that a bottom-up inventory using NEI activity data
would help mitigate the possibility of allocating CO2 emissions based on fuel sales that may not
directly translate to fuel usage by state. They noted that it would be helpful to present
transportation sector emissions using multiple approaches.

EPA Response: The EPA appreciates this comment regarding bottom-up accounting of
vehicle emissions. The EPA continues to base vehicle estimates on point-of-fuel sales as the
primary approach for allocating vehicle emissions to states, as it is consistent with the IPCC
guidelines and ensures no double-counting or gaps. The EPA will continue making
comparisons with the other approaches of allocating vehicle emissions.

4. Mobile source non-CCh emissions are allocated across states based on vehicle-miles-
traveled data while mobile source CO2 emissions were allocated based on fuel sales, as
mentioned above. Is there an issue with using two different methodologies for mobile
source CO2 vs. non-CCh state splits?

Regarding on-road mobile missions, reviewers from Texas CEQ recommended consistent
methods for estimating CO2 and non-C02 emissions. The activity factors that have already been
developed for non- CO2 pollutants can be used for CO2. Reviewers from Rhode Island's DEM
had no concerns about the difference in methodology but wondered why the EPA decided to take
two different approaches to estimate mobile emissions. Reviewers from Arizona's DEQ did not
think there is a concern with using two different methodologies as long as the fuel type and
quantity of fuel burned is noted. Reviewers from Alaska's DEC had no concerns with using two
different methodologies.

EPA Response: The EPA appreciates the comments. The national- and state-level inventories
use different methodologies for calculating CO2 and non-C02 emissions for vehicles. CO2
emissions are derivedfrom fuel consumption data, while non-C02 emissions account for fleet

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

information (i.e., age distribution/vehicle technology breakout, VMT) in the calculation of
non-CO2 emissions. More detail on this is provided in the report, and further clarity can be
provided in future releases as needed.

5. Several fuels have variable C factors over time including coal, natural gas, gasoline, and
diesel fuel. Those fuels might also have variable C factors across areas/states. Are data
available to build out state-specific C factors for the fuels with variable C contents? If
so, could it be done in a way that the state-level total emissions still matched up to the
national total emissions for those fuels?

For Texas CEQ, the Texas CEQ Commission on Environmental Quality (TCEQ) utilizes
multiple data sources to develop state-specific fuel property profiles for the six fuel regions in
Texas. The two data sources are: a TCEQ-sponsored statewide triennial fuel study and fuel
compliance data submitted to the EPA for federal reformulated gasoline. The two data sources
are used in conjunction with regulatory information and default data to develop Texas-specific
fuel property profiles for the six Texas fuel regions for both historical and future year
assessments, according to reviewers from Texas. Reviewers from Rhode Island's DEM
commented that Rhode Island uses SIT defaults for carbon factors and does not have additional
information on state specific factors. Reviewers from Arizona's DEQ were not aware of any data
available to build out state-specific C factors for the fuels with the variable C content.

EPA Response: The EPA appreciates the comment on state-level C factors. The EPA
continues to use national average C factors to develop the state-level inventory values but will
continue to investigate the possibility of using state-specific C factors in future reports,
specifically for liquid transportation fuels. The EPA will also consider the possibility of
applying state-specific factors as part of the national Inventory development as part of
potential future improvements.

6. Geothermal emissions could be allocated by the type of geothermal production per state
(because different types have different emissions factors) if that data are available. Is
there more information on state-level geothermal emission factors and production?

No reviewers were familiar with up-to-date geothermal datasets or commented that it was not
relevant for their state. Reviewers from Alaska's DEC noted that there is significant volcanic
geothermal activity throughout Alaska. USGS and University of Alaska-Fairbanks have large
databases of information on geothermal power options that should be reviewed and used if
possible, they commented. They noted although these are natural sources, these data would be
valuable to include in the report.

EPA Response: The EPA appreciates the comment on possible sources of geothermal
emissions. The EPA also notes that although volcanic activity may result in emissions, the
focus of the report as per IPCC guidelines is on anthropogenic sources. The term
"anthropogenic," in this context, refers to greenhouse gas emissions and removals that are a

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

direct result of human activities or are the result of natural processes that have been affected
by human activities.

2.2.1.2 NEU

1. For petrochemical feedstocks, non-energy use (NEU) of natural gas is allocated across
states based on petrochemicals emissions data per state from the IPPU adjustments,
while other fuels are allocated based on the underlying SEDS data. Allocating across
states based on the underlying SEDS data ensures there are no states where NEU use is
larger than original SEDS data and there are no zeros associated with subtracting NEU
(it is not an issue for natural gas because use is so high overall compared with NEU
use). Could different approaches be used or can the petrochemical data be used without
resulting in negative use?

Reviewers were unaware of alternate approaches. Reviewers from North Dakota's Department of
Environmental Quality (DEQ) noted that based on the methodology described in the State
Methods report which included coal used to produce SNG at the Eastman gas plant for chemical
feedstock and therefor accounted for under NEU, the Dakota Gasification Company's Great
Plains Synfuels Plant in North Dakota should be included in NEU as it also uses coal to produce
SNG.

Reviewers from Arizona's DEQ noted that different approaches to allocating NEU across states
should be considered and compared to the current method of using a combination of data from
IPPU adjustments and SEDS. They also thought that the current methodology could be sufficient
considering the lack of use for most fuels.

Reviewers from Iowa DNR noted that there is a potential gap in their inventory as NEU has not
been included. They wondered if this sector will be included in the SIT. Reviewers from
Alaska's DEC trusted the SEDS data and noted it is probably the best source of data for Alaska.
They also noted that there is not a great use for NEU of fuels.

EPA Response: The EPA appreciates the comments and has provided more information in the
report on the data sources used and the step of adjusting energy values to account for NEU of
fuels. In terms of the SNG produced at the Eastman plant, they are included as NEU because
they are used directly as petrochemical feedstocks. The SNG produced at the Great Plains
plant is assumed to be used as supplemental natural gas and, therefore, is accounted as fuel
combustion emissions. The EPA accounts for the CO2 that is exported to Canada for CCS and
it is excluded from the Inventory. More information on these adjustments has been added to
the report.

2.2.1.3 Incineration of Waste

1. Waste incineration emissions are calculated based on the combustion of fossil

components of both municipal solid waste (MSW) and tires. However, emissions are
disaggregated to states based only on MSW tonnage. Are there approaches or data

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

available to disaggregate emissions based on waste category (e.g., MSW combustion vs.
tire combustion)?

Reviewers from Alaska's DEC noted that burning tires is prohibited Alaska except at approved
incarnation facilities. Large incineration facilities are permitted, and their emissions are reported
in the NEI, but smaller rural incinerator facilities are permitted so NEI data are limited. They
suggested developing a scaling technique for village incinerators and assuming they burn tires.

Reviewers from Texas CEQ expressed doubt for values in Table A-l 17 in Appendix A which
shows MSW incineration by state and indicates zeroes for 2001 through 2019 for Texas. For
Area Sources, they use tonnage derived from EPA and were unaware of other data sources to
disaggregate emissions.

EPA Response: EPA data from the GHGRP shows intermittent MSW combustion for the
years 2011-2015 but none after that. As noted below the methodology for this source is being
updated at the national level to include GHGRP data and the results will be reflected in the
next release of state-level data.

Reviewers from Iowa DNR calculated the amount of CH4 mitted from power plants burning
MSW to produce electricity using data reported annually by individual facilities to the Iowa
DNR's Air Quality Bureau on their annual Title V air emissions inventories. The only facility in
Iowa generating electricity from burning MSW reported burning a total of 12,763 tons of refuse
derived waste in 2019. Site-specific proportions of discards that are plastics, synthetic rubber,
and synthetic fibers were used instead of SIT default values to calculate CO2 emissions from
MSW combustion using SIT. These values came from the 2017 Iowa Statewide Waste
Characterization Study. They noted that there may be a gap in Iowa DNR's inventory as they had
not previously considered emissions from other incinerators in the state, nor had they included
site-specific emissions from tire combustion as a separate value, and they expressed uncertainty
that this activity data is easily available. These differences in methodology account for
significant differences in Iowa's state-level estimates compared to State Methods results; the
Iowa DNR's estimates for CO2 are +14%, for CH4 are +9625%, and for N2O are -3%.

EPA Response: The EPA appreciates the comments on updating the sources for waste
incineration. For the current state-level report, the EPA has maintained the current approach
but is planning updates to the national Inventory approach for waste incineration, including
the use of GHGRP data as part of the 1990-2020 report. These updates at the national level,
when finalized, will be implemented in the next state-level inventory report, likely for the
1990—2020 state-level estimates.

2.2.1.4 International Bunker Fuels

1. The approach used to allocate jet fuel bunker fuels by state is currently based on the
total amount of jet fuel used by state, which could potentially lead to an over- or under-
estimation for some states' bunker fuel emissions. Are there other more accurate
approaches to allocate jet fuel bunker data across states as opposed to the percentage of
jet fuel total use? For example, using Federal Aviation Administration flight-level data

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

on departures and destinations or assuming based on states with international airports
and flights?

Reviewers from Alaska's DEC used EIA data for their International Bunker fuels and noted that
Alaska has lots of international travel. They suggested using NEI landing and takeoff data.

Reviewers from Texas CEQ noted that the TCEQ develops non-road emissions from airport
sources on a per-facility approach, using the Federal Aviation Administration's Aviation
Environmental Design Tool (AEDT), and entering aircraft and engine types based on collected
facility activity data. AEDT then applies appropriate emission factors, fuel types used, etc. based
on the aircraft and engine types selected. The TCEQ does not collect or track specific fuel usage
data from any airport facilities as part of the emissions inventory development process.

Reviewers from Rhode Island's DEM noted that Rhode Island has one airport (PVD) that reports
GHG emissions estimates on an annual basis. They know of no breakout of international fuel in
the report provided to RIDEM and no known information about marine bunker fuels.

Reviewers from Iowa DNR noted that they used the total amount of jet fuel used by the state per
SEDS and the SIT module and report emissions in the transportation sector. No international
flights depart from Iowa municipal airports.

EPA Response: The EPA appreciates the comments on state-level bunker fuel data. The EPA
will continue to investigate approaches and data for allocating international bunker fuel data
at the state level, including, for example, flight-level data. The EPA has also provided more
detail on how transportation emissions are developed at the national and state levels in the
report to clarify where emissions from different sources are accounted for.

2.2.1.5 Wood Biomass and Biofuels Consumption

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.2.2 Fugitive
2.2.2.1 Coal Mining

1. Do you have any comments specific to the methodology and emission estimates for
active coal mines and abandoned coal mines?

Reviewers from Arizona's DEQ found this method description appropriate. Reviewers from
Alaska's DEC noted that Alaska's coal mining activity is limited compared to its historical
activity thanks in large part to the end of coal exports to East Asia in the last decade. Alaska has
only one active coal mine presently. Other coal mines have been closed for several years under

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

state and federal reclamation rules. They commented that for the 2020 Alaska Greenhouse Gas
Report, the Alaska Department of Environmental Conservation (ADEC) conducted an inventory
of abandoned coal mines and found the average year these mines were abandoned was 1941.
They also commented that from initial research at the state level, it does not appear that there is
any ongoing methane capture or recovery from abandoned mines in Alaska. Due to the age of
most of these abandoned mines, it is highly unlikely that they would be a viable target for
methane capture or recovery, they argued. Thus, they suggested that any generic methane
recovery emissions should not be applied to Alaska.

EPA Response: The EPA thanks the state reviewers from Alaska's DEC for their feedback
and additional information related to coal mining and abandoned mines in Alaska,

2. Are you aware of any state datasets that may be useful in helping to refine emission
estimates for abandoned coal mines, including state-level datasets addressing recovery
of methane from abandoned mines?

Reviewers from Alaska's DEC recommended data from the Alaska Department of Mining, Land,
and Water and Alaska DNR which outline all identified remaining abandoned mines in the state.
They noted additional research was needed from ADEC to verify the age of these mines and the
dates of mine closure. In instances where the date of the mine closure could not be verified or
was within a five-year period, the earliest estimated date of mine closure was used. These data
can be forwarded to EPA to be used for future NEI and GHG estimates, but reviewers from
Alaska's DEC noted that the database does not have emissions information which needs to be
added in by EPA.

Reviewers from Texas CEQ suggested the Railroad Commission of Texas (RRC) for data on
permitted surface mining, acreage, and abandoned mine programs.6 Reviewers from Arizona's
DEQ acknowledged that a comprehensive inventory of abandoned coal mines in Arizona does
not currently exist.

EPA Response: The EPA thanks the state reviewers from Alaska's DEC, Arizona's DEQ, and
Texas CEQ for their feedback and the state data sources identified in their comments.

2.2.2.2 Abandoned Underground Coal Mines

1. Are you aware of any state datasets that may be useful in helping to refine emission
estimates for abandoned coal mines, including state-level datasets addressing recovery
of methane from abandoned mines?

Reviewers provided no comments.

6 RRC. n.d. "Mining Regions/Fields and Sites." Available at: https://www.rrc.texas.gov/surface-mining/historical-
coal-mining/mining-regions-fields-and-sites/

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2.2.2.3 Petroleum Systems and Natural Gas Systems

1. Are there relevant dataset(s) that could be used to replace or supplement the data

currently used to allocate petroleum and natural gas system emissions to the state level?
Particularly, state or detailed location information on gathering and boosting stations,
processing plants, and transmission and storage stations?

Reviewers from Alaska's DEC commented that they have emission fee data from these sources
as well as NEI information. In 2000, they asked all small nonpoint sources to submit NEI
emission information. Once submitted to EPA, it will be the most complete NEI on these
sources. They noted, however, that NEI does not include fugitive emissions unless it is a
permitted emission unit. Permit emission fee information has more fugitive information. These
data are only available for CAPS but could be used to calculate GHG emissions in some cases,
reviewers from Alaska's DEC suggested.

Reviewers from Texas CEQ recommended information from EIA's triennial EIA-757 Schedule
A, Natural Gas Processing Plant Survey, which tracks the country's population of natural gas
plants and has basic location data. They found the current methodology to allocate petroleum and
natural gas system emissions to the state level reasonable. Reviewers from North Dakota's DEQ
noted that it is unclear how the condensate emissions are estimated, especially for North Dakota
which has substantial GHG emissions associated with condensate. Additionally, they noted that
is unclear if fugitive emissions associated with compressor stations (natural gas-drive and
electric-drive) were accounted for. This would account for many GHG emissions in North
Dakota. Furthermore, they noted that the reported volume of gas production in Appendix B
appears low for 2019. The NG 3 - Annual Gas Production tab (851,750 MMScf) and the NG 2 -
Gas Well Gas Production tab (1,155,856 MScf) totals 852,905,856 MScf of gas produced in
North Dakota in 2019. The North Dakota Industrial Commission numbers estimate
927,804,888 MScf of natural gas produced. Finally, they noted it was unclear how emissions
from storage vessels were estimated. They suggested North Dakota may have more storage
vessels relative to pipeline miles in comparison to other states, which would underestimate
storage tank GHG emissions in North Dakota.

Reviewers from Arizona's DEQ found that the current data set appeared to accurately reflect
permitted natural gas compressor stations in Arizona. Reviewers from Iowa DNR noted that
Iowa has four liquid natural gas (LNG) storage compressor stations, eighteen gas transmission
compressor stations, and four gas compressor stations. They also noted that the SIT Natural Gas
and Oil system module only calculates CH4 emissions for Iowa from transportation and
distribution. However, EPA's disaggregated Iowa values show CO2 and N2O emissions from this
sector, likely from flaring, even though natural gas production or petroleum systems are not
present in Iowa. In addition, the SIT does not include any activity data for flaring in Iowa, so
they were uncertain as to why EPA shows CO2 and N2O emissions for this sector in its
disaggregated Iowa emissions. Finally, they remarked that is also interesting that DNR calculates
CH4 emissions using the SIT values that are 18% higher than the State Methods values for Iowa
for 2019.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: Emissions from compressor stations are included in the state-level inventory
estimates.

As suggested by the commenters, we are considering alternate activity data sources (such as
processing plant data) to develop state-level estimates for future state inventories. We are also
assessing the potential use of additional GHGRP data to develop state-level estimates that
incorporate additional state-specific information on emission sources such as tanks.

We note that there is much coordination between the EPA's NEI and GHG Inventory teams,
and we often use the same input data (e.g., activity data) to calculate emissions. Although the
NEI does quantify CH4 emissions for some sources in the NEI O&G Tool as an intermediate
step to quantifying VOCs, those estimates are not available for all sources. For future state
inventories, the EPA will consider comparing the GHG Inventory state values with other state
data sources to identify potential improvements.

Regarding production data for North Dakota, the value used for production did not include all
gas production from wells classified as oil wells. This will be corrected for future state-level
GHG Inventories.

Emissions from storage vessels are allocated to states based on production volumes, not
pipelines. We will continue to consider update approaches to improve the state-level tank
estimates.

Regarding production segment emissions from Iowa, those emissions resulted from a
spreadsheet error, which has since been corrected.

The EPA notes that there are a number of differences between the current SIT tool that can
result in different emissions between the SIT tool and the state GHG Inventory. As both
products continue to be updated, it is expected that the values will align better. The EPA has
also provided a fact sheet with information cross walking the state-level GHG estimates with
the SIT methods. The fact sheet is available online here:
https://www.epa.gov/ghgemissions/state-ghg-emissions-and-removals.

2. Are there additional Greenhouse Gas Reporting Program (GHGRP) data that could be
used to allocate natural gas and petroleum emissions to each state?

Reviewers from Texas CEQ noted it is difficult to allocate emissions to individual states for
sources that report basin-level data to the EPA's GHGRP since many basins cross state
boundaries. They suggested using county-level data reported to the GHGRP where possible to
use to allocate emissions to individual states. Reviewers from Rhode Island's DEM noted that
data from the Narragansett Electric Company on the GHGRP are used to estimate natural gas
distribution leakage in Rhode Island. Reviewers from Arizona's DEQ noted that there are no
additional GHGRP data that could be used.

Reviewers from Alaska's DEC again recommended NEI data on point source emissions for
allocation.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: We are assessing the potential use of additional GHGRP data to develop state-
level estimates for future versions of the state GHG inventory.

3. Are there particular sources for which state-level regulatory or voluntary programs
result in large differences in emission rates between states? Are state-specific datasets
available for those sources?

Reviewers from Alaska's DEC mentioned the Alaska Bureau of Land Management (BLM)
abandoned well program (also known as legacy wells or orphaned wells) which carries multiple
datasets. Furthermore, they mentioned the existence of an AOGCC dataset for all wells ever
drilled in Alaska. This dataset also lists the well status. They noted that the AOGCC is strict on
plug and abandonment and there are fewer abandoned wells in Alaska than other states. DNR
Division of Oil and Gas estimated there are 20 wells.

Reviewers from Texas CEQ noted that RRC has detailed data on newer wells (generally, those
drilled after 1950). This includes a well status data element that could be used to identify
abandoned and plugged wells. For older historical wells drilled prior to 1950, they were unaware
of a detailed dataset of Texas abandoned wells. The RRC does have some limited statewide well
counts for older wells as noted in the State Methods, but it does not include data at the county
level.

EPA Response: Datasets used to develop abandoned well estimates in the national Inventory
are documented in the national Inventory and in a 2018 memo, Inventory of U.S. Greenhouse
Gas Emissions and Sinks 1990-2016: Abandoned Oil and Gas Wells found online here:
https://www. eya. sov/sites/default/files/2018-

04/documents/shsemissions abandoned wells, ydf. The Enverus/Drillinglnfo reference used
in the national Inventory does incorporate data from AOGCC and RRC. We will consider
assessing other datasets (e.g., from Alaska BLM) for potential updates to the estimates.

4. Are there particular sources for which state-level regulatory or voluntary programs
result in large differences in emission rates between states? Are state-specific datasets
available for those sources?

Reviewers from Texas CEQ noted sites located in ozone non-attainment areas are subject to
additional emissions and control requirements, and those sites' emission rates can differ from
sites located in attainment areas. Reviewers from Arizona's DEQ were unaware of sources for
which state-level regulatory or voluntary programs result in large differences in emission rates
between states.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: The EPA will continue to review information on state-level programs for
potential updates to future annual state-level estimates of inventories.

2.2.2.4 Abandoned Oil and Gas Wells

1. What other relevant data sources could be included? If data are used at the national
level, are you aware of other comparable data sources at the state level?

Reviewers provided no comments.

2.3 Industrial Processes and Product Use (IPPIJ)

1. What are your overall impressions of the clarity and transparency of this section?

Reviewers from Arizona's DEQ thought that the selection of emissions factors for Cement
Production was clear. However, they noted that the improvement of modernized production
methods ought to be considered. Furthermore, there are several new technologies that can reduce
CO2 emissions or even absorb them (such as CarbonCure) which should be considered for the
cement production sector, they noted.

EPA Response: The EPA thanks the reviewer for their feedback. The EPA will assess this as a
potential update for future annual publications of this report.

2. What recommendations do you have to add to the overall completeness and accuracy of
this chapter?

Reviewers from Arizona's DEQ commented that Cement production includes several chemical
processes. It will be more helpful to include different chemical reactions and list the CO2
emissions from each chemical reaction process. Also, improvement technologies and control
measures that are recently used by the cement industry should be considered. Since some cement
facilities have installed continuous emission monitoring systems (CEMS), the CO2 emission data
calculated by the activity method can be correlated and adjusted by CEMS results. Therefore,
some CEMS adjustment methods can be considered.

EPA Response: The EPA thanks the reviewer for their feedback. The State Methodologies
document has details on how the GHGRP data was used to allocate cement emissions to the
different states, including how adjustments were made to facilities that report using CEMS
data. The report is available online at https://www.epa.gov/ghgemissions/state-ghg-
emissions-and-removals. More detail on the chemical process and process equations
associated with cement production can be found in the national Inventory report.

3. Data Availability. Please address the following questions for each inventory source:
a. For each of the categories, are there additional relevant data sources that are not

currently included but could be incorporated into this analysis?

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

b. For national-level datasets that are currently used, are you aware of other

comparable datasets of activity, emission factor, or emissions data that are available
at the state, county, or zip-code level?

Reviewers from Arizona's DEQ expressed interest in CO2 emissions from Cement Production
for after 2005 or 2010 would be helpful because production has increased since 2015 but is still
lower than in 2005.

EPA Response: The EPA thanks the reviewer for their feedback.

4. Uncertainty. Currently uncertainty ranges are not included for the state-level estimates.
Please provide feedback on what qualitative and quantitative information would be
useful. Timeseries Coverage. Currently state data cover 1990-2019 consistent with the
2021 National Inventory and are inclusive of most known baseline periods for climate
policy. Subsequent publications of this data will also strive to maintain this consistency
with the national Inventory. As state-specific input datasets are not always available
over the entire time series, understanding which years may be more important and can
help us better prioritize our backcasting and methodological efforts across the time
series. The EPA appreciates feedback on which, if any, years should be prioritized for
future state-level estimates (e.g., 2000 and later, 2005 and later, 2010 and later, or the
full time series).

Reviewers provided no comments.

Reviewers from Arizona's DEQ suggested a KCA for Cement Production because the CO2
emissions from this sector represents 20% of the total emissions from IPPU sectors in Arizona
and it shows a higher increase trend in the state. They also suggested a KCA would be beneficial
for industrial combustion sector because it is one of the major sources of CO2 emissions in
Arizona.

EPA Response: The EPA thanks the reviewers for their feedback. Consistent with the
national Inventory, in future annual publications of the GHG Inventory by U.S. State, the
EPA plans to provide state-level analysis of key categories in accordance with Volume 1,
Chapter 4 of the 2006IPCC guidelines. This analysis will provide information on significant
emission and sinks source categories in the latest year of the GHG Inventory by U.S. State and
also categories influencing trends for each state.

6. Data Presentation and Usability.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could be
helpful?

iii. What data format would best facilitate the use of the state-level emissions data
(e.g., .xlsx download)?

b. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

Reviewers from Arizona's DEQ suggested disaggregation to the facility level for the Cement
Production sector would be helpful because there are large cement production facilities in
Arizona.

EPA Response: The EPA thanks the reviewers for their feedback. The GHGRP FLIGHT
provides information about GHG emissions from large facilities in the United States. These
facilities are required to report annual data about GHG emissions to the EPA as part of the
GHGRP. FLIGHT is available online at https://shsdata,epa,sov.

2.3.1 Minerals
2.3.1.1 Cement Production

1. Are you aware of data on state-level clinker production for the full 1990-2019 time
series? If not, is there any surrogate data that could be used (e.g., facility production
capacity, utilization rates by facility or state) for 1990-2019 that could refine this state
inventory calculation to enhance accuracy and consistency of state GHG emissions and
trends?

Reviewers from Texas CEQ suggested the USGS National Mineral Information Center.7
Reviewers from Arizona's DEQ mentioned the Arizona clinker production data from SLEIS
since 2010 which could be useful for creating a state inventory.

Reviewers from Iowa noted that Iowa has two Portland cement manufacturing facilities currently
operating. Both facilities are required to report their CO2 emissions to GHGRP, so they used this
GHGRP data for their inventory instead of using clinker production data to calculate emissions.
They noted that the combined CO2 emissions reported by the two facilities for 2019 were 72%
higher than EPA's estimated value for Iowa.

EPA Response: The EPA appreciates Iowa noting this difference. Consistent with IPCC
methodological guidance and international transparency reporting standards, only process

7 United States Geological Survey. n.d. "Cement Statistics and Information" Available at:
https://www.rrc.texas.gov/surface-mining/historical-coal-mining/mining-regions-fields-and-sites/

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

emissions are reported for cement production in this IPPU category, while combustion
emissions are accountedfor under the Energy sector. Some cement facilities report emissions
to GHGRP using CEMS data that includes process and fuel combustion related activities. For
this state level estimate, emissions data from facilities using CEMS to report to GHGRP were
adjusted to differentiate between combustion and process emissions. That could account for
the lower emissions indicated in the GHG Inventory by U.S. State report.

2.3.1.2 Lime Production

1. Are you aware of state-level data on lime production (activity data) by type (e.g., high-
calcium quicklime; dolomitic quicklime, high-calcium, hydrated; dolomitic, hydrated;
dead-burned dolomite; CO2 captured for use in onsite processes) for some or all of the
1990-2019 time series? If not, is there any surrogate data (e.g., facility production
capacity, utilization rates by facility or state) for 1990-2019 that could refine this state
inventory calculation to enhance accuracy and consistency of state GHG emissions and
trends?

Reviewers from Texas CEQ suggested the USGS National Mineral Information Center.8 They
also requested a contact name and email address to share production data from the TCEQ point
source emission inventory on Texas lime production sites.

Reviewers from Iowa noted that Iowa has one lime production facility currently operating, and it
is required to report its CO2 emissions to GHGRP. They used this GHGRP data for their
inventory instead of using Lime Production data. Their estimate of CO2 emissions was 19%
higher than EPA's value for Iowa's Lime Production.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA will follow up with
the reviewers from Texas CEQ to explore the data mentioned in their comments.

2. Based on analysis of Greenhouse Gas Reporting Program data, it appears that most
facilities that manufacture beet sugar and lime, as well as a few lime manufacturing
facilities, capture CO2 for use in on-site processes. Are you aware of any information
why lime-producing facilities capture CO2 for use in on-site processes, any trends in this
practice during the 1990-2019 time series (e.g., have facilities increased or decreased
adoption of this practice during the time series), or whether the amount of CO2
captured is proportional to the amount of lime produced or some other metric? Are you
aware of any data on the amount of CO2 captured on site per facility or state for 1990-
2009?

8 United States Geological Survey. n.d. "Lime Statistics and Information" Available at:
https://www.usgs.gov/centers/nmic/lime-statistics-and-information

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Texas CEQ noted that based on information provided in the TCEQ point source
emissions inventory, Texas lime plants do not appear to employ CO2 capture. They were
unaware of surrogate data that can be used.

EPA Response: The EPA thanks the reviewers for their feedback.

3. For some states and years (Colorado for 2010-2015, Idaho for 2011 and 2019, and
Nebraska for 2010-2014), calculations using GHGRP data on emissions and CO2
captured for on-site processes yielded small but erroneous negative emissions. The EPA
zeroed emissions for those states and years and plans to adjust calculations so that state
emissions totals match national emissions. Do you have any general feedback on this
approach?

Reviewers provided no comments.

2.3.1.3 Glass Production

1. Are you aware of state-level data on glass production or the amount of carbonate (i.e.,
limestone, dolomite, soda ash) consumed for glass production by state (activity data) for
some or all of the 1990-2019 time series? If not, can you share any state-level surrogate
data (e.g., more complete data on glass facilities by state, amount of glass products by
type [i.e., containers, flat (window) glass, fiber glass, and specialty glass]) for 1990-2019
that could refine this state inventory calculation to enhance accuracy and consistency of
state GHG emissions and trends?

Reviewers from Texas CEQ requested a contact name and email address to share production data
from the TCEQ point source emission inventory. Reviewers from Iowa noted that they did not
identify emissions from glass production separately in their inventory. Rather, they calculated
total emissions from limestone and dolomite use for industrial consumption per the SIT
Industrial Processes Module. They were unaware of any state-level activity data that may
enhance the state inventory calculation.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA will follow up with
the reviewers from Texas CEQ to explore the data mentioned in their comments.

2.3.1.4 Other Process Uses of Carbonates

1. Are you aware of state-level data on the consumption of limestone and dolomite for the
iron and steel sector for the 1990-2019 time series? If not, can you share any state-level
surrogate data for 1990-2019 that could refine this state inventory calculation to
enhance accuracy and consistency of state GHG emissions and trends from carbonate
consumption by the iron and steel sector?

Reviewers were unaware of state-level data on consumption of limestone and dolomite for the

Iron and Steel sector for the specified period.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: The EPA thanks the reviewers for their feedback.

2. Are you aware of state-level data on the consumption of soda ash (not associated with
glass manufacturing) for the 1990-2019 time series?

Reviewers were unaware of other data sources on consumption of soda ash for the specified
period. Reviewers from Iowa calculated the total emissions from limestone and dolomite use for
industrial consumption per the SIT Industrial Processes Module. They suggested including the
use of soda ash by the corn wet milling industry because it commonly uses soda ash for pH
control, ion exchange regeneration, and other operations. They noted that Iowa's DNR may be
able to survey corn wet milling facilities to obtain this annual data.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA will assess the use
of soda ash by the corn wet milling industry to understand if it is an emissive activity and then
include as a potential update for future annual publications of this report.

3. Are you aware of any state-level data on limestone and dolomite consumption for flux
stone, flue gas desulfurization systems, chemical stone, mine dusting or acid water
treatment, acid neutralization, and sugar refining activities for the 1990-2019 time
series?

Reviewers were unaware of other state-level data sources on limestone and dolomite
consumption for flux stone, flue gas desulfurization systems, chemical stone, mine dusting or
acid water treatment, acid neutralization, and sugar refining activities for the specified period.

EPA Response: The EPA thanks the reviewers for their feedback.

2.3.1.5 CO: Consumption

1. Are you aware of other data on the consumption of CO2 by state for the 1990-2019 time
series?

Reviewers were unaware of other data sources of CO2 consumption data. Reviewers from Iowa
noted that this sector is currently not included in the Iowa DNR inventory. They noted that CO2
is used in the food and beverage industry, and Iowa meat-packers use CChto stun livestock.
Typically, meat-packers purchase biogenic CCtefrom ethanol plants. They noted that Iowa DNR
may be able to survey the industry to determine the amount of CO2 from ethanol production that
is used in these industries.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA will assess the use
of CO2 captured by ethanol facilities and used by the meat-packing industry as a potential
update for future iterations of the state GHGI.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2.3.2 Chemicals
2.3.2.1 Ammonia Production

1. Currently, production capacity is used as a surrogate for state-level ammonia
production for 1990-2009. In the absence of ammonia production by state in more
recent years, are you aware of other surrogate data (e.g., facility utilization rates by
state) that could refine this state inventory calculation to enhance accuracy and
consistency of state GHG emissions and trends?

Reviewers from Texas CEQ referenced national Ammonia Production data for the U.S. from
USGS and Energy Star to estimate state-level ammonia production.9'10

Reviewers from Iowa reported that three facilities in Iowa currently produce ammonia and report
their CO2 emissions to the GHGRP, so the Iowa DNR uses this GHGRP data in its inventory.
The total emissions reported by the two facilities for 2019 were 69% higher than the State
Methods data. They wondered whether they should remove land-applied ammonia from the
ammonia production emissions.

EPA Response: Consistent with IPCC methodological guidance and international
transparency reporting standards, under the IPPU sector, the EPA is aggregating only
process-related emissions from ammonia feedstock use rather than total emissions from these
facilities. Combustion-related emissions are accounted for under the Energy sector and
emissions associated with application of ammonia-basedfertilizers, along with all other
nitrogen inputs to agricultural soils, are reported under Agricultural Soil Management in the
Agriculture sector. A difference in sector allocation of the combustion and process related
emissions could be leading to the differences in emissions. The EPA will reach out to Iowa
NDR to better understand available state data and any further potential differences between
emissions reported to GHGRP resulting from allocation differences EPA '.s national and state
Inventories and DNR's inventory. The EPA will continue to assess this issue for future annual
publications of the GHG Inventory by U.S. State data.

Urea Consumption for Nonagricultural Purposes

1. Are you aware of state-level data on urea consumption for nonagricultural purposes
(activity data) for some or all of the 1990-2019 time series?

Reviewers from Texas CEQ suggested that HIS Markit may have data regarding urea
consumption.11 Reviewers from Iowa identified this sector as a potential gap as it had not been

9 USGS. n.d. "Nitrogen Statistics and Information," Available at: https://www.usgs.gov/centers/national-minerals-
information-center/nitrogen-statistics-and-information

10 Worrell, E., Phylipsen, D., Einstein, D., & Martin, N. (2000). "Energy Use and Energy Intensity in the U.S.
Chemical Industry," Lawrence Berkeley National Laboratory.

https://www.energvstar.gov/sites/default/files/buildings/tools/industrial LBNL-44314.pdf

11 HIS Markit. (2020). "Urea Chemical Economics Handbook," Available at: https://ihsmarkit.com/products/urea-
chemical-economics-handbook.html

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

included in their state-level inventory calculations and wondered if this sector will be added to
the SIT.

EPA Response: The EPA thanks the reviewers for their feedback.

2.3.2.2 Nitric Acid Production

1. Are you aware of state-level data on nitric acid production (activity data) for some or
all of the 1990-2009 time series? We currently use production capacity as a surrogate
for nitric acid production by state for 1990-2009. We know that the production
capacity data used for this state inventory calculation are incomplete for 1990-2009.
Are you aware of more complete data on facility production capacity by state?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.

EPA Response: The EPA notes the reviewers have no additional feedback.

2. Are you aware of state-level data other than facility production capacity (e.g.,
utilization rates by facility or state, information about abatement technology
installations and use per facility) for 1990-2009 that could refine this state inventory
calculation to enhance accuracy and consistency of state GHG emissions and trends?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.
Reviewers from Iowa reported that there are two nitric acid production facilities in Iowa. They
are required to report their N2O emissions to the GHGRP and the Iowa DNR use this data in its
inventory. The values for Iowa DNR's inventory and State Methods values for 2019 were within

0.13%. Reviewers from Alaska's DEC noted that there are no surrogate state-level data to
replace national level data but cautioned not to apply generic national emissions to Alaska.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA found no data
showing that nitric acid production occurred in Alaska during any part of the time series.

2.3.2.3 Adipic Acid Production

1. Are you aware of any other state-level data on adipic acid production (activity or
emissions data) for some or all of the 1990-2019 time series?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.

EPA Response: The EPA notes the reviewers have no additional feedback..

2.3.2.4 Caprolactam, Gtyoxai and Gtyoxytic Acid Production

1. Are you aware of state-level data on caprolactam production or emissions for some or
all of the 1990-2009 time series? We currently use production capacity as a surrogate

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

for caprolactam production by state. Are you aware of more complete data on facility
production capacity or actual production by state? Are you aware of better surrogate
data other than facility production capacity (e.g., utilization rates by facility or state,
information about abatement technology installations and use per facility) that could
refine this state inventory calculation to enhance accuracy and consistency of state
GHG emissions and trends?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.
Reviewers from Alaska's DEC noted that there are no surrogate state-level data to replace
national level data but cautioned not to apply generic national emissions to Alaska.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA found no data
showing that caprolactam production occurred in Alaska during any part of the time series.

2.3.2.5 Carbide Production and Consumption

1. Are you aware of state-level data on SiC production (activity data) for the 1990-2019
time series? Are you aware of other data to refine accuracy of the estimation of SiC
consumption by state for the 1990-2019 time series?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.
Reviewers from Iowa noted that they do not account for Carbide Production and Consumption
because it is not included in the SIT Industrial Processes model. They will assess if this is a gap
in their inventory if Iowa data are available.

EPA Response: The EPA thanks the reviewers for their feedback. The only emissions
attributed to Iowa are related to silicon carbide consumption.

2. Are you aware of information that can help us improve the accuracy of production in
the two states where SiC facilities are located?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.

EPA Response: The EPA notes the reviewers have no additional feedback..

2.3.2.6 Titanium Dioxide (TiO:) Production

1. Are you aware of data on TiCh production (activity data) by state for the 1990-2009
time series? Please share any other surrogate data than facility production capacity or
more data by state (e.g., facility utilization rates by facility or state) for 1990-2009 that
could refine this state inventory calculation to enhance accuracy and consistency of
state GHG emissions and trends.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Texas CEQ referenced national annual production in the U.S. from USGS.12
Reviewers from Alaska's DEC noted that there are no surrogate state-level data to replace
national level data but cautioned not to apply generic national emissions to Alaska.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA found no data
showing that titanium dioxide production occurred in Alaska during any part of the time
series.

2.3.2.7 Petrochemical Production

1. Are you aware of data on petrochemical production by type by state for the 1990-2019
time series? Is there any other surrogate data by state or facility (e.g., facility
production capacity; utilization rates by facility or state; timing of facility expansions,
openings, and temporary or permanent closures) for the full 1990-2019 time series that
could address data gaps and refine this state inventory calculation to enhance accuracy
and consistency of state GHG emissions and trends?

Reviewers from Texas CEQ suggested state- or region-level petrochemical data from EIA.
Reviewers from Iowa noted that they do not account for Petrochemical Production because it is
not included in the SIT Industrial Processes model. They will assess if this is a gap in their
inventory if Iowa data are available.

EPA Response: The EPA thanks the reviewers for their feedback.

2.3.2.8 Phosphoric Acid Production

1. Are you aware of state-level data on phosphoric acid production (activity data) for the
1990-2009 time series? Is there any other surrogate data or information (e.g., timing of
facility expansions and temporary or permanent closures, origin of phosphate rock used
in facilities) by state or facility for 1990-2019 that could refine this state inventory
calculation to enhance accuracy and consistency of state GHG emissions and trends?

Reviewers were unaware of other data sources beyond those referenced in the national Inventory.

EPA Response: The EPA notes the reviewers have no additional feedback..

2.3.2.9 HCFC-22 Production

1. For the years 1990-2009, there are significant uncertainties in the allocation of

national-level U.S. emissions to individual facilities and states, particularly for the five
HCFC-22 production facilities that closed before 2003 and for which production

12 USGS. n.d. "Titanium Statistics and Information," Available at: https://www.usgs.gov/centers/national-minerals-
information-center/titanium-statistics-and-information

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

capacity data are therefore not available. Are you aware of any more complete sources
of production capacity or other relevant historical data?

Reviewers were unaware of state-level data.

EPA Response: The EPA notes the reviewers have no additional feedback..

2. Do you have recommendations for how to refine the methodology to more accurately
estimate emissions from HCFC-22 production over the time series?

Reviewers provided no comments.

2.3.3 Metals

2.3.3.1 Iron and Steel and Metallurgical Coke Production

1. Are you aware of state-level data on iron and steel production (activity data) by
category (i.e., sinter production, iron production, pellet production, steel production,
other activities) for some or all of the 1990-2019 time series? In the absence of steel
production by state, are you aware of better surrogate data that could refine this state
inventory calculation to enhance accuracy and consistency of state GHG emissions and
trends?

Reviewers were unaware of state-level data, but reviewers from Texas CEQ referenced
production for the U.S. from USGS.13

Reviewers from Minnesota's PC A noted that the emissions from Iron and Steel Production stood
out as an anomaly. They commented that an obvious jump in emissions seemed to be due to the
change in data sources with the GHGRP data became available, not a change in production that
created an actual increase in emissions. Additionally, they noted that their estimates use taconite
production collected by the Department of Revenue and energy consumption data that are
reported to the state and federal data sources for electricity sold to the grid in order to allocate
some emissions to the electricity sector.14

Reviewers from Iowa reported that there are currently no metallurgical coke production facilities
or pig iron mills operating in Iowa. Furthermore, they reported that all three steel production
facilities currently operating in Iowa use electric arc furnaces to produce steel from scrap. They
noted that the Iowa DNR inventory used the CO2 emissions that were reported to the GHGRP.
The CO2 emissions in the Iowa DNR inventory were 7% higher than the State Methods value for
2019.

13 Tuck, C. C. (2021). "Iron and Steel," U.S. Geological Survey, Mineral Commodity Summaries, Available at:
https://pubs.uses. gov/periodicals/mcs202 l/mcs2021 -iron-steel.pdf

14 Minnesota Taconite Production Summary (1950-2020) data available at:
https://www.revenue.state.mn.us/sites/default/files/2021-

10/Minnesota%20Taconite%20Production%20Summarv%20%281950-2020%29.xlsx

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: The EPA thanks the reviewers for their feedback.

The EPA thanks reviewers from Minnesota's PCA for the publicly available data on taconite
production and is aware of the time series consistency issues. The EPA notes that the
commenter is correct that this is an issue with use of different data sources over time. The
EPA will assess this for future annual publications of this report and will also need to
consider data across other states as well.

The EPA thanks Iowa for their feedback. Their description of the iron & steel facilities
operating in Iowa in 2019 match what is included in this report.

2. Are you aware of information to better allocate basic oxygen furnace and electric arc
furnace (EAF) production by state for 1990-2009?

Reviewers were unaware of state level or facility information data. However, reviewers from
Texas CEQ noted that 100% of iron and steel production in Texas occurs using EAFs and almost
none or none through basic oxygen furnaces, based on a conversation with a representative of the
Steel Manufacturing Association and confirmed by a contact for Nucor Steel.

EPA Response: The EPA thanks the reviewers for their feedback. The EPA thanks Texas
CEQ for confirming that all iron & steel facilities currently in operation in Texas use EAF.

2.3.3.2 Ferroalloy Production

1. Are you aware of state- or facility-level data on ferroalloy production (activity data) for
the 1990-2019 time series? Is there any other surrogate data (e.g., facility production
capacity, utilization rates by facility or state) for 1990-2019 that could refine this state
inventory calculation to enhance accuracy and consistency of state GHG emissions and
trends?

Reviewers were unaware of relevant datasets beyond those referenced in the national Inventory.

EPA Response: The EPA notes the reviewers have no additional feedback..

2.3.3.3 Aluminum Production

1. Are you aware of data available to incorporate differences in emissions between
smelters based on technology type? Is there any other surrogate data or emission
sources that could be used to allocate national total aluminum production emissions
across states?

Reviewers from Texas CEQ referenced USGS annual national data but were unaware of existing
state or facility level data.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: The EPA thanks the reviewers for their feedback. The EPA already
incorporates USGS data as neededfor the aluminum sector.

2.3.3.4 Magnesium Production and Processing

1. Are you aware of state- or facility-level magnesium production or capacity data (or
surrogate data) for the 1990-2019 time series?

Reviewers from Texas CEQ referenced USGS annual national data but were unaware of existing
state or facility level data.

EPA Response: The EPA thanks reviewers for their feedback. The EPA already incorporates
USGS data as needed for the Magnesium sector.

2. Are you aware of information on the location (by state) of magnesium production and
processing facilities or information on the location (by state) of magnesium production
and processing facilities by process type?

Reviewers from Texas CEQ referenced USGS annual national data but were unaware of existing
state or facility level data.

EPA Response: The EPA thanks reviewers for their feedback. The EPA already incorporates
USGS data as needed for the Magnesium sector.

2.3.3.5 Lead Production

1. Are you aware of state- or facility-level data on primary or secondary lead production
(activity data) for the 1990-2019 time series? Is there any other surrogate data (e.g.,
primary or secondary production capacity by facility or state) for 1990-2009 that could
refine this state inventory calculation to enhance accuracy and consistency of state
GHG emissions and trends?

Reviewers from Texas CEQ referenced USGS annual national data but were unaware of existing

state or facility level data.

EPA Response: The EPA thanks reviewers for their feedback.

2.3.3.6 Zinc Production

1. Are you aware of state- or facility-level data on zinc production (activity data) by unit
type (i.e., electrothermic furnace, Waelz kiln, other furnaces, and flame reactor units)
for the 1990-2019 time series? Is there any other surrogate data (e.g., total number of
zinc facilities by state, production capacity by unit type and by facility or state) or other
data by state (e.g., utilization rates by facility or state) for 1990-2009 that could refine

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

this state inventory calculation to enhance accuracy and consistency of state GHG
emissions and trends?

Reviewers from Texas CEQ referenced USGS annual national data but were unaware of existing
state or facility level data.

EPA Response: The EPA thanks reviewers for their feedback.

2.3.4 Product Use
2.3.4.1 Electronics Industry

1. Are you aware of state- or facility-level capacity data or other state-level surrogate data
(e.g., sales data) for photovoltaic (PV) manufacturing for 1990-2006 that could be used
to refine the allocations of emissions by state? Is there any surrogate data (e.g., sales
data by state) by state for semiconductor or micro-electro-mechanical systems (MEMS)
manufacturing for 1990-2007 that could be used to refine the allocations of emissions
by state?

Reviewers were unaware of sales, capacity, or other surrogate data for PV manufacturing.
Reviewers from Iowa reported that the 2017 Economic Census identifies eleven businesses in
Iowa under the North American Industry Classification System (NAICS) for code 33441 -
Semiconductor and Other Electronic Manufacturing (U.S. Census 2019), but EPA's
disaggregated Iowa values do not include data for Iowa. They also reported that emissions in
2019 from semiconductor manufacturing were calculated by DNR by assuming that Iowa
emissions were 0.96% of national emissions as Iowa's population is 0.96% of the total U.S.
population.

EPA Response: The EPA thanks the reviewers for their feedback and identification of
potential sources of data. Emissions estimated from the electronics manufacturing sector are
estimatedfor the manufacture of semiconductors, MEMS, photovoltaics, and LCDs using
fluorinated GHGs in etching, wafer cleaning and chamber cleaning processes and from the
use of N2O and heat transfer fluids. Not all companies that have a NA ICS code of33441
manufacture electronic devices and use fluorinated GHGs. However, the EPA does not
currently estimate emissions from MEMS manufacturers that do not report to the GHGRP
and is continuing to review data sources for estimates of electronics manufacturing
production capacity by state.

2.3.4.2 Substitution of Ozone-Depleting Substances

1. Are you aware of bottom-up modeling data that are available by state? Is there any
surrogate data other than population data that could be used to disaggregate the
emissions of substitutes for ozone-depleting substances?

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers were unaware of modeling data for ozone-depleting substances or surrogate data other
than population. Reviewers from Iowa found their 2019 estimate of emissions to be within 1% of
EPA's State Inventory data. They noted they used the SIT which reports emissions in CChe and
does not report disaggregated emissions from various GHGs.

EPA Response: The EPA notes most reviewers provided no additional feedback. The EPA
appreciates Iowa's notes from comparing the data and notes the new fact sheet cross-walking
GHG Inventory by U.S. State methods and data with the SIT available online at

https://www.epa.gov/ghgemissions/state-ghg-emissions-and-removals.

2.3.4.3 Electrical Transmissions and Distribution

2. Are you aware of state-level electrical transmission and distribution equipment data
(e.g., nameplate capacity by state) or other data for 1990-2019 (or part of the time
series) that could refine this state inventory calculation to reflect state trends in
emissions more closely? Is there any other surrogate data (e.g., state population data) to
enhance accuracy and consistency of state GHG emissions and trends than the current
data being used (transmission mile data by state)?

Reviewers from Texas CEQ noted that data may be available from either the Electric Reliability
Council of Texas (ERCOT) which oversees the state of Texas electrical grid or the Public Utility
Commission of Texas which oversees ERCOT. Reviewers from Rhode Island's DEM
commented that they request SF6 emissions from National Grid on an annual basis and use this
estimate in their annual GHG inventory.

Reviewers from Iowa noted they calculated emissions from electric power transmission and
distribution by using the most current national emissions data adjusted for Iowa retail electricity
sales compared to US retail electricity sales. Reviewers from Iowa reported their estimate for
2019 emissions is 24% lower than the disaggregated national Inventory value. They attributed
this to 2019 data not being available at the time that the Iowa DNR completed its inventory and
they used 2018 as a proxy.

Reviewers from Alaska's DEC noted the state has information in its permit system for nameplate
capacity which can be provided to EPA.

EPA Response: The EPA thanks the state reviewers for their feedback and identification of
potential sources of data. The EPA '.s state-level estimates use transmission miles and GHGRP
data to allocate emissions by state. Differences can be expected if retail electricity sales are
used as a proxy instead of transmission miles.

2.3.4.4 N?0 from Product Use

1. Are you aware of state-level data on N2O usage for medical and dental anesthesia, food
processing propellant and aerosols, sodium azide production, or other applications (e.g.,
fuel oxidant in auto racing, oxidizing agent in blowtorches) for some or all of the 1990-
2019 time series? Is there any other surrogate data (e.g., state population data) that

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

could be used to enhance accuracy and consistency of state GHG emissions and trends
other than the current data (transmission mile data by state)?

Reviewers from Texas CEQ were unaware of state level N2O usage data for the medical and
dental industries. Only through industry specific surveying efforts could this data be acquired.
Reviewers from Iowa noted this sector is not in their inventory as it is not included in the SIT
Industrial Processes module and wondered whether it will be added to the SIT. They reported
they will research if this a gap in their inventory and if Iowa activity data are available.

EPA Response: The EPA thanks reviewers for their feedback. As additional state-level data
and/or methodological approaches become available through the national Inventory
disaggregation, they will be used to supplement or improve the embedded calculations and
defaults in SIT as appropriate. SIT users will retain the ability to customize the tool with their
own data in lieu of using defaults.

2.4 Agriculture

1. What are your overall impressions of the clarity and transparency of this section?

Reviewers provided no comments.

2. What recommendations do you have to add to the overall completeness and accuracy of
this chapter?

Reviewers from Alaska's DEC suggested data from the Alaska Department of Agriculture for
state agriculture to supplement USD A information and 2020 NEI as well.

EPA Response: The EPA thanks the reviewers from Alaska's DEC for their feedback and
suggested data sources. The EPA will review and assess how best to reflect the data in the
agriculture chapter.

3. Data Availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are not
currently included, but could be incorporated into this analysis?

Reviewers provided no comments.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

4. Uncertainty. Currently, uncertainty ranges are not included for the state-level

estimates. Please provide feedback on what qualitative and quantitative information
would be useful. Time Series Coverage. Currently, state data cover 1990-2019
consistent with the 2021 National Inventory and are inclusive of most known baseline
periods for climate policy. Subsequent publications of this data will also strive to
maintain this consistency with the national Inventory. As state-specific input datasets
are not always available over the entire timeseries, understanding which years may be
more important can help us to better prioritize our backcasting and methodological
efforts across the time series. The EPA appreciates feedback on which, if any years
should be prioritized for future state-level estimates (e.g., 2000 and later, 2005 and
later, 2010 and later, or the full time series).

Reviewers from Virginia's DEQ compared Inventory data to SIT results. They found that
emissions derived from SIT appears to be far apart from what is given in the draft, as the results
for agriculture soil management were not yet completed at the time Virginia DEQ received
results. States and organizations received State Methods results in Fall of 2021, though some
sectors in the Agriculture and LULUCF chapters were not available until January 2022. For
more detail, see comments from Virginia's DEQ in Section 3.9.

EPA Response: The EPA thanks the reviewers for their feedback and notes state-level data on
agricultural soil management was shared with States on January 18 and 19, 2022.

Reviewers provided no comments.

6. Data Presentation and Usability.

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could be
helpful?

iii. What data format would best facilitate the use of the state-level emissions data
(e.g., .xlsx download)?

b. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

Reviewers provided no comments.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2.4.1 Livestock
2.4.1.1 Enteric Fermentation

1. Are there other/newer data sources or methods, particularly at the state level, that the
EPA should be aware of and consider in calculating these emissions? Especially for:

• Dry matter/gross energy intake;

• Annual data for the digestible energy (DE) values (expressed as the percentage of
gross energy intake digested by the animal), CH4 conversion rates (Ym) (expressed
as the fraction of gross energy converted to CH4), and crude protein values of
specific diet and feed components for foraging and feedlot animals;

• Monthly beef births and beef cow lactation rates;

• Weights and weight gains for beef and dairy cattle.

Reviewers from Iowa commented that their enteric emissions are calculated using the SIT. Both
the State Methods and the SIT use emission factors from the Cattle Enteric Fermentation Model
(CEFM), and the most recent animal population data from either USDA's NASS Quick Stats or
USDA's Iowa Agricultural Summary. They noted that since the Iowa DNR's inventory is for the
previous calendar year and the national Inventory is for two years prior, DNR and EPA are not
using the same animal populations. This may be the reason for DNR's emissions being 18%
higher than the State Methods for 2019, they commented.

EPA Response: The EPA thanks the reviewers from Iowa DNR for their feedback and
additional information on why the state-level estimates compiled by the EPA and Iowa DNR's
state-level GHG Inventory differ.

2. Are state-specific diet data available to the EPA to enhance characterization of diet
differences across livestock types and U.S. states?

Reviewers from Alaska's DEC suggested that the Department of Natural Resources, Division of
Agriculture or the University of Alaska Agriculture Research Station might have useful data.

EPA Response: The EPA thanks the reviewers from Alaska's DEC for their feedback and
suggested data sources. The EPA will review available data and assess how best to reflect the
data in methods to estimate state-level emissions from livestock.

3. For the enteric fermentation source category and the Cattle Enteric Fermentation
Model (CEFM), are the various regional designations of U.S. states (as presented in
Annex 3.10 of the GHG Inventory) used for characterizing the diets of foraging cattle
appropriate? The CEFM is used to estimate cattle CH4 emissions from enteric
fermentation and incorporates information on livestock population, feeding practices,
and production characteristics.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers were unaware of newer sources of state-specific livestock emission factors and
production. Reviewers from Alaska's DEC noted that Alaska has little cattle production and
USDA information is sufficient.

EPA Response: The EPA thanks the state reviewers for their feedback.

2.4.1.2 Manure Management

1. Are there other/newer data sources, particularly at the state level, that the EPA should
be aware of and consider in calculating these emissions? Especially for the following:

• waste management system data, particularly seasonal changes in emissions from
different waste management systems;

• maximum methane-producing capacity;

• volatile solids and nitrogen excretion rates; and

• measured emission estimates (by waste management system) to help refine estimates
of methane conversion factors.

Reviewers from Texas CEQ were unaware of newer sources of Texas-specific livestock emission
factors and production. Reviewers from Iowa noted their manure management emissions are
calculated using the SIT and the same animal population sources as used for enteric
fermentation. To that end, Iowa specific data are more recent than in the State Methods which is
likely responsible for their emissions being 4% higher than the State Methods for 2019. The Iowa
DNR calculated emissions for 2017 through 2019 are 60% to 73% higher than the State Methods
and reviewers from Iowa will further research this discrepancy.

EPA Response: The EPA thanks the reviewers from Texas CEQ and Iowa DNR for their
feedback. In particular, if Iowa DNR has state-specific information on manure management
they are able to provide EPA, such as distribution and usage or different waste management
systems, it could be useful for the EPA to assess where the EPA and Iowa emissions estimate
differ.

2.4.2 Agricultural Soil Management

1. What are your overall impressions of the clarity and transparency of this section?

Reviewers from Iowa commented that emissions from agricultural soil management in 2019
were 20.97 MMT CChe, accounting for 16% of Iowa's total GHG emissions in their estimates,
and they look forward to seeing the State Methods values when they become available.

EPA Response: The EPA thanks the Iowa DNR reviewers for their feedback. State-level data
on N2O emissions from agricultural soil management estimates were shared on January 18,
2022.

2. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory?

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Alaska's DEC noted that Alaska has a very long winter season and thus it makes
sense to have an adjustment for the season where agricultural soils are frozen.

EPA Response: The EPA thanks the Alaska DEC reviewers for this feedback. As noted in the
State Methods document, for Agricultural Soil Management, only N2O emissions from
mineral fertilizer and PRP N additions are estimated for Alaska, The EPA will assess how best
to reflect state-specific circumstances in future state estimation efforts.

2.4.3 Other Charge Questions

2.4.3.1 Rice Cultivation

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.4.3.2 Liming

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers from Iowa DNR noted that their CO2 emissions from liming are calculated using the
total amount of limestone CO2 that is emitted when acidic agricultural soils are neutralized by
adding limestone or dolomite. They also reported that the Iowa DNR used the total annual
amount of limestone produced for agricultural use as reported by their members to the Iowa
Limestone Producers Association. However, producers do not report the percentage of limestone
that is dolomitic. The Iowa DOT tracks general information for active aggregate sources used for
construction, including whether the material is limestone or dolomite. They do not track that
information for limestone produced for agricultural purposes. The Iowa DOT indicated that some
areas of the state have 100% dolomite, some have 100% limestone, and some areas are mixed.
Therefore, the Iowa DNR assumed that 50% of the material produced in Iowa for agricultural use
is dolomite and 50% is limestone. Reviewers from Iowa commented that the Iowa DNR's
emissions are quite different than the disaggregated national Inventory values for Iowa. Finally,
they noted that the disaggregated national Inventory values are not provided for 2017 or 2019.

EPA Response: The EPA thanks Iowa DNR for their feedback. The EPA shared updated data
with state experts on January 18, 2022, which included values for 2017—2019. In the national
Inventory, the EPA applied a Tier 2 approach to estimate emissions from liming. The EPA
obtains state-level data from the U.S. Geological Survey (USGS') on limestone and dolomite
applied for agricultural purposes and uses a country-specific emission factor to estimate
emissions. The USGS data could be used as a quality assurance/verification measure to assess
the ratio of limestone to dolomite applied each year, and how that compares to Iowa's 50%
assumption.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2.4.3.3 Urea Fertilization

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers from Minnesota PCA referenced and shared a report from the Minnesota Department
of Agriculture.15 Reviewers from Iowa DNR reported their estimated emissions from Urea
Fertilization are significantly lower than the State Methods values for Iowa for 2018 (-24%) and
2019 (-30%) but are higher than the State Methods values for Iowa for 2015 (+7%), 2016
(+20%), and 2017 (+6%). They believed this corresponds to a change in the availability of state-
specific fertilizer data. Prior to 2018, the Iowa Department of Land and Agriculture Stewardship
published a report of fertilizer sales that the Iowa DNR used for its inventory. The report was not
available for 2018 or 2019, so the Iowa DNR used the amount of 2018 urea applied from the
USDA's Iowa Agricultural Statistics Bulletin for both 2018 and 2019.

EPA Response: The EPA thanks Iowa DNR for their feedback and additional information on
why the state-level estimates compiled by the EPA and Iowa DNR's state-level GHG Inventory
differ. Fertilizer sales data is used by the EPA to calculate emissions from urea fertilization.
These reports are generally reported in fertilizer years; therefore, the EPA converts these
estimates to calendar year to align with national Inventory reporting needs. However,
fertilizer sales data for the 2016 through 2019 fertilizer years were not available so application
was estimated using a linear, least squares trend of consumption over the data from the
previous five years (2011 through 2015) at the state scale. This could be another reason for
differences in emissions.

2.4.3.4 Field Burning of Agricultural Residues

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.5 Land Use, Land-Use Change, and Forestry

1. What are your overall impressions of the clarity and transparency of this section?

Reviewers provided no comments.

15 Minnesota Department of Agriculture. n.d. "Fertilizer Use, Nutrient Management Survey," Available at:
https://www.mda.state.mn.us/pesticide-fertilizer/fertilizer-use-sales-data

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2. What recommendations do you have to add to the overall completeness and accuracy of
this chapter?

The USCA comments suggested that states have appropriate time to review disaggregated
Agricultural Soil Management and Croplands and Grasslands Data. They also suggested giving
the opportunity to states to contribute state-level data with the opportunity to review
disaggregated data and amended methodologies once state input and data have been
incorporated.

Reviewers from Iowa DNR reported their estimates of carbon sequestration in the LULUCF
sector are significantly higher than the disaggregated national Inventory state values (between
+76% to +135%). They were unsure what is driving this difference as they used the SIT to
calculate the amount of carbon sequestered and used the SIT default values for forest carbon
flux, urban forests, and agricultural soils. In addition, the SIT seems to use the same method as
the national Inventory.

EPA Response: The EPA appreciates the feedback from Iowa DNR on the differences, and
these will be reviewedfor the next annual publication of this report.

The EPA notes that there are a number of differences in methods, in particular, cropland and
grassland, between the current SIT tool and the GHG Inventory by U.S. State that can result
in different emissions. As both products continue to be updated, it is expected that the values
will align better. The EPA has also provided a fact sheet with information cross walking the
state-level GHG estimates with the SIT methods. The fact sheet is available online here:
https://www.epa.gov/ghgemissions/state-ghg-emissions-and-removals.

3. Data Availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are not
currently included, but could be incorporated into this analysis?

Reviewers from Alaska's DEC suggested USGS data on Land Use and Land Use Change on
Carbon sequestration Capacity that should be integrated. These data are more complete than EPA
datasets which are limited to Southcentral and Southeast Alaska.

EPA Response: The EPA thanks the reviewers for their feedback. EPA has previously
reviewed the referenced USGS dataset and conducted a preliminary assessment using these
data. Please see Box 6-5 (page 6-34) of the Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990—2016, found here: https://www. eya. sov/sites/default/files/2018-
01/documents/2018 complete report, ydf. Based on this review, EPA found that the USGS
data was not directly applicable to the Managed Land Proxy approach used for the Land
Representation analysis in the national Inventory, consistent with IPCC methodological
guidance. The data EPA currently uses covers areas outside of Southcentral and Southeast
Alaska, USFS began an operational inventory in interior Alaska in 2016 with a reduced plot
density that is being used in combination with remote sensing which allows us to include

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

estimates of carbon stock changes in interior Alaskan forests. Other datasets used for the
lower 48 states (e.g., NRI) are limited in Alaska, so some land use, land use change categories
(Croplands, Grasslands, Settlements) are not estimated at this time. EPA notes these as
planned improvements.

4. Uncertainty. Currently, uncertainty ranges are not included for the state-level

estimates. Please provide feedback on what qualitative and quantitative information
would be useful. Time Series Coverage. Currently, state data cover 1990-2019
consistent with the 2021 National Inventory and are inclusive of most known baseline
periods for climate policy. Subsequent publications of this data will also strive to
maintain this consistency with the national Inventory. As state-specific input datasets
are not always available over the entire time series, understanding which years may be
more important can help us to better prioritize our backcasting and methodological
efforts across the time series. The EPA appreciates feedback on which, if any years
should be prioritized for future state-level estimates (e.g., 2000 and later, 2005 and
later, 2010 and later, or the full time series).

EPA Response: The EPA thanks the reviewers for their feedback. Please see response to
comment 2.5.3, above.

Reviewers from Alaska's DEC commented that LULUCF should be viewed as separate from the
national average and analyzed separately from nationwide land use.

EPA Response: The EPA presents emissions consistent with international reporting guidelines
and internationally accepted methodological guidance from the IPCC. Consistent with these
guidelines, the EPA presents both national totals and state-level totals, including and
excluding the L UL UCF sector.

6. Data Presentation and Usability.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could be
helpful?

iii. What data format would best facilitate the use of the state-level emissions data
(e.g., .xlsx download)?

b. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

Reviewers provided no comments.

2.5.1 Forest Lands and Lands Converted to Forest Land

2.5.1.1 Forest Land Remaining Forest Land

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers from Minnesota's PC A noticed that their method for sequestration is highly variable
on an annual basis and noticeably different from the State Methods estimate. They recommended
discussion with inventory staff and experts at the Minnesota Department of Natural Resources to
help them understand the method used by the EPA, to share approaches, and further develop
accounting methods.

EPA Response: The EPA works closely with USFS to develop the FLRFL estimates. The EPA
will follow up with Minnesota DNR andfacilitate a discussion on this topic.

2.5.1.2 Land Converted to Forest Land

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.5.2 Agricultural Lands (Croplands and Grasslands)

1. What are your overall impressions of the clarity and transparency of this section?

Reviewers found the methodology for this category sound. Reviewers from Alaska's DEC noted
that the methods were clear but inadequate as the data covers only a small part of Alaska.

EPA Response: Currently, for the national Inventory and the State Methods report, EPA does
not estimate GHG fluxes from Croplands and Grasslands in Alaska due to insufficient data
(e.g., NRI has no coverage in Alaska). Consistent with IPCC methodological guidance and

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

international reporting guidelines we have identified the managed lands in Alaska, consistent
with the Managed Land Proxy approach, and are working on assembling the necessary data to
quantify N2O emissions and carbon stock changes for Croplands. Grasslands are more
challenging due to limited available data but EPA is working to include these GHG fluxes in a
future national Inventory.

See response to comment 2.5.3 regarding use of USGS data.

2. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory?

Reviewers from Alaska's DEC recommended using the data for croplands and grasslands in the
USGS dataset and to coordinate with them to integrate findings into future modeling. They also
suggested the University of Alaska.16

EPA Response: The EPA thanks the reviewers for their feedback. See response to comment
2.5.3 regarding use of USGS data. EPA will review data from the University of Alaska to
assess how it can potentially be incorporated into the Inventory.

2.5.3 Wetlands and Lands Converted to Wetlands

2.5.3.1 Wetlands Remaining Wetlands

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.5.3.2 Land Converted to Wetlands

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.5.3.3 Peatlands Remaining Peatlands

1. Are there state-level data available on the application ("consumption") of peat,
including the state of use and the horticultural/landscaping use?

16 University of Alaska Division of Agriculture, n.d. "Alaska," https://economic-impact-of-ag.uada.edu/alaska/

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Alaska's DEC noted all data for peatlands and wetlands is in USGS dataset
developed for GHG emissions for the state. They suggested coordinating with USGS to integrate
findings into future modeling.17

EPA Response: The EPA thanks the reviewers for their feedback and will review the data
from USGS to assess how it can potentially be incorporated into the Inventory for Peatlands.
See response to comment 2.5.3 regarding use of USGS data.

2. Are there data sources that could support the EPA in determining the quantity of peat
harvested per hectare and the total area undergoing peat extraction?

Reviewers were unaware of sources of quantity of peat harvested or total area undergoing peat
extraction.

2.5.4 Settlements and Lands Converted to Settlements

2.5.4.1 Settlements Remaining Settlements

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5).
NB: Emissions estimates not yet available.

Reviewers were unaware of sources of emission factors and production.

2.5.4.2 Changes in Carbon Stocks in Settlements Trees

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.5.4.3 N?Q Remissions from Settlement Soils

Reviewers provided no comments.

17 Jewell, S., & Kimball, S. M. (2016). "Baseline and Projected Future Carbon Storage and Greenhouse-gas Fluxes
in Ecosystems of Alaska," U.S. Geological Survey. Professional Paper 1826. Reston, Virginia.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

2.5.4.4 Changes in Yard Trimmings and Food Scrab Carbon Stocks in Landfills

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers provided no comments.

2.5.4.5 Land Converted to Settlements

Reviewers provided no comments.

2.5.5 Other Lands and Lands Converted to Other Lands

1. What recommendations do you have to add to ensure high-quality state-level estimates
are consistent with the national Inventory? (Cf. General Chapter Charge Questions 1-5)

Reviewers were unaware of sources of LULUCF emission factors and production.

2.6 Waste

1. What are your overall impressions of the clarity and transparency of this section?

Reviewers provided no specific comments.

2. What recommendations do you have to add to the overall completeness and accuracy of
this chapter?

Reviewers from Iowa commented that for Industrial sectors, industrial wastewater emissions
were calculated using the emission reported by facilities to GHGRP. In 2019, eleven ethanol
plants and five food processors reported emissions to the GHGRP. They believed this to be more
accurate than calculating emissions based on red meat or ethanol production values. They also
noted that they have been unable to find state-specific data to calculate emissions from pulp &
paper, fruits & vegetables, or breweries, and the SIT wastewater module does not include default
activity data for Iowa.

EPA Response: The EPA appreciates the insight on the industries within Iowa, and as noted
in the Planned Improvements section, will determine if additional data sources can be used for
all states, including GHGRP data. The EPA is planning to update SIT to better align default

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

data with the GHG Inventory by U.S. State data. In the interim, SIT does allow states to enter
categories not currently included so this new data could be entered directly into SIT by state
users as a starting point for including wastewater emissions from these industries.

3. Data Availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are not
currently included, but could be incorporated into this analysis?

Reviewers from Alaska's DEC suggested NEI data to include PW Treatment permitted facilities.

EPA Response: EPA appreciates the commenters offer to either provide or point EPA to a
potential data source to further refine emissions estimates.

4. Uncertainty. Currently, uncertainty ranges are not included for the state-level
estimates. Please provide feedback on what qualitative and quantitative information
would be useful. Time Series Coverage. Currently, state data cover 1990-2019
consistent with the 2021 National Inventory and are inclusive of most known baseline
periods for climate policy. Subsequent publications of this data will also strive to
maintain this consistency with the national Inventory. As state-specific input datasets
are not always available over the entire time series, understanding which years may be
more important can help us to better prioritize our backcasting and methodological
efforts across the time series. The EPA appreciates feedback on which, if any years
should be prioritized for future state-level estimates (e.g., 2000 and later, 2005 and
later, 2010 and later, or the full time series).

Reviewers from Arizona's DEQ expressed interest in prioritizing results from the last decade to
investigate how climate policy has changed and if that has caused trends in landfill emissions.

EPA Response: The EPA appreciates the comments. National trends in landfill emissions are
covered in the national Inventory chapter on waste available online here:
https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks-1990-
2019.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Alaska's DEC commented that state landfills are not calculated using
continental US decomposition rates, and they need to work with DEC on solid waste
decomposition for future emissions modeling. They suggested this as decomposition is different
in cold regions, especially areas with permafrost.

EPA Response: The EPA appreciates the comments and data provided. The EPA has added
examining Alaska-specific landfill modeling to planned improvements for MSW landfills.

6. Data Presentation and Usability.

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could be
helpful?

iii. What data format would best facilitate the use of the state-level emissions data
(e.g., .xlsx download)?

b. What additional datasets or information could be provided to help increase the
usability of the state-level emissions data?

Reviewers provided no comments.

2.6.1 Waste

2.6.1.1 MSW Landfills
1. Data Questions

• Are there datasets for individual states' landfill gas (LFG) recovery activity?

• Are there data available for open dumpsites in the U.S. territories?

Reviewers from Alaska's DEC noted that Alaska has landfill gas processing for Anchorage
landfill and Merril field. All other landfills are open air. They also noted the Alaska DEC
Environmental Health Division may have data available for open landfills and dumpsites that are
not permitted by Air Quality.

From reviewers from Texas CEQ, "Datasets for Texas LFG recovery activity can be found in the
TCEQ MSW annual report at the following link: annual report. Pages 99 - 102 of the 2020 report
show a List of Facilities Recovering Landfill Gas for Beneficial Use (2020), including
information on Gas Processed (ft3). The annual report is available for several years.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Explanation of Municipal Solid Waste Data Fields for more information.

Data on open dumpsites may be found at: nisw-uiiuni-texas.xls. A spreadsheet of historical
information listing old, closed unnumbered MSW landfills that were operated before permits
were required, as well as unauthorized landfills and miscellaneous illegal dumps and disposal
sites. See the TCEQ Inventory of Closed Municipal Solid Waste Landfills page for information
about why this information was collected and how it is used. See Explanation of Municipal Solid
Waste Data Fields for more information."

Reviewers from Minnesota's PC A commented that NPCA is responsible for closed landfills, and
they collect data. They can provide additional data on closed landfill activities to the EPA if
requested.

Reviewers from Rhode Island's DEM noted that in Rhode Island, there is only one MSW landfill
which is the RI Resource Recover Corporation (RIRRC).18 They noted that "the landfill gas
collection system consists of an intricate network of trenched horizontal piping and several
strategically placed negative-pressure gas collection wells. All of the gas collection systems
come together at a single common gas extraction point. The gas is then treated by a gas
conditioning and compression site (GCC). After treatment, the landfill gas is sent as fuel to the
combined cycle combustion turbine generator (CTG) power plant operated by RI LFG Genco,
LLC." Regarding data, RIRRC's "permits require that they monitor and calculate quantities of
landfill gas generated. They are also required to measure gauge pressure in the gas collection
system monthly for each individual well and trench to demonstrate that the gas collection system
flow rate is sufficient. Testing is performed on the actual gas itself, as well, including methane
content and VOC content. These records are kept on-site at Central Landfill. RIDEM review the
records during inspections. RI LFG Genco must also continuously monitor the amount of landfill
gas that flows to the engines. In summary, no landfill gas data is submitted to RIDEM."

Reviewers from Arizona's DEQ noted that there are no data on open dumpsites in Appendix F
Table F-3 for MSW landfills methane recovery. However, Section 6.1.1.4 of the State Methods
mentions the inclusion of emissions from all waste management practice, including open
dumpsites, as a potential refinement to landfill estimation methods.

Reviewers from Iowa commented that they used emissions reported by MSW landfills to the
GHGRP, which are calculated based on the characteristics of each individual report. They also
noted that EPA requires MSW landfills that emit 25,000 metric tons CChe or more to report their
emissions. This included twenty-four Iowa landfills in 2019. An additional twenty-two Iowa
MSW landfills were not required to report to the GHGRP. To calculate emissions for those that
did not report to the GHGRP, they calculated the potential methane emissions using EPA's
Landfill Gas Emissions Model (LandGEM) version 3.02.

18 They noted that there are two others, but they are very small and for inventory purposes they are not included.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: See above response for Alaska's DEC comments.

Regarding comments from Iowa: The EPA will confirm that the 24 nonreporting landfills in
Iowa are captured in our nonreporter database of landfills used to develop the scale-up factor.
More information on how the scale-up factor is used to account for non-GHGRP reporters
and to develop the national MSW landfill emissions is in the waste chapter of the national
Inventory available online here: https://www.epa.gov/ghgemissions/inventory-us-greenhouse-
gas-emissions-and-sinks-1990-2019.

Regarding comments from other states: The EPA appreciates the data suggestions and will
continue to evaluate the feasibility of including site-specific data from each state.

2. The current method makes some simplifying assumptions and includes uncertainties in
the allocation of national-level U.S. emissions to states (e.g., recovery rates are the same
for all states and match the national recovery rate). Are there alternative assumptions
or different datasets that would improve the accuracy of MSW landfill estimates? Do
you have recommendations to refine the methodology to estimate emissions over the
time series more accurately?

Reviewers from Alaska's DEC commented that Alaska has many landfills, but they are permitted
with the waste division. They noted using the national methods may not be applicable to Alaska
due to the lengthy time of frozen conditions.

EPA Response: See above response for Alaska's DEC comments.

2.6.1.2 Industrial Landfills

1. Do you have recommendations to refine the methodology to estimate emissions over the
time series more accurately?

Reviewers from Iowa commented that this sector is not included in their inventory and will
research if this is a gap and if Iowa activity data are available.

2.6.1.3 Composting
1. Data Questions

• Is the assumption that Alaska has no commercial composting operations correct?

• Are there any data about composting in U.S. territories?

• Are there any state-level data sources that describe composting activity over time?

Regarding the first bullet point, reviewers from Alaska's DEC agreed that this assumption is
correct. Alaska has no commercial composting operations presently. Limited Anchorage Muni

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

experimental composting is set-up at present. They also noted some villages have very small
local composting operations, but these are not commercial arrangements.

Reviewers from Rhode Island's DEM noted that RIDEM collects annual solid waste data from
permitted composting facilities from 2016 through 2020. This includes tonnages of the waste
incoming and outgoing from the facilities. In addition, other than solid waste tonnage data, the
facility name & contacts, address and permit capacities are maintained. They cautioned that since
this data collection effort is relatively new, RIDEM does not have 100% of the data for years
before 2019.

Reviewers from Arizona DEQ noted that due to the large area and quantities of fish waste, it may
not be correct to assume that there is no commercial composting in Alaska, though they were
unaware of any state-level data sources that describe composting activity over time. Reviewers
from Iowa commented that this sector is not included in their inventory and will research if this
is a gap and if Iowa activity data are available.

EPA Response: The EPA appreciates the data suggestions and will continue to evaluate the
feasibility of including site-specific data from each state while ensuring consistency with
national methodologies and datasets.

2.6.1.4 Stand-Alone Anaerobic Digestion

1. Do you have or know of any state-level data for counts of operational anaerobic
digesters (processing food waste) by year?

Reviewers were unaware of sources of operational anaerobic digester emission factors, counts, or
production. Reviewers from Iowa commented that this sector is not included in their inventory
and will research if this is a gap and if Iowa activity data are available.

EPA Response: The EPA appreciates the feedback acknowledging limited sources of state-
level data on operational anaerobic digesters.

2. Are there any facility-specific data sources we could use to fill data gaps on the quantity
of waste processed by stand-alone digesters for any and all years of the 1990-2019 time
series?

Reviewers from Rhode Island's DEM noted that in Rhode Island there is one stand-alone
facility, but they are not fully operational at this time. The owners are still going through their
shakedown period and stack testing has not yet been completed. After testing, RIDEM will
revise the permit to include updated emissions limits and conditions. After that, they will be
considered fully operational. In the future, RIDEM will receive data on quantity of waste
processed via Air Inventory Forms. Reviewers from Arizona's DEQ noted that there are sources
that break down digester type for specific estimate such as biogas data, but they could not find
any facility-specific data sources for quantity of waste processed by stand-alone digesters.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers from Alaska's DEC were unaware of such information, but think it could be found
after some research.

EPA Response: The EPA appreciates the commenters offer to either provide or point the EPA
to a potential data source to further refine emissions estimates.

2. 6.2 Wa stew at er

1. The following national average parameters were used to estimate emissions by state,
with state populations used to proxy the distribution of domestic emissions and state-
level production data (if available) used to proxy the distribution of industrial
emissions. Please comment if you believe states would differ significantly from the
national averages for the following parameters and, if so, whether there are state-
specific data sources for the EPA to consider:

• wastewater outflow

• biological oxygen demand (BOD), total N, and chemical oxygen demand (COD)
concentration in untreated wastewater

• BODrCOD ratios for industrial wastewater

• wastewater treatment unit operations in use at centralized domestic treatment
plants or at industrial plants

Reviewers from Alaska's DEC suggested DEC Water Division likely contains information
through the wastewater permits and water well program. The DEC also may have some air
quality permit information for permitted facilities, they noted.

EPA Response: The EPA appreciates the commenters offer to either provide or point the EPA
to a potential data source to further refine emissions estimates.

2. Are there domestic or industrial wastewater treatment operations present on other
Pacific islands for industrial sectors included in the national Inventory?

Reviewers provided no comments.

3. For each of the wastewater treatment and discharge subcategories listed for this

category, is there any information that was not considered on available state-level data
sources with regional or other disaggregated information on emissions?

Reviewers were unaware of additional data sources. Reviewers from Iowa reported that their
estimates were 64% lower than EPA's estimate for 2019. They attributed this to being unable to
find state-specific activity data to calculate emissions from several sources.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

EPA Response: The EPA recognizes that complete state-level data have not been included
and therefore relied on a national value which is based on more recently available data. The
EPA is continuing to determine state-level sources for type of wastewater treatment (in the
Planned Improvements section) that may inform the point of centralized versus onsite
treatment.

2.6.2.1 Domestic

1. The following national average parameters were used to estimate domestic wastewater
treatment emissions by state. Please comment on whether you think that states would
differ significantly from the national averages for the following parameters and, if so,
are there state-specific data sources for the EPA to consider:

• discharge of POTWs to impaired waterbodies and nonimpaired waterbodies

• discharge of POTWs to reservoirs, lakes, and estuaries

• consumed protein

• percentage of the population on septic (versus centralized treatment)

Reviewers were unaware of significant changes or additional data sources and suggested the
national averages for these parameters are acceptable. Reviewers from Iowa calculated municipal
wastewater emissions using the SIT wastewater module along with the fraction of Iowa's
population without septic systems, 76%.19 They noted that this value has not been updated since
2002 and the fraction value has not included in the US Census of Housing since 1990. They also
noted they used the most recent protein value from the national Inventory.

Reviewers from Alaska's DEC noted Alaska DEC Water Division has this information and
possibly permitted facilities in the air program for the discharge of POTWs to reservoirs, lakes,
and estuaries. To address consumed protein, they also suggested calculating a percentage of the
population on septic (versus centralized treatment) as septic systems are common in rural Alaska
except in Western and Northern Alaska.

EPA Response: The EPA appreciates the commenters offer to either provide or point the EPA
to a potential data source to further refine emissions estimates. The EPA also thanks the
commenter for their thoughts on how the national value for the percentage ofpopulation
served by centralized treatment may not be appropriate for the state of Alaska, The EPA plans
to update the Planned Improvements to reflect this potential improvement.

2.6.2.2 Industrial—Pulp and Paper

1. Pulp and paper wastewater flows were estimated using the EPA's ECHO datasets. Do
you have any reason to believe that states' pulp and paper wastewater information is

19 US Environmental Protection Agency. (2002). "Onsite Wastewater Treatment Systems Manual," Available at:
https://www.epa.gov/sites/default/files/2015-06/documents/2004 07 07 septics septic 2002 osdm all.pdf

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

underrepresented in ECHO? If so, do you have an alternative, publicly available pulp
and paper wastewater dataset by state?

Reviewers from Texas CEQ suggested the Technical Association for the Pulp and Paper
Industry20 and the National council for Air and Stream Improvement.21

EPA Response: The EPA appreciates the commenters offer to either provide or point the EPA
to a potential data source to further refine emissions estimates.

2. Currently, a single year, 2019, is used to estimate the distribution of national estimates
to each state and territory for every year of the time series. Is there reason to believe
states' pulp and paper manufacturing operations have changed significantly since
1990? If so, are there data sources to quantify those changes?

Reviewers from Texas CEQ suggested 1990 data should not be used to determine 2020
emissions and referenced sources listed in the question above.

EPA Response: The EPA appreciates Texas CEQfeedback which suggests operations may
have changes since 1990. See the EPA response to previous question.

3. Data for pulp and paper manufacturing for U.S. territories are limited in the ECHO
dataset. Are there resources to help estimate a time series of production data for pulp
and paper wastewater flows? Or are there territory-level data on the number of pulp
and paper plants in each U.S. territories?

Reviewers provided no comments.

2.6.2.3 Industrial—Meat and Poultry

1. Currently, a single year, 2019, is used to estimate the distribution of national estimates
to each state and territory for every year of the time series. Is there reason to believe
states' meat and poultry processing operations have changed significantly since 1990? If
so, are there data sources to quantify those changes?

Reviewers were unaware of significant changes to this category or additional data sources.

2. Data for meat and poultry processing for U.S. territories are not captured in the USDA
dataset. Are there resources to help estimate a time series of production data for

20 Available at: https://www.tappi.org/

21 Available at: https://www.ncasi.org/

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

poultry (broilers, turkeys, chicken), beef and calves, hogs, and sheep (lamb and
mutton), for example, live weight killed, number of head slaughtered?

Reviewers were unaware of significant changes to this category or additional data sources.

2.6.2.4 Industrial—Fruits and Vegetables

1. Currently, a single year, 2017, is used to estimate the distribution of national estimates
to each state and territory for every year of the time series. Is there reason to believe
states' fruit and vegetable processing operations have changed significantly since 1990?
If so, are there data sources to quantify those changes?

Reviewers were unaware of significant changes to this category or additional data sources.

2. Data for fruit and vegetable processing for U.S. territories are not captured in the
USDA dataset. Are there resources to help estimate a time series of territory-level
production data for fruits and vegetables, for example, canned and frozen processed
vegetables, potato production, noncitrus fruits, and citrus production?

Reviewers were unaware of additional data sources.

2.6.2.5 Industrial—Ethanol

1. Ethanol production for each state was estimated using the Energy Information
Administration (EIA) SEDS dataset. Do you have any reason to believe that states'
information is underrepresented in the SEDS dataset? If so, do you have an alternative,
publicly available ethanol production dataset by state?

Reviewers from Texas CEQ referenced the Nebraska Department of Environment and Energy
which lists Ethanol Facilities Capacity by state and Plant which can be used for cross reference.22

EPA Response: The EPA appreciates the commenters offer to either provide or point the EPA
to a potential data source to further refine emissions estimates.

2. Data for ethanol production for U.S. territories are limited in the SEDS dataset. Are
there resources to help estimate a time series of production data for ethanol
production?

22 Nebraska Department of Environment and Energy. (2020). "Ethanol Facilities Capacity by State and Plant,"
https ://neo ,ne. gov/pro grams/stats/inf/122. htm

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

Reviewers provided no comments.

3. What other relevant data sources could be included? If data are used at the national
level, are you aware of other comparable data sources at the state level?

Reviewers provided no comments.

2.6.2.6 Industrial—Petroleum

1. Petroleum production for each state was estimated using EIA's Petroleum

Administration for Defense Districts production and state-level operating capacity
datasets. Do you have any reason to believe that states' information is underrepresented
in the EIA datasets? If so, do you know of an alternative, publicly available petroleum
refining production dataset by state?

Reviewers from Alaska's DEC suggested the Department of Revenue likely has this information.

EPA Response: The EPA appreciates the commenters offer to either provide or point the EPA

to a potential data source to further refine emissions estimates.

2. Do you have any concerns about using operating capacity to estimate petroleum
production by state is not a good method? If so, would you suggest an alternative
method?

Reviewers provided no comments.

3. Data for petroleum refining for U.S. territories are limited in the EIA dataset. Are there
resources to help estimate a time series of territory-level production data for petroleum
production?

Reviewers provided no comments.

2.6.2.7 Industrial—Breweries

1. Brewery production, and by extension brewery production emissions, for each state was
estimated using the Alcohol and Tobacco Tax and Trade Bureau (TTB) taxable
production dataset.

• The TTB dataset is based on taxable production/volume. Is there any reason why
taxable production from breweries may be underrepresented by state and therefore

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

potentially underrepresent total emissions? If so, do you know of an alternative
dataset or assumption?

• The TTB dataset provides production data from 2008 to the present. The 2008
values were used as a proxy for 1990-2007 values. Is there reason to believe states'
brewery production has significantly changed over that time period? If so, do you
have an alternative, publicly available state-level brewery production dataset (i.e.,
barrels produced) or suggestions for alternative data to use as a proxy?

Reviewers from Texas CEQ found the methodology sound and were unaware of other proxy,
datasets, or assumptions. Reviewers from Rhode Island's DEM had no reason to believe that the
taxable production/volume is not accurate. Reviewers from Alaska's DEC also thought TBB
production data are reliable. Reviewers from Rhode Island's DEM commented that there is no
reason to believe that Rhode Island's brewing production has significantly increased during 1990
through 2007.

EPA Response: The EPA thanks the state reviewers for their feedback on TBB data.

2. Data for brewery production for U.S. territories are limited in the TTB dataset. Are
there resources to help estimate a time series of territory-level production data for
brewery production (i.e., barrels produced)?

Reviewers provided no comments.

EPA Response: The EPA notes that no feedback was provided on this question.

2.7 Comments from Independent Organizations

The United States Climate Alliance (USCA) submitted comments of behalf of alliance Sates that
emphasized the importance of consistency between the state level results and other state level
sources such as the SIT and from the national Inventory and other data available at the federal
level, e.g., the Energy Information Administration's (EIA) State Energy Data System (SEDS)
and Greenhouse Gas Reporting Program (GHGRP). Data consistency across federal and state
tools is important because states rely on SIT as the starting point for their own official GHG
inventories and conflicting datasets requires justification of one source over another. When this
is the case, the USCA comments recommended making the methodology explicit to identify and
explain methodological or data differences for each state. The USCA comments suggested
releasing downscaled inventory data after EPA's SIT is released for the same year as several
Alliance states rely on SIT for their state-specific greenhouse gas inventory (GHGI).

EPA Response: The EPA has provided further information and clarity on the data sources
used in the state-level estimates, in particular for the use of SEDS and GHGRP data. This
effort will also provide updated information for EPA '.s existing State Inventory Tool.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

The EPA also recognizes that a number of states rely on the SIT as a starting point for
preparing their state GHG inventory estimates. The SIT includes default activity data and
estimates that states can use as a starting point for compiling a state-level GHG inventory. The
default data included in SIT are largely consistent with the EPA's state-level inventory
because the data are based on methodologies, emissions factors, and other data from the
national Inventory. However, some differences exist between default data in SIT and the
GHGI by state estimates due to differences in methods, data, and level of completeness. As
additional state-level data and/or methodological approaches become available through the
national Inventory disaggregation, they will be used to supplement or improve the embedded
calculations and defaults in SIT as appropriate. SIT users will retain the ability to customize
the tool with their own data in lieu of using any defaults. Consistent with IPCC
methodological guidance, for more significant emission sources, it is good practice to apply
higher-tiered methods which generally reflect more detailed characterization of activities and
involve use of data and emission factors that are country, region, state or technology specific.

Regarding uncertainty, the USCA provided comments on behalf of alliance states and suggested
highlighting individual state inventories as the preferred source of state-specific GHG emissions
when relevant. When state inventories are not available, they thought downscaled data could be a
useful source but in cases where bottom-up inventory data are not available, they suggested
including appropriate caveats and disclaimers to make it clear that state inventories take priority
for policy making and tracking. They suggested explaining any remaining differences in
methodologies between state GHGI results and SIT. Furthermore, the USCA comments
suggested uncertainty for at least the current year and common baseline years, e.g., 1990, 2005.

EPA Response: The EPA recognizes that there will be differences between the EPA '.s state-
level estimates and some inventory estimates developed independently by individual state
governments. The EPA will add a caveat to the data to indicate that inventory data presented
here should not be viewed as official data of any state government. The EPA will also point to
additional information on official state data, where that data exist, including information on
potential areas of difference between EPA '.s data and official state data.

At this time, the uncertainty provided in the GHG Inventory by U.S. State report reflects that
of the national Inventory and includes a qualitative discussion of state level uncertainties. The
EPA continues to assess how best to analyze and present qualitative uncertainty estimates
associated with state-level estimates but will do so in the context of prioritizing improvements
in other areas.

The USCA comments suggested adding additional caveats explaining the lack of standardized
allocation methodology for bunker fuel. As a source category, many Alliance states are actively
working to improve, and Alliance states would welcome continued discussion on the topic.

-------
U.S. Environmental Protection Agency
STATE TECHNICAL REVIEW SUMMARY OF COMMENTS AND RESPONSES REPORT

detail on how transportation emissions are developed at the national and state levels in the
report to clarify where emissions from different sources are accountedfor.

The USCA comments on landfill emissions reported, "It appears that the oxidation was removed
after recovery for years 1990-2004 (flat 10%). For years 2005-2019, the oxidation was removed
before the recovery (flat 19.5%) which implies that oxidation may be overestimated for years
2005-2019 (Section 6.1.1 in draft methodology report, with the data in Appendix F). EPA should
clarify if oxidation is removed after recovery or why the oxidation appears to be removed after
recovery for one period and before recover for another period."

EPA Response: The EPA has removed the MSW activity estimates from the GHG Inventory by
U.S. State for the 1990—2019 report because presenting them was not consistent with the
national Inventory. The forthcoming national Inventory report for 1990—2020 will include
these, and the full methodology for developing the estimates will be presented.

65

-------
3. Individual State Comments

This section provides comments from States presented in alphabetical order.

3.1 Alaska Department of Environmental Conservatic

Charge to State GHG Inventory Experts for the Draft Methodology Report: Inventory of

U.S. Greenhouse Emissions and Sinks by State

General Questions

a) What are your overall impressions of the clarity of the methods described in this report?
• The methods are clearly laid out and established.

b) What recommendations do you have to add to or improve the overall
transparency, completeness, consistency and accuracy of this report?

The accuracy of the methodology needs to be adjusted for the State of
Alaska when using national averages. Some of Alaska's differences from
national averages include Land Use/Change, decomposition rates in
landfills, population using waste treatment systems and electrical grid,
agriculture, and many other sectors. Our responses in this questionnaire
will identity data availability as well as comments on methodology.

c) Data availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that
are notcurrently included, but could be incorporated into this analysis?

• Agriculture - Alaska Department of Agriculture has available data
for state Ag to supplement USD A information; reference 2020
NEI as well.

• Oil and Gas flaring - Alaska Oil and Gas Conservation Council
(AOGCC) puts out yearly reports that have flaring data (over 1
hour) and fills in the gaps on routine practices and flaring less than
one hour.

• Energy Fugitive Emissions: State of Alaska NEI

• Waste water- NEI to include PW Treatment permitted facilities

• Land Use/Change: U.S. Geological Survey (USGS) has Land
Use/Change data and Carbon Sequestration Capacity data that
should be integrated. It is far more complete than the EPA data
sets that are limited to Southcentral and Southeast Alaska.

b. For national level datasets that are currently used, are you aware of other
comparabledatasets of activity, emission factor, or emissions data that are
available at the State, county, or zip-code levels?

• NEI should be used for point and non-point sources

d) For the oil and gas sector, the state contractor used the NEI to generate GHG
emissions that we believe is a good representative of how that can be achieved

-------
(attached)

e)	Key Category Analysis. EPA anticipates prioritizing methodological refinements for
more significant categories to make efficient use of available resources over time. EPA
appreciatesfeedback on which categories are more relevant for further refining for your
State.

a.	Given that the emissions profile of some states will be different from the
national average, which categories that are more significant in terms of
absolute emissions, orhave changing emission trends (e.g., increasing,
variable)?

¦	Land use/change should be viewed as separate from national average
and analyzed separately from nationwide land use.

¦	State landfills are not calculated using continental US decomposition
rates; need to work with DEC on solid waste decomposition for future
emissions modeling. Decomposition is different in cold regions,
especially with permafrost.

¦	Pipeline compressor stations, pipeline, and tank emissions. Oil and gas
rigs are different on the North Slope. The infrastructure for the lower
48 does not match what we have in AK.

b.	The national Inventory includes a key category analysis (KCA) consistent with
2006 IPCC Guidelines. Would it be useful for States if a key category analysis
(KCA) was completedfor each state? Yes

•

f)	Data Presentation and Usability.

a.	Are there other ways the state-level emissions data could be presented to
facilitate theiruse (e.g., in the EPA GHG Inventory Data Explorer available
online at: https://cfpub.epa.gov/ghgdata/inventoryexplorer/)?

i.	Related to the level of category/gas aggregation or disaggregation?

ii.	Are there specific categories where further data disaggregation
could behelpful? No comment

b.	What data format would best facilitate the use of the state-level emissions data

(eg-,

.xlsx download, etc.)? any spreadsheet format works.

c.	What additional datasets or information could be provided to help increase the
usability of the state-level emissions data? Alaska Oil and Gas Conservation
Commission maintains data on all wells drilled that is complete.

d.	EPA plans to provide users additional information on where they can find
official State data, where it exists. Do you have suggestions on how we should
direct users to official state data (i.e., section in methods report including links
to State data)?

• Include links to state websites. Include link to updated EPA NE1 data
Energy - Combustion Emissions Questions

-------
Some fuels have differences in consumption data between the aggregated State-level
totals andnational totals. The current approach is to use data from the national Inventory
in those cases. Are there other approaches that could be taken? Has this come up in
developing estimates at the State level?

NEI data is very helpful but the annual data sets are for the largest facilities
only. It may be worth while to calculate a percentage that the annual data
represents the rest of the state emissions and update that scaling factor every
three years.

• The SEDS data has been beneficial and does account for small facilities that
don't report emissions through the permit program

• Other comments for consideration: For the development of state-level
inventories for Alaska, what is needed is significantly more engagement with
state-level planners for generating aggregate totals. AK has a unique
relationship with both aviation and maritime transportation due to the state's
infrastructure arrangements. Fuel consumption habits vary in the state compared
to utilization in the continental United States. There is continued use of large
amounts of light aircraft to bring passengers and cargo to and from remote
communities, some of which still burn leaded AvGas and non-standard fuels.
Alaska also receives a large amount of international marine traffic via the Great
Circle Route which needs to be considered for long-term fuel consumption data.
Many international vessels are arriving in state waters using IMO-designated
low sulfur content fuel which generates other CAP and GHG emissions due to
fuel chemistry from prior bunker fuel. The state's adjacent location to major
developing trans-Arctic shipping routes could also change how the state's fuel
consumption habits appear on national-level inventories.

Rather than relying on generic nationwide data, ADEC strongly recommends
coordination with the NEI team, specifically the teams working on the aviation
and marine inventories. Both have developed specific activity data which
incorporates satellite location-driven emissions data and actual engine activity
data. This would be far more accurate, as CAP emissions have already been
generated by this data. All that would be needed at that stage would be the
application of emissions factors for GHG emissions. The same applies for the
state's aviation inventory, which is already generated by LTO data from the NEI
team. This dataset already includes activity and landing and take-off data from
all categories of aircraft.

As both the aviation and marine inventories have already been generated by the
NEI team, ADEC encourages the EPA GHG team to coordinate with the NEI
team for the purposes of GHG inventory generation,
b) Consistent with the IPCC Guidelines, we have adjusted fuel consumption totals
in the energy sector to account for consumption in the IPPU sector. In some cases, this

-------
step could lead to a negative emission total for a State if the subtracted amount (as
determined from the assumed distribution) was greater than consumption data from the
State Energy Data System (SEDS). Thisoutcome was corrected to zero if that was the
case, but are there other approaches for correcting for that difference? No comment
•

c) Consistent with the national Inventory, the default approach taken here was to allocate
transportation sector C02 emissions based on FHWA fuel use/sales by State. For some
States, this may not be accurate because fuel sold in a State may be combusted in other
States. Anotheroption is to use vehicle miles traveled (VMT) data by State but that
approach does not factor in vehicle fuel economy. Are there other alternative or
complementary approaches to allocate transportation fuel across States, including
VMT data and other sources (e.g., NEI - based on county-level fleet and activity data
to generate a bottom-up inventory) that EPA should

consider? If so, what data sources exist to help with that alternative approach? Would it be
helpful to present transportation sector emissions using multiple approaches in future
inventories?

• The existing methodology works for AK since we don't have adjacent states.

• The SEDS data set is reliable and should be used - as it is with the SIT tool.

• ADEC recommends looking at the NEI dataset for mobile sources combustion,
especially on- and off-road, maritime, and aviation emissions.

Note: FHWA does not have complete data on registered vehicles because native
populations don't have to register vehicles that don't travel on the highway system.
In the triennial NEI, the state uses an adjustment of 10% to compensate for all the
unregistered vehicles.

d) Mobile source non-C02 emissions are allocated across States based on vehicle-miles-
traveled data while mobile source C02 emissions were allocated based on fuel sales,
as mentioned above. Do you have any concerns with using two different
methodologies for mobile source C02vs. non-C02 State splits? No

e) Several fuels have variable C factors over time including coal, natural gas, gasoline,
and diesel fuel. Those fuels might also have variable C factors across areas/States.
Are data available to build out State-specific C factors for the fuels with variable C
contents? If so, could it be done ina way that the State-level total emissions still
matched up to the national total emissions for those fuels?

•

f) Geothermal emissions could be allocated by the type of geothermal production per
State (because different types have different emissions factors) if that data is
available. Is there moreinformation on State-level geothermal emission factors and
production?

• No geothermal data is available.

-------
• Note: As Alaska has significant amounts of volcanic geothermal activity
throughout the state. Rather than verifying with ADEC, this is a question which needs
to be taken to the USGS and University of Alaska-Fairbanks. Both USGS and UA-F
have large databases of information on geothermal power options that should be
reviewed and used if possible. Although these are natural sources, it would be
valuable to include in the report.

Non-Energy Uses (NEU) of Fossil Fuels:

a) For petrochemical feedstocks, non-energy use (NEU) of natural gas is allocated across
States based on petrochemicals emissions data per State from the IPPU adjustments,
while other fuelsare allocated based on the underlying SEDS data. Allocating across
States based on the underlying SEDS data ensures there are no States where NEU use
is larger than original SEDS data and there are no zeros associated with subtracting
NEU (it is not an issue for natural gas because use is so high overall compared with
NEU use). Could different approaches be used or can the petrochemical data be used
without resulting in negative use?

AK trusts the SEDS data and it is probably the best source of data for AK. In Alaska, there is
not a lot of use for NEU of fuels.

Incineration of Waste:

a) Waste incineration emissions are calculated based on the combustion of fossil components
of both municipal solid waste (MSW) and tires. However, emissions are disaggregated to
Sates based only on MSW tonnage. Are there approaches or data available to disaggregate
emissionsbased on waste category (e.g., MSW combustion vs. tire combustion)?

• In AK, burning tires is prohibited except at approved incineration facilities.
Normally, large incineration facilities are permitted, and their emissions are
reported in the NEI, however few small rural incinerator facilities are permitted so
NEI data is limited. It may be worthwhile to develop a scaling technique for
village incinerators and assume they burn tires.

International Bunker Fuels:

a) The approach used to allocate jet fuel bunker fuels by State is currently based on the total

amount of jet fuel used by State which could potentially lead to an over- or under-estimation for
some States of bunker fuel emissions. Are there other more accurate approaches to allocate jet
fuel bunker data across States as opposed to the percentage of jet fuel total use? For example,
using Federal Aviation Administration flight level data on departures and destinations or
assuming based on States with international airports and flights?

• DEC relies on the EI A data. AK does have a lot of international travel. Perhaps, the
NEI landing/takeoff data could be used.

-------
Energy - Fugitive Emissions Questions
Coal mining:

a)	Do you have any comments specific to the methodology and emission estimates for
active coalmines and abandoned coal mines?

Alaska's coal mining activity is limited compared to its historical activity
thanks in large part to the end of coal exports to East Asia in the last decade. The
state only has one active coal mine at present. All other coal mines have been closed
for several years under state and federal reclamation rules. . For the 2020 Alaska
Greenhouse Gas Report, the DEC conducted its first inventory of abandoned coal
mines and included them in the state report. The average year these coal mines were
abandoned was 1941.

Why is the focus on methane recovery from abandoned coal mines. From initial
research at the state level, it does not appear that there is any ongoing methane
capture or recovery from any abandoned mines in the state. Due to the age of most
of these abandoned mines, it is highly unlikely that they would be a viable target for
methane capture or recovery. Any application of generic methane recovery
emissions Should not be applied to Alaska.

b)	Are you aware of any State datasets that may be useful in helping to refine emission
estimatesfor abandoned coal mines, including State-level datasets addressing recovery
of methane from abandoned mines?

The only state datasets that would be of use are those from the AK Department of
Mining, Land, and Water and Alaska Department of Natural Resources which
outlines all identified remaining abandoned mines in the state. Additional research
was needed from ADEC to verify age of these mines and date of mine closure. In
some instances, the date of mine closure could not be verified or was within a five-
year period. In those instances, earliest estimated date of mine closure was used.
Should EPA require access, ADEC can forward the verified dataset on for use
with future NE1 and GHG estimates. However, database does not have emissions
information which needs to be added in by EPA.

Petroleum Systems and Natural Gas Systems:

a) Are there relevant dataset(s) that could be used to replace or supplement the data
currently used to allocate petroleum and natural gas system emissions to the state
level? Particularly, state or detailed location information on gathering and boosting
stations, processing plants, andtransmission and storage stations?

• DEC has emission fee data from these sources as well as NE1 information. In 2000
we asked all small nonpoint sources to submit NEI emission information. Once

-------
submitted to EPA, It will be the most complete NEI on these sources; however,
NEI does not include fugitive emissions unless it is a permitted emission unit.
Permit emission fee information has more fugitive information. This data is only
for CAPS data but could be used to calculate GHGs in some cases.

b)	Are there additional Greenhouse Gas Reporting Program (GHGRP) data that could
be used toallocate natural gas and petroleum emissions to each State?

•	see above

c)	Are you aware of any State datasets that may be useful in helping to refine emission
estimatesfor abandoned wells, including State-level datasets addressing plugging
status of abandoned wells?

•	The Alaska BLM abandoned well program (aka legacy wells or orphaned
wells) carries numerous datasets.

•	Also, the Alaska oil and Gas Conservation Commission has a data set for all
wells ever drilled that lists well status. Please note that the Commission is very
strict on plug and abandonment practices. There are not as many abandoned wells
in AK as in other states. DNR Division of Oil and Gas estimates there are 20 well.

d)	Are there particular sources for which State-level regulatory or voluntary programs
result inlarge differences in emission rates between states? Are state-specific data
sets available forthose sources? Not applicable

IPPU - Minerals Emissions Questions

Cement Production:

) Are you aware of data on clinker production by all States for the full 1990-2019
time series? Please share a surrogate data that could be used (e.g., facility
production capacity, utilization rates by facility or State) for 1990-2019 that could
refine this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends. No cement production in AK.

Lime Production:

a)	Are you aware of data on State-level lime production (activity data) by type (e.g.,
high-calcium quicklime; dolomitic quicklime, high-calcium, hydrated; dolomitic,
hydrated; dead-burned dolomite; CO2 captured for use in onsite processes) for some
or all of the 1990-2019 time series? If not, is there any surrogate data (e.g., facility
production capacity, utilization rates by facility or State) for 1990-2019 that could
refine this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends?

•

b)	Based on analysis of Greenhouse Gas Reporting Program data, it appears that most
but not all beet sugar manufacturing facilities that also produce lime and a few lime

-------
manufacturing facilities capture CO2 for use in onsite processes. Are you aware of any
information on why and how facilities producing lime capture CO2 for use in onsite
processes (e.g., purification), and any trends in this practice during the 1990-2019
time series (e.g., have facilities increased or decreased adoption of this practice during
the time series), or whether the amount of CO2 captured is proportional to the amount
of lime produced or some other metric?

• This does not apply to Alaska.

Glass Production:

a) Are you aware of state-level data on glass production or the amount of carbonate
(i.e., limestone, dolomite, soda ash) consumed for glass production by State (activity
data) for some or all of the 1990-2019 time series? If not, can you share any state-
level surrogate data (e.g., more complete data on glass facilities by State, amount of
glass products by type [i.e., containers, flat (window) glass, fiber glass, and specialty
glass]) for 1990-2019 that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends?

• This does not apply to Alaska.

Other Process Uses of Carbonates:

a) Are you aware of state-level data on the consumption of limestone and dolomite for
the iron and steel sector for the 1990-2019 time series? If not, can you share any
state-level surrogate data for 1990-2019 that could refine this State inventory
calculation to enhance accuracy and consistency of State GHG emissions and trends
from carbonate consumption by the iron and steel sector? This does not apply to
Alaska.

b) Are you aware of state-level data on the consumption of soda ash (not associated
with glass manufacturing) for the 1990-2019 time series? This does not apply to
Alaska.

c) Are you aware of any state-level data on limestone and dolomite consumption for
flux stone, flue gas desulfurization systems, chemical stone, mine dusting or acid
water treatment, acid neutralization, and sugar refining activities for the 1990-2019
time series? This does not apply to Alaska.

CO2 Consumption:

a) Are you aware of other sources of data on the consumption of CChby State or
region for the 1990-2019 time series? This does not apply to Alaska.

IPPU - Chemicals Emissions Questions

Ammonia Production:

a) Currently, production capacity is used as a surrogate for state-level ammonia

production for 1990-2009. In the absence of ammonia production by State in more
recent years, are you awareof other surrogate data (e.g., facility utilization rates by

-------
State) that could refine this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends?

• This does not apply to Alaska to this time.

Urea Consumption for Nonagricultural Purposes:

a) Are you aware of state-level data on urea consumption for nonagri cultural purposes
(activity data) for some or all of the 1990-2019 time series? No

Nitric Acid Production:

a) Are you aware of state-level data on nitric acid production (activity data) for some or
all of the 1990-2009 time series? We currently use production capacity as a
surrogate for nitric acid production by State for 1990-2009. We know that the
production capacity data used for this State inventory calculation are incomplete for
1990-2009. Are you aware of more complete data on facility production capacity by
State?

• Alaska does not have Nitric Acid production at this time.

•

b) Are you aware of surrogate state-level data other than facility production capacity
(e.g., utilization rates by facility or State, information about abatement technology
installations anduse per facility) for 1990-2009 that could refine this State inventory
calculation to enhance accuracy and consistency of State GHG emissions and
trends?

• Currently, there is no surrogate state-level data to replace national level data.
However, do not apply generic national emissions to Alaska.

Adipic Acid Production:

a) Are you aware of any other state-level data on adipic acid production (activity or emissions data)
for some or all of the 1990-2019 time series?

• No other state level data for adipic acid production in Alaska at present. Alaska
does not have this production. Do not apply generic national emissions to Alaska.

Caprolactam, Glyoxal, and Glyoxylic Acid Production:

a) Are you aware of state-level data on caprolactam production or emissions for some or
all of the 1990-2009 time series? We currently use production capacity as a surrogate
for caprolactam production by State. Are you aware of more complete data on facility
production capacity or actual production by State? Are you aware of better surrogate
data other than facility production capacity (e.g., utilization rates by facility or State,
information about abatement technology installations and use per facility) that could
refine this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends?

-------
• Currently, there is no surrogate state-level data to replace national level data.
However, do not apply generic national emissions to Alaska.

Carbide Production and Consumption:

a)	Are you aware of state-level data on SiC production (activity data) for the 1990-2019
time series? Are you aware of other data to refine accuracy of the estimation of SiC
consumption by State for the 1990-2019 time series?

• Alaska does not have carbide manufacturing or production

b)	Are you aware of information that can help us improve the accuracy of production
in the two States where SiC facilities are located?

Titanium Dioxide (TiCh) Production:

a) Are you aware of state-level data on TiCh production (activity data) for the 1990-2009
time series? Is there any surrogate data other than facility production capacity (e.g.,
facility utilization rates by facility or State) for 1990-2009 that could refine this State
inventory calculation to enhance accuracy and consistency of State GHG emissions
and trends?

• Currently, there is no surrogate state-level data to replace national level data.

However, do not apply generic national emissions to Alaska.

Petrochemical Production:

This section is not applicable to AK

a) Are you aware of state-level data on petrochemical production by type for the
1990-2019 timeseries? Is there any other surrogate data by State or facility (e.g.,
facility production capacity; utilization rates by facility or State; timing of facility
expansions, openings, and temporary or permanent closures) for the full 1990-2019
time series that could address data gaps and refine this State inventory calculation to
enhance accuracy and consistency of State GHG emissions and trends?

Phosphoric Acid Production:

a) Are you aware of state-level data on phosphoric acid production (activity data) for
the 1990- 2009 time series? Is there any other surrogate data or information (e.g.,
timing of facility expansions and temporary or permanent closures, origin of phosphate
rock used in facilities) by State or facility for 1990-2019 that could refine this State
inventory calculation to enhance accuracy and consistency of State GHG emissions and
trends?

HCFC-22 Production:

-------
a)	For the years 1990-2009, there are significant uncertainties in the allocation of
national-level U.S. emissions to individual facilities and States, particularly for the
five HCFC-22 production facilities that closed before 2003 and for which production
capacity data are therefore not available. Are you aware of any more complete sources
of production capacity or other relevant historical data?

b)	Do you have recommendations for how to refine the methodology to more accurately
estimate emissions from HCFC-22 production over the time series?

IPPU - Metals Emissions Questions

This section is not applicable to AK

Iron and Steel and Metallurgical Coke Production:

a)	Are you aware of State-level data on iron and steel production (activity data) by
category (i.e., sinter production, iron production, pellet production, steel production,
other activities) for some or all of the 1990-2019 time series? In the absence of steel
production by State, are you aware of better surrogate data that could refine this State
inventory calculation to enhance accuracy and consistency of State GHG emissions
and trends?

b)	Are you aware of state or facility-specific information to better allocate basic
oxygen furnace and electric arc furnace production by State for 1990-2009?

Ferroalloy Production:

a) Are you aware of state or facility-level data on ferroalloy production (activity data)
or facility for the 1990-2019 time series? Please share any other surrogate data (e.g.,
facility production capacity, utilization rates by facility or State) for 1990-2019 that
could refine this State inventory calculation to enhance accuracy and consistency of
State GHG emissions and trends.

Aluminum Production:

a) Are you aware of state or facility-level data available to incorporate differences in
emissions between smelters based on technology type? Is there any other
surrogate data or emission sources that could be used to allocate national total
aluminum production emissions across States?

Magnesium Production and Processing:

a)	Are you aware of state or facility-level magnesium production or capacity data (or
surrogate data) or facility for the 1990-2019 time series?

b)	Are you aware of information on the location (by State) of magnesium production
and processing facilities or information on the location (by State) of magnesium
production and processing facilities by process type?

Lead Production:

-------
a) Are you aware of state or facility-level data on primary or secondary lead production
(activity data) or facility for the 1990-2019 time series? Is there any other surrogate
data (e.g., primary or secondary production capacity by facility or State) for 1990-
2009 that could refine this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends?

Zinc Production:

a) Are you aware of data on zinc production (activity data) by unit type (i.e.,
electrothermic furnace, Waelz kiln, other furnaces, and flame reactor units) by State or
facility for the 1990- 2019 time series? Is there any other surrogate data (e.g., total
number of zinc facilities by State, production capacity by unit type and by facility or
State) or other data by State (e.g., utilization rates by facility or State) for 1990-2009
that could refine this State inventory calculation to enhance accuracy and consistency
of State GHG emissions and trends?

IPPU - Product Use Emissions Questions

This section is not applicable to AK

Electronics Industry:

a) Are you aware of State- or facility-level capacity data or other surrogate data (e.g.,
sales data) by State for PV manufacturing for 1990-2006 that could be used to refine
the allocations of emissions by State? Please share any surrogate data (e.g., sales data
by State) by State for semiconductor or MEMS manufacturing for 1990-2007 that
could be used to refine the allocations of emissions by State.

Substitution of Ozone-Depleting Substances:

a) Are you aware of bottom-up modeling data that are available by State? Is there any
surrogate data other than population data that could be used to disaggregate the
emissions of substitutes for ozone-depleting substances? No

Electrical Transmissions and Distribution:

a) Are you aware of State-level electrical transmission and distribution equipment
data (e.g., nameplate capacity by State) or other data by State for 1990-2019 (or part
of the time series) that could refine this State inventory calculation to reflect State
trends in emissions more closely? Is there any other surrogate data (e.g., State
population data) to enhance accuracy and consistency of State GHG emissions and
trends than the current data being used (transmission mile data by State)?

• The state has information in its permit system for nameplate capacity which could be
provided to EPA

N2O from Product Use:

-------
a) Are you aware of any State-level data on N2O usage for medical and dental
anesthesia, food processing propellant and aerosols, sodium azide production, or
other applications (e.g., fuel oxidant in auto racing, oxidizing agent in blowtorches)
by State for some or all of the 1990-2019 time series? Is there share any other
surrogate data (e.g., State population data) to enhance accuracy and consistency of
State GHG emissions and trends than the current data being used (transmission mile
data by State)? Not available.

Agriculture - Livestock Emissions Questions

Enteric Fermentation:

a) Are there other/newer data sources or methods, particularly at the State level, that
EPA shouldbe aware of and consider in calculating these emissions? Especially for:

• Dry matter/gross energy intake;

• Annual data for the digestible energy (DE) values (expressed as the percent of
gross energy intake digested by the animal), CH4 conversion rates (Ym)
(expressed as the fraction of gross energy converted to CH4), and crude
protein values of specific diet andfeed components for foraging and feedlot
animals;

• Monthly beef births and beef cow lactation rates;

• Weights and weight gains for beef and dairy cattle.

• The State does not have this information.

b) Are State-specific diet data available to EPA to enhance characterization of diet
differences across livestock types and U.S. States?

• Data may exist with the Department of Natural resources, Division of Agriculture
or the University of Alaska Agriculture Research Station

c) For the enteric fermentation source category and the Cattle Enteric Fermentation
Model (CEFM), are the various regional designations of U.S. States (as presented in
Annex 3.10 of theGHG Inventory) used for characterizing the diets of foraging cattle
appropriate? The CEFM is used to estimate cattle CH4 emissions from enteric
fermentation and incorporates informationon livestock population, feeding practices,
and production characteristics.

• Alaska has little cattle production therefore, USDA information is sufficient.

Manure Management:

a) Are there other/newer data sources, particularly at the State-level, that EPA should be
aware of and consider in calculating these emissions? AK does not have this
information at this time. Especially for the following:

• waste management system data, particularly seasonal changes in
emissions from different waste management systems;

• maximum methane-producing capacity;

-------
• volatile solids and nitrogen excretion rates; and

• measured emission estimates (by waste management system) to help refine
estimates of methane conversion factors.

Agriculture - Rice Cultivation, Urea Fertilization, Liming and Field Burning of
Agricultural Residues Questions

No category-specific questions, see general questions.

Agriculture - Agricultural Soil Management Emissions Questions
As described in the methodology section, EPA is currently compiling the state-level
emissions estimates from Agricultural Soil Management. EPA plans to provide these
estimates in coming weeks. The methods used to compile state-level estimates are the same
as those in the national Inventory, described in Chapter 5.4 and Annex 3.12.

a) What are your overall thoughts of the clarity and transparency of these methods? The
methods are clear.

b) What recommendations do you have to add to ensure high-quality state-level
estimates consistent with the national Inventory? Alaska has a very long winter
season. It makes sense to have an adjustment for the season where ag soils are
frozen.

LULUCF - Forest Lands and Lands Converted to Forest Land.

No category-specific questions, see general questions.

LULUCF - Croplands and Grasslands Questions
As described in the methodology section, EPA is currently compiling the state-level emissions
estimates from Croplands and Grasslands. EPA plans to provide these estimates in coming
weeks. The methods used to compile state-level estimates are the same as those in the national
Inventory, described in Chapters 6.4 through 6.7 and Annex 3.12. To view an example state-
table currently available, please seetable A-201 in Annex 3.12.

a) What are your overall thoughts of the clarity and transparency of these methods? The
methods are clear but inadequate for Alaska as the data covers only a small part of
Alaska.

b) What recommendations do you have to add to ensure high-quality state-level
estimates consistent with the national Inventory? Alaska recommends using the

• All data for croplands and grasslands is in USGS dataset; please coordinate with

them to integrate findings into future modeling.

• See also the University of Alaska, https://economic-impact-of-ag.uada.edu/alaska/

LULUCF - Wetlands and Lands Converted to Wetlands Questions

Peatlands Remaining Peatlands

-------
a)	Are there state-level data available on the application ("consumption") of peat,
including the state of use and the horticultural/landscaping use?

b)	Are there data sources that could support EPA determining the quantity of peat
harvested perhectare and the total area undergoing peat extraction?

• All data for peatlands and wetlands is in USGS dataset developed for greenhouse
gas emissions for the state. Please coordinate with them to integrate findings into
future modeling. An attached copy of this report is provided with our responses.

LULUCF - Settlements and Lands Converted to Settlements Questions

No category-specific questions, see general questions.

LULUCF - Other Lands and Lands Converted to Other Lands Questions

No category-specific questions, see general questions.

Waste - Solid Waste Disposal and Management Emissions Questions

MSW Landfills:

a) Data Questions

• Are there datasets for individual States' landfill gas (LFG) recovery activity?

Yes, Alaska has land fill gas processing for Anchorage landfill, and
Merril field. All other landfills are open air.

• Are there data available for open dumpsites in the U.S. territories?

• No. Alaska has numerous dumpsites. The Alaska DEC Environmental
Health Division may have data available for open landfills and
dumpsites that are not permitted by Air Quality.

b) The current method makes some simplifying assumptions and includes uncertainties
in the allocation of national-level U.S. emissions to States (e.g., recovery rates are
the same for all States and match the national recovery rate). Are there alternative
assumptions or different datasets that would improve the accuracy of MSW landfill
estimates? Do you have recommendations to refine the methodology to estimate
emissions over the time series more accurately?

As noted, AK has many landfills, but they are permitted with the waste
division. Using the national methods may not be applicable to AK due to the
lengthy time of frozen conditions.

Industrial Landfills:

-------
a) Do you have recommendations to refine the methodology to estimate emissions
over the time series more accurately?

Composting:

a) Data Questions

•	Is it correct to assume that Alaska has no commercial composting operations
correct?

•	This is correct, AK has no commercial composting operations at present.
Limited Anchorage Muni experimental composting set-up at present. Some
villages have very small local composting operations, but these are not
commercial arrangements.

•	Are there any datasets about composting in U.S. territories?

•	Are there any State-level data sources that describe composting activity over
time? No

Stand-Alone Anaerobic Digestion:

a)	Do you have or know of any State-level data for counts of operational anaerobic
digesters (processing food waste) by year? No.

b)	Are there any facility-specific data sources we could use to fill data gaps on the
quantity of waste processed by stand-alone digesters for any and all years of the
1990-2019 time series? No. There may be some information, but it would take
research.

Waste - Wastewater Treatment and Discharge Emissions Questions

Overall:

a) The following national average parameters were used to estimate emissions by State,
with Statepopulations used to proxy the distribution of domestic emissions and State-
level production data (if available) used to proxy the distribution of industrial
emissions. Please comment if you believe States would differ significantly from the
national averages for the following parameters and, if so, whether there are State-
specific data sources for EPA to consider:

•	wastewater outflow

•	biological oxygen demand (BOD), total N, and chemical oxygen
demand (COD)concentration in untreated wastewater

•	BOD:COD ratios for industrial wastewater

•	wastewater treatment unit operations in use at centralized domestic treatment
plantsor at industrial plants

•

• There is likely some information available through DEC Water
division in waste water permits and through the water well program.
DEC may have some air quality permit information for permitted
facilities.

-------
b)	Are there domestic or industrial wastewater treatment operations present on other
Pacificislands for industrial sectors included in the national Inventory? N/A

c)	For each of the wastewater treatment and discharge subcategories listed for this
category, isthere any information that was not considered on available State-level
data sources with regional or other disaggregated information on emissions?

d)	For each of the subcategories, what relevant data sources could be included? If data
are used atthe national level, are you aware of other comparable data sources at the
State level?

Domestic:

a) The following national average parameters were used to estimate domestic

wastewater treatment emissions by State. Please comment on whether you think that
States would differ significantly from the national averages for the following
parameters and, if so, are there State-specific data sources for EPA to consider:

•	discharge of publicly owned treatment works (POTWs) to impaired
waterbodies andnonimpaired waterbodies Alaska has no impaired water
bodies. There is likely information on permitted facilities.

•	discharge of POTWs to reservoirs, lakes, and estuaries. Alaska DEC Water
division will have this information and perhaps permitted facilities in the air
program.

•	consumed protein

•	percentage of the population on septic (versus centralized treatment): Septic
systems are common in rural Alaska except in Western and Northern Alaska.

Industrial - Pulp and Paper:

•	No Pulp and Paper in AK

a)	Pulp and paper wastewater flows were estimated using EPA's Enforcement and
Compliance History Online (ECHO) datasets. Do you have any reason to believe that
States' pulp and paper wastewater information is underrepresented in ECHO? If so, do
you have an alternative, publicly available pulp and paper wastewater dataset by State?

b)	Currently, a single year, 2019, is used to estimate the distribution of national
estimates to each State and territory for every year of the time series. Is there reason
to believe States' pulp and paper manufacturing operations have changed
significantly since 1990? If so, are there data sources to quantify those changes?

c)	Data for pulp and paper manufacturing for U. S. territories are limited in the ECHO
dataset. Are there resources to help estimate a time series of production data for pulp
and paper wastewater flows? Or are there territory-level data on the number of pulp
and paper plants in each U.S. territories?

Industrial - Meat and Poultry:

a) Currently, a single year, 2019, is used to estimate the distribution of national

estimates to each State and territory for every year of the time series. Is there reason

-------
to believe States' meat and poultry processing operations have changed significantly
since 1990? If so, are there data sources to quantify those changes? No.

b) Data for meat and poultry processing for U.S. territories are not captured in the
USD A dataset. Are there resources to help estimate a time series of territory-level
production data for poultry (broilers, turkeys, chicken), beef and calves, hogs, and
sheep (lamb and mutton), for example, live weight killed, number of head
slaughtered?

• AK accepts USDA information.

Industrial - Fruits and Vegetables:

a) Currently, a single year, 2017, is used to estimate the distribution of national
estimates to each State and territory for every year of the time series. Is there reason
to believe States' fruit and vegetable processing operations have changed
significantly since 1990? If so, are there data sources to quantify those changes? No.

b) Data for fruit and vegetable processing for U.S. territories are not captured in the
USDA dataset. Are there resources to help estimate a time series of territory-level
production data for fruits and vegetables, for example, canned and frozen processed
vegetables, potato production, noncitrus fruits, and citrus production? N/A

Industrial - Ethanol:

a) Ethanol production for each State was estimated using the Energy Information
Administration (EIA) SEDS dataset. Do you have any reason to believe that States'
information is underrepresented in the SEDS dataset? If so, do you have an
alternative, publicly available ethanol production dataset by State? AK accepts EIA
data.

b) Data for ethanol production for U.S. territories are limited in the SEDS dataset.
Are there resources to help estimate a time series of production data for ethanol
production? N/A

Industrial - Petroleum:

a) Petroleum production for each State was estimated using EIA's Petroleum
Administration for Defense Districts production and State-level operating capacity
datasets. Do you have any reason to believe that States' information is
underrepresented in the EIA datasets? If so, do you know of an alternative, publicly
available petroleum refining production dataset by State?

• The Department of Revenue likely has this information.

b) Do you have any concerns about using operating capacity to estimate petroleum
production by State is not a good method? If so, would you suggest an alternative
method? AK is ok with this.

c) Data for petroleum refining for U.S. territories are limited in the EIA dataset. Are there
resources to help estimate a time series of territory-level production data for petroleum
production?

Industrial - Breweries:

-------
a)	Brewery production, and by extension brewery production emissions, for each
State wasestimated using the Alcohol and Tobacco Tax and Trade Bureau (TTB)
taxable productiondataset.

•	The TTB dataset is based on taxable production/volume. Is there any reason
why taxableproduction from breweries may be underrepresented by State and
therefore potentiallyunderrepresent total emissions? NO If so, do you know of
an alternative dataset or assumption?

•	The TTB dataset provides production data from 2008 to the present. The
2008 values were used as a proxy for 1990-2007 values. Is there reason to
believe States' breweryproduction has significantly changed over that time
period? If so, do you have an alternative, publicly available State-level
brewery production dataset (i.e., barrels produced), or suggestions for
alternative data to use as a proxy?

No state-level data; TTB production data is reliable for AK.

b)	Data for brewery production for U.S. territories are limited in the TTB dataset.
Are thereresources to help estimate a time series of territory-level production
data for brewery production (i.e., barrels produced)?

-------
3.2 Arizona Department of Environmental Quality Air Quality Division
Technical Analysis Unit

Arizona's Response to General and Charge Questions on the Draft
MethodologyReport: Inventory of U.S Greenhouse Emissions and

Sinks by State

Prepared by Arizona Department of Environmental
Quality Air Quality Division
Technical Analysis Unit

General Questions

a. What are your overall impressions of the clarity of the methods described in this report?

The methodology and emissions estimate for Fossil Fuel Combustion sectors are
described clear, especially the description of why some fuel use reported in SEDS
may be different from reportingin the national inventory. The tables provided in
appendices are also helpful to understand the adjustments made to different sectors.
However, more bottom up data collection would be helpful.

R: For the Non-Energy Uses of Fossil Fuels, the emissions calculation method is
clearly laid out.The steps taken to reconciliate the disaggregated inventory on a state
level with emission on anational level was impressing.

For the cement production sector, the selection of emission factors is clear. However,
the improvement of modernized production methods needs to be considered. Also,
there are severalnew technologies that can reduce C02 emissions or even absorb
C02 emissions (such as CarbonCure). These credits should be considered for cement
production sector.

b. What recommendations do you have to add to or improve the overall transparency,
completeness, consistency and accuracy of this report?

R: In general, that would be helpful to include more details on the calculation of the
emissions foreach sector.

Control measures or any existing best practice aimed at reducing GHG emissions
should be considered in the emission estimates where possible. If the activity data
incorporates these measures, it should be specified in the report. For example, for
Electric Power sector, is it possibleto consider the heat rate improvement actions?

Many power plants have applied that to reduce emissions but it seems that these
improvements are not considered in this report.

Cement production includes several chemical processes. It will be more helpful to
include differentchemical reactions and list the C02 emissions from each chemical

-------
reaction process. Also, improvement technologies and control measures that are recently
used by the cement industry should be considered. Since some cement facilities have
installed CEMS, the C02 emission data calculated by the activity method can be
correlated and adjusted by CEMS results. Therefore, someCEMS adjustment methods
can be considered.

Section 2.6 of the methodology report describes that the total amount of distillate fuel
and motorgasoline used in Transportation sector was taken from the national inventory
(based on FHWA data). The totals are said to be based on multiple factors to determine
transportation sector fuel use. A more specific description of the mentioned factors
would be useful. It would also be useful to know how EPA plans to investigate why the
total fuel use on the FHWA forms used to allocatetransportation sector gasoline and
diesel fuel use across states doesn't match with the EIA statistics. More details
regarding how FHWA determines fuel consumption would be helpful as well.

The transportation section provided information on the carbon content of the fuel,
among other details. Some discussion of carbon content in MSW fossil
components/emission factors would behelpful.

Data availability. Please address the following questions for each inventory source:
a. For each of the categories, are there additional relevant data
sources that arenot currently included, but could be
incorporated into this analysis? Providing facility-level data
(such as C02 emissions reported to SLEIS from each facility)
would be useful.

R: Weights and Measures might be able to help provide local data on

transportation fuel consumption.

EPA's Facility Level Information on Green House gases Tool

(FLIGHT) database includes statewide C02 emissions for the year

2019.	The EPA FLIGHT databasewas newly updated on SEP. 26th,

2020.	htt ps://www.eia.gov/environment/emissions/state

b. For national level datasets that are currently used, are you
aware of other comparable datasets of activity, emission
factor, or emissions data that are available at the State,
county, or zip-code levels?

R: Statewide C02 data from the Energy Information Agency's (EIA) energy
related C02emissions database was updated to 2018 on Mar. 2nd, 2021.
https://ghgdata.epa.gov/ghgp/main.do

-------
d. Uncertainty. Currently uncertainty ranges are not included for the state level estimates.
Please provide feedback on what qualitative and quantitative information would be useful.
Timeseries Coverage. Currently State data covers 1990-2019 consistent with the 2021
National GHG Inventory, and inclusive of most known baseline periods for climate policy.
Subsequent publications of this data will also strive to maintain this consistency with the
National Inventory. As state-specific input datasets are not always available over the entire
timeseries, understanding which years may be more important can help us to better
prioritize our backcasting and methodological efforts across the time series. EPA
appreciates feedback on which, if any years should be prioritized for future State-level
estimates (e.g., 2000 and later, 2005 and later, 2010 and later, or the full time series).

R: It would be useful to investigate how climate policy has changed in the last
decade to see ifany changes have caused major trends in landfill emissions and
prioritize those years.

C02 emissions from Cement production in Arizona have increased after 2015 but
is much lower back in 2005 (-800 tons vs -300 tons). A dataset for after 2005 or
after 2010 would behelpful. Since the methods after 2010 are consistent,
prioritizing emissions data for 2010 or later is recommended.

For Transportation sector, El 5 fuel was first approved in 2012, so it might be
useful to prioritize emission data beyond this point. This will be helpful for future
projections and disaggregating data even further.

For Electric Power sector, emissions for 2010 and later should be prioritized for
Arizona because there was a significant increase followed by a decrease in
emissions, which is not thesame trend observed for other states.

For Non-Energy Use of Fossil Fuels, emissions in 2008 and later should be a
priority. With NEIdata starting from 2008, this GHG inventory can complement the
NEI database for some analyses.

For Stationary Combustion sector, emissions in 2005 and later should be prioritized
for Arizona. Since the peak of using coal and natural gas for stationary combustion
occurred in 2006 and started to decrease after 2010.

e. Key Category Analysis. EPA anticipates prioritizing methodological refinements for
moresignificant categories to make efficient use of available resources over time. EPA
appreciates feedback on which categories are more relevant for further refining for your
State.

a. Given that the emissions profile of some states will be different
from the national average, which categories that are more
significant in terms of absolute emissions,or have changing emission
trends (e.g., increasing, variable)?

-------
R: Based on the results of the GHG inventory and other
available data, thefollowing sectors in Arizona are
recommended to be prioritized for more refinement and KCA:

The C02 emissions from the Cement Production sector represents 20%
of the totalC02 emissions from the industrial processes sectors in
Arizona. It also shows a higher increase trend in the state. Therefore, a
KCA should be considered for this sector.

Based on the state-level GHG emissions estimation results, industrial
combustionsector is one of the major sources of C02 emissions in
Arizona. A KCA would be beneficial for this sector.

The emissions from the Electric Power sector has had a significant
change over time and has decreased in the past years. So, a KCA is
recommended for this sector. Also a KCA for each state would be
useful because energy structure for power plants are different
(coal/natural gas/nuclear/solar/other new energies).

The Stationary Combustion Sector for Arizona shows a significant
difference between national average level and state level emissions
inventory. So, a KCA would be needed.

Natural Gas Systems is the fourth largest GHG sector for AZ, so
some priorityfor further refinement may be appropriate.

A KCA within the mobile source sector could be useful as well.

b. The national Inventory includes a key category analysis (KCA)
consistent with 2006IPCC Guidelines. Would it be useful for States
if a key category analysis (KCA) was completed for each state?

R: Yes, a KCA for states would be useful.

f. Data Presentation and Usability.

a. Are there other ways the state-level emissions data could be
presented to facilitatetheir use (e.g., in the EPA GHG Inventory
Data Explorer available online at:
https://cfpub.epa.gov/ghgdata/inventoryexplorer/)?

i.	Related to the level of category/gas aggregation or
disaggregation?

ii.	Are there specific categories where further data
disaggregation could be helpful?

-------
R: Disaggregation to facility level for the Cement Production sector
would be helpful because there are couple large cement production
facilities in Arizona.

For the Transportation Sector, further data disaggregation in terms of
vehicle ageis helpful when looking at light-duty vehicles emissions.
Breaking vehicles out by Tier type (Tier II or Tier III) illustrates how
low emission rates are for newer vehicles.

For the Electric Power sector, more data disaggregation and providing
facility-level data are useful.

More data disaggregation for the Stationary Combustion sector will
be useful too. If it is too much work to provide facility-level data, a
county or non- attainment level data will be helpful.

b.	What data format would best facilitate the use of the state-level emissions
data (e.g., .xlsx download, etc.)?

R: Excel download would be ideal for data format. Text and CVS downloads
and FTP sitewill be useful too.

Monthly data disaggregation could be useful (if possible), especially for sectors
with seasonal variability. The ability to browse the data by state, year, fuel,
sector, facility level could be useful.

c.	What additional datasets or information could be provided to
help increase theusability of the state-level emissions data?

R: Facility-level data and location information would be useful.

d. EPA plans to provide users additional information on where they
can find official State data, where it exists. Do you have suggestions
on how we should direct usersto official state data (i.e., section in
methods report including links to State data)?

R: Similar to how references are mentioned at the end of each section, a
separate section containing all necessary links would be useful (perhaps in
a table with shortdescriptions of link content). Disseminating information
via work groups EPA is involved with is also helpful. Sending out emails
via official EPA electronic mailing lists works as well.

It would be very helpful if EPA can provide the details of emissions
calculations methods.

-------
Energy - Combustion Emissions Questions
Fossil Fuel Combustion:

a)	Some fuels have differences in consumption data between the aggregated State-
level totals and national totals. The current approach is to use data from the
national Inventory in those cases. Are there other approaches that could be
taken? Has this come up in developing estimates at the State level?

R: For most of the sectors, this approach sounds appropriate and accurate enough.
However,for some sectors such as Electric Power, a bottom-up method to collect
activity data or CEMSdata from the facility would be more accurate. Applying this
method may not be easy for the Stationary Combustion sector but it can be
considered an alternative way to collect data.

Arizona has not developed the estimates at the State level for the sectors mentioned
above.

b)	Consistent with the IPCC Guidelines, we have adjusted fuel consumption totals
in the energy sector to account for consumption in the IPPU sector. In some
cases, this step could lead to a negative emission total for a State if the
subtracted amount (as determined from the assumed distribution) was greater
than consumption data from the State Energy Data System (SEDS). This
outcome was corrected to zero if that was the case, but are there other
approaches for correcting for that difference?

R: The approach is generally reasonable. However, whenever a negative value is
increased to 0, probably that difference should be deducted somewhere else. Maybe
the total of the negative values should be tabulated, converted to zero, and then all
state totals should be scaled down uniformly by the appropriate amount.

c)	Consistent with the national Inventory, the default approach taken here was to
allocate transportation sector CO2 emissions based on FHWA fuel use/sales by
State. For some States, this may not be accurate because fuel sold in a State may
be combusted in other States. Anotheroption is to use vehicle miles traveled
(VMT) data by State but that approach does not factor in vehicle fuel economy.
Are there other alternative or complementary approaches to allocate
transportation fuel across States, including VMT data and other sources (e.g.,
NEI - based on county-level fleet and activity data to generate a bottom-up
inventory) that EPA should consider? If so, what data sources exist to help with
that alternative approach? Would it behelpful to present transportation sector
emissions using multiple approaches in future inventories?

R: A bottom-up inventory using NEI activity data would help mitigate the
possibility of allocating C02 emissions based on fuel sales that may not directly

-------
translate to fuel usage by State. It would be helpful to present transportation
sector emissions using multiple approaches.

d)	Mobile source non-CCh emissions are allocated across States based on
vehicle-miles-traveleddata while mobile source CO2 emissions were
allocated based on fuel sales, as mentioned above. Do you have any concerns
with using two different methodologies for mobile source CChvs. non-CCh
State splits?

R: As long as the fuel type and quantity of fuel burned is noted, we don't think there
is aconcern with using two different methodologies.

e)	Several fuels have variable C factors over time including coal, natural gas,
gasoline, and diesel fuel. Those fuels might also have variable C factors
across areas/States. Are data available to build out State-specific C factors
for the fuels with variable C contents? If so, could it be done in a way that
the State-level total emissions still matched up to the nationaltotal emissions
for those fuels?

R: At this time, we are not aware of any data available to build out State-
specific C factorsfor the fuels with variable C content.

f) Geothermal emissions could be allocated by the type of geothermal
production per State (because different types have different emissions
factors) if that data is available. Is there more information on State-level
geothermal emission factors and production?

R: We are not aware of any information on state-level geothermal emission factors.
Non-Energy Uses (NEU) of Fossil Fuels:

a) For petrochemical feedstocks, non-energy use (NEU) of natural gas is allocated
across States based on petrochemicals emissions data per State from the IPPU
adjustments, while other fuelsare allocated based on the underlying SEDS
data. Allocating across States based on the underlying SEDS data ensures
there are no States where NEU use is larger than original SEDS data and there
are no zeros associated with subtracting NEU (it is not an issue for natural gas
because use is so high overall compared with NEU use). Could different
approaches be used or can the petrochemical data be used without resulting in
negative use?

R: Different approaches to allocating NEU across states should be considered and
compared tothe current method of using a combination of data from IPPU

-------
adjustments and SEDS. From theperspective of NEU of fossil fuels in Arizona, the
current methodology could be sufficient considering the lack of use for most fuels.

Incineration of Waste:

a) Waste incineration emissions are calculated based on the combustion of
fossil components of both municipal solid waste (MSW) and tires. However,
emissions are disaggregated to Sates based only on MSW tonnage. Are
there approaches or data available to disaggregateemissions based on
waste category (e.g., MSW combustion vs. tire combustion)?

R: The Biocycle, EIA, and EPA GHGRP data for Arizona indicates MSW
incineration is negligible, and EPA's FLIGHT tool has zero emissions for
Arizona. Therefore, this question doesn't apply to Arizona.

International Bunker Fuels:

a) The approach used to allocate jet fuel bunker fuels by State is currently based
on the total amount of jet fuel used by State which could potentially lead to an
over- or under-estimationfor some States of bunker fuel emissions. Are there
other more accurate approaches to allocate jet fuel bunker data across States
as opposed to the percentage of jet fuel total use? For example, using Federal
Aviation Administration flight level data on departures and destinations or
assuming based on States with international airports and flights?

R: At this time, we are not aware of other approaches to allocate jet fuel bunker data
across states.

Energy - Fugitive Emissions Questions
Coal Mining:

a) Do you have any comments specific to the methodology and emission
estimates for active coalmines and abandoned coal mines?

R: The methods described seem appropriate.

b) Are you aware of any State datasets that may be useful in helping to refine
emission estimates for abandoned coal mines, including State-level datasets
addressing recovery of methane from abandoned mines?

-------
R: A comprehensive inventory of abandoned coal mines in Arizona does not currently
exist.

Petroleum Systems and Natural Gas Systems:

a)	Are there relevant dataset(s) that could be used to replace or supplement the
data currentlyused to allocate petroleum and natural gas system emissions to
the state level? Particularly, state or detailed location information on
gathering and boosting stations, processing plants, and transmission and
storage stations?

R: The current data set appears to accurately reflect permitted natural gas
compressorstations in Arizona.

b)	Are there additional Greenhouse Gas Reporting Program (GHGRP) data that
could be used to allocate natural gas and petroleum emissions to each State?

R: No.

c)	Are you aware of any State datasets that may be useful in helping to refine
emission estimates for abandoned wells, including State-level datasets
addressing plugging status ofabandoned wells?

R: No state dataset of abandoned wells currently exists.

d)	Are there particular sources for which State-level regulatory or voluntary
programs resultin large differences in emission rates between states? Are
state-specific data sets availablefor those sources?

R: No.

IPPU - Minerals Emissions Questions
Cement Production:

a) Are you aware of data on clinker production by all States for the full 1990-
2019 time series? Please share a surrogate data that could be used (e.g.,
facility production capacity,utilization rates by facility or State) for 1990-
2019 that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends.

R: We are not aware of such dataset but we have the Arizona clinker
production data fromSLEIS since 2010, which is probably useful for creating
state inventory.

Waste - Solid Waste Disposal and Management Emissions Questions
MSW Landfills:

-------
a)	Data Questions:

•	Are there datasets for individual States' landfill gas (LFG) recovery
activity?

•	Are there data available for open dumpsites in the U.S. territories?

R: There are datasets for MSW landfills methane recovery estimates by state in
Appendix FTable F-3. There is no data on open dumpsites. However, section
6.1.1.4 mentions the inclusion of emissions from all waste management practice,
including open dumpsites, as apotential refinement to landfill estimation methods.

b)	The current method makes some simplifying assumptions and includes
uncertainties in theallocation of national-level U.S. emissions to States (e.g.,
recovery rates are the same for all States and match the national recovery
rate). Are there alternative assumptions or different datasets that would
improve the accuracy of MSW landfill estimates? Do you have
recommendations to refine the methodology to estimate emissions over the
time series more accurately?

R: No recommendations to refine the methodology at this time.

Industrial Landfills:

a) Do you have recommendations to refine the methodology to estimate emissions
over thetimeseries more accurately?

R: No recommendations to refine the methodology at this time.

Composting:

a) Data Questions

•	Is it correct to assume that Alaska has no commercial composting
operations correct?

•	Are there any datasets about composting in U.S. territories?

•	Are there any State-level data sources that describe composting activity
over time?

R: Due to the large area and quantities of fish waste, it may not be correct to
assume thatthere is no commercial composting in Alaska. We are not aware of
any State-level data sources that describe composting activity over time.

Stand-Alone Anaerobic Digestion:

a) Do you have or know of any State-level data for counts of

operational anaerobicdigesters (processing food waste) by year?

-------
R: Other than publicly available data on EPA's website, we are not aware
of any otherstate-level counts for operational anaerobic digesters.

b) Are there any facility-specific data sources we could use to fill data gaps on
the quantity ofwaste processed by stand-alone digesters for any and all
years of the 1990-2019 time series?

R: There are sources that break down digester type for specific estimates such
as biogasdata, but we could not find any facility-specific data sources for
quantity of waste processed by stand-alone digesters.

-------
Iowa Department of Natural Resources

iowa Department of Natural resources govi***™ Km r^moids

It. GovtONOD Adam Gregg

L J " DIRECTOR KAV1A LVOM

October 27, 2021

Greenhouse Gas (GHGJ Inventory Staff
US EPA

G HGI n ventorv@e pa. a o v
Dear EPA GHG Inventory Staff,

Thank you for this opportunity to review and comment on EPA's draft inventory of U.S. Emissions and Sinks by
State (referred to in this comment letter as "EPA's disaggregated Iowa emissions or values"). This review process
has helped the Iowa Department of Natural Resources (DNR) identify several possible gaps in its inventory as
well as identify several categories where EPA's results and DNR's results differ significantly. The DNR also
appreciated the pre-meeting with Andrea Denny and Mausami Desai on July 2,2021, and DNR looks forward to
discussing these comments further. We have also greatly benefitted from the assistance of Andrea, Mausami,
and Kong Chiu over the past decade as DNR has developed its GHG inventory program.

I was the GHG inventory lead for the 2010 - 2018 DNR inventories, but I have been transitioning the inventory
over to Krysti Mostert starting with the 2019 inventory. Please send any responses, questions, or comments to
both Krysti (krysti. mostertf5)dnrjowa.eov) and me.

Sincerely,

(k/uj SfrCM^

Mamie Stein
Air Quality Bureau

Supervisor - Operating Permits & Emissions inventory
515-725-9525

marnie.stelnffldnr. iowa.gov

Attachments:

Iowa Inventory Comparison I0-27.202lj
-------
DNR's Annual Statewide GHG Inventory for Iowa

The Iowa DNR has prepared a full top-down statewide emissions inventory annually starting with emissions year
2010 as required by Iowa Code 4SSB.1Q4. The inventory reports and technical support documents are posted at
https://wwW-iowadnr.gov/Environmental-Protection/Air-Qualitv/Greenhouse-Gas-Emis.sions. and DNR is happy
to provide our SIT module files upon request. For these comments, DNR staff compared its calculated emissions
for 2015 - 2019 as published in the 2019 GHG Inventory Report & Technical Support Document (TSD) to the
disaggregated Iowa emissions provided by EPA in the file "State GHG_Trends_Efni£Sions_&_Stnks_By_Gas". This
comparison is provided in the attached spreadsheet titled "Iowa Inventory Comparison 10.27.2021Because
DNR and EPA do not use the exact same sector names in their inventories, DNR has matched its data to the EPA
sectors as closely as possible. Please note that DNR was not able to review the EPA's Draft Methodology Report
in its entirety or answer all of the charge questions due to limited staff resources. Instead, DNR is providing
comments on sectors with the biggest differences in emissions or sectors that are potential gaps In DNR's
inventory.

Method

DNR calculates annual GHG emissions using the most recent version of the EPA's State Greenhouse Gas
Inventory Tool (SIT) and updating it with any available lowa-specific activity data. The energy and industrial
processes sectors are also supplemented with GHG emissions data submitted by individual Iowa facilities to the
federal GHG reporting program (GHGRP). The versions of the SIT modules used and their corresponding
chapters in the TSD are listed in Table 1. The coal module was not used, as no coal mines currently operate In
Iowa. The DNR uses the global warming potentials from the Intergovernmental Panel on Climate Change's (IPCC)
Fourth Assessment Report (AR4).

Table 1: DNR 2019 TSD Chapters and Corresponding SIT Modules

TSD Chapter

SIT Module

Release Date

Agriculture

10/06/20

Energy

COjFFC

11/05/18

Stationary Combustion

11/05/18

Industrial Processes

10/06/20

Natural Gas Transmission & Distribution

Natural Gas and OH

11/05/18

Transportation

Mobile Combustion

10/06/20

Waste

Solid Waste

11/05/18

Wastewater

11/05/18

Land Use, Land Use Change, and Forestry

LULUCF

10/06/20

It is worth noting that the Iowa Code requires that "by December 31 of each year, the department shall submit a
report to the governor and the general assembly regarding the greenhouse gas (GHG) emissions in the state
during the previous calendar year..." This makes DNR's Inventory out-of-sync with EPA's national inventory and
the SIT because they both calculate emissions For two years before the current year. This can lead to differences
In emissions because DNR may have been Forced to use an earlier year as a proxy for the previous calendar year,
Forecast emissions for the previous calendar year, or may be use more recent activity data than EPA.

-------
OMR Comments and Response to EPA's Charge Questions

General OuattiBfti

DM* is concerned about how to explain the differences between EPA's disaggregated Iowa values and OMR's
annual emissions Inventories. The Inventory method is wy complex and not easy for public and other data
users to understand. Overall, EPA has far mors resources than ONR does and Dili is using EPA's method to
create the annual inventory- OMR is doing the best it an with less than ISO hours devoted to the inventory per
year and it may be burdensome for OMR to respond to questions.

DNR recommentls that EPA carefully consider the wording of Its announcement of publicaif releasing the data.
DNft also encourages EPA. to include information on the webpage housing the dala that helps to clarity the data
that EPA Is posting, EPA should also inctytfi a Unite to tow.; ' -j r r ihvO'-.i \

ips in its inventory. 1

in blue font on the attached torn Inventory Comparison i0.27.2021 spreadsheet

I in Table 2 below and

Ijbit: I: POisibU: .iwiturv Gaps

S*±k.to'f

NOn'tfleigy Use Or rUtlS

Natural Gas Systems

CO*. NjO

Glass Production

CO,

Carbon Dioxide Consumption

COi

Urea tamumpflaa for Non-Agricuttural Purposes

CO*

Petrochemical Production

CO,

Cat tide Production and Goman&m

COi

intentaUQMl Bunker fuels

Composting

CI!,, N.,0

Anaerobic Digestion at Stof as Facfflttts

CH*

NjO from Product Use*

M,0

iOJfJOM spreadsheet,

*,iblc i Vr' U.M 'Ahrre DNf( J» t PA 1 V i iV'. Ui'Km Su,"nlu .tnlly

aetiftf

PolU;t:im(sl

wn?ni

v,v;

tiirsr P'oiiin lira*

CO*

C»>H ( l>H\ 1'-,, 11 i» ..1 i ,Hl»nh,ltC

Cn.

lnelfiii.il«-i..i v.m

i ' I • 1, N i, >

Ammei >'ni
-------
Mobile Combustion

til.. N.O

Natural 6a> Systems

Enteric Fermentation

CM;

Manure Management

N;0

Wastewater Treatment

CHi, N rj

Stationary Combustion

M.U

Electronics Industry

P IPCs

Electrical Transmission »').| 0 '=tnuj' on

LULU a

i, -1

Ewatgy ~ Combustion fcwilsaltttis Questions
i i on m,-it.

Dttft's estimate for fossil fuel combustion from the industrial sector are sigmitomttf hijjjter than EPA'i
disaggregated towa wall* (wanes from 2954 . 56* from 2015 - 20JSJ. It is likely that OMR's inventory teclutles
emissions in this sector that may be double-counted in the IPPlt sectors for the production of cement, lime, iron
& steel, ammonia, etc.. it appears that in the national inventory, IPA reallocated portions of fuel eonsumftttort
data for several fuel categories leaking coal, other coal, natural gas, residual fuel, and distillate fuel) to IPPU, as
these portions were consumed as raw materials Awing non-energy related industrial processes. This is definitely
am mm of future Improvement for Dili's inventory.

Fossil l;uel Combustion - F'r asportation:

DMft calculates highway transportation emissions using the SIT mobile combustion module, n-or tm and ffaQ,
DM* uses the actual total annual VMT provided from the towa Oepartment of Transportation (DOT) • VMT BY

The towa OOT does not ha* vehicle miles traveled (VMT) data by
model pais, so DNR allocates Iowa VMT wing the default national on-road dfeWbutton by vehicle/fuel type.
DM* also usb tli« annual vehicle mite accumulation, ate distribution, and control technology values from thf
most recent national Inventory, DNR calculates CO* emissions from highway vehicles and all mMon from non«
highway whiclfs using fuel sates data from either S€0$» FHMM, m the SIT default value. This hybrid calculation
method results In Iowa emissions that are lower than the disaggregated ie»§ virtues provided by IM ICOi »f*»
t,irt« -M, NjO -14* foe 2019), and may have a higher level of uncertainty because Of* Is not using the tint
vehicle/fuel distribution tor Iowa.

Regarding CPA's question* about non-CO» emissions toeing caiculateil by allocated VMT white C02 emissions are
based on fuel salts, OKI has no specific concerns p#r se as tt seems to hi the test method available. However, it
would be Interesting to see the difference between COj emissions based on VMT versus fuel sate.

Regarding IWi questions about state-specific carbon factors and geothermal emission factors, Oii it not
aware of §my data sources currently available to develop tbese factors.

¦ : '1' .v..'' ¦ 1 •¦:¦¦¦¦ ¦¦¦' i' :

This is i potential pp In Bit's Inventory as n his not been previously Included. (t appears this i calculation

method for this, sector Is not included in any of the iff modules. Will ti»A oe adoing this sector to the SIT?

-------
Incineration of Waste:

DNR calculates the amount of CH* emitted from power plants burning municipal solid waste (MSW) to produce
electricity using data reported annually by individual facilities to the DNR's Air Quality Bureau on their annual
Title V air emissions inventories. The only facility in Iowa generating electricity from burning MSW reported
burning a total of 12,763 tons of refuse derived waste in 2019. The DNR uses state-specific proportions of
discards that are plastics, synthetic rubber, and synthetic fibers instead of SIT default values to calculate CCh
emissions from MSW combustion using SIT. These state-specific proportion values are from the 2017 Iowa
Statewide Waste Characterization Study. A previous study was done in 2011. The state-specific proportions of
discards used are shown in Table 4.

Table 4: Proportions of Discards used by DNR

Material

SIT Default Value

2011 lowa Study

2017 lowa Study

Plastics

17.0 - 18.0%

16.7%

18.3%

Synthetic Rubber

2.3 - 2.6%

1.0%

1.2%

Synthetic Fibers

S.6 - 6.3%

4.1%

4.S%

There may be a gap in DNR's inventory as DNR has not previously considered emissions from other incinerators
in the state, nor has DNR Included site-specific emissions from tire combustion as a separate value. It is unclear
If this activity data is easily accessible. The calculation methods used by DNR as described above result in quite
different emissions than EPA's disaggregated Iowa values. As shown in the attached spreadsheet, for 2019,
DNR's results are COi +14%, Cm +9625%, and NjO -3%.

International Bunker Fuels:

The DNR uses the total amount of jet fuel used by the state per SEDS and the SIT module and reports emissions
in the transportation sector. No International flights depart from lowa municipal airports.

Energy-fugitive Emissions Questions

Coal Mining:

lowa does not have any active coal mines, so this sector is not Included in the DNR's inventory.

Petroleum Systems and Natural Gas Systems:

lowa does not have natural gas production or petroleum systems, so there Is not GMGRP data available for this
category, lowa has four LNG storage compressor stations, eighteen gas transmission compressor stations, and
four gas compressor stations. The SIT natural gas and oil system module only calculates CM< emissions for lowa
from transportation and distribution. However, EPA's disaggregated lowa values show COj and N.>0 emissions
from this sector, likely from flaring, even though natural gas production or petroleum systems are not present In
lowa. In addition, the SIT does not include any activity data for flaring In lowa, so DNR is uncertain as to why EPA
shows C0j and NjO emissions for this sector in Its disaggregated lowa emissions, it Is also interesting that DNR
calculates CH4 emissions using the SIT that are higher (+18% in 2019) than EPA's disaggregated lowa values,

IPPU - Minerals Emissions Questions

Cement Production:

lowa has two Portland cement manufacturing facilities currently operating, Both facilities are required to report
their COj emissions to the GMGRP, so DNR uses this 6HGRP data In Its inventory Instead of using clinker

-------
production data calculate om.ssiem. As shown in tt-c attache:! sor^adiicc:, the ccrvbinec CO: sessions

rracned &y The t«& facilities tor 2019 v»ere 72 .1 h.ghsrfian CPA's d s.ifgrc^ated icwa va'uc.

Lime Production:

Iowa his or»e Him production facility currently operating, and it is required to report its CO, emissions to the
GH6RP. DNR uses this GHGRP data in its inventory instead of using lime production data. As shown in the
attached spreadsheet, the CQj emissions reported by the facility for 2019 mm iSX higher than EPA's
disaggregated Iowa wall*.

Class Production;

ON ft does not identify emissions from glass production separately in its Inventory. Instead, DNR orientates the
total emissions from limestone andl dolomite use tor industrial consumption per the SIT Mistrial Processes
Module, OMR Is not aware of any state-level activity data that may enhance the state inventory calculation.

Other Process Uses of Carbonates:

OMR calculates the total emissions from limestone and dolomiti use for industrial consumption per the SU
Industrial Processes Module. One area of enhancement may be the use of soda ash by the corn wet milling
industry. In Iowa, this industry commonly uses soda ash fctr pH control, ion exchange regeneration, and other
operations. DNR may I* able to survey is corn wet milling facilities to obtain the actual annual amount of soda
ash used by the Industry.

CO., Consumption:

This * a potential««lor enhancement in 011*'$ inventory as ft is not currently incfcidod. CO* It i* well in the
food i beverage Industry, and Iowa m#it*pacters use significant amounts of COj to stun animals (poultry, pip,
cattle, etc.) before the* are miled. Typically, meat'packers purchase biogenic CO, from ethanol plants. DNR may
be able to mrv#y the industry to determine the amount of COi from ethanol production that is ustd In these
Industries.

~P(>U thomitalsttnijrtonaflmttllctn

Arnmofsta Production:

Three facilities I n km* currently produce ammonia and report 1,1*1 r C0» emissions to the GHG RP, so OMR uses

this GHGRP data In IK inventory. M shown In lit* attached spreadsheet, tht total CO:? emissions re ported by the
two faculties for 2019 are fI* higher thin IPA's disaggregated lows data.

lows farmers land-apply § targe amount of ammonia to agricultural land e*cti 'par. Should DM* be subtracting
the land*apptled ammonia from ft# ammonia production emissions?

i r.-u I'-i'ji- i.,.; 'J ! i. i!:': ¦

This Is a potential gap in WW's Inventory as It has not been previously inciuoed. n apoears this • calculation
method for this sector Is net included In wy of the SIT modules. WUIPA bf adding this sector to the SIT?

!, ;¦ i; J:ii1 ¦. :'¦¦¦¦

Two nitric add productionfacilities currently In low®. TNry are raqMrtd to report their NaO emissions lo tl*
GHGRP, and DM* usesthis data in lit Inventory. At shewn in the attached sprtad'Sftiw. me DNR's inventory and

-------
EPA's disaggregated Iowa data agree within 0.13% for 2019.

Adiptc Acid Production; Caprolactarn, Glyoxal, and Glyoxyltc Acid Production; Titanium Dioxide ITiOjj
Production, Phosphoric Acid Production; HOC-22 Production;

These sectors are not operating in Iowa and art riot included In either OMR's Inventory or EPA's disaggregated

Carbide Production and Consumption:

This sector is not in OMR's inventory as it Is not inducted In the SIT industrial Ptoamsm module. DM* wH further
research If this is a gap in its inventory and If Iowa activity data Is available.

Petrochemical Production:

This sector is not in OMR's inventory as It is not Included in the SIT I ndustiM Processes module. EMM ™. ,»*her
research if this Is a gap in its inventory and if Iowa activity data is available.

IPPU- Metals Lmrssions Questions

Iron arid Steel and Metallurgical Coke Production:

There art currently no metallurgical coke production facilities or pig iron mills operating In Iowa. Alt' three steel
production facilities currently operating In lows use electric arc furnaces to produce steel from scrap. 0*1 uses
the CO* emissions they report to the GHGRP in its inventory. As shown in the attached spreadsheet, the COi
emissions In OMR's Inventory lot 201S are m higher than fpft's disaggregated Iowa value.

ferroalloy Production, Aluminum Production, Magnesium Production & Processing, lead Production.

CI: C C .1,1 h .(I.

These sectors we not present in Iowa and are not included in either OMft's inventory or disaggregated
iowa data-

M'Pij -- i?(u.duU \J-iV I

The 201? Economic Census identifies eleven businesses in iowa tinder the North American Industry Classification
System pftiCS} for code 13441 - Semiconductor an J Other Electronic ManuiactMrinf |UJ>. census 2011), but
liVs disaggregated Iowa values tie not Include data tor iowa. Emissions m 2019 from semiconductor
manufacturing: were calculated Df* by assuming thai towa emissions were 0.96% of national emissions as
lowa's population is OJfi* of the total U.S. population

¦¦¦l. . 11 ; 1 • , ,

The .1818 emissions calculated by OM are within i* of !W'i dtsanregated tome values. However, the ONi
uses the SIT, which reports emissions to units of COj« and does not identify NfCs and PFCs separately as Wi

Bectr icjI T> jncmiMiofw and Distribution

-------
for 2019 is 24% lower than tPA's disaggregated Iowa value, but this is likely because 2019 data was not available
at the time that DNR completed its inventory, and thus used 2018 is a proxy,

NjO from Product Use:

This sector is not in OMR's inventory as it is not Included in the SIT Industrial Processes module. Wilt EPA be
adding it to the S1TT? OMR will further research if this is a gap in its inventory and if Iowa activity data is awaRaMe.

Agrfcultura - llwtftacfc EmtataM Quwtkxa
Enteric I ermentation:

DUR calculates enteric emissions using the SIT. It appears that EPA's disaggregated state values and the SH both
use emission factors from the same source, CEFM, and the most recent animal population data from either
USDA's MASS Quick Stats or USDA's Iowa Agricultural Summary. Because OMR's Inventory is for tt* previous
calendar fear and Em's national inventory is for tmo fears before that, OMR mm! If* are not using the same
animal papulations. 'This may lead to DMi's emissions being higher than EPA's (+18* for 2019) as shown on the
attached spreadsheet.

Manure Management;

Dili calculates manure management emissions using the SIT and the same animal population sources as for
enteric fermentation. Again, DNR uses mom current animal populations thai EPA, likely leading to umts
emissions being sJighfly higher than CPA's f+4% for 2019). Oddly, OMR's calculated emissions for 101? - 2019
are SOU ¦ im higher than EPA's disaggregated state values. DNR will farther research this discrepancy and
will advise EPA of our findings.

Wet Is not cultivated in lowa.

Urea I-ertih.#atiori;

As the attached spreadsheet shows, OMR's nttmM emissions from urea fiirtiiiattioii art significantly lower
than !Wi disaggregated lowa values for HU (*24%) and 20i§ (-J&*| but are higher than ftt'j
disaggregated towa values for 2015 (+7*1,1016 (+20*1, and 201? ONR believes this corresponds to a
change In the availability of state-specific ftrtliitr data. Wer to 2®II» the Iowa Department of land and
Agriculture Stewardship published a report of fertilizer sites that DNR used for Its Inventory. The report was
not available tor 2018 or 20IS, so ONI used the amount of 2018 una applied from the USDA's towa
Agricultural Statistics Bulletin for tmtfi 20S8 and 2019.

DNR calculates CO} emissions from liming using the total amount of limestone COj Is emitted when icicle
agricultural soils .are neutralited by adding limestone or dolomite. DNR used the total annual amount of
Imestone produced foe agricultural ust m reported by their members to the towa limestone Producers
Association, However, producers do not report the percentage of Kmettone that Is dotomMc The towa dot
tracks funeral information for active aggregate sources used for construction, including whether the material
Is Hint stent or dolomite, They do not track that information for limestone produced foe agricultural purposes.
The Iowa DOT indicated that torn areas of the sine have WO* dolomite, some have i®0% limestone, and

-------
son* areas are mixed. Therefore, the DM Ft assumed that 50K of the material produced In Iowa for % m

use "is dolomite and 50% is limestone. As shown in the attached spreadsheet, DNIfi emissions are quite
different than EPA's disaggregated Iowa values and EPA does not provide a value for 2011 or 2019.

held Burning of Agricultural Residues:

Burning of agricultural fields is not a common practice in Iowa.

Agrlwltuw - Agteultoril Sol M»niwaiwt MmJohh OwmUmw

Emissions from a|rlcuitural soil management In 2013 were 20.97 MMtC02e, accounting for i§* of Iowa's total
GHG emissions in OMR's Inventory. DNR looks forwari to seeing EPA's disaggregated town values when they
become available.

WitiOF

OMR's estimates of cartoon sequestration to the LilLOCF sector are significantly higher than EPA's itsaggmsgmed
state values (tetww +?§* to ~USX) as shown in the attached spreadsheet. MHt is unsure what is driving this
difference as ut»;R uses the SIT to calculate the amount of carton sequestered and uses the SIT default tabes for
forest carbon flux, urban forests, and agricultural soils. In addition, the SIT seems to use the same method is
EPA's national inventory.

The DNft used emissions reported by MSW lariciifls to the GHGRP, which are calculated taseef on the
characteristics of such individual report EPA requires MSW landfills that emit 25,000 metric tons COie or mere
to report their emissions. This included twenty-four Iowa landfills in 2019. An additional twenty-two Nww MSW
landfills wert not required to- report to the GHGRP. To calculate emissions for those that
-------
state-specific actMty data to calculate emissions from several sources,

Municipal:

DMS. calculates municipal wastewater emissions using' the SIT wastewater module along with the tows fraction
of population without septic systems, 76%, from EPA's Cinsite Wastewater Systems Treatment Manual
Unfartunat«lvr the fraction has not been upiited by EPA since 2002, and the fraction value has not been
included in the U.S. Census of Housing since 1990. DNR also uses the most recent protein value (kg/person/year)
from the national Inwntory.

industrial;

OMR calculates industrial wastewater emissions using the emission reported by facilities to M's SHGRP. In
2019, eleven cthanof plants and ft* fowl processors reported emissions to the GHGRP. MM believes this to be
more accurate than calculating emissions based on red meat or ethanol production values. DNR has been
unable id find state-specific data to calculate emissions from pulp I paper, fruits 4 vefieiaIsles, or breweries,
anil the SII wastewater module dots not Include default activity data for lowi.

-------
.4 Maine Department of Environmental Protection

STATE OF MAINE

Department of Environmental Protection

JANET T. MILLS
GCWEftNQfl

MC.LAMC LOY2IM
COMMISSIONER

November 1, 2021

U.S. Environmental Protection Agency
GHGInventory@epa.gov

Re: Charge to State GHG Inventory Experts for the Draft Methodology Report.1 Inventory of US.
Greenhouse Emissions and Sinks by State

Dear Greenhouse Gas Inventory Team:

The Maine Department of Environmental Protection (Department) appreciates the opportunity to
provide feedback on both the Draft Methodology Report: Inventory of U.S. Greenhouse Gas Emissions
and Sinks by State, which describes the disaggregation of the national greenhouse gas (GHG) inventory
across the SO states, and the resulting state data.

While the Department was unable to address the 100+ questions listed in the 14-page Charge outline,
the Department would like to express qualitative overarching concerns with the proposed release of the
disaggregated dataset.

The Department has three primary concerns:

1. Data is inconsistent between various EPA GHG datasets;

2. Methods are inconsistent between the National GHG Inventory, the disaggregated state GHG
Inventories, and the State Inventory Tool (SIT); and

3. Stakeholder confusion and associated decreased confidence are likely to result from comparison
of EPA*developed state GHG inventories and SIT based state-developed GHG inventories.

Data consistency is critical for confidence in a dataset and any strategies or policies that are based on
that foundational data. Without consistency, stakeholders will question which dataset is accurate.
Unfortunately, the current SIT default data differ significantly from the draft state disaggregated data
under review. While the reasons for variability between the national inventory and the disaggregated
state data are described In the Methodology Report, the differences between the disaggregated data
and the SIT default data are not defined, which is concerning since these are the two datasets likely to
be compared as many states currently rely on the SIT for development of state GHG inventories.

Although EPA has stated they plan to ensure users understand that the EPA disaggregated state datasets
should not be viewed as official data of any state government, stakeholders will almost certainly
compare the datasets, and state governments will be faced with numerous questions about the
differences in both the data and methodologies. In Maine, these stakeholders already compare SIT
default data with the Maine-produced GHG inventory. Currently, these data and methodology
questions are relatively easily answered. The Department purposefully augments the SIT when more

Al «.l\TA

I'SIVU (IIHM STATION
AlHtl VI A, .MAIM! IM.UMIIH?
(HIT) 2H?-?4IM FAX <»lf) 1H1.7M6

BANtK >K

111, II* KIAN IMIAtl, sl'ITI'. IV
H'\N4«OK, M VIM 1*44111
<30T) 'Ml I AX <2ilT) "Ml

IPOHTILANt)

UJ t AM i > R<>AI>

I'OKTl.ANI). \lAIM iH 1(1.1
(2l>7) H22-Althi MX; (2l>?) N22-6ll»1

IWlMjUli I M.I.

12** i INTRA! t) It J VI1., W PARK
Pftl NQl I IM I . MAIN!

(2ll?) ?A4 4»4?* PAX; <2ll?)

wi.li.if. n\4\*v u»»v yjkp

-------
accurate state-specific data is available. These changes are tracked and are transparent in the revised
SIT modules, allowing such improvements to be clearly explained to stakeholders, Hi contrast, the
specific methodology and input data differences between the SIT and the disaggregated tiataset are not
clear

One of the primary benefits of the SIT is full transparency. Input data and data sources are dear in each
module, and the input data is easily augmented when more accurate state-spedfic data is available. Ike
proposed disaiggregated state datasets do not have the necessary level of transparency- While some
data are available »the appendices to the Methodology Re-f»»t. not input data are accessible, and
for data that are, where and exactly how Owe data are usei to calculate the final data is very difficult
to determine or understand, even with the explanations in the Methodology Report,

The inconsistencies In metNKtotogy ami data are numerous, and such inconsistencies wit inevitably lead
to stakeholder confusion and questions that cannot be readily answered. The Department recommends
EPA hold the release of the disaggregated state data until the SIT default data far each stile tetter
matches the slate data disaggregated from the EPA National GHG Inventory. Stales rely heawiff on the
Si i as a foundation to produce state GHG inventories, and regular updates at them GHG inventories are
important for tracking emissions trends as well as far strategy and policy development For these
reasons, ll* Department recommends the annual release of the SIT not be oetaptf and that the
dijajsgregattii data net be mleased until EPA more closely aH§« these two datasets.

Sincerely,

Slaty R *napp

Emissions iwwtti* Section Manager, Bureau of Air Quality

(207} 2«7> 2235

i»»t» J si 1

-------
Minnesota tion Control Agency

MINNESOTA PCUUTION
111 CONTROL AGfcMO

520 LffaprftE Road North f St. Pan*. 55ISS-4IW | 65I-2M-4300

HOW*# | Urt jour prepared relay serwte ( mllifxMmmmmtn | fi|0»I IRiplt**

November 1, 2021

US. Environmental Protection Agency
Climate Change Division
Office of Atmospheric Programs (MC*6202A)

120® Pennsylvania km NW
Washington, DC 20460

Submitted decsranically via mMmmmimsmMm

M: u;: _. on the Draft Methodology :„r ::„... Greenhouse

Emissions and Sinks by Stale

Dear Colleagues:

Stiff at the Minnesota Pot lotion Control Agency (MPCA) have reviewed the methods used to create the
scaled-down inventory far the stale and estimates of greenhouse gas finissionj In Minnesota. There sine
significant differences betwemn the boundaries, of the U.S. Environmental Protection Agency {EPA} State
Inventory and the inventory pr®£li»«cl by the MPCA. This 4ms create challenges for us to explain the
different results, but m also rtcoinii# that the complexity of estimating emission* for different
purposes will always producedistinct, but complementary results, While we have general concerns
about the compatibility and comparability of this inventory with others, our comments here will focus
en technic#! Issues where we can offer Information that Is unique or of special interest to Mini*sol*.

from an exploration of the EPA data, emissions from Iron and Steel Production stood out as an anomaly.
An obvious lump in emissions seems to be due to the change In data sources when the GHGRP data
became available. There was not an increase or change in production that created an actual Increase in
emissions. for the Minnesota GHG inventory, ow estimates use aconite production data collected by
the Department of Revenue, citation tut low. We alio use energy consumption data that i* reported to
the state and federal clsta sources lor electricity soM to the grid in order to allocate tome emissions to
the electricity sector.

MiMWiQfii T&coM-i? PfO€$u€ikM fl9$Q->%Q20h

httos;//www.revenue.st9te.mn.us/sftes/delault/files/20ai'

lQ/Mfflnesota%20Ta«>mte%20Productton?«0Summary%20%2lJi9S0-2020%29.jrts*

¦< i,,'i iV.'M . v .ir.'i : Ki

Minnesota has i group of technical experts working on emissions from and carbon sequestration on
natural and working tends. we are working to Improve accounting In our own inventory and may have
lnforniatio.fi that li also useful for fit if»A inventory. Collaboration with experts would be mutually
beneficial, and Marty a «wt efficient way te discuss available data.

-------
1J S Fnyirnnmpnt^I Prafprtirm Agonry

Office of Atmospheric Programs (MO6202A)
Page J

November 1, 2021

There are a w?w areas that Minnesota Mute within natural and working limits that are of special
interest In the state inventory. We include an estimate of methane emissions from inland lakes and
rivers because human influences have Increased emissions from those areas beyond natural tew*. Also,
farming In peatlands and cultivation of wild rice occur in Minnesota and emissions are accounted far in
our inventory.

The MN Department of Agriculture publishes a reowt of fertfteer sold within Minnesota, which may be
useful:

Forestry is a large store of sequestered cartoon in Minnesota. Our anient method, which uses federal
data, creates mi estimate that ts highly variable on an annual basis and noticeably Afferent from the fPA
estimate. Discussion with inventory staff and experts at the Minnesota Department of natural
Resources wouli be useful to understand the method used by the EPA, share our approach and further
develop accounting methods that reflect the best science.

The MPCA is responsible far closed landfills, which provides us with the opportunity to manage landfill
gas and collect data. We on provide additional data on closed landfill activities to the EM If requested.

Finaliy^ we want ® convey owr hop# that ongoing collaboration with IP* will help the MPCA Improve
the estimates in our 6HG Inventory. there are many similarities between our inventories but there »rt
areas where we simply do not lave disaggregated date available. Fu» transparency and access to input
date Is important ier our abiRv to compare our methods and estimates to 1*A estimates and to
Incorporate infofirrialicm inloowr inventory.

Think you to the opportunity to comment on this draft, we took forward to our future collaboration.
Mease contact Anne iMirn {MPCA Research Scientist) at anne.ciafm r>.us with an* technical

Kari Palmer

Air Assessment Section

KP/AH:Je

-------
3.6 North Dakota Department of Environmental Quality

North Dakota's comments for the Draft Methodology Report: Inventory of U.S. Greenhouse

Emissions and Sinks by State

We were not able to readily find the units in Appendix A, which made it difficult to check how
accurate the data were for North Dakota specifically.

In Step 5: Subtract Consumption forNEU (Page 18 of DRAFT-

StateGHG_MethodologyReport_09172021.pdf), it is discussed that the coal used to produce
SNG at the Eastman gas plant was assumed to be used for chemical feedstock and therefore was
accounted for under NEU. It appears that the Dakota Gasification Company's Great Plains
Synfuels Plant in North Dakota should be included in this since it also uses coal to produce
SNG.

It is not clear how condensate emissions were estimated for each state, specifically in North
Dakota, which has substantial GHG emissions associated with condensate.

It is not clear if fugitive emissions associated with compressor stations (natural gas-driven and
electric-driven) were accounted for. This would account for many GHG emissions in North
Dakota.

The volume of gas production appears low in Appendix B for 2019. The NG 3 - Annual Gas
Production tab (851,750 MMScf) and the NG 2 - Gas Well Gas Production tab
(1,155,856 MScf) totals 852,905,856 MScf of gas produced in North Dakota in 2019. The North
Dakota Industrial Commission numbers estimate 927,804,888 MScf of natural gas produced.

It is unclear how emissions were estimated from storage vessels. Was this only based off pipeline
data? We expect that North Dakota may have more storage vessels relative to pipeline miles in
comparison to other states, which would underestimate storage tank GHG emissions in North
Dakota.

-------
3.7 Rhode Island Department of Environmental Management (DEM)

10/26/21

Comments prepared by Dena Gonsalves (dena.gonsalves@demri. gov)
and Owen Jackson (owen.i ackson@dem.ri. gov)

RI Department of Environmental management (RIDEM), Office of Air Resources

General comments - RI like several Northeast states uses the State Inventory Tool (SIT) for
calculating RI's GHG Emissions Inventory. In general, we agree with EPA's methodology,
however, several sectors do not apply to RI (see below). For example, RI does not have active
industrial landfills, however, using EPA's approach, RI has emissions associated with industrial
landfills.

General Questions -

a) What are your overall impressions of the clarity of the methods described in this report?

• Clear and thorough description of methods

• References at end of each Chapter helpful

b) What recommendations do you have to add to or improve the overall transparency,
completeness, consistency and accuracy of this report?

• Would be extremely helpful to have a section comparing the methodology
used with EPA's SIT tool since many states especially in the Northeast use
the SIT to calculate state GHG emissions. Perhaps use a table to compare?

• In Energy section, could the transportation C02 emissions from mobile
combustion be discussed in one section.

• Appendix A Table A-l in the "FCC C02" tab - provides details on where
State level data (based on SEDS) was used to make adjustments and
disaggregate National numbers across fuel types and sectors - where is
Appendix A?

• RIDEM uses MOVES to calculate transportation emissions and we also
calculate electricity consumption and not generation.

• We agree with methodology for determining carbon content of fuels.

• Including sources and estimates of any state level data used in emissions
estimates.

• Specific data sources and estimates for states. Including the data used to
estimate GHGs in Rhode Island.

c) Data availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are
not currently included, but could be incorporated into this analysis?

• See below

-------
b. For national level datasets that are currently used, are you aware of other
comparable datasets of activity, emission factor, or emissions data that are
available at the State, county, or zip-code levels?

• See below

d) Uncertainty. Currently uncertainty ranges are not includedfor the state level estimates.
Please provide feedback on what qualitative and quantitative information would be
useful. Timeseries Coverage. Currently State data covers 1990-2019 consistent with the
2021 National GHG Inventory, and inclusive of most known baseline periods for climate
policy. Subsequent publications of this data will also strive to maintain this consistency
with the National Inventory. As state-specific input datasets are not always available
over the entire timeseries, understanding which years may be more important can help us
to better prioritize our backcasting and methodological efforts across the time series.
EPA appreciates feedback on which, if any years should be prioritizedfor future State-
level estimates (e.g., 2000 and later, 2005 and later, 2010 and later, or the full time
series).

• Timeseries of 1990-2019 is consistent with RI GHG Emissions Inventory and
have no additional feedback on any years which should be prioritized.

e) Key Category Analysis. EPA anticipates prioritizing methodological refinements for
more significant categories to make efficient use of available resources over time. EPA
appreciates feedback on which categories are more relevant for further refining for your
State.

a. Given that the emissions profile of some states will be different from the
national average, which categories that are more significant in terms of absolute
emissions, or have changing emission trends (e.g., increasing, variable)?

• In RI, transportation, electricity consumption and residential
heating are the largest sources of GHG emissions.

• Why does EPA have electricity emissions estimated at 0.660
MMTC02e in 1990? Rhode Island's estimated electricity
consumption in 1990 is 2.82 MMTC02e. There is a 327%
difference in these estimates. Do you have any idea why these
estimates vary so significantly?

• In every year, electricity emissions are higher in Rhode
Island's greenhouse gas inventory when compared to EPA's
state level inventory. Emissions vary by about 10-20% (0.30 -
0.64 MMTC02e difference, 2015-2018).

b. The national Inventory includes a key category analysis (KCA) consistent with
2006IPCC Guidelines. Would it be useful for States if a key category analysis
(KCA) was completedfor each state?

• Yes.

•

f) Data Presentation and Usability.

-------
a. Are there other ways the state-level emissions data could be presented to
facilitate their use (e.g., in the EPA GHG Inventory Data Explorer available
online at: httys://cfyub. eya. sov/shsdata/inventoryexylorer/) ?

•

i. Related to the level of category/gas aggregation or disaggregation?

ii. Are there specific categories where further data disaggregation could be
helpful?

•

c. What data format would best facilitate the use of the state-level emissions
data (e.g., .xlsx download, etc.)?

• .xlsx or .zip

d. What additional datasets or information could be provided to help increase
the usability of the state-level emissions data?

• Any data used to estimate GHG emissions in Rhode Island
(energy sales, population, per capita emissions factors used)

d. EPA plans to provide users additional information on where they can find
official State data, where it exists. Do you have suggestions on how we should
direct users to official state data (i.e., section in methods report including links to
State data)?

• RIDEM GHG Emissions Inventory webpage found at

http://www.dem.ri.gov/programs/air/ghg-emissions-inventory.php

• Energy - Combustion Emissions Questions

• Fossil Fuel Combustion:

a) Some fuels have differences in consumption data between the
aggregated State-level totals and national totals. The current
approach is to use data from the national Inventory in those
cases. Are there other approaches that could be taken? Has
this come up in developing estimates at the State level?

• No other approaches have been identified in Rhode
Island. Rhode Island uses SIT default fuel
consumption data when developing Rhode Island's
annual greenhouse gas inventory.

b) Consistent with the IPCC Guidelines, we have adjustedfuel
consumption totals in the energy sector to account for
consumption in the IPPU sector. In some cases, this step
could lead to a negative emission total for a State if the
subtracted amount (as determinedfrom the assumed
distribution) was greater than consumption data from the
State Energy Data System (SEDS). This outcome was

-------
corrected to zero if that was the case, but are there other
approaches for correcting for that difference?

• No other approaches have been identified in Rhode
Island.

Consistent with the national Inventory, the default approach
taken here was to allocate transportation sector C02
emissions based on FHWA fuel use/sales by State. For some
States, this may not be accurate because fuel sold in a State
may be combusted in other States. Another option is to use
vehicle miles traveled (VMT) data by State but that approach
does not factor in vehicle fuel economy. Are there other
alternative or complementary approaches to allocate
transportation fuel across States, including VMT data and
other sources (e.g., NEI - based on county-level fleet and
activity data to generate a bottom-up inventory) that EPA
should consider? If so, what data sources exist to help with
that alternative approach? Would it be helpful to present
transportation sector emissions using multiple approaches in
future inventories?

• Rhode Island has very rough transportation data
provided by the Rhode Island Department of
Transportation. Current transportation data has
significant limitations such as county level VMT
data and speed distributions, but updates to
transportation data collection are being considered.
Rhode Island is in the process of creating MOVES
inputs using in state data. For MOVES inputs prior
to 2017, NESCAUM and other contractors provided
Rhode Island with the inputs required. No MOVES
inputs have been created for 2018-2020. The
transportation data necessary for MOVES is in the
process of being created for 2020 ahead of the NEI
submission.

Mobile source non-C02 emissions are allocated across
States based on vehicle-miles-traveled data while mobile
source C02 emissions were allocated based on fuel sales, as
mentioned above. Do you have any concerns with using two
different methodologies for mobile source C02 us. non-C02
State splits?

• No concerns about the difference in methodology.
Why did EPA decide to take two different
approaches to estimate mobile emissions?

Several fuels have variable C factors over time including
coal, natural gas, gasoline, and diesel fuel. Those fuels
might also have variable C factors across areas/States. Are
data available to build out State-specific C factors for the

-------
fuels with variable C contents? If so, could it be done in a
way that the State-level total emissions still matched up to
the national total emissions for those fuels?

• No coal used in Rhode Island. Rhode Island uses
SIT defaults for carbon factors and does not have
additional information on state specific factors.

f) Geothermal emissions could be allocated by the type of
geothermalproduction per State (because different types
have different emissions factors) if that data is available. Is
there more information on State-level geothermal emission
factors and production?

• No data has been identified in Rhode Island.

Non-Energy Uses (NEU) of Fossil Fuels:

a) For petrochemical feedstocks, non-energy use (NEU) of

natural gas is allocated across States based on petrochemicals
emissions data per State from the IPPU adjustments, while
other fuels are allocated based on the underlying SEDS data.
Allocating across States based on the underlying SEDS data
ensures there are no States where NEU use is larger than
original SEDS data and there are no zeros associated with
subtracting NEU (it is not an issue for natural gas because use
is so high overall compared with NEU use). Could different
approaches be used or can the petrochemical data be used
without resulting in negative use?

• No different approaches have been identified in
Rhode Island.

Incineration of Waste: Not applicable to RI

a) Waste incineration emissions are calculated based on the
combustion of fossil components of both municipal solid waste
(MSW) and tires. However, emissions are disaggregated to
Sates based only on MSW tonnage. Are there approaches or
data available to disaggregate emissions based on waste
category (e.g., MSW combustion us. tire combustion)?

International Bunker Fuels:

a) The approach used to allocate jet fuel bunker fuels by State is
currently based on the total amount ofjet fuel used by State
which could potentially lead to an over- or under-estimation
for some States of bunker fuel emissions. Are there other more
accurate approaches to allocate jet fuel bunker data across
States as opposed to the percentage ofjet fuel total use? For
example, using Federal Aviation Administration flight level

-------
data on departures and destinations or assuming based on
States with international airports andflights?

• Rhode Island has one airport (PVD) that reports
greenhouse gas emissions estimates on an annual
basis. No breakout of international fuel in the report
provided to RIDEM. No known information about
marine bunker fuels in Rhode Island.

-------
Coal Mining: Not applicable RI a) Do you have any comments specific to the methodology and
emission estimates for active coal mines and abandoned coal mines?

• b) Are you aware of any State datasets that may be useful in
helping to refine emission estimates for abandoned coal mines,
including State-level datasets addressing recovery of methane from
abandoned mines?

•

• Petroleum Systems and Natural Gas Systems:

a) Are there relevant dataset(s) that could be used to replace or
supplement the data currently used to allocate petroleum and
natural gas system emissions to the state level? Particularly,
state or detailed location information on gathering and
boosting stations, processing plants, and transmission and
storage stations?

• The RIDEM does not have any state level data at
this time.

b) Are there additional Greenhouse Gas Reporting Program
(GHGRP) data that could be used to allocate natural gas and
petroleum emissions to each State?

• Data from the "Narragansett Electric Company" on
the GHGRP (FLIGHT) are used to estimate natural
gas distribution leakage in Rhode Island.

c) Are you aware of any State datasets that may be useful in
helping to refine emission estimates for abandoned wells,
including State-level datasets addressing plugging status of
abandoned wells?

• Not applicable to RI

d) Are there particular sources for which State-level regulatory
or voluntary programs result in large differences in emission
rates between states? Are state-specific data sets available for
those sources?

• No.

•

• IPP U - Minerals Emissions Questions

• Cement Production: Not applicable to RI

b) Are you aware of data on clinker production by all States for
the full 1990-2019 time series? Please share a surrogate data
that could be used (e.g., facility production capacity, utilization
rates by facility or State) for 1990-2019 that could refine this
State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends.

•

• Lime Production: Not applicable to RI

• a) Are you aware of data on State-level lime production (activity
data) by type (e.g., high-calcium quicklime; dolomitic quicklime,
high-calcium, hydrated; dolomitic, hydrated; dead-burned

-------
dolomite; C02 capturedfor use in onsite processes) for some or
all of the 1990-2019 time series? If not, is there any surrogate
data (e.g., facility production capacity, utilization rates by facility
or State) for 1990-2019 that could refine this State inventory
calculation to enhance accuracy and consistency of State GHG
emissions and trends?

b) Based on analysis of Greenhouse Gas Reporting Program data,
it appears that most but not all beet sugar manufacturing facilities
that also produce lime and a few lime manufacturing facilities
capture C02 for use in onsite processes. Are you aware of any
information on why and how facilities producing lime capture C02
for use in onsite processes (e.g., purification), and any trends in
this practice during the 1990-2019 time series (e.g., have facilities
increased or decreased adoption of this practice during the time
series), or whether the amount of C02 captured is proportional to
the amount of lime produced or some other metric? Is this C02
ultimately released to the atmosphere? Are you aware of any data
on the amount of C02 captured onsite per facility or State for
1990-2009? C) For some States and years (Colorado for 2010-
2015, Idaho for 2011 and 2019, and Nebraska for 2010-2014),
calculations using GHGRP data on emissions and CO2 captured
for onsite processes yielded small but erroneous negative
emissions. EPA zeroed emissions for those States and years and
plans to adjust calculations so that State emissions totals match
national emissions. Do you have any general feedback on this
approach?

• No feedback from RI.

Glass Production: Not applicable to RI

a) Are you aware of state-level data on glass production or the
amount of carbonate (i.e., limestone, dolomite, soda ash)
consumed for glass production by State (activity data) for some or
all of the 1990-2019 time series? If not, can you share any state-
level surrogate data (e.g., more complete data on glass facilities by
State, amount of glass products by type [i.e., containers, flat
(window) glass, fiber glass, and specialty glass]) for 1990-2019
that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends?

Other Process Uses of Carbonates:

a) Are you aware of state-level data on the consumption of
limestone and dolomite for the iron and steel sector for the
1990-2019 time series? If not, can you share any state-level
surrogate data for 1990-2019 that could refine this State
inventory calculation to enhance accuracy and consistency of
State GHG emissions and trends from carbonate consumption
by the iron and steel sector?

-------
• Not applicable to RI

b) Are you aware of state-level data on the consumption of soda
ash (not associated with glass manufacturing) for the 1990
2019 time series?

• RIDEM is not aware of any state-level data on soda
ash.

c) Are you aware of any state-level data on limestone and
dolomite consumption for flux stone, flue gas de sulfur ization
systems, chemical stone, mine dusting or acid water treatment,
acid neutralization, and sugar refining activities for the 1990
2019 time series?

• RIDEM is not aware of any state-level data on the
activities listed above.

C02 Consumption:

a) Are you aware of other sources of data on the consumption of
C02 by State or region for the 1990-2019 time series?

• RIDEM is not aware of any other data sources for
C02 consumption.

IPPU - Chemicals Emissions Questions
Ammonia Production: Not applicable to RI

a) Currently, production capacity is used as a surrogate for state-
level ammonia production for 1990-2009. In the absence of
ammonia production by State in more recent years, are you aware
of other surrogate data (e.g., facility utilization rates by State) that
could refine this State inventory calculation to enhance accuracy
and consistency of State GHG emissions and trends?

Urea Consumption for Nonagricultural Purposes:

a) Are you aware of state-level data on urea consumption for
nonagricultural purposes (activity data) for some or all of the
1990-2019 time series?

• RIDEM is not aware of any state-level data on urea
consumption.

Nitric Acid Production: Not applicable to RI

a) Are you aware of state-level data on nitric acid production
(activity data) for some or all of the 1990-2009 time series? We
currently use production capacity as a surrogate for nitric acid
production by State for 1990-2009. We know that the production
capacity data usedfor this State inventory calculation are
incomplete for 1990-2009. Are you aware of more complete data
on facility production capacity by State?

b) Are you aware of surrogate state-level data other than facility
production capacity (e.g., utilization rates by facility or State,

-------
information about abatement technology installations and use per
facility) for 1990-2009 that could refine this State inventory
calculation to enhance accuracy and consistency of State GHG
emissions and trends?

Adipic Acid Production: Not applicable to RI

a) Are you aware of any other state-level data on adipic acid
production (activity or emissions data) for some or all of the 1990
2019 time series?

Caprolactam, Glyoxal, and Glyoxylic Acid Production: Not

applicable to RI

a) Are you aware of state-level data on caprolactam production or
emissions for some or all of the 1990-2009 time series? We
currently use production capacity as a surrogate for caprolactam
production by State. Are you aware of more complete data on
facility production capacity or actual production by State? Are you
aware of better surrogate data other than facility production
capacity (e.g., utilization rates by facility or State, information
about abatement technology installations and use per facility) that
could refine this State inventory calculation to enhance accuracy
and consistency of State GHG emissions and trends?

Carbide Production and Consumption:

a) Are you aware of state-level data on SiC production (activity
data) for the 1990-2019 time series? Are you aware of other
data to refine accuracy of the estimation of SiC consumption by
State for the 1990-2019 time series?

• RIDEM is not aware of any data that exists.

b) Are you aware of information that can help us improve the
accuracy ofproduction in the two States where SiC facilities are
located?

• No.

Titanium Dioxide (Ti02) Production: Not applicable to RI

a) Are you aware of state-level data on Ti02 production (activity
data) for the 1990-2009 time series? Is there any surrogate data
other than facility production capacity (e.g., facility utilization
rates by facility or State) for 1990-2009 that could refine this State
inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends?

Petrochemical Production: Not applicable to RI

-------
a) Are you aware of state-level data on petrochemical production
by type for the 1990-2019 time series? Is there any other
surrogate data by State or facility (e.g., facility production
capacity; utilization rates by facility or State; timing offacility
expansions, openings, and temporary or permanent closures) for
the full 1990-2019 time series that could address data gaps and
refine this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends?

Phosphoric Acid Production: Not applicable to RI

a) Are you aware of state-level data on phosphoric acid
production (activity data) for the 1990-2009 time series? Is there
any other surrogate data or information (e.g., timing of facility
expansions and temporary or permanent closures, origin of
phosphate rock used in facilities) by State or facility for 1990
2019 that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends?

HCFC-22 Production: Not applicable to RI

a) For the years 1990-2009, there are significant uncertainties in
the allocation of national-level U.S. emissions to individual
facilities and States, particularly for the five HCFC-22 production
facilities that closed before 2003 and for which production
capacity data are therefore not available. Are you aware of any
more complete sources of production capacity or other relevant
historical data?

b) Do you have recommendations for how to refine the
methodology to more accurately estimate emissions from HCFC-
22 production over the time series?

IPPU - Metals Emissions Questions

Iron and Steel and Metallurgical Coke Production: Not

applicable to RI

a) Are you aware of State-level data on iron and steel production
(activity data) by category (i.e., sinter production, iron production,
pellet production, steel production, other activities) for some or all
of the 1990-2019 time series? In the absence of steel production
by State, are you aware of better surrogate data that could refine
this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends?

b) Are you aware of state or facility-specific information to better
allocate basic oxygen furnace and electric arc furnace production
by State for 1990-2009?

Ferroalloy Production: Not applicable to RI

a) Are you aware of state or facility-level data on ferroalloy

production (activity data) or facility for the 1990-2019 time

-------
series? Please share any other surrogate data (e.g., facility
production capacity, utilization rates by facility or State) for 1990
2019 that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends.

Aluminum Production: Not applicable to RI

a) Are you aware of state or facility-level data available to
incorporate differences in emissions between smelters based on
technology type? Is there any other surrogate data or emission
sources that could be used to allocate national total aluminum
production emissions across States?

Magnesium Production and Processing: Not applicable to RI

a) Are you aware of state or facility-level magnesium production
or capacity data (or surrogate data) or facility for the 1990-2019
time series?

b) Are you aware of information on the location (by State) of
magnesium production and processing facilities or information on
the location (by State) of magnesium production and processing
facilities by process type?

Lead Production: Not applicable to RI

a) Are you aware of state or facility-level data on primary or
secondary lead production (activity data) or facility for the 1990
2019 time series? Is there any other surrogate data (e.g., primary
or secondary production capacity by facility or State) for 1990
2009 that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends?

Zinc Production: Not applicable to RI

a) Are you aware of data on zinc production (activity data) by unit
type (i.e., electrothermic furnace, Waelz kiln, other furnaces, and
flame reactor units) by State or facility for the 1990-2019 time
series? Is there any other surrogate data (e.g., total number of zinc
facilities by State, production capacity by unit type and by facility
or State) or other data by State (e.g., utilization rates by facility or
State) for 1990-2009 that could refine this State inventory
calculation to enhance accuracy and consistency of State GHG
emissions and trends?

IPPU - Product Use Emissions Questions
Electronics Industry: Not applicable to RI

a) Are you aware of State- or facility-level capacity data or other
surrogate data (e.g., sales data) by State for PV manufacturing
for 1990-2006 that could be used to refine the allocations of
emissions by State? Please share any surrogate data (e.g.,

-------
sales data by State) by State for semiconductor or MEMS
manufacturing for 1990-2007 that could be used to refine the
allocations of emissions by State.

Substitution of Ozone-Depleting Substances:

a) Are you aware of bottom-up modeling data that are available
by State? Is there any surrogate data other than population
data that could be used to disaggregate the emissions of
substitutes for ozone-depleting substances?

• RIDEM is not aware of any data that exists. Default
data derived from state population is used in Rhode
Island's greenhouse gas inventory.

Electrical Transmissions and Distribution:

a) Are you aware of State-level electrical transmission and

distribution equipment data (e.g., nameplate capacity by State)
or other data by State for 1990-2019 (or part of the time
series) that could refine this State inventory calculation to
reflect State trends in emissions more closely? Is there any
other surrogate data (e.g., State population data) to enhance
accuracy and consistency of State GHG emissions and trends
than the current data being used (transmission mile data by
State)?

• Rhode Island requests SF6 emissions from National
Grid on an annual basis. This estimate is used in the
annual greenhouse gas inventory.

N20 from Product Use:

a) Are you aware of any State-level data on N20 usage for

medical and dental anesthesia, food processing propellant and
aerosols, sodium azideproduction, or other applications (e.g.,
fuel oxidant in auto racing, oxidizing agent in blowtorches) by
State for some or all of the 1990-2019 time series? Is there
share any other surrogate data (e.g., State population data) to
enhance accuracy and consistency of State GHG emissions and
trends than the current data being used (transmission mile data
by State)?

• RIDEM is unaware of any data at this time.

Agriculture - Livestock Emissions Questions

Enteric Fermentation:

a) Are there other/newer data sources or methods, particularly at

the State level, that EPA should be aware of and consider in

-------
calculating these emissions? Especially for: • Dry matter/gross
energy intake;

• Annual data for the digestible energy (DE) values (expressed as
the percent of gross energy intake digested by the animal), CH4
conversion rates (Ym) (expressed as the fraction of gross energy
converted to CH4), and crude protein values of specific diet and
feed components for foraging andfeedlot animals;

• Monthly beef births and beef cow lactation rates;

• Weights and weight gains for beef and dairy cattle.

• RIDEM is not aware of any data that currently
exists.

b) Are State-specific diet data available to EPA to enhance
characterization of diet differences across livestock types and
U.S. States?

• RIDEM is not aware of any data that currently
exists.

c) For the enteric fermentation source category and the Cattle
Enteric Fermentation Model (CEFM), are the various regional
designations of U.S. States (as presented in Annex 3.10 of the
GHG Inventory) usedfor characterizing the diets of foraging
cattle appropriate? The CEFM is used to estimate cattle CH4
emissions from enteric fermentation and incorporates
information on livestock population, feeding practices, and
production characteristics.

• Rhode Island has no better way to designate cattle
diets, this approach is appropriate and there is not
much agriculture in Rhode Island.

Manure Management:

a) Are there other/newer data sources, particularly at the State-
level, that EPA should be aware of and consider in calculating
these emissions? Especially for the following:

-------
• waste management system data, particularly seasonal changes in
emissions from different waste management systems;

• No data sources.

• maximum methane-producing capacity;

• No data sources.

• volatile solids and nitrogen excretion rates; and

• No data sources.

• measured emission estimates (by waste management system) to
help refine estimates of methane conversion factors.

• No data sources.

Agriculture - Rice Cultivation, Urea Fertilization, Liming and
Field Burning of Agricultural Residues Questions

No category-specific questions, see general questions.

Agriculture - Agricultural Soil Management Emissions Questions
As described in the methodology section, EPA is currently
compiling the state-level emissions estimates from Agricultural
Soil Management. EPA plans to provide these estimates in coming
weeks. The methods used to compile state-level estimates are the
same as those in the national Inventory, described in Chapter 5.4
and Annex 3.12.

a) What are your overall thoughts of the clarity and transparency
of these methods?

• Rhode Island has nothing to add. Agriculture is a
very small part of our overall emissions.

b) What recommendations do you have to add to ensure high-
quality state-level estimates consistent with the national
Inventory?

• No recommendations.

LULUCF - Forest Lands and Lands Converted to Forest Land.
No category-specific questions, see general questions.

LULUCF - Croplands and Grasslands Questions
As described in the methodology section, EPA is currently
compiling the state-level emissions estimates from Croplands and
Grasslands. EPA plans to provide these estimates in coming
weeks. The methods used to compile state-level estimates are the
same as those in the national Inventory, described in Chapters 6.4
through 6.7 and Annex 3.12. To view an example state-table
currently available, please see table A-201 in Annex 3.12.
a) What are your overall thoughts of the clarity and
transparency of these methods?

-------
• TBD

b) What recommendations do you have to add to ensure high-
quality state-level estimates consistent with the national
Inventory?

• TBD

•

• LULUCF - Wetlands and Lands Converted to Wetlands Questions

• Peatlands Remaining Peatlands

a) Are there state-level data available on the application
("consumption ") of peat, including the state of use and the
horticultural/landscaping use ?

• No data available.

b) Are there data sources that could support EPA determining the
quantity ofpeat harvested per hectare and the total area
undergoing peat extraction?

• No data available at this time.

• LUL UCF - Settlements and Lands Converted to Settlements
Questions

• No category-specific questions, see general questions.

• LUL UCF - Other Lands and Lands Converted to Other Lands
Questions

• No category-specific questions, see general questions.

Waste - Solid Waste Disposal and Management Emissions Questions

• MSW Landfills:

• a) Data Questions

• • Are there datasets for individual States' landfill gas (LFG)
recovery activity?

• In RI, there is really only one MSW landfill
which is RI Resource Recovery Corporation
(Central Landfill). FYI, there are two others but
they are very small and for inventory purposes
are not included. Background: The landfill gas
collection system consists of an intricate network
of trenched horizontal piping and several
strategically placed negative-pressure gas
collection wells. All of the gas collection systems
come together at a single common gas extraction
point. The gas is then treated by a gas
conditioning and compression site (GCC). After
treatment, the landfill gas is sent as fuel to the
combined cycle combustion turbine generator
(CTG) power plant operated by RI LFG Genco,
LLC

-------
• In regards to data, RIRRC's/Central Landfill
permits require that they monitor and calculate
monthly quantities of landfill gas generated.
They are also required to measure gauge
pressure in the gas collection system monthly for
each individual well and trench to demonstrate
that the gas collection system flow rate is
sufficient. Testing is performed on the actual gas
itself, as well, including methane content and
VOC content. These records are kept on-site at
Central Landfill. RIDEM review the records
during inspections. RI LFG Genco must also
continuously monitor the amount of landfill gas
that flows to the engines. In summary, no
landfill gas data is submitted to RIDEM.

• Are there data available for open dumpsites in the
U.S. territories?

o N/A to Rhode Island.
c) The current method makes some simplifying assumptions and
includes uncertainties in the allocation of national-level U.S.
emissions to States (e.g., recovery rates are the same for all
States and match the national recovery rate). Are there
alternative assumptions or different datasets that would
improve the accuracy ofMSW landfill estimates? Do you have
recommendations to refine the methodology to estimate
emissions over the time series more accurately?

• No recommendations.

Industrial Landfills: Not applicable to RI

a) Do you have recommendations to refine the methodology to

estimate emissions over the time series more accurately?

Composting:

a) Data Questions • Is it correct to assume that Alaska has no
commercial composting operations correct? N/A

• Are there any datasets about composting in U.S. territories?

• Unknown

• Are there any State-level data sources that describe composting
activity over time?

• RIDEM collects annual solid waste data from
permitted composting facilities from 2016-2020.
This includes tonnages of the waste incoming
and outgoing from the facilities. In addition,
other than solid waste tonnage data, the facility
name & contacts, address and permit capacities

-------
are maintained. Please note, that since this data
collection effort is relatively new, RIDEM does
not have 100% of the data for years before 2019.

Stand-Alone Anaerobic Digestion:

a) Do you have or know of any State-level data for counts of
operational anaerobic digesters (processing food waste) by
year?

• No known data exists in Rhode Island.

b) Are there any facility-specific data sources we could use to fill
data gaps on the quantity of waste processed by stand-alone
digesters for any and all years of the 1990-2019 time series?

• In RI, there is one stand-alone facility, however,
they are not fully operational at this time. They are
still going through their shakedown period and
stack testing has not yet been completed. After
testing, RIDEM will revise the permit to include
updated emissions limits and conditions. After that,
they will be considered fully operational. In the
future, RIDEM will receive data on quantity of
waste processed via Air Inventory Forms.

Waste - Wastewater Treatment and Discharge Emissions

Questions

Overall:

a) The following national average parameters were used to
estimate emissions by State, with State populations used to
proxy the distribution of domestic emissions and State-level
production data (if available) used to proxy the distribution of
industrial emissions. Please comment if you believe States
would differ significantly from the national averages for the
following parameters and, if so, whether there are State-
specific data sources for EPA to consider:

• • wastewater outflow

• • biological oxygen demand (BOD), total N, and chemical
oxygen demand (COD) concentration in untreated wastewater

• • BOD:COD ratios for industrial wastewater

• wastewater treatment unit operations in use at centralized
domestic treatment plants or at industrial plants

b) Are there domestic or industrial wastewater treatment
operations present on other Pacific islands for industrial
sectors included in the national Inventory? c) For each of the
wastewater treatment and discharge subcategories listedfor
this category, is there any information that was not considered

-------
on available State-level data sources with regional or other
disaggregated information on emissions?

• Rhode Island has nothing to add.

d) For each of the subcategories, what relevant data sources could
be included? If data are used at the national level, are you aware of
other comparable data sources at the State level?

• Rhode Island is not aware of comparable data sources.
Domestic:

a) The following national average parameters were used to
estimate domestic wastewater treatment emissions by State. Please
comment on whether you think that States would differ
significantly from the national averages for the following
parameters and, if so, are there State-specific data sources for
EPA to consider: • discharge of publicly owned treatment works
(POTWs) to impaired waterbodies and nonimpaired waterbodies

• discharge of POTWs to reservoirs, lakes, and estuaries

• consumed protein

•percentage of the population on septic (versus centralized
treatment)

Industrial — Pulp and Paper:

a) Pulp and paper wastewater flows were estimated using EPA 's
Enforcement and Compliance History Online (ECHO) datasets.
Do you have any reason to believe that States 'pulp and paper
wastewater information is underrepresented in ECHO? If so, do
you have an alternative, publicly available pulp and paper
wastewater dataset by State?

b) Currently, a single year, 2019, is used to estimate the
distribution of national estimates to each State and territory for
every year of the time series. Is there reason to believe States 'pulp
and paper manufacturing operations have changed significantly
since 1990? If so, are there data sources to quantify those
changes?

c) Data for pulp and paper manufacturing for U.S. territories are
limited in the ECHO dataset. Are there resources to help estimate
a time series ofproduction data for pulp and paper wastewater
flows? Or are there territory-level data on the number of pulp and
paper plants in each U.S. territories?

Industrial — Meat and Poultry:

a) Currently, a single year, 2019, is used to estimate the

distribution of national estimates to each State and territory for
every year of the time series. Is there reason to believe States'
meat and poultry processing operations have changed

-------
significantly since 1990? If so, are there data sources to
quantify those changes?

• No reason to believe meat and poultry processing
operations have changed significantly. No data in
Rhode Island supports any significant change.

b) Data for meat and poultry processing for U.S. territories are not
captured in the USDA dataset. Are there resources to help estimate
a time series of territory-level production data for poultry
(broilers, turkeys, chicken), beef and calves, hogs, and sheep (lamb
and mutton), for example, live weight killed, number of head
slaughtered?

• N/A to Rhode Island - use SIT default values
Industrial — Fruits and Vegetables:

a) Currently, a single year, 2017, is used to estimate the distribution of
national estimates to each State and territory for every year of the time
series. Is there reason to believe States 'fruit and vegetable processing
operations have changed significantly since 1990? If so, are there data
sources to quantify those changes?

b) Data for fruit and vegetable processing for U.S. territories are not
captured in the USDA dataset. Are there resources to help estimate a
time series of territory-level production data for fruits and vegetables,
for example, canned and frozen processed vegetables, potato
production, noncitrus fruits, and citrus production?

• No reason to believe fruits and vegetables processing
operations have changed significantly. No data in
Rhode Island supports any significant change.

Industrial — Ethanol:

a) Ethanol production for each State was estimated using the Energy
Information Administration (EIA) SEDS dataset. Do you have any
reason to believe that States' information is underrepresented in the
SEDS dataset? If so, do you have an alternative, publicly available
ethanol production dataset by State?

b) Data for ethanol production for U.S. territories are limited in the
SEDS dataset. Are there resources to help estimate a time series of
production data for ethanol production?

• Unsure if this data exists in Rhode Island.

Industrial — Petroleum:

a) Petroleum production for each State was estimated using EIA 's
Petroleum Administration for Defense Districts production and State-
level operating capacity datasets. Do you have any reason to believe
that States' information is underrepresented in the EIA datasets? If so,
do you know of an alternative, publicly available petroleum refining
production dataset by State?

-------
b) Do you have any concerns about using operating capacity to
estimate petroleum production by State is not a good method? If so,
would you suggest an alternative method?

• No concerns with the methodology used in Rhode
Island.

c) Data for petroleum refining for U.S. territories are limited in the
EIA dataset. Are there resources to help estimate a time series of
territory-level production data for petroleum production?

• N/A in Rhode Island.

Industrial — Breweries:

a) Brewery production, and by extension brewery production
emissions, for each State was estimated using the Alcohol and
Tobacco Tax and Trade Bureau (TTB) taxable production dataset.
• The TTB dataset is based on taxable production/volume. Is there
any reason why taxable production from breweries may be
underrepresented by State and therefore potentially
underrepresent total emissions? If so, do you know of an
alternative dataset or assumption?

• No reason to believe taxable production/volume is not
accurate in Rhode Island.

• The TTB dataset provides production data from 2008 to the present.
The 2008 values were used as a proxy for 1990-2007 values. Is there
reason to believe States' brewery production has significantly changed
over that time period? If so, do you have an alternative, publicly
available State-level brewery production dataset (i.e., barrels
produced), or suggestions for alternative data to use as a proxy?

• No reason to believe Rhode Island's brewing production
has significantly increased during 1990-2007.

b) Data for brewery production for U.S. territories are limited in the
TTB dataset. Are there resources to help estimate a time series of
territory-level production data for brewery production (i.e., barrels
produced) ?

• N/A to Rhode Island.

-------
3.8 Texas CEQ

Responses to Greenhouse Gas Emissions and Sinks Related Questions

General Questions:

a. What are your overall impressions of the clarity of the methods described
in this report?

RESPONSE: We do not have overall comments on the methods.

b. What recommendations do you have to add to or improve the overall
transparency, completeness, consistency, and accuracy of this report?

RESPONSE: Provide an executive summary.

c) Data availability. Please address the following questions for each inventory source:

a. For each of the categories, are there additional relevant data sources that are
not currently included, but could be incorporated into this analysis?

RESPONSE: See responses below.

b. For national level datasets that are currently used, are you aware of other
comparable datasets of activity, emission factor, or emissions data that are
available at the State, county, or zip-code levels?

RESPONSE: See responses below.

d. Uncertainty. Currently uncertainty ranges are not included for the state
level estimates. Please provide feedback on what qualitative and quantitative
information would be useful. Timeseries Coverage. Currently State data covers
1990-2019 consistent with the 2021 National Greenhouse Gas (GHG) Inventory,
and inclusive of most known baseline periods for climate policy. Subsequent
publications of this data will also strive to maintain this consistency with the
National Inventory. As state-specific input datasets are not always available over
the entire timeseries, understanding which years may be more important can
help us to better prioritize our backcasting and methodological efforts across
the time series. United States Environmental Protection Agency
(EPA) appreciates feedback on which, if any years should be prioritized for
future State-level estimates (e.g., 2000 and later, 2005 and later, 2010 and later,
or the full time series).

RESPONSE: No comment.

e) Key Category Analysis. EPA anticipates prioritizing methodological refinements for
more significant categories to make efficient use of available resources over time.
EPA appreciates feedback on which categories are more relevant for further
refining for your State.

a. Given that the emissions profile of some states will be different from the
national average, which categories that are more significant in terms of absolute
emissions, or have changing emission trends (e.g., increasing, variable)?
RESPONSE: No comment.

-------
b. The national Inventory includes a key category analysis (KCA) consistent with
2006 Intergovernmental Panel on Climate Change (IPCC) Guidelines. Would it be
useful for States if a KCA was completed for each state?

RESPONSE: No comment.

f) Data Presentation and Usability.

a. Are there other ways the state-level emissions data could be presented to
facilitate their use (e.g., in the EPA GHG Inventory Data Explorer available online
at: https://cfpub.epa.gov/ghgdata/inventorvexplorer/)?

RESPONSE: No comment.

i.Related to the level of category/gas aggregation or disaggregation?
RESPONSE: No comment.

ii.Are there specific categories where further data disaggregation
could be helpful?

RESPONSE: No comment.

b. What data format would best facilitate the use of the state-level emissions
data (e.g., .xlsx download, etc.)?

RESPONSE: No comment.

c. What additional datasets or information could be provided to help increase
the usability of the state-level emissions data?

RESPONSE: No comment.

Energy - Combustion Emissions Questions
Fossil Fuel Combustion:

a. Some fuels have differences in consumption data between the aggregated
State-level totals and national totals. The current approach is to use data from
the national Inventory in those cases. Are there other approaches that could be
taken? Has this come up in developing estimates at the State level?

RESPONSE:

Point Source: We recommend developing electric utility fuel consumption at the
state or site level using Energy Information Administration (EIA) data. For
electric utilities, please see attached file (EIA annual_consumption_state.xls)
from EIA website.

Area Source: For area source categories such as residential fuel
combustion, adjusting the state-level totals to match the national totals for
consistency is reasonable since the adjustments are mostly small (less than 5%).

-------
b. Consistent with the IPCC Guidelines, we have adjusted fuel consumption totals
in the energy sector to account for consumption in the IPPU sector. In some cases, this
step could lead to a negative emission total for a State if the subtracted amount (as
determined from the assumed distribution) was greater than consumption data from
the State Energy Data System (SEDS). This outcome was corrected to zero if that was the
case, but are there other approaches for correcting for that difference?

RESPONSE:

Point Source: In-house energy generation occurs within some industrial sites
(e.g., petrochemical production sites). Adjusting fuel consumption in the energy
sector may not be needed for those industrial sites. EIA collects consumption
and generation data on all generators at sites greater than one megawatt on the
Annual Power Plant Operations Report.

Area Source: For some industrial, commercial, and institutional combustion
sources, we have seen an approach where the remaining negative emissions are
re-allocated among the other non-zero states to fully account for the adjustment.
However, making the adjustment to zero without re-allocating the remaining
negative emissions is simpler and easier to understand.

c. Consistent with the national Inventory, the default approach taken here
was to allocate transportation sector carbon dioxide (C02) emissions based

on Federal Highway Administration fuel use/sales by State. For some States, this
may not be accurate because fuel sold in a State may be combusted in other
States. Another option is to use vehicle miles traveled (VMT) data by State but
that approach does not factor in vehicle fuel economy. Are there other
alternative or complementary approaches to allocate transportation fuel across
States, including VMT data and other sources (e.g., National Emissions
Inventory - based on county-level fleet and activity data to generate a bottom-up
inventory) that EPA should consider? If so, what data sources exist to help with
that alternative approach? Would it be helpful to present transportation sector
emissions using multiple approaches in future inventories?

RESPONSE: Consistent with EPA requirements, on-road emissions inventories are
developed using emissions factors and corresponding activity levels. The same
method can be used to generate both C02 and non-C02 emissions rates for on-
road mobile sources. Using the C02 emissions rates from the national emission
factor model combined with the corresponding activity levels is the preferred
method for producing emissions estimates consistent with other pollutants.

d. Mobile source non-C02 emissions are allocated across States based

on VMT data while mobile source CO2 emissions were allocated based on fuel
sales, as mentioned above. Do you have any concerns with using two different
methodologies for mobile source CO2 vs. non-C02 State splits?

RESPONSE:

On-road Mobile: Consistent methods for estimating C02 and non-C02 emissions
are recommended. The activity factors that have already been developed for
the non-C02 pollutants can be used for C02.

e. Several fuels have variable C factors over time including coal, natural gas,
gasoline, and diesel fuel. Those fuels might also have variable C factors across
areas/States. Are data available to build out State-specific C factors for the fuels

-------
with variable C contents? If so, could it be done in a way that the State-level total
emissions still matched up to the national total emissions for those fuels?
RESPONSE: The Texas Commission on Environmental Quality (TCEQ) utilizes
multiple data sources to develop state-specific fuel property profiles for the six
fuel regions in Texas. The two data sources are: a TCEQ-sponsored statewide
triennial fuel study and fuel compliance data submitted to the EPA for federal
reformulated gasoline. The two data sources are used in conjunction with
regulatory information and default data to develop Texas-specific
fuel property profiles for the six Texas fuel regions for both historical and
future year assessments.

f. Geothermal emissions could be allocated by the type of geothermal
production per State (because different types have different emissions factors)
if that data is available. Is there more information on State-level geothermal
emission factors and production?

RESPONSE: We are not aware of any sources of Texas-specific geothermal
emission factors and production.

Non-Energy Uses (NEU) of Fossil Fuels:

For petrochemical feedstocks, NEU of natural gas is allocated across States based
on petrochemicals emissions data per State from the IPPU adjustments, while other
fuels are allocated based on the underlying SEDS data. Allocating across States
based on the underlying SEDS data ensures there are no States where NEU use is
larger than original SEDS data and there are no zeros associated with subtracting
NEU (it is not an issue for natural gas because use is so high overall compared with
NEU use). Could different approaches be used or can the petrochemical data be
used without resulting in negative use?

RESPONSE: We are not aware of alternate approaches.

Incineration of Waste:

Waste incineration emissions are calculated based on the combustion of fossil
components of both municipal solid waste (MSW) and tires. However, emissions are
disaggregated to Sates based only on MSW tonnage. Are there approaches or data
available to disaggregate emissions based on waste category (e.g., MSW combustion
vs. tire combustion)?

RESPONSE: Table A ll7 in Appendix A of this report shows MSW incineration by
state and indicates zeros for 2001 through 2019 for Texas. This does not seem
correct, is there a way to verify this?

For area sources, we use tonnage also derived from EPA and do not know of
other data sources to disaggregate emissions.

International Bunker Fuels:

The approach used to allocate jet fuel bunker fuels by State is currently based on
the total amount of jet fuel used by State which could potentially lead to an over- or
under-estimation for some States of bunker fuel emissions. Are there other more
accurate approaches to allocate jet fuel bunker data across States as opposed to the
percentage of jet fuel total use? For example, using Federal Aviation Administration
flight level data on departures and destinations or assuming based on States with
international airports and flights?

-------
RESPONSE: The TCEQ develops non-road emissions from airport sources on a
per-facility approach, using the Federal Aviation Administration's Aviation
Environmental Design Tool (AEDT), and entering aircraft and engine types
based on collected facility activity data. AEDT then applies appropriate emission
factors, fuel types used, etc. based on the aircraft and engine types selected. The
TCEQ does not collect or track specific fuel usage data from any airport facilities
as part of the emissions inventory development process.

Energy - Fugitive Emissions Questions
Coal Mining:

a. Do you have any comments specific to the methodology and emission
estimates for active coal mines and abandoned coal mines?

RESPONSE: No comment.

b. Are you aware of any State datasets that may be useful in helping to
refine emission estimates for abandoned coal mines, including State-level
datasets addressing recovery of methane from abandoned mines?

RESPONSE: The Railroad Commission of Texas (RRC) has data on permitted
surface mining, acreage, and abandoned mine

programs: https://www.rrc.texas.gov/surface-mirring/historical-coal-
mining/mining-regions-fields-and-sites/

Petroleum Systems and Natural Gas Systems:

a. Are there relevant dataset(s) that could be used to replace or supplement
the data currently used to allocate petroleum and natural gas system emissions
to the state level? Particularly, state or detailed location information on
gathering and boosting stations, processing plants, and transmission and
storage stations?

RESPONSE: Detailed location information is available for sites reporting to
the Texas point source emissions inventory. EIA's triennial EIA-757 Schedule
A, Natural Gas Processing Plant Survey, tracks the country's population of
natural gas plants and has basic location data.

The current methodology to allocate petroleum and natural gas system
emissions to the state level are reasonable (e.g., using oil, condensate, and
natural gas production and oil and gas well counts to ratio the national
emissions to individual states).

b. Are there additional Greenhouse Gas Reporting Program (GHGRP) data
that could be used to allocate natural gas and petroleum emissions to each
State?

RESPONSE: For sources that report basin-level data to the EPA's GHGRP, it is
difficult to allocate the emissions to individual states since many basins cross
state boundaries. As a result, some of the GHGRP data is for multiple states. For
sources that report county-level data to the GHGRP, it may be possible to use
that data to allocate emissions to individual states.

-------
c. Are you aware of any State datasets that may be useful in helping to
refine emission estimates for abandoned wells, including State-level datasets
addressing plugging status of abandoned wells?

RESPONSE: The RRC has detailed data on newer wells (generally, those drilled
after 1950). This includes a well status data element that could be used to
identify abandoned and plugged wells. For older historical wells drilled prior to
1950, we are not aware of a detailed dataset of Texas abandoned wells. The
RRC does have some limited statewide well counts for older wells as noted in
the EPA methodology document, but it does not include data at the
county level.

d. Are there particular sources for which State-level regulatory or voluntary
programs result in large differences in emission rates between states? Are state-
specific data sets available for those sources?

RESPONSE: Sites located in ozone nonattainment areas are subject to additional
emissions and control requirements, and those sites' emission rates can differ
from sites located in attainment areas. The TCEQ does not inventory GHG
emissions.

IPPIJ - Minerals Emissions Questions
Cement Production:

Are you aware of data on clinker production by all States for the full 1990-2019
time series? Please share a surrogate data that could be used (e.g., facility
production capacity, utilization rates by facility or State) for 1990-2019 that could
refine this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends.

RESPONSE: See the USGS National Mineral Information

Center: https://www.usgs.gov/centers/nmir/rement-statistics-and-information
T ime Production:

a. Are you aware of data on State-level lime production (activity data) by
type (e.g., high-calcium quicklime; dolomitic quicklime, high-calcium, hydrated;
dolomitic, hydrated; dead-burned dolomite; CO2 captured for use in onsite
processes) for some or all of the 1990-2019 time series? If not, is there any
surrogate data (e.g., facility production capacity, utilization rates by facility or
State) for 1990-2019 that could refine this State inventory calculation to
enhance accuracy and consistency of State GHG emissions and trends?

RESPONSE: Please see the USGS National Mineral Information
Center: https://www.usgs.gov/centers/nmir/lime-statistics-and-information. As
part of the annual TCEQ point source emission inventory, some Texas lime
production sites may provide non-confidential source level activity data. Please
provide a contact name and email address so we can share this production data.
We are not aware of surrogate data that can be used.

b. Based on analysis of Greenhouse Gas Reporting Program data, it appears
that most but not all beet sugar manufacturing facilities that also produce lime
and a few lime manufacturing facilities capture CO2 for use in onsite processes.
Are you aware of any information on why and how facilities producing lime
capture CO2 for use in onsite processes (e.g., purification), and any trends in this

-------
practice during the 1990-2019 time series (e.g., have facilities increased or
decreased adoption of this practice during the time series), or whether the
amount of CO2 captured is proportional to the amount of lime produced or
some other metric? Is this C02 ultimately released to the atmosphere? Are you
aware of any data on the amount of C02 captured onsite per facility or State for
1990-2009?

RESPONSE: Based on information provided in the TCEQ point source emissions
inventory, Texas lime plants do not appear to employ C02 capture. We are not
aware of surrogate data that can be used.

c. For some States and years (Colorado for 2010-2015, Idaho for 2011 and
2019, and Nebraska for 2010-2014), calculations using GHGRP data on
emissions and CO2 captured for onsite processes yielded small but erroneous
negative emissions. EPA zeroed emissions for those States and years and plans
to adjust calculations so that State emissions totals match national emissions.
Do you have any general feedback on this approach?

RESPONSE: No comment at this time.

Glass Production:

a. Are you aware of state-level data on glass production or the amount of
carbonate (i.e., limestone, dolomite, soda ash) consumed for glass production by
State (activity data) for some or all of the 1990-2019 time series? If not, can you
share any state-level surrogate data {e.g., more complete data on glass facilities
by State, amount of glass products by type [i.e., containers, flat (window) glass,
fiber glass, and specialty glass]} for 1990-2019 that could refine this State
inventory calculation to enhance accuracy and consistency of State GHG
emissions and trends?

RESPONSE: As part of the annual TCEQ point source emission inventory, some
Texas glass production sites may provide non-confidential source level activity
data. Please provide a contact name and email address so we can share this
production data.

Other Process Uses of Carbonates:

b. Are you aware of state-level data on the consumption of limestone and
dolomite for the iron and steel sector for the 1990-2019 time series? If not, can
you share any state-level surrogate data for 1990-2019 that could refine this
State inventory calculation to enhance accuracy and consistency of State GHG
emissions and trends from carbonate consumption by the iron and steel sector?

RESPONSE: We are not aware of state-level data on the consumption of limestone
and dolomite for the iron and steel sector for the specified period.

c) Are you aware of state-level data on the consumption of soda ash (not associated
with glass manufacturing) for the 1990-2019 time series?

RESPONSE: We are not aware of state-level data on the consumption of soda ash
for the specified period or any other period.

d. Are you aware of any state-level data on limestone and dolomite
consumption for flux stone, flue gas desulfurization systems, chemical stone,

-------
mine dusting or acid water treatment, acid neutralization, and sugar refining
activities for the 1990-2019 time series?

RESPONSE: We are not aware of state-level data on limestone and dolomite
consumption for flux stone, flue gas desulfurization systems, chemical stone,
mine dusting or acid water treatment, acid neutralization, and sugar refining
activities for the specified period or any other period.

CO2 Consumption:

Are you aware of other sources of data on the consumption of CCteby State or
region for the 1990-2019 time series?

RESPONSE: We are not aware of other sources of C02 consumption data.

IPPIJ - Chemicals Emissions Questions
Ammonia Production:

Currently, production capacity is used as a surrogate for state-level ammonia
production for 1990-2009. In the absence of ammonia production by State in more
recent years, are you aware of other surrogate data (e.g., facility utilization rates by
State) that could refine this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends?

RESPONSE: We are not aware of a current source of ammonia production in
Texas.

Annual ammonia production for the U.S., not by state, can be found at the
following USGS web site: https://www.usgs.gov/centers/nmic/nitrogen-
statistics-and-information. It may be possible to combine the information from the
USGS website with the information in Appendix B of the_Energy Use and Energy
Intensity of the U.S. Chemical

Industry report (https://www.energystar.gov/sites/default/files/buiidings/tools/
industrial_LBNL-44314.pdf) to estimate state-level ammonia production.

Urea Consumption for Non-agricultural Purposes:

Are you aware of state-level data on urea consumption for non-agricultural
purposes (activity data) for some or all of the 1990-2019 time series?

RESPONSE: We are not aware of publicly available resources regarding urea
consumption. It appears that IHS Markit may have data regarding urea
consumption for non-agricultural

purposes: https://ihsmarkit.com/products/urea-rhemical-economirs-
handbook.html

Nitric Acid Production:

a. Are you aware of state-level data on nitric acid production (activity data)
for some or all of the 1990-2009 time series? We currently use production
capacity as a surrogate for nitric acid production by State for 1990-2009. We
know that the production capacity data used for this State inventory calculation
are incomplete for 1990-2009. Are you aware of more complete data on facility
production capacity by State?

RESPONSE: We are not aware of state-level data.

b. Are you aware of surrogate state-level data other than facility production
capacity (e.g., utilization rates by facility or State, information about abatement

-------
technology installations and use per facility) for 1990-2009 that could refine
this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends?

RESPONSE: We are not aware of surrogate state-level data.

Adipic Acid Production:

Are you aware of any other state-level data on adipic acid production (activity or
emissions data) for some or all of the 1990-2019 time series?

RESPONSE: We are not aware of state-level data.

Caprolactam, Glyoxal, and Glyoxylic Acid Production:

Are you aware of state-level data on caprolactam production or emissions for some
or all of the 1990-2009 time series? We currently use production capacity as a
surrogate for caprolactam production by State. Are you aware of more complete
data on facility production capacity or actual production by State? Are you aware of
better surrogate data other than facility production capacity (e.g., utilization rates
by facility or State, information about abatement technology installations and use
per facility) that could refine this State inventory calculation to enhance accuracy
and consistency of State GHG emissions and trends?

RESPONSE: We are not aware of state-level data.

Carbide Production and Consumption:

a. Are you aware of state-level data on SIC production (activity data) for the
1990-2019 time series? Are you aware of other data to refine accuracy of the
estimation of SIC consumption by State for the 1990-2019 time series?

RESPONSE: We are not aware of state-level data.

b. Are you aware of information that can help us improve the accuracy of
production in the two States where SIC facilities are located?

RESPONSE: We are not aware of state-level data.

Titanium Dioxide (T1O2) Production:

Are you aware of state-level data on T1O2 production (activity data) for the 1990-
2009 time series? Is there any surrogate data other than facility production
capacity (e.g., facility utilization rates by facility or State) for 1990-2009 that could
refine this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends?

RESPONSE: Annual production for the U.S., not by state, can be found at the
following USGS web site: https://www.usgs.gov/centers/nmir/titaniiim-
statistics-and-information

Petrochemical Production:

Are you aware of state-level data on petrochemical production by type for the
1990-2019 time series? Is there any other surrogate data by State or facility (e.g.,
facility production capacity; utilization rates by facility or State; timing of facility
expansions, openings, and temporary or permanent closures) for the full 1990-
2019 time series that could address data gaps and refine this State inventory
calculation to enhance accuracy and consistency of State GHG emissions and
trends?

-------
RESPONSE: State- or region-level petrochemical data is available from the EIA:
https://www.eia.gov/

Phosphoric Acid Production:

Are you aware of state-level data on phosphoric acid production (activity data) for
the 1990-2009 time series? Is there any other surrogate data or information (e.g.,
timing of facility expansions and temporary or permanent closures, origin of
phosphate rock used in facilities) by State or facility for 1990-2019 that could
refine this State inventory calculation to enhance accuracy and consistency of State
GHG emissions and trends?

RESPONSE: We are not aware of state-level data.

HCFC-22 Production:

a. For the years 1990-2009, there are significant uncertainties in the
allocation of national-level U.S. emissions to individual facilities and States,
particularly for the five HCFC-22 production facilities that closed before 2003
and for which production capacity data are therefore not available. Are you
aware of any more complete sources of production capacity or other relevant
historical data?

RESPONSE: We are not aware of state-level data.

b. Do you have recommendations for how to refine the methodology to
more accurately estimate emissions from HCFC-22 production over the time
series?

RESPONSE: No comment at this time.

IPPIJ - Metals Emissions Questions

Iron and Steel and Metallurgical Coke Production:

a. Are you aware of State-level data on iron and steel production (activity
data) by category (i.e., sinter production, iron production, pellet production,
steel production, other activities) for some or all of the 1990-2019 time series?
In the absence of steel production by State, are you aware of better surrogate
data that could refine this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends?

RESPONSE: We are not aware of existing state-level data on iron and steel
production for the sinter, iron, pellet, odr steel production categories for the
specified period or any other period. Annual production for the U.S., not by
state, can be found at the following USGS web

site: https://pubs.usgs.gov/periodicals/mcs2021/mcs2021-iron-steel.pdf

b. Are you aware of state or facility-specific information to better allocate
basic oxygen furnace and electric arc furnace (EAF) production by State for
1990-2009?

RESPONSE: We are not aware of state-level or facility information data to better
allocate production between EAFs and other oxygen furnaces. Based on a direct
conversation with a representative of the Steel Manufacturing Association, 100%
of iron and steel production in Texas occurs using EAFs, and almost none or
none through basic oxygen furnaces. This information was confirmed
by a contact for Nucor Steel.

-------
Ferroalloy Production:

Are you aware of state or facility-level data on ferroalloy production (activity data)
or facility for the 1990-2019 time series? Please share any other surrogate data
(e.g., facility production capacity, utilization rates by facility or State) for 1990-
2019 that could refine this State inventory calculation to enhance accuracy and
consistency of State GHG emissions and trends.

RESPONSE: We are not aware of existing state-level data on ferroalloy
production.

Aluminum Production:

Are you aware of state or facility-level data available to incorporate differences in
emissions between smelters based on technology type? Is there any other surrogate
data or emission sources that could be used to allocate national total aluminum
production emissions across States?

RESPONSE: We are not aware of existing state-level, facility, or surrogate data.
The USGS has annual national data available

here: https://www.usgs.gov/centers/nmic/commoditv-statistics-and-
information.

Magnesium Production and Processing:

a. Are you aware of state or facility-level magnesium production or capacity
data (or surrogate data) or facility for the 1990-2019 time series?

RESPONSE: We are not aware of existing state-level, facility, or surrogate data.
The USGS has annual national data available

here: https://www.usgs.gov/centers/nmic/commoditv-statistics-and-
information.

b. Are you aware of information on the location (by State) of magnesium
production and processing facilities or information on the location (by State) of
magnesium production and processing facilities by process type?

RESPONSE: We are not aware of existing state-level, facility, or surrogate data.
The USGS has annual national data available

here: https://www.usgs.gov/centers/nmic/commodity-statistics-and-
information.

Lead Production:

Are you aware of state or facility-level data on primary or secondary lead
production (activity data) or facility for the 1990-2019 time series? Is there any
other surrogate data (e.g., primary or secondary production capacity by facility or
State) for 1990-2009 that could refine this State inventory calculation to enhance
accuracy and consistency of State GHG emissions and trends?

RESPONSE: We are not aware of existing state-level, facility, or surrogate data.
The USGS has annual national data available

here: https://www.usgs.gov/centers/nmic/commoditv-statistics-and-
information.

Zinc Production:

-------
Are you aware of data on zinc production (activity data) by unit type (i.e.,
electrothermic furnace, Waelz kiln, other furnaces, and flame reactor units) by State
or facility for the 1990-2019 time series? Is there any other surrogate data (e.g.,
total number of zinc facilities by State, production capacity by unit type and by
facility or State) or other data by State (e.g., utilization rates by facility or State) for
1990-2009 that could refine this State inventory calculation to enhance accuracy
and consistency of State GHG emissions and trends?

RESPONSE: We are not aware of existing state-level, facility, or surrogate data.
The USGS has annual national data available here:

https://www.usgs.gov/centers/nmic/commodity-statistics-and-information.

IPPIJ - Product Use Emissions Questions
Electronics Industry:

Are you aware of State- or facility-level capacity data or other surrogate data (e.g.,
sales data) by State for PV manufacturing for 1990-2006 that could be used to
refine the allocations of emissions by State? Please share any surrogate data (e.g.,
sales data by State) by State for semiconductor or micro-electromechanical
system manufacturing for 1990-2007 that could be used to refine the allocations of
emissions by State.

RESPONSE: We are unaware of any sales data, capacity data, or other surrogate
data for PV manufacturing in Texas.

Substitution of Ozone-Depleting Substances:

Are you aware of bottom-up modeling data that are available by State? Is there any
surrogate data other than population data that could be used to disaggregate the
emissions of substitutes for ozone-depleting substances?

RESPONSE: We are unaware of any Texas-specific modeling for ozone-depleting
substances or surrogate data other than population.

Electrical Transmissions and Distribution:

Are you aware of State-level electrical transmission and distribution equipment
data (e.g., nameplate capacity by State) or other data by State for 1990-2019 (or
part of the time series) that could refine this State inventory calculation to reflect
State trends in emissions more closely? Is there any other surrogate data (e.g., State
population data) to enhance accuracy and consistency of State GHG emissions and
trends than the current data being used (transmission mile data by State)?
RESPONSE: We do not have access to data on electrical transmission and
distribution equipment data. This data may be available from either the Electric
Reliability Council of Texas (ERCOT) which oversees the State of Texas electrical
grid or the Public Utility Commission of Texas which oversees ERCOT.

N20 from Product Use:

Are you aware of any State-level data on N20 usage for medical and dental
anesthesia, food processing propellant and aerosols, sodium azide production, or
other applications (e.g., fuel oxidant in auto racing, oxidizing agent in blowtorches)
by State for some or all of the 1990-2019 time series? Is there any other surrogate
data (e.g., State population data) to enhance accuracy and consistency of State GHG
emissions and trends than the current data being used (transmission mile data by
State)?

-------
RESPONSE: We are unaware of state level N20 usage data for the medical and
dental industries. Only through industry specific surveying efforts could this
data be acquired.

Atrial It lire - Livestock Emissions Questions
Enteric Fermentation:

a. Are there other/newer data sources or methods, particularly at the State
level, that EPA should be aware of and consider in calculating these emissions?
Especially for: • Dry matter/gross energy intake;

• Annual data for the digestible energy (DE) values (expressed as the percent of
gross energy intake digested by the animal), CH4 conversion rates (Ym) (expressed
as the fraction of gross energy converted to CH4), and crude protein values of
specific diet and feed components for foraging and feedlot animals;

• Monthly beef births and beef cow lactation rates;

• Weights and weight gains for beef and dairy cattle.

RESPONSE: We are not aware of any newer sources of Texas-specific livestock
emission factors and production.

b. Are State-specific diet data available to EPA to enhance characterization
of diet differences across livestock types and U.S. States?

RESPONSE: We are not aware of any sources of Texas-specific livestock emission
factors and production.

c. For the enteric fermentation source category and the Cattle Enteric
Fermentation Model (CEFM), are the various regional designations of U.S. States
(as presented in Annex 3.10 of the GHG Inventory) used for characterizing the
diets of foraging cattle appropriate? The CEFM is used to estimate cattle

CH4 emissions from enteric fermentation and incorporates information on
livestock population, feeding practices, and production characteristics.
RESPONSE: We are not aware of any sources of Texas-specific livestock emission
factors and production.

Manure Management:

Are there other/newer data sources, particularly at the State-level, that EPA should
be aware of and consider in calculating these emissions? Especially for the
following:

• waste management system data, particularly seasonal changes in emissions from
different waste management systems;

• maximum methane-producing capacity;

• volatile solids and nitrogen excretion rates; and

• measured emission estimates (by waste management system) to help refine
estimates of methane conversion factors.

RESPONSE: We are not aware of any newer sources of Texas-specific livestock
emission factors and production.

Agriculture - Rice Cultivation, Urea Fertilization, liming and Field
Burning of Agricultural Residues Questions
No category-specific questions, see general questions.

-------
Agriculture - Agricultural Soil Management Emissions Questions
As described in the methodology section, EPA is currently compiling the state-level
emissions estimates from Agricultural Soil Management. EPA plans to provide these
estimates in coming weeks. The methods used to compile state-level estimates are the
same as those in the national Inventory, described in Chapter 5.4 and Annex 3.12.

a. What are your overall thoughts of the clarity and transparency of these
methods?

RESPONSE: We do not have any comments.

b. What recommendations do you have to add to ensure high-quality state-
level estimates consistent with the national Inventory?

RESPONSE: We do not have any recommendations.

Land use, land-use change and forestry (LULIJCF) - Forest Lands and

Lands Converted to Forest Land.

No category-specific questions, see general questions. (No Response Needed From
TCEQ)

LULUCF - Croplands and Grasslands Questions

As described in the methodology section, EPA is currently compiling the state-level
emissions estimates from Croplands and Grasslands. EPA plans to provide these
estimates in coming weeks. The methods used to compile state-level estimates are the
same as those in the national Inventory, described in Chapters 6.4 through 6.7 and
Annex 3.12. To view an example state-table currently available, please see table A-201
in Annex 3.12.

a. What are your overall thoughts of the clarity and transparency of these
methods?

RESPONSE: The methodology for this category seems sound.

b. What recommendations do you have to add to ensure high-quality state-
level estimates consistent with the national Inventory?

RESPONSE: We have no further recommendations.

LULUCF - Wetlands and Lands Converted to Wetlands Questions
Peatlands Remaining Peatlands

a. Are there state-level data available on the application ("consumption") of
peat, including the state of use and the horticultural/landscaping use?

RESPONSE: We are not aware of any sources of Texas-specific application of
peat/consumption.

b. Are there data sources that could support EPA determining the quantity
of peat harvested per hectare and the total area undergoing peat extraction?

RESPONSE: We are not aware of any sources of Texas-specific quantity of peat
harvested or total area undergoing peat extraction.

LULUCF - Settlements and Lands Converted to Settlements Questions
No category-specific questions, see general questions.

-------
RESPONSE: We are not aware of any sources of Texas-specific LULUCF emission
factors and production.

LULUCF - Other Lands and Lands Converted to Other Lands Questions
No category-specific questions, see general questions.

RESPONSE: We are not aware of any sources of Texas-specific LULUCF emission
factors and production.

Waste - Solid Waste Disposal and Management Emissions Questions
MSW Landfills:

a. Data Questions

• Are there datasets for individual States' landfill gas (LFG) recovery activity?

• Are there data available for open dumpsites in the U.S. territories?

RESPONSE:

• Datasets for Texas LFG recovery activity can be found in

the TCEQ MSW annual report at the following link: annual report.

Pages 99 - 102 of the 2020 report show a list of Facilities Recovering
Landfill Gas for Beneficial Use (2020), including information on Gas
Processed (ft3)- The annual report is available for several years.

Additional information can be found by filtering the 'physical type' column
for 9GR, which are the landfill gas recovery facilities of the spreadsheet at
the following link: msw-lacilities-texas.xls— A spreadsheet listing issued or
acknowledged permits and other authorizations as well as pending
applications for MSW landfills and processing facilities that are active,
inactive, or not yet constructed. Data fields include facility name and type;
permit, registration, or notification number; authorization status; facility
physical status; and location information. See Explanation of Municipal Solid
Waste Data Fields for more information.

• Data on open dumpsites may be found at: msw-unum-texas.xls. A
spreadsheet of historical information listing old, closed unnumbered MSW
landfills that were operated before permits were required, as well as
unauthorized landfills and miscellaneous illegal dumps and disposal sites.
See the TCEQ Inventory of Closed Municipal Solid Waste Landfills page for
information about why this information was collected and how it is used.
See Explanation of Municipal Solid Waste Data Fields for more
information.

b. The current method makes some simplifying assumptions and includes
uncertainties in the allocation of national-level U.S. emissions to States (e.g.,
recovery rates are the same for all States and match the national recovery rate).
Are there alternative assumptions or different datasets that would improve the
accuracy of MSW landfill estimates? Do you have recommendations to refine the
methodology to estimate emissions over the time series more accurately?

RESPONSE: We are not aware of alternative assumptions or other datasets that
would improve the accuracy of MSW landfill estimates.

Industrial Landfills:

-------
Do you have recommendations to refine the methodology to estimate emissions

over the time series more accurately?

RESPONSE: We do not have any recommendations.

Composting:

Data Questions: Is it correct to assume that Alaska has no commercial composting
operations correct?

• Are there any datasets about composting in U.S. territories?

• Are there any State-level data sources that describe composting activity over
time?

RESPONSE: We are not aware of any sources of Texas-

specific composting datasets or data sources in regards to composting activity
over time.

Stand-Alone Anaerobic Digestion:

a. Do you have or know of any State-level data for counts of operational
anaerobic digesters (processing food waste) by year?

RESPONSE: We are not aware of any sources of Texas-specific operational
anaerobic digester emission factors, counts, or production.

b. Are there any facility-specific data sources we could use to fill data gaps
on the quantity of waste processed by stand-alone digesters for any and all
years of the 1990-2019 time series?

RESPONSE: We are not aware of any sources of Texas-specific data.

Waste - Wastewater Treatment and Discharge Emissions Questions
Overall:

a. The following national average parameters were used to estimate
emissions by State, with State populations used to proxy the distribution of
domestic emissions and State-level production data (if available) used to proxy
the distribution of industrial emissions. Please comment if you believe States
would differ significantly from the national averages for the following
parameters and, if so, whether there are State-specific data sources for EPA to
consider:

• wastewater outflow

• biological oxygen demand (BOD), total N, and chemical oxygen demand (COD)
concentration in untreated wastewater

• BOD:COD ratios for industrial wastewater

• wastewater treatment unit operations in use at centralized domestic treatment
plants or at industrial plants

RESPONSE: We do not have any comments.

b. Are there domestic or industrial wastewater treatment operations present
on other Pacific islands for industrial sectors included in the national Inventory?

RESPONSE: Not applicable to Texas.

c. For each of the wastewater treatment and discharge subcategories listed
for this category, is there any information that was not considered on available

-------
State-level data sources with regional or other disaggregated information on
emissions?

RESPONSE: We are not aware of any additional data.

d. For each of the subcategories, what relevant data sources could be
included? If data are used at the national level, are you aware of other
comparable data sources at the State level?

RESPONSE: We are not aware of any additional data.

Domestic:

The following national average parameters were used to estimate domestic
wastewater treatment emissions by State. Please comment on whether you think
that States would differ significantly from the national averages for the following
parameters and, if so, are there State-specific data sources for EPA to consider:

• discharge of publicly owned treatment works (POTWs) to impaired waterbodies
and nonimpaired waterbodies

• discharge of POTWs to reservoirs, lakes, and estuaries

• consumed protein

• percentage of the population on septic (versus centralized treatment)

RESPONSE: The TCEQ knows of no significant changes to this category or
additional data sources and believe the national averages for the parameters
listed are acceptable.

Industrial - Pulp and Paper:

a. Pulp and paper wastewater flows were estimated using EPA's
Enforcement and Compliance History Online (ECHO) datasets. Do you have any
reason to believe that States' pulp and paper wastewater information is
underrepresented in ECHO? If so, do you have an alternative, publicly available
pulp and paper wastewater dataset by State?

RESPONSE: The following web sites may help estimate pulp and paper
wastewater activity data: the Technical Association for the Pulp and Paper
Industry: https://www.tappi.org/ and the

National Council for Air and Stream Improvement: https://www.ncasi.org/ .

b. Currently, a single year, 2019, is used to estimate the distribution of
national estimates to each State and territory for every year of the time series. Is
there reason to believe States' pulp and paper manufacturing operations have
changed significantly since 1990? If so, are there data sources to quantify those
changes?

RESPONSE: Yes. 1990 data should not be used to determine 2020 emissions.
Please see the data sources listed in a) above.

c. Data for pulp and paper manufacturing for U.S. territories are limited in
the ECHO dataset. Are there resources to help estimate a time series of
production data for pulp and paper wastewater flows? Or are there territory-
level data on the number of pulp and paper plants in each U.S. territories?

RESPONSE: Not applicable to Texas.

Industrial - Meat and Poultry:

-------
a. Currently, a single year, 2019, is used to estimate the distribution of
national estimates to each State and territory for every year of the time series. Is
there reason to believe States' meat and poultry processing operations have
changed significantly since 1990? If so, are there data sources to quantify those
changes?

RESPONSE: We are unaware of significant changes to this category or additional
data sources.

b. Data for meat and poultry processing for U.S. territories are not captured
in the United States Department of Agriculture (USDA) dataset. Are there
resources to help estimate a time series of territory-level production data for
poultry (broilers, turkeys, chicken), beef and calves, hogs, and sheep (lamb and
mutton), for example, live weight killed, number of head slaughtered?

RESPONSE: We are unaware of significant changes to this category or additional
data sources in regard to time series territory level production data.

Industrial - Fruits and Vegetables:

a. Currently, a single year, 2017, is used to estimate the distribution of
national estimates to each State and territory for every year of the time series. Is
there reason to believe States' fruit and vegetable processing operations have
changed significantly since 1990? If so, are there data sources to quantify those
changes?

RESPONSE: We are unaware of significant changes in the fruit and vegetable
processing and no additional data sources.

b. Data for fruit and vegetable processing for U.S. territories are not
captured in the USDA dataset. Are there resources to help estimate a time series
of territory-level production data for fruits and vegetables, for example, canned
and frozen processed vegetables, potato production, noncitric fruits, and citrus
production?

RESPONSE: We are unaware of additional data sources to estimate time series of
territory-level production.

Industrial - Ethanol:

a. Ethanol production for each State was estimated using the EIA SEDS
dataset. Do you have any reason to believe that States' information is
underrepresented in the SEDS dataset? If so, do you have an alternative, publicly
available ethanol production dataset by State?

RESPONSE: The Nebraska Department of Environment and Energy lists Ethanol
Facilities Capacity by State and Plant provides a data set that can be used for
cross-reference: https://neo.ne.gov/programs/stats/inf/122.htm

b. Data for ethanol production for U.S. territories are limited in the SEDS
dataset. Are there resources to help estimate a time series of production data
for ethanol production?

RESPONSE: Not applicable to Texas.

Industrial - Petroleum:

-------
a. Petroleum production for each State was estimated using EIA's Petroleum
Administration for Defense Districts production and State-level operating
capacity datasets. Do you have any reason to believe that States' information is
underrepresented in the EIA datasets? If so, do you know of an alternative,
publicly available petroleum refining production dataset by State?

RESPONSE: No comments.

b. Do you have any concerns about using operating capacity to estimate
petroleum production by State as not a good method? If so, would you suggest
an alternative method?

RESPONSE: No comments.

c. Data for petroleum refining for U.S. territories are limited in the EIA
dataset. Are there resources to help estimate a time series of territory-level
production data for petroleum production?

RESPONSE: Not applicable to Texas.

Industrial - Breweries:

a. Brewery production, and by extension brewery production emissions, for
each State was estimated using the Alcohol and Tobacco Tax and Trade Bureau
(TTB) taxable production dataset.

• The TTB dataset is based on taxable production/volume. Is there any reason why
taxable production from breweries may be underrepresented by State and therefore
potentially underrepresent total emissions? If so, do you know of an alternative
dataset or assumption?

• The TTB dataset provides production data from 2008 to the present. The 2008
values were used as a proxy for 1990-2007 values. Is there reason to believe States'
brewery production has significantly changed over that time period? If so, do you
have an alternative, publicly available State-level brewery production dataset (i.e.,
barrels produced), or suggestions for alternative data to use as a proxy?

RESPONSE: For Area Source, the methodology used above seems sound and

we are unaware of other proxy, datasets, or assumptions.

b. Data for brewery production for U.S. territories are limited in the TTB
dataset. Are there resources to help estimate a time series of territory-level
production data for brewery production (i.e., barrels produced)?

RESPONSE: Not applicable to Texas.

-------
3.9 Virginia Department of Environmental Quality (DEQ)

K! \ II \\
of

Draft Inventory of 1,1 .S. bMm and Sinks by Slate: 1990-2019
Prepared by hi' \ J»»r Virginia

Protection Agency's (EPA) aon'tial Inventory o/1/.t Greenhouse Cm Emimtmn mdSitih (lieMaler
national M'mtmyl titled the htvemo/y nf US, Emissions mtiSinks by Staim Iti piepm-Mtmi for ins. EPA
has circulated a draft inventory for 1990.2019. disaggregated to different stales seeking a. cevkw on the
draft applicable ten the respective stales.

The review provided in this document is. applicable to the state of Virginia, i m review has oeert
oriented. primarily. to txmpme the emiisiaiis, being estimated annually few the km 30 yeais (a bottoms up
approach) with that derived for the slate on die basts of National Inventory Iji up down approach). Bated
on the review result*, any inconsistency that is observed in the process is addressed

In Virginia. it »s mot mandatory to maintain annual GHC inventory. A onetime CIIIG inventory
was prepared to assist a Committee set up by the Governor w 2001. From then on GHG inventoiy is
being compiled annually with m Virguita DEQ, Far this purpose, State Inventory Tool (SIT), sponsored
by EPA and encouraged tor use by the state* has been the basic, by augmenting additional input tea as
available for the purpose. The latest annual inventory till 2819 is built m the results of KPA's CiHCiRI*
program to Ike extent available and further expanded to cover other sources using SIT. He reasons for
using GHGR.P reports in the development of the inventory

(2) Source report* are scrutinized % El* A and validated before publicised;

limit 50 percent of emissions from industry ot" si V iiginia «i saims; tend

(4) GHURP program that is 10 vein* old is likely to encompass more wmm«« future.

The disaggregated state level drill inventory has three versions, covering all the emissions
attributable to ihe state and one version accounting for sinks. Version 3, in which the emissions are
classified o* the basis of economic mam is usedintins review, a.* translating suite inventoiy to the draft
ft* line item comparison is band to be wawewem, There »« 19 in* items «lit* draft inventory,
iiceotmiitig for a small proportion of the total emissions f less than 5%) that are not identifiable it» the mate
inventory. fur comparison to hi on asm* fowl, these unique 19 l#» items .*» also included appropriately

>(>«•' "'m., - iin.iitofy. ti'il, \ .i in I.i «.u v , ¦ which no slate comparison is
made and disaggregated data only is considered for the purpose of u t\> In >J «> n(I"
default and the other 1***1 on OHGKP reports for 2® HI W 2019. Eni«wio«» cm the basis of OllCilf
Pcp^ii'tei! imluci. orr bfcklrnck^d tor ciilier year* on ilw basis ot aitfiiml thfoi.iglipi.ii

IK#!!! liwi 4fe'pm1.iitcttlii1 ciitiiisiitRs database fC'fJJSl,

-------
Figure 1 and Figure 2 show a companion of gross and net emissions for the state between the
disaggregated national and the state values. Net emissions are based on the sink data from the draft.

Figure 1; Gross State Emissions ( oniparismi

]:
-------
Above charts primarily indicate, that emissions disaggregated on the basis of National Inv entory
(White Line) are the least in recent times compared to values reckoned by the state, while the trends are
very similar. State emissions computed for the purpose, has three scenarios:

(1) SIT Default - As it results from all die modules contained in SIT and energy input values
derived from State Energy Data System (SbDS);

(2) All as in item (1 >, except, energy input data for power generation derived from the operating
data of individual plants that feed die national grid, obtained from the Electricity Section,
odier wing of the Energy Information Administration (ElA), of die Department of die Energy
{DOE). If the inventory has to reckon impact of every kwh of power that goes to national
grid, tins is closer in contrast to SEDS that only take coal, oil, natural gas &. biomass into
account; and

(3) Oriented to GHGRP reports based on actual values for 2010 to 2011 (mostly from 2011)
projected to earlier years on the basis of their performance linked to throughput values of
inputs contributing GHG.

In the above assessment, computed state emissions for Agriculture and Residential Sectors are
only SIT based. Whereas, Commercial Sector has a GHGRP component, since the draft includes
emissions from municipal landfills.

Net emissions are lower to the extent discounted by sinks attributed to the stale in the draft.

Disparities observed in emissions between different approaches, sector wise, are further
addressed in the following pages.

2,(1 Transportation:

71,000

B fcS.000

r*. fcO.OOO

B MjOOO

J 41.000

U 4S.0M
|

€ .id.ooo

2 is.ooo
Mjm

(WO IW 1I1UH 2M5 >19111 Jllis IU2IJ

UAH

Figure 2.3: Comparison of Gross Emission* from Transportation

-------
Figure 2.3 shows a comparison of gross emissions from Transportation between the disaggregated and the
state computed values resulting from SIT. SIT provides two alternatives:

{1) C02 emissions are computed on the basis of carbon contained in the fuels used with the minor
components of CH4 & N20 on the basis of vehicle miles travelled (VMT) and a variety of
vehicle specifics; and

(2) C02 emissions computed on the basis of VMT along with minor components as above-

As can be seen from the Figure 2.1, alternative (2) above provides greater consistency & convergence
with the disaggregated values dian that relates fuel combustion for CCJ2.

Steady decline in the emissions in recent years, is attributable to both improved fuel efficiency of
miles per gallon as well as relative changes in the proportion of greenhouse gases in the total emissions as
supported by Figure 2.4.

100%

I 96%
«

95*

94%

91%

92%

SOXUIt

CARBON DIOMDt.

-CQ2 MMt 002 E
-N 20 MMT CO 2 E
CH4 MMT C02 E

1990

199$

2000

2005
YEAR

2010

201S

30 ul

Ifl

to £

Figure 2.4; Trend in the Distribution of (¦!!(> Component* in Transportation I'missions

Response to charges to state* in this context: SIT based approach to assess C02 emissions nls«
indirectly on the basis of VMT and other changing vehicle specifies than to depend only on the
combustion of fuel consumed by the sector, appears a better approach.

3,11 Power Industry; Figure 3-5 shows a comparison between disaggregated emissions & computed
state emissions. Different scenarios considered for computation* arc:

(I) SIT Default (Fuel data based on SliDS);

(2) SIT Default with fuel data based on individual plant performance {Based on EIA Database)

(3 ) Oriented to GHGRP Reports (for 2010 lo 201') with extrapolation based on appropriate
throughput data based on Departmental database that feeds NIT) and

-------
(4) Based on power consumption based on SIT evaluation {perhaps outside the scope of reference).

DLvaggrcgital llrac-ulnm
SIT Default
Biwd on Individual Plant Dila |EIA Oaten
Oriienltd ta (*ll(iRP departs
Had I w Ptiwtr ( iniiump<«in (SIT)

Figure 3.5; Comparison of (iross (JH( ¦ K missions from Power Industry

From Figure 3.5 ii is observed, besides similarity in trends irrespective of method for computing
emissions, the spread converges specially in recent years supporting near consistency. However, CHORP
report oriented data, the highest of all the assessments is slightly higher than disaggregated values. It may
be pointed out that total emissions reported by GHGRP between 2010 & 2019 account for above AO
percent of ail emissions from power industry in the state.

Emissions based on actual consumption of total power in the state based on sales resulting from
provisions available in SIT Module is also compared in the same figure. The estimated emissions are far
above that attributed to generation within the state since the Virginia imports significant amount of power
from across the matt, as is the case for many other states. If at all there is any need for state specific GHG
inventory, it is more to serve as an awareness of the lewd on the state. Regional emissions cannot be
related to climate change. It is also interesting to be aware that indirect CG2 emissions based on
consumption gives the lowest emissions for all the power consumed in the state because it is difficult to
attribute specifically emissions for each kwh of power imported. This is show n in Figure 3.6,

Accurate assessment of GHG emissions will also he useful in some economic programs like
Regional Greenhouse Gas Initiative (RGGI), that Virginia along with many other states are participaling.

-------
75

2 60

Emissions Based on Heat Input & Emission Factors
Emissions Based on GHGRP Report
Emissions Based on indirect CQ2 Emissions

2011 2012 2013 2014 2015 2016 2017 201B 2019 2020 2021 2012 2021 2024 2025 202S 2027 202* 202-9 2010 2031 2032 2033

YEAR

Figure 3.6: Estimated emission* for Total Power Usage in Virginia

4JI Industry: Figure 4.7 shows a comparison of emissions between disaggregated and state level
computed values.

DnnkglErrfalrd Kmihutrnt
SIT OeUuit

<.l|{,kl* t fruaifil

Figure 4.7: Comparison of Gross Industries Emission

-------
Different scenarios considered are:

(!) SIT Default; and

(2) GHGRP Oriented.

GHGRP emissions for fossil fuel combustion are arrived at on the basis of 2010 to 2019 reported
emissions for all the 44 sources, with their emissions assessed for other years based on the corresponding
annua! fuel throughput from the Departmental Point Sources database. Similar estimations are made for
all other point sources not covered by GHGRP.

There is a good convergence between disaggregated values
-------
Only difference between SIT & GHGRP oriented is that due to emissions from municipal solid
wastes. A large number of solid waste facilities reporting to GHGRP makes a difference. This is the only
sector, showing significant divergence in recent years between disaggregated & state emissions.

Figure (>M: ( ompiirittin of hums ions for ( omnierciitl Sector

7,0 Residential Sector: Figure 7.10 show the comparison of emissions between disaggregated values
and that obtained from SIT. Emissions from two sets are consistent and almost identical.

illBI jm*

run.ni.nr wars

YEAR

l-igure 7,10; Comparison of Emissions for Residential Sector

-------
8,0 Review Conclusions: On the basis of individual sectoral analysis, difference between the
disaggregated individual sectoral emissions & gross values from the state is computed.

Gross as well as the sectoral emissions in the draft inventory disaggregated from National
Inventory being the least compared to any state values, and mindful of the statement,

"Differences in the accounting decisions do not necessarily indicate that one of the estimates
is accurate, or "correct", but it will make comparability more difficult".

The extent of disparity in emissions is evaluated by the difference between the draft report and
the least of the corresponding state computed value.

Figure 8.11 shows the results of the evaluation. There are only two points beyond +10
percent error in recent years, after 2005. White line represents gross emissions.

Disaggregated values indicated in the draft report being reviewed are consistent with state
emissions obtained from SIT and/or GHGRP oriented methods with +10% accuracy. More
accurate values for emissions for Transportation arc obtained based on VMT travelled & vehicle
specifics, obtainable with the use of SIT Module.

M.IMI
H.IM
M.M

31.1H
IMC

| 19.IMI

j «.IMI
cc
-

-20.M
¦M.W
-4«.M
-SO.M

Figure 8,11; Compulation of Consistency of Disaggregated Emissions with State Values

Kotur S. Narostmhan

Point Source Inventory Specialist

Air Data Analysis

Virginia Department of Environmental Quality
Richmond VA

-------
4. Independent Organization Responses

4.1 U.S. Climate Alliance on Behalf of Alliance Member States

MEMO

To:
From:
Date:
Subject:

U.S. Environmental Protection Agency, GHGInventory@epa.gov
U.S. Climate Alliance Secretariat
November 5, 2021

Charge to State GHG Inventory Experts for the Draft Methodology Report: Inventory of
U.S Greenhouse Emissions and Sinks by State

Context

The U.S. Climate Alliance is a bipartisan coalition of two dozen U.S. governors collaborating to address
the climate crisis. Through its Emissions Working Group, Alliance states work with one another to
identify and overcome challenges towards regularly measuring and publicly reporting progress towards
their own climate goals.

On September 17, 2021, the U.S. Environmental Protection Agency requested feedback on its Draft
Methodology Report: Inventory of U.S Greenhouse Emissions and Sinks by State. The below set of
comments, intended to inform the final methodology report and future data updates, were developed
in consultation with Member States and do not represent an official U.S. Climate Alliance position.

Some Climate Alliance states raised concerns related to the timing, purpose, and the accuracy of data
relied upon in the Draft Methodology Report. Therefore, we request EPA address the following prior to
public release:

Ensure data consistency across federal data sources and tools

• State-level data should match with other federal sources, including EPA's State Inventory
Tool (SIT), National Emissions Inventory, and Greenhouse Gas Reporting Program, and
the Energy Information Administration's State Energy Data System. EPA should also
confirm that they have not omitted industrial process emissions erroneously for some of
the states as initial review suggests that there is an inconsistency between this assessment
and the list of facilities that have been reporting to the GHGRP.

• Because many states rely on SIT as the starting point for their own official GHG
inventories, having SIT match EPA's state inventory data would allow states to see the
input data and make adjustments based on their own methodologies and data sources as
relevant. Having two conflicting data sets from EPA can cause confusion and would
require states to then have to justify why they are using one over another.

• Where EPA's final state inventory data are unable to be consistent with other federal data
sources, EPA's methodology document needs to clearly identify and discuss the specific
differences for each state.

Emphasize important caveats

-------
• For states that develop their own inventory, EPA should highlight these individual state
GHG inventories as the preferred source of state-specific GHG emissions.

• For states that do not develop their own inventory, this down scaled data can be a useful
source of broad trends and major GHG emissions source identification. However, an
important benefit of a bottom-up inventory is being able to track whether policies are
working or not. There is concern that, in some cases, EPA state inventory estimates will
not be able to reflect targeted state policy and therefore should not be used in that
manner. EPA should make sure to include appropriate caveats and disclaimers that speak
to this point and make it clear that state inventories take priority for policymaking and
tracking. EPA should consider undertaking a more thorough review with outside
reviewers as well as ensuring that there is sufficient information provided that states are
able to repeat EPA's analysis.

• EPA is preparing to publish 2019 state GHG inventory data; however, due to the timing
and availability of different data sources (including updates to EPA's SIT), many official
state inventories will lag. To avoid confusion and potential conflicts with official state
inventories, EPA should make it very clear why it is able to provide more recent
estimates than states.

• EPA should consider adding uncertainty ranges for all source categories for at least the
current year and common baseline years like 1990 and 2005, which would allow states to
compare uncertainty to their own inventory data.

Align publication of the EPA's final inventory data with EPA's SIT

• Because many Alliance states rely on EPA's SIT as the basis for their own state-specific
GHG inventory, EPA should consider releasing its downscaled inventory data only after
EPA's SIT is released for the same data year.

• Understanding that it is EPA's intention to reflect the downscaled state data in the next
iteration of SIT, EPA should consider providing a detailed description and quantification
of all data updates to help states describe potential differences in previous GHG
inventories. EPA should also be available for direct consultation with states.

• EPA should consider thoroughly explaining any remaining differences in methodologies
between the final state inventory data and the SIT in addition to identifying all source
categories that do not align yet

Sector-specific notes:

• Aviation. Due to the cross-boundary and allocation issues with aviation, EPA should
consider adding additional caveats explaining the lack of a standardized allocation
methodology . As a source category many Alliance states are actively working to
improve, Alliance states would welcome continued discussion on this topic.

• MSW landfill calculations. It appears that the oxidation was removed after recovery for
years 1990-2004 (flat 10%). For years 2005-2019, the oxidation was removed before the
recovery (flat 19.5%) which implies that oxidation may be overestimated for years 2005-
2019 (Section 6.1.1 in draft methodology report, with the data in Appendix F). EPA
should clarify if oxidation is removed after recovery or why the oxidation appears to be
removed after recovery for one period and before recover for another period.

-------
Agriculture & Land Use, Land Use Change, and Forestry (LULUCF). EPA should
ensure that for (1) Agricultural Soil Management and Croplands and Grasslands, states
are given an appropriate amount of time to review disaggregated data once it is released
and (2) Wetlands and Livestock Emissions, states are given the opportunity to contribute
state-level data with the opportunity to review disaggregated data and amended
methodologies once state input and data has been incorporated.

-------