1. EPA’s Nonpoint Oil and Gas Emission Estimation Tool for the 2014 NEI Regi Oommen and Mike Pring, Eastern Research Group Jennifer Snyder, U.S. EPA

2. 2 Overview of Presentation NEI Basics Introduction/Timeline of the Tool Development 2011 NEI Oil and Gas Tool Development of the 2014 NEI Oil and Gas Tool Case Studies from Oklahoma Next Steps/Q&A

3. National Emissions Inventory (NEI) The full NEI is on a 3-yr cycle (e.g. 2008, 2011, 2014) Point sources (facility-process for ~100,000 facilities) Nonpoint and mobile sources (county-process) Fires (daily/point) Biogenic soil and vegetation (county) States, locals, and tribes are required to submit CO, SO X, NO X, VOC, PM 10, PM 2.5, NH 3, and Lead. Basis is National Ambient Air Quality Standards (NAAQS) parts of the Clean Air Act (Sections 110 and 172) Use CAA-based emissions thresholds for “point”. States can go lower. Hazardous Air Pollutants (HAPs) and GHGs can also be voluntarily submitted EPA augments the data to make HAPs more complete 3

4. Role of States versus EPA The presumption of the NEI program is that states are responsible for the emissions estimates EPA methods are assumptions about activity and emissions rates that can be improved with local understanding While EPA can create and promote emissions factors & methods, states choose the methods to apply We allow states to meet their obligations by accepting EPA estimates where we have them While we ask for some “method” information from states, we do not always get it (e.g., activity and emission factors) 4

5. The NEI is a composite 5

6. NEI Uses A critical input for many EPA analyses Inputs for detailed air quality and risk modeling at national, regional, and local levels EPA NAAQS-related rules and Regulatory Impact Analyses (RIAs) Planning ambient monitoring network locations Factor in designations of nonattainment areas Large scale summaries and trends assessments A resource to many outside EPA Upholding international reporting treaties Research 6

7. Examples of 2011 NEI Uses Ozone NAAQS review proposal RIA and final RIA National Air Toxics Assessment (NATA) (OAQPS) Transport analysis (OAP) State modeling platforms for Ozone SIPs (State/regional air planning) Heavy Duty GHG, Phase 2 Rule (OTAQ) Report on the Environment (ORD) 2013 Canada-US Transboundary PM Science Assessment (ORD) 7

8. 8

9. Timeline for 2014 Tool Version 1 of 2014 Oil & Gas Tool will be posted on CHIEF website by November 30, 2015 http://www3.epa.gov/ttn/chief/net/2014inventory.html http://www3.epa.gov/ttn/chief/net/2014inventory.html States have until January 15, 2016 to submit to the NEI for version 1 Corrections to version 1 can be submitted by February 3, 2017 to be incorporated into version 2 of the NEI 9

10. 10 2011 NEI Oil and Gas Tool Prior to the development of the 2011 EPA Oil & Gas Tool, oil and gas emission inventories were sparse within the NEI: Few states had the capacity to develop estimates Calculations were complex Obtaining information was difficult, even within state agencies Source category and pollutant coverage were incomplete nationwide

11. 11 2008 NEI Nonpoint Oil and Gas

12. 12

13. 2011 NEI Oil and Gas Tool Established National Oil and Gas Committee (2011) OAQPS/EIAG, Westar/WRAP, MARAMA, CenSARA, LADCO, states Prepared white paper regarding need for complete estimates EPA funding/resources became available to develop Tool in early 2012 EPA Inspector General Report (February 2013) “EPA Needs to Improve Air Emissions Data for the Oil and Natural Gas Production Sector” Concluded that “…the NEI likely underestimates oil and gas emissions” 13

14. Collaborative Efforts 14 National Oil and Gas Emissions Committee (meets monthly) Internal agencywide EPA Oil and Gas Team that includes regional experts, reg writers, EF developers, modelers Reviewing data from new studies from EDF, NOAA, UT and others Working closely with OAP to incorporate GHG EI and subpart W GHGRP data and methods into the Tool

15. What is the EPA Nonpoint Oil & Gas Emission Estimation Tool? Access ® -based emissions estimation calculator County-level activity data County-level process characterization data Emission factors down to the county/basin level Generates county-level emission estimates by SCC Exports to Emission Inventory System (EIS) format for NEI State/Local/Tribal (SLT) uses States, local agencies and tribes can input local parameters For NEI or State Implementation Plan work EPA uses For NEI, where SLTs do not submit nonpoint O&G emissions To augment HAPs from criteria pollutants 15

16. 16 How Does the Tool Work? “Bottom up” inventory Based on estimation methodologies developed by CenSARA Engineering equations, ideal gas law, mass balances, emission factors Estimates on a county level using activity factors (e.g., well counts, oil production, gas production) Refer to “2011 Nonpoint Oil and Gas Emission Estimation Tool” (November, 2014) for details2011 Nonpoint Oil and Gas Emission Estimation Tool ftp://ftp.epa.gov/EmisInventory/2011nei/doc/ Tool_and_Report112614.zip New version of the tool for estimating 2014 emissions will be posted on CHIEF by end of November 2015

17. 17 Source Category Coverage Exploration Sources:  Drilling  Mud Degassing  Hydraulic Fracturing  Well Completions

18. 18 Source Category Coverage Production Sources:  Artificial Lifts  Associated Gas  Condensate Tanks  Crude Oil Tanks  Dehydrators  Fugitives  Gas-Actuated Pumps  Heaters  Lateral/Gathering Compressors  Liquids Unloading  Loading Operations  Pneumatic Devices  Produced Water  Wellhead Compressors

19. where: E engine are emissions from a particular type (rich vs. lean) of compressor engine [ton/year/engine] EF i is the emissions factor of pollutant i [g/hp-hr] HP is the horsepower of the engine [hp] LF is the load factor of the engine t annual is the annual number of hours the engine is used [hr/yr] F controlled is the fraction of compressors of a particular type (rich vs. lean) that are controlled CF i is the control factor for controlled engines for pollutant i 907,185 is the unit conversion factor g/ton Example Calculation 19 Rolled up to the county level based on well counts

20. 20 Pollutant Coverage Criteria Pollutants CO, NH 3, NO x, PM 10, PM 2.5, SO 2, VOC HAPs BTEX, formaldehyde, and Other HAPs Other Pollutants GHGs (intermediate calculations) Hydrogen sulfide

21. 21 Data Sources Methodologies CenSARA Texas calculation tools Activity Data HPDI State OGC databases EIA RigData Emission Factors EPA AP-42 API Climate Registry Basin Factors CenSARA WRAP State feedback

22. 22 Data Flow

23. Does the Tool Take into Account Different Geological Formations? Can specify input parameters at a county/basin level Because of the different geologies of each basin, the gas or oil can look very different Example: New Mexico has a lot of coal bed methane, NE Pennsylvania has dry gas, SW Pennsylvania has very wet gas Equipment differences (conventional, fractured, CBM) While some states and RPOs have supplied specific information, we could use more 23

24. 24 2011 Results Source category coverage 54 SCCs from 18 source categories Pollutant coverage 51 pollutants Geographic coverage 33 states 1168 counties 64 basins

25. 25

26. 26 Re-engineering for 2014 NEI Why re-design? Based on user feedback… Access can be “intimidating” Use of Forms can alleviate Data files are big Created 2 separate modules (exploration and production) It’s hard to find my data 2014 Tool focuses on states

27. 27 Re-engineering for 2014 NEI What’s new? Use of data entry forms Embedded Excel import/export utility Side-by-side EPA defaults along with data used Use of “Dashboard” view with buttons and tabs Filters and large-scale updates

28. 28 Dashboard View

29. 29 Dashboard View

30. 30 Activity Data View

31. 31 Basin Factor View

32. 32 Emission Factor View

33. 33 2014 Tool Updates Activity Data HPDI and state OGC data obtained Fall 2015 Basin Factors/Gas Composition Data Incorporate new/updated data from EPA/states/RPOs Inputs recently available from 2014 subpart W Emission Factors Incorporate new/revised factors Basin Factors Incorporate new/updated data from EPA/states/RPOs

34. 34 GHG EI and NEI Though different pollutants, whole gas can be compared Use same activity data, emission factors and equipment counts, where possible, e.g.: HPDI Subpart W We are actively working to understand the effects of incorporating the newly released 2014 GHGRP data.

35. 35 2014 GHGRP Subpart W Process characterization Equipment counts (per well) Control characterization Emissions characterization Liquids unloading Well completions Synchronization of data used in the National GHG EI to be coordinated with OAP

36. 36 2014 GHGRP Subpart W Associated gas venting Flaring % Dehydrators Flaring % Per well counts Condensate/Oil tanks Flaring % Vapor recovery % Equipment leaks Component counts Heaters Per well counts Pneumatic devices Per well counts of high, low, and intermittent bleed devices

37. Case Studies from Oklahoma Refining tool inputs using industry surveys and permit data 37

38. Ongoing challenges 38 Controls New regulations (i.e., 2012 OOOO, state regulations) need to be accurately accounted for Improving modeling results Gas speciation profiles The tool can be greatly improved by states providing gas composition data and other basin-specific parameters.

39. and even more challenges… Some studies using top-down measurements (e.g. airplane, tower) and ambient air measurements (reverse modeling) suggest we’re underestimating VOC Reconciliation work on a basin level needed Fat tail analysis should be conducted Many users outside of CenSARA commented that using CenSARA defaults overestimates emissions in their regions States and MJOs have subsequently updated with locally appropriate data in many areas However, many areas still use the CenSARA average basin defaults 39

40. EPA Priority Data Needs 40 Data to help represent super-emitters (the fat tail problem) Review of new studies (EDF, NOAA, others) Abandoned wells—population and emission rates? (ORD) Overall need for local-specific inputs and emission factors Gas composition data Engine fleet data Storage tanks - emission rates per barrel throughput

41. 41 Develop defensible methods for agencies to use for air quality planning and management At the formation level in each basin, across exploration & production With respect to current and changing future practices integrate: Existing reservoir decline curves vs. new formations & production methods Expectations for future production Commodity prices Commodity distribution and waste/by-product mgmt. systems Episodic and regular practices Evolution in source type distributions and regulations Projecting Future O&G Emissions

42. 42 Discussion/Q&A How you can help Q&A