Regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions

Transcription

1 Regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Thomas Lauvaux, Zach Barkley Kenneth Davis, Natasha Miles, Scott Richardson, Douglas Martins Department of Meteorology, Pennsylvania State University Yanni Cao, Eun-Kyeong Kim, Guido Cervone, Alan Taylor Department of Geography, Pennsylvania State University Colm Sweeney, Anna Karion, Kathryn McKain ESRL/NOAA GMD, CIRES, Boulder Tom Murphy Marcellus Center for Outreach and Research, Pennsylvania State University Penn State Extension webinar series, July 20 th, 2017

2 Regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Project sponsored by the Department of Energy National Energy Technology Laboratory

3 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Natural gas production in the United States from the major shale gas formations since 2008 (EIA data)

4 Natural Gas Leak Rate (%) Natural Gas Leak Rate (%) TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Production Processing Bottom up estimates for the US suggests that emissions from production and processing of natural gas account for less than 0.8% of total production. 10 NOAA Mass Balance estimates Year 0.8% 5 0 Karion et al. (2013;2015) Petron et al. (2012;2014) Peischl et al. (2015; 2016) Smith et al. (2016) The NOAA mass balance estimates from 9 different basins suggest that the EPA estimate is too low. The smaller producing fields are driving the average leak rate up.

5 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Project goals and objectives Overall objective: quantify fugitive emissions of CH4 from the Marcellus gas production region of north-central Pennsylvania over a period of two years ( ) 1. Generate maps of CH4 emissions over the northeastern region of the Marcellus shale formation 2. Break down of the relative contribution of various regional sources (animal agriculture, wetlands, landfills, gas production, other regional sources identified) 3. Compare to emissions-factor, activity-data based inventory regional CH4 emissions estimates, evaluating both the accuracy of the inventory methods, and our ability to detect changes.

6 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Two new elements in this approach: 1. First long-term deployment (i.e. 2 years) of several instruments to measure continuously the CH4 emissions from the natural gas production activities (only short aircraft or automobile campaign so far) 2. Use of 13CH4 to identify the contributions from biogenic sources and thermogenic sources (no calibration protocol to measure 13CH4 at high accuracy)

7 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Upwind Upwind CH 4 CH 4 CH 4 Downwind Objectives: - Regional attribution of spatial and temporal variability - Separation of source contributions using continuous 13 CH 4 /CH 4 measurements

8 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Inventory of CH4 emissions (Natural Gas Prod.) Other sources of CH4 (Farming, wetlands, industrial) Atmospheric model (1km resolution) Atmospheric concentrations of CH4 (modeled) Optimization (inversion system) Atmospheric concentrations of CH4 (observations)

9 Collection of activity data for the preparation of a bottom-up CH 4 emission map, including biogenic and thermogenic sources - Collection of inventory data to construct the a priori CH 4 emissions 1. US EPA gridded product from Maasakkers et al., (2016) 2. Updated production rates for conv. & unconv. wells for selected months 3. Compressor stations as point sources (based on air permits) from Barkley et al., under review

10 Collection of activity data for the preparation of a bottom-up CH 4 emission map, including biogenic and thermogenic sources Unconventional Production Conventional Production Distribution Animal Agriculture Coal mines/beds Landfill / Industry from Barkley et al., under review

11 Collection of activity data for the preparation of a bottom-up CH 4 emission map, including biogenic and thermogenic sources Map of the CH 4 emissions inventory for the extended Pennsylvania study domain in mol.km -2.hour -1 and our domain of interest (red box)

12 Deployment of calibrated CRDS instruments at the four identified tower locations Network design of the four instrumented towers: Two towers instrumented to measure the background mixing ratios (independent of wind direction) Background: Towers N and S Two towers downwind of the production wells, nearby the area of interest (maximum amplitude) Downwind: Towers C and E Installation of instruments - Negotiation of lease agreement for the selected towers - Installation of sheds for the duration of the deployment (no space for instruments indoor) - Tubing installed to the top of the towers, - Deployment before the aircraft campaign

13 Deployment of calibrated CRDS instruments at the four identified tower locations Definitive tower locations of the 4 towers called North (N), East (E), South (S), and Central (C). Unconventional wells are plotted in the background. Coordinates, elevations, and sampling heights of the 4 towers Photo of temporary shed (upper) and tube inlet at tower N, 46m AGL (lower)

14 Deployment of calibrated CRDS instruments at the four identified tower locations Calibration of the four CRDS CH 4 / 13 CH 4 instruments: - Four standards for 13Ch4 measurements with various levels of enrichment/depletion and concentrations - Dry air (use of Nafion) with automated valve (flow path selector) between calibration standards and outdoor inlet tube Field sampling flow schematic with automatic calibration system using four gas-phase standards of unique isotopic 13 CH 4 values ( ) and CH 4 (X) mixing ratios at ambient (amb) and heavy and high (HI) and low (Lo1, Lo2) mixing ratios, respectively. from Miles et al., in prep.

15 Deployment of calibrated CRDS instruments at the four identified tower locations Comparison between continuous CRDS mixing ratio measurements and flask samples for CO 2 (in ppm), CH 4 (in ppb), and 13 CH 4 (in permil) from Miles et al., in prep.

16 Deployment of calibrated CRDS instruments at the four identified tower locations CH 4 mixing ratio measurements over from the four CRDS CH 4 / 13 CH 4 instruments (in ppb)

17 Deployment of calibrated CRDS instruments at the four identified tower locations CH 4 mixing ratio enhancements over from the four CRDS CH 4 / 13 CH 4 instruments (in ppb)

18 Aircraft campaign to quantify CH 4 emissions over the entire northern Marcellus region NOAA Twin Otter on the tarmac of the Williamsport airport List of flights, conditions, and durations for the 12 flights of the aircraft campaign

19 Aircraft campaign to quantify CH 4 emissions over the entire northern Marcellus region Flight Show patterns flight paths for the 12 flights of the aircraft campaign with the mean wind direction during the flight (red arrows). The flight on 14 May shows highly variable winds across the domain. Colm Sweeney, NOAA Anna Karion, NIST

20 Aircraft campaign to quantify CH 4 emissions over the entire northern Marcellus region Example of flight measurements for May 24th, 2015 with vertical profiles of meteorological and greenhouse gas measurements (upper left), wind direction and spedd (upper right), methane (lower left), and ethane (lower right)

21 Quantification of emissions: Data Assimilation System Inventory of CH4 emissions (Natural Gas Prod.) Other sources of CH4 (Farming, wetlands, industrial) Atmospheric model (1km resolution) Atmospheric concentrations of CH4 (modeled) Optimization (inversion system) Atmospheric concentrations of CH4 (observations)

22 Atmospheric modeling system: WRF-CH 4 WRF-CH 4 atmospheric modeling system: - WRF-Chem, modified for CH 4, v3.5.1, - Physics configuration based on previous studies (Lauvaux et al., 2012; Deng et al., in prep.) - 3 grids in nested mode (one-way): 9km/3km/1km resolution, - Emission fields from inventory data coupled to the 3km and 1km grids, - 3 modes developed for the project: - Forecast: 9km/3km resolutions,12-hour assimilation, 24-hour forecast. - Historical: initial and boundary conditions from North American Regional Reanalysis (NARR) - Data assimilation: Four Dimensional Data Assimilation (FDDA) mode with nudging of WMO surface stations and radiosondes Evaluation Aircraft campaign of May Simulations of flight days at 1km resolution in the three different modes

23 Atmospheric modeling system: WRF-CH 4 Modeling domain to simulate the atmospheric conditions during the deployment period ( ) using the Weather Research and Forecasting model Atmospheric CH 4 enhancement from the various sectors of emissions over PA (in ppm)

24 Unconventional Production/Gathering Coal Mines May 24 th 2015 Total Enhancement NG Transmission/Distribution Enteric Fermentation Conventional Wells Landfills and Other Modeled Methane Enhancement (in ppm)

25 May 29 th 2015: A good day. Wind Speed 4.8 m/s Atmospheric CH 4 enhancement (in ppm)

26 Aircraft emissions estimate on May 29 th 2015 Enhancement (ppm) B C A D Observed CH 4 Enhancement (ppm) A B C D Observed CH 4 Enhancement measured during the flight (in ppm)

27 Aircraft emissions estimate on May 29 th 2015 Enhancement (ppm) B C A D Non-NG CH 4 Enhancement (ppm) A B C D Observed and modeled Non-Natural Gas CH 4 enhancement for the May 29 th flight (in ppm)

28 Aircraft emissions estimate on May 29 th 2015 Enhancement (ppm) B C A D Natural Gas CH 4 Enhancement (ppm) A B C D Observed CH 4 enhancement for the May 29 th flight (in ppm)

29 Aircraft emissions estimate on May 29 th 2015 Enhancement (ppm) B C A D Natural Gas CH 4 Enhancement (ppm) Emission Rate = 0.13% A B C D Observed and modeled Natural Gas CH 4 Enhancement for the May 29 th flight (in ppm)

30 Aircraft emissions estimate on May 29 th 2015 Enhancement (ppm) B C A D Natural Gas CH 4 Enhancement (ppm) Emission Rate = 0.26% A B C D Observed and optimized Natural Gas CH 4 enhancement for the May 29 th flight (in ppm)

31 CASE STUDY: MAY 24 th, 2015 The Importance of a Good Methane Inventory 31

32 Aircraft emissions estimate on May 24 th 2015 Observed CH 4 enhancement for the May 24 th flight at 20z (in ppm)

33 Aircraft emissions estimate on May 24 th 2015 Coal Plume Final Result: All Emissions UNG Emission Rate=0.29% Importance of the inventory with the projected atmospheric enhancement from Coal mines/beds explaining the observed spatial gradients

34 Best-guess upstream emission estimates Optimal mean leakage rate based on 10 flights in May 2015: 0.39% of production

35 Marcellus Mass Balance: valid over 4 days out of 10 May 22 nd, 2015 May 29 th, 2015 May 23 nd, 2015 May 29 th, 2015

36 Best-guess upstream emission estimates Optimal mean leakage rate based on 10 flights in May 2015: 0.39% of production

37 TOWER INVERSION: June to December, 2015 Time-dependent flux estimate for NE PA

38 Tower-based inversion for June-December 2015 Inversion window Diagram of the different flux sub-regions used in the inversion. Towers (green pins) and wells (pink dots) are plotted overtop Map of the percent change in the posterior flux compared to the prior flux using tower observations from Oct-Dec 2015

39 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Conclusions Successful calibration of 13 CH 4 continuous CRDS sensors (about 0.2 permil) Leakage rate equivalent to 0.4% of production (wells and gathering compressors) Site-level estimate (about 15 wells in PA), Omara et al. (2016): 0.64% of production (wells only) EPA-derived estimate for NE PA (Maasakkers et al., 2016): 0.13% of production Aircraft mass-balance (1 flight only) for NE PA (Peischl et al., 2013): 0.3% of production Initial tower-based inversion results suggest lower leakage rates over June to December 2015 Future work ACT-America study on unconventional and coal mines in SW PA and WV ongoing Perform tower-based inversion over June 2015 to December 2016

40 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Publications and references Barkley, Z. R., Lauvaux, T., Davis, K. J., Deng, A., Cao, Y., Sweeney, C., Martins, D., Miles, N. L., Richardson, S. J., Murphy, T., Cervone, G., Karion, A., Schwietzke, S., Smith, M., Kort, E. A., and Maasakkers, J. D.: Quantifying methane emissions from natural gas production in northeastern Pennsylvania, Atmos. Chem. Phys. Discuss., in review, Cao, Y., Cervone, G., Barkely, Z., Lauvaux, T., Deng, A., and Taylor, A.: Influence of Geographic Coordinate System on Weather Simulations of Atmospheric Emissions, Geosci. Model Dev. Discuss., in review, Miles, N., D. K. Martins, S. J. Richardson, C. Rella, C. Arata, T. Lauvaux, K. J. Davis, Z. Barkley, K. McKain, C. Sweeney: Calibration and Field Testing of Cavity Ring-Down Laser Spectrometers Measuring Methane Mole Fraction and Isotopic Ratio Deployed on Towers in the Marcellus Shale Region, in prep. Data from towers data archive for CH 4 and 13 CH 4 continuous tower measurements in northeastern PA: Data from the 2015 aircraft campaign Aircraft Campaign over the Northeastern Marcellus Shale (May 2015):

41 TA [2] Continuous, regional methane emissions estimates in northern Pennsylvania gas fields using atmospheric inversions Publications in preparation Barkley, Z. R., et al.: Regional inversion of methane emissions from natural gas production in northeastern Pennsylvania, in prep. Barkley, Z. R., et al.: Methane emissions from coal mines and natural gas production in southwestern PA, in prep. Project Website