CLIFF NOTES to this blog…
“….in using the .01 tpy per well method, that 250 tons per year Barnett Shale wide Benzene leaks is dead on when backing into Arlington’s 1.4% Benzene contribution per the Truman Chesapeake padsite’s PBR of .01 tpy Benzene per well method.”
and comments from Mr Ramon Alvraez in response…
“I think you math is correct but I think you are extrapolating too far from a single ratio of benzene-to-methane content and a single emission estimate from a permit”.
I am thankful that Ecowatch reported of the Center for Public Integrity’s study that (I boldfaced for emphasis)…”At its peak, the SoCal Gas leak emitted 58,000 kilograms of methane per hour. By comparison, researchers with universities in Colorado and Michigan, partnering with the Environmental Defense Fund, estimate around 60,000 kilograms are spewed every hour by more than 25,000 natural gas wells in operation on the Barnett Shale—with the Dallas/Fort Worth Metroplex at the center. This amounts to around 544,000 tons of methane every year.
“UPDATE there is an agreement that an underestimate of 33 times exists, but it is only in the Benzene ratio to Methane proportion that I brought up NOT THE METHANE leaks”.…. I boldfaced for emphasis on the Mr Alvarez’s responses…
“Then, 0.03% is 30 times higher than the BSEEC estimate of 0.001%. That difference of 30x is what I was alluding to when I said it was in the ballpark of your 33x error in our methane estimate. Once you account for the different benzene to methane ratio, the 33x error you allude to no longer exists.”
From: Ramon Alvarez <RAlvarez@edf.org>
To: “email@example.com” <firstname.lastname@example.org>
Cc: David Lyon <email@example.com>; Daniel Zavala-Araiza <firstname.lastname@example.org>; “Steve Pacala (email@example.com)” <firstname.lastname@example.org>; Steven Hamburg <email@example.com>
Sent: Tuesday, March 1, 2016 8:07 AM
Subject: RE: Correction ? to Reconciling divergent estimates of oil and gas methane emissions
Dear Ms. Feil,
Steve Hamburg asked me to respond to your recent email. Please see below.
We reviewed your piece that uses our recent estimate of Barnett Shale oil/gas methane emissions as the basis for an estimate of benzene emissions in the region. The ratio of benzene to methane is expected to vary greatly across the Barnett region, given the spatial variability in raw gas composition. Processed natural gas also has much of the benzene removed, so that the ratio of benzene to methane will differ for upstream and downstream sources. Such spatial heterogeneity was observed in the Barnett Shale for the ethane-to-methane content in natural gas emitted across the region.
Because of this variability, there is very large uncertainty in selecting a single, “correct” value for the ratio of benzene to methane in the emissions of natural gas in the Barnett. We would not consider the value from BSEEC to be reliable, since the reported value was for dry gas wells and since the BSEEC might tend to claim a lower benzene content in natural gas to divert attention away from the benzene issue. EPA has developed an estimation tool for oil/gas emissions that includes regional gas composition data.* Again, we want to emphasize that using a single regional value carries a high degree of uncertainty. Nevertheless, we note that the EPA estimation tool reports ratios of methane and benzene (relative to total organic compounds, TOC) that yield a benzene to methane fraction of about 0.03(%). Fortuitously, this is in the ballpark of the 33x discrepancy that you noted in your calculation.
Gas Wells – Weight fraction Methane /TOC
Gas Wells – Weight fraction Benzene /TOC
Due to such great uncertainty in the relationship of methane to benzene emissions, we would not place much confidence in benzene estimates that were not derived from direct, regionally explicit measurements.
*United States Environmental Protection Agency, Oil and Gas Emission Estimation Tool 2014 Version 1. (2015). Available atftp://ftp.epa.gov/EmisInventory/2014/doc/OIL_GAS_TOOL_2014_NEI_PRODUCTION_V1_0_PACKAGE.zip
END EDF RESPONSE————————————————-
Begin edited blog to reflect EDF response…
Agreeing with the number that 544,000 tons of Methane leaks from the Barnett Shale annually (which could be **5.44 tpy associated Benzene per low balling using industry estimates) would mean buying into that Arlington TX contributes over *HALF (3.5 tpy) of the associated Benzene with those numbers….wrong!
- Per the EDF response, the 544,000 tpy Methane leaking includes upstream (end user-ready-most-of-the-Benzene-removed) and so I will attempt to get Barnett Shale wide Benzene reading based on a “per well” calculation so as to NOT include the upstream methane leaks that EDF says does not contain as much Benzene.
- Per the EDF response of the variability of wet/vs/dry gas in the Barnett Shale, I’ll be conservative and use only the dry gas estimate (less Benzene than in wet gas) of the Truman/Cowboy Stadium drill site to figure the Barnett Shale wide Benzene.
Per the recent Truman and GM Chesapeake padsites I had the TCEQ recently audit, I’ll use the Truman padsite to be more representative of Barnett Shale padsites with 4 gaswells and two lift compressors on site. Chesapeake reported on the PBR/permit to TCEQ their Benzene to be .04 tpy. Divide that by 4 wells= .01 ave. tpy/Benzene per (dry gas) well.
So .01 tpy Benzene per well times 25,000 Barnett Shale wide wells = 250 tons per year Benzene.
Take 350 gaswells in Arlington times.01 tpy and that is 3.5 tpy Benzene city wide.
If I use the EPA .03% Benzene to methane ratio for Total Organic Compounds, then the Barnett Shale is said to have (544,000 tpy x .0003 Benzene ratio) = 163 tons per year (but we still have to pull out the upstream/purified methane leaks NOT containing as much Benzene). So to be clear using the (non advised) EPA .03% to arrive at 163 tpy for Barnett Shale wide Benzene is risky as well as overstated with upstream methane numbers that have had most of the Benzene purified/removed.
So my Barnett Shale Benzene 250 tpy leak number I arrived at based on “benzene per dry well” is 87 tons higher than the EPA’s risky/overstated 163 tpy number if we use .03%. But it is CLEAR that the BSEEC’s 5.44 tpy Barnett Shale wide number (if we use their .001%) is grossly understating the Benzene floating around in our arished.
Arlington’s 350 gas wells only represent 1.4% of the 25,000 Barnett Shale wells, to check my work, take 3.5 tpy (Arlington’s Benzene contribution) and divided by 250 tpy (Barnett wide using the .01 tpy/per well method) =.014 which EQUALS Arlington’s 1.4% representation.
Likewise take 3.5 tpy (Arlington’s Benzene contribution) and divided by 163 tpy (risky/overstated EPA .03% Benzene Barnett wide) =.02 or 2% which OVER represents Arlington’s 1.4% Arlington representation/contribution to the Barnett Shale’s emissions, therefore 163 tpy is TOO LOW.
This tells me that in using the .01 tpy per well method, that 250 tons per year Barnett Shale wide Benzene leaks is dead on when backing into Arlington’s 1.4% Benzene contribution per the Truman Chesapeake padsite’s PBR of .01 tpy Benzene per well method.
Per….https://www.michigan.gov/documents/deq/deq-ess-sara-releasecalcs_306022_7.pdf “Benzene (CAS number 71-43-2) is a CERCLA hazardous substance listed in the “List of Lists.” The reportable quantity (RQ) for benzene under CERCLA is 10 pounds.
Have a safe breathing, drinking, and farming experience in the “Barnett Shale of unknown emissions contributing largely to the worlds run-away climate change”.