123
Sierra Club also compares NETL’s 1.2% leakage rate to the 2.01% leakage rate
calculated by Burnham et al.
268
Again, a boundary difference explains why the two leakage rates
are not directly comparable. Burnham et al.’s leakage rate includes natural gas distribution,
which is an additional transport step beyond transmission. Natural gas distribution moves
natural gas from the “city gate” to small scale end users (commercial and residential consumers).
NETL’s leakage rate ends after natural gas transmission, the point at which natural gas is
available for large scale end users such as power plants. The natural gas distribution system is a
highly-branched network that uses vent-controlled devices to regulate pressure. This boundary
difference explains why Burnham et al.’s leakage rate is higher than NETL’s rate. Sierra Club
also compares NETL’s leakage rate to a shale gas analysis conducted by Weber et al.
269
We
have reviewed Weber et al.’s work and do not see any mention of leakage rate.
It is also important to note that leakage rate is not an input to NETL’s life cycle model.
Rather, it is calculated from the outputs of NETL’s life cycle model. NETL uses an approach
that assembles all activities in the natural gas supply chain into a network of interconnected
processes. The emissions from each process in this model are based on engineering relationships
and emission factors from the EPA and other sources. This method is known as a “bottom-up”
approach. Researchers are trying to discern why “top-down” studies such as Pétron’s
measurements in northeast Colorado
270
do not match the bottom-up calculations by NETL and
other analysts. We believe that inconsistent boundaries ( i.e., bottom-up models that account for
long term emissions at the equipment level in comparison to top-down measurements that
268
Burnham, Andrew, et al. Life-cycle greenhouse gas emissions of shale gas, natural gas, coal, and petroleum.
Environmental Science & Technology 46.2 (2011): 619-627.
269
Weber, Christopher L., and Christopher Clavin. Life cycle carbon footprint of shale gas: Review of evidence and
implications. Environmental science & technology 46.11 (2012): 5688-5695.
270
Pétron, G., Frost, et al. (2012). Hydrocarbon emissions characterization in the Colorado Front Range: A pilot
study. Journal of Geophysical Research: Atmospheres (1984–2012), 117(D4).
124
encompass an entire region with more than one type of industrial activity over a narrow time
frame) partly explain the differences between bottom-up and top-down results. As research
continues, however, we expect to learn more about the differences between bottom-up and top-
down methods.
Zimmerman and Associates references a recent study by Ingraffea et al. that assessed
failure rates of well casings for oil and gas wells in Pennsylvania.
271
However, Ingraffea et al.
do not calculate a methane leakage rate in their analysis; rather, they calculate the rate at which
wells develop leaks. The rate at which leaks develop in well casings is a different phenomenon
than the rate at which methane leaks from the natural gas supply chain. The former is a
measurement of failure rates (the number of wells in a group that have leaks) and the latter is a
measurement of the magnitude of total leakage (the amount of methane in extracted natural gas
that is released to the atmosphere).
The breakeven analysis shown in Section 6 of the LCA GHG Report models hypothetical
scenarios that increase the natural gas leakage rate to the point where the life cycle emissions
from natural gas power are the same as those from coal power. The breakeven points between
natural gas and coal systems are illustrated in Figures 6-8 and 6-9 of the Report. These results
are based on the most conservative breakeven point, which occurs between the high natural gas
cases ( i.e., lowest power plant efficiency, longest transport distance, and highest methane
leakage) with the low coal case ( i.e., highest power plant efficiency and shortest transport
distance). These graphs show that on a 100-year GWP basis, methane leakage would have to
increase by a factor of 1.7 to 3.6, depending on the scenario, before the breakeven occurs. The
271
Ingraffea, A. R., Wells, M. T., Santoro, R. L., & Shonkoff, S. B. (2014). Assessment and risk analysis of casing
and cement impairment in oil and gas wells in Pennsylvania, 2000–2012. Proceedings of the National Academy of
Sciences, 111(30), 10955-10960.
125
breakeven methane leakage is lower for the 20-year GWP basis and, for some scenarios, is lower
than the modeled leakage rate.
6.
The Uncertainty Bounds of the LCA GHG Report
a.
Comments
Concerned Citizens claim that the LCA GHG Report has significant uncertainty, and
contend that “poor modeling is not a reason to dismiss impacts.”
b.
DOE/FE Analysis
The results of the LCA GHG Report are based on a flexible model with parameters for
natural gas extraction, processing, and transport. Uncertainty bounds are assigned to three key
parameters: well production rates, flaring rates, and transport distances. These uncertainty bars
are not an indication of poor modeling. To the contrary, they are used to account for variability
in natural gas systems. If the analysis did not account for uncertainty, the results would imply
that the GHG emissions from natural gas systems are consistently a single, point value, which
would be inaccurate. We therefore believe the chosen uncertainty bounds strengthen the LCA
model, as opposed to indicating any weakness in modeling.
7.
The LCA GHG Report and the NEPA Approval Process
a.
Comments
Several commenters, including Citizens Against LNG, Dominion Cove Point LNG,
Susan Sakmar, and Americans Against Fracking et al., note that the LCA GHG Report does not
fulfill the requirements of an EIS as defined by NEPA. These commenters maintain that the
LCA GHG Report should not be used as a basis for approving proposed LNG export terminals.
b.
DOE/FE Analysis
We agree that the LCA GHG Report does not fulfill any NEPA requirements in this
proceeding, nor has DOE/FE made any suggestion to that effect. The LCA GHG Report
Dostları ilə paylaş: |