Methane in Soil Gas and Its Migration to the Atmosphere in the Dawanqi Oilfield, Tarim Basin, China
Methane microseepage is the result of natural gas migration from subsurface hydrocarbon accumulations to the Earth’s surface, and it is quite common in hydrocarbon-prone basins. In this study, by analyzing gas concentrations and isotope composition of soil gas, the potentials of CH4 gas transferred to the surface were studied at three measurement transects in Dawanqi oilfield, Tarim Basin, China. It was found that CH4 from deep-buried reservoirs could migrate upwards to the surface through faults, fissures, and permeable rocks, during which some CH4 was oxidized and the unoxidized methane remained in the soil or was emitted into the atmosphere. Soil gas samples had mean concentrations of 907.1, 62.3, 21.7, 11.0, and 5.8 ppmv for CH4, C2H6, C3H8, C4H10, and C5H12, respectively. The C1/C2+ (13.3 for soil gas and 3.75 for absorbed gas) and gas wetness ratio (12% for soil gas and 26% for absorbed gas) suggested that the hydrocarbons were derived from a thermogenic process. According to isotope composition analysis, the δ13CCO2, δ13CCH4, and δDCH4 values for the soil gas from Dawanqi oilfield varied from -15.5 to -17.2‰, -11‰ to -17‰, and -150 to -189‰, respectively. The extreme 13C enrichment in CH4 is possibly because of the fractionation effects of diffusional migration and methanotrophic oxidation. Soil gas and absorbed gas showed high CH4 concentrations at the edge of the fault block, which indicated that fault was conductive to gas migration. Also, gas migrated from the surface to the atmosphere in the center region of the fault block because of the high permeability and shallow depth of the reservoir in Dawanqi oilfield.
Methane (CH4) is a principal greenhouse gas, and its concentration in the atmosphere has risen from ~0.7 ppmv to ~1.9 ppmv now in the past 300 years . CH4 is currently responsible for approximately 20% of the direct radiative forcing from all long-lived greenhouse gases (2.3 W/m2) and for one-third of total radiative forcing, including the indirect effects of CH4 emissions (3 W/m2) such as changes in ozone and stratospheric water vapour concentrations . The estimation and prediction of its future concentration in the atmosphere have been a key aspect in the study of global environmental changes.
Due to lack of seepage flux data, geological CH4 seepage was considered to be a negligible or minor contributor to atmospheric CH4 concentrations than other natural sources, such as wetlands, termites, wild animals, and the ocean [3–6]. In the recent years, numerous studies have shown that geologic CH4 emission could play an important role in the atmospheric CH4 budget, mainly due to CH4 emissions from petroleum seepage through faults, fissures and permeable rocks, mud volcanism, marine seeps, and geothermal manifestations (Klusman et al. 1998b, [7–10]). Gas seepage includes macroseeps, which is the visible gas manifestations such as oil and gas seeps and mud volcanoes (either onshore or shallow/coastal offshore), and microseepage, which is the invisible, diffuse exhalation of gas from the ground, typically occurring in correspondence with petroleum fields. The published research has shown that microseepage in sedimentary basins can overcome the methanotrophic consumption occurring in dry soil, especially in the winter season [11, 12]. The CH4 fluxes in sedimentary basins normally are a few units or tens of mg m-2 day-1 and may be at the hundreds level in active tectonic and faulted regions [13–17]. Global microseepage emission in hydrocarbon-prone basins is about 10–28 Tg y-1, which cannot be ignored in the atmospheric CH4 budget [18, 19].
In this work, the CH4 flux and soil gas from a petroleum field in China (Dawanqi field) are presented. This study is aimed at identifying the unusual CH4 sources and the potential of CH4 gas transferred from the ground to the atmosphere by analyzing soil CH4 concentration and δ13CCH4 of soil gas and by comparing CH4 in soil gas with CH4 flux survey data.
2. Geology of the Study Area
The Dawanqi oilfield, located in the Baicheng sag of the Kuche Depression (Figure 1), northern Tarim Basin, is a distinctive oil accumulation area and characterized by an arid climate, strong salt-base reaction in soils, and rare vegetation. The type of soil is defined as “Gobi” (Luo et al., 2014) because of wind erosion and land desertification. Soil is mainly composed by brown and yellow sand. The biochemical and human interference is minimal in Dawanqi because this kind of soil is not suitable to cultivation. It is a typical petroleum system of the Tarim Basin with sandstone reservoirs located in Quaternary and Neogene (Kangcun) formations. These reservoirs are relatively shallow (170–700 m) and have high permeability (1–; [20, 21]). The hydrocarbons, oil and gas, are generated in Triassic and Jurassic coal-bearing formations .
Paleogene gypsum, located between source rocks and reservoirs, acts as an impermeable barrier for vertical oil and gas migration, but a regional fault, the Tuzimazha fracture (Yang et al. 2006), allows for fluid transfer to the upper permeable strata, as schematized in Figure 2. Oil and gas generated in the deep Triassic and Jurassic formations migrate to the Quaternary and Neogene (Kangcun) formations along the faults. Subsequently, oil and gas will horizontally move to higher strata along the reservoirs. The Tuzimazha structure is one of many fault systems that characterize the area, and the fault systems are well developed in the Dawanqi area (Figure 3).
The Dawanqi oilfield is an ideal site for investigating geologic CH4 seepage. The main reasons include the following: (1) Existing faults in the Dawanqi oilfield provide channels for the vertical migration of oil and gas to the surface. (2) Reservoirs are characterized by shallow burial depths. Gas migrates to the surface easily because of the short migration distance, permeable stratigraphy, and faults. (3) The Dawanqi oilfield is at hydrostatic pressure. Even if the field is under active production, the reservoir pressure may be less than the hydrostatic pressure provisionally, but the formation water will charge the reservoir pore space by water injection in order to maintain production. (4) The climate is arid, and the surface evaporation is strong, which is favorable to oilfield water diffusion and evaporation. (5) The near surface is thick gravel and sandy soil, so the soil aeration is propitious to measurement. (6) The biological and human interferences are few because of the arid climate in the Dawanqi area.
3.1. Flux Measurements
The microseepage measurements were performed in August 2014. According to the geological structure and surface geochemical database, fault blocks 105 and 109 were chosen as the study area, which were the main oil and gas resource development region in Dawanqi oilfield . Gas fluxes were measured at 51 points, with positioning recorded by a handheld GPS, along three transects as shown in Figure 3. Each measurement was based on accumulation times of approximately 20 min. The sampling distance varied from 50 to 300 m, depending on the presence of suitable ground conditions for closed flux chamber installation. As control and for comparison, two measurements were done outside the oilfield, about 50 km from the field boundary. Gas fluxes were measured by using a polyethylene closed chamber (radius of 37 cm and a height of 12 cm) connected to portable CH4 and CO2 sensors (915-0011, Los Gatos Research). Transparent polyethylene chambers, with a cylindric base, were inserted 4-5 cm into the soil. A translucent polyethylene cap sealed the flux chamber from the surrounding atmosphere, as described by Tang et al.  (Tang et al. 2008). Soil temperatures were measured with an Hg-thermometer, which was pushed into the soil to the depth of 5 cm. As control and for comparison, two flux measurements were done outside the oilfield, about 50 km from the field boundary.
3.2. Soil Gas and Absorbed Gas Samples
The soil gas and absorbed gas samples were also collected at 51 points along three measurement transects. The soil and gas samples were collected from 3 m below the ground surface by use of a hollow core drill machine. The soil gas samples were collected from the interior tube of a drill pipe using a gas-tight syringe. Before the soil gas was collected, the air inside the pipe was purged to avoid atmospheric contamination. There were 59 soil gas samples in total, with 51 samples for gas concentration measurement and eight samples for isotopic composition measurement. The gas samples were stored in a 500 ml Pyrex bottle. The absorbed gas samples were collected (51 samples in total) from the soil on the spiral drill pipe and stored in aluminum foil bags. Background samples were collected outside the oilfield, about 50 km from the field boundary. The background samples were collected at five sites, regularly distributed throughout the background area.
3.3. Laboratory Measurements
Laboratory measurements were determined in the Wuxi Research Institute of Petroleum Geology, Research Institute of Petroleum Exploration and Production. Absorbed gas samples were processed by pyrolysis desorption hydrocarbon technology. The soil samples were placed in a vacuum and held at 200°C for 45 minutes, releasing the gases absorbed by soil. The releasing gases were analyzed for gas composition after NaOH treatment to remove CO2.
Concentrations of C1-C4 hydrocarbons in soil gas and absorbed gas samples were analyzed by using an HP-5890 gas chromatograph equipped with a flame ionization detector (FID). Gas separation was on a stainless steel column packed with a 13x molecular sieve of a 60/80 mesh. The temperatures of the column and FID were 55°C and 200°C, respectively. Analytical error was 0.11–0.25%. To ensure comparability of the CH4 data, all concentrations were converted to standard temperature and pressure conditions (STP: 0°C, 101.325 kPa).
The δCCH4, δCCO2, and δDCH4 were determined on eight samples from the Dawanqi oilfield. The isotope measurements were conducted with an on-line system interfaced with a Thermo Quest GC/TC (Thermo Finnigan, Bremen, Germany) and an isotope ratio mass spectrometer (IRMS) (Thermo Quest Delta plus XP, Thermo Finnigan, Bremen, Germany). Isotopic ratios were determined in the Wuxi Research Institute of Petroleum Geology, Research Institute of Petroleum Exploration and Production, as described in detail by Tang et al. . The working condition of the mass spectrometer was adjusted to an ion source pressure of 3.0 k V, an ion source current of 1.5 mA, and an ion source heating current of 6.0 mA. The chromatographer interfaced with the mass spectrometer was operated with a C-2000 column (). He (99.999%) was used as carrier gas with a flow of 12.0 ml/min. For the purpose of separating CH4, CO2, and CO, a program of temperature was as follows: 30°C for 10 min, ramping up to 200°C at 15°C/min and holding for 2 min at 200°C with an (NiO/CuO/Pt) temperature of 9°C. The measure error was less than 0.4‰.
4. Results and Discussion
4.1. Methane Fluxes
Table 1 shows the CH4 and CO2 flux data. CH4 fluxes ranged from -1.4 to 329.9 mg m-2 d-1, with a mean value of 17.5 mg m-2 d-1. CO2 fluxes ranged from 0 to 76.8 g m-2 d-1, which were in the normal range of soil respiration fluxes. Of the 51 individual chamber measurements, 40 produced positive flux values (78% of the total) and 11 had negative fluxes. The microseepage database of Etiope and Klusman  shows that, statistically, positive fluxes occur in less than 50% of a petroleum field’s area, and there are three possible levels of microseepage. Based on this, total methane emissions to the atmosphere in the Dawanqi field were evaluated be on the order of 60 t/yr . Negative fluxes are a consequence of methanotrophic activity when the speed of diffusion and infiltration of methane from the oil reservoir is lower than that of its methanotrophic oxidation activity. Negative fluxes are typical, normal in dry soil, as a consequence of methanotrophic activity, and for this reason, soil is a net sink for atmospheric CH4 in most environments . Negative fluxes (both 1.4 mg m-2 d-1), as expected, were measured in the background, the control area, located 50 km from the petroleum field boundary. The positive fluxes obtained in correspondence with the petroleum field indicate that methanotrophic oxidation is not able to consume all microseeping CH4, resulting in substantial loss of CH4 into the atmosphere even at summer temperatures when methanotrophic activity is highest.
4.2. Gas Composition Analysis
The soil gas and absorbed gas samples at the test site were analyzed for C1–C5 hydrocarbons (Tables 2 and 3). The gas concentration of background samples are shown in Table 4. In soil gas samples, the average concentrations were 907.1, 62.3, 21.7, 11.0, and 5.8 ppmv for CH4, C2H6, C3H8, C4H10, and C5H12, respectively. Absorbed gas samples had average concentrations 5.2, 1.8, 0.3, 0.2, and 0.1 μl/kg soil for CH4, C2H6, C3H8, C4H10, and C5H12, respectively. The standard deviation and range show significant variations. The hydrocarbon concentrations in the Dawanqi oilfield were significantly higher than the background area. The hydrocarbon concentrations of the Dawanqi field sites were 10 to 100 times higher than the concentrations in the background area. In background soil gas samples, there were few samples where C2H6, C3H8, and C4H10 were detected.
The soil gas and absorbed gas samples had many anomalously high values of CH4, C2H6, C3H8, C4H10, and C5H12 concentrations. Methane may be derived from either thermogenic or bacterial processes. The wet gases (ethane, propane, butane, and pentane) are believed to be derived from only thermogenic sources . The mean C1/C2+ ratio for soil gas and absorbed gas was 9.1 and 3.75 (Table 3), which is in the range of thermogenesis natural gas (). The gas wetness ratio , which includes not only ethane and propane but also butane and pentane, is a common parameter to help evaluate bacterial vs. thermogenic origins . Samples with a wet gas fraction greater than 5.0% are most likely derived from a thermogenic process . The gas wetness ratio of all 51 soil gas samples are greater than 5% in soil gas and absorbed gas (Table 3). The C1/C2+ and gas wetness ratio suggest that the hydrocarbons are derived from a thermogenic process.
4.3. Isotope Composition Analysis of Soil Gas
Table 2 shows isotope composition analysis of the soil gas at Dawanqi oilfield. The δ13CCO2 (-15.5‰ to -17.2‰) indicates that CO2 is derived from organic matter, possibly with some contribution from methane oxidation [29–31]. The δ13CCH4 and δDCH4 values for the soil gas from Dawanqi oilfield varied from -11‰ to -17‰ and -150 to -189‰, respectively. Research has shown that δ13C of CH4 in the reservoir at Dawanqi oilfield was distributed between -17.9‰ to -38.18‰, which is in the range of thermogenic natural gas . The microseepage feature in Dawanqi oilfield shows a 13C enrichment in CH4, due to the fractionation effects during diffusion migration and oxidative degradation near surface. Huaiping et al.  suggested that the enriched trend is mainly due to the fractionation effect of diffusion during migration. The CH4 molecules with depleted carbon isotopes would be dissolved during diffusion migration (the water table is 4.2 m underground). As a result, the carbon isotopes would be fractionated or become enriched upward along the stratigraphic section. Furthermore, the methanotrophic activity of CH4 near surface is associated with a fractionation that enriches the residual CH4 in the heavier isotopes of both carbon and hydrogen because methanotrophs preferentially consume 12C [15, 34]. The methanotrophic oxidation could not consume all microseepage CH4, leaving the residual CH4 isotopically enriched. And higher soil porosity in the thick gravel and sandy soil near surface in Dawanqi oilfield enhance also permeability for the diffusional migration of natural gas and provide a suitable environment for methanotrophs. Therefore, it is possible that fractionation in the diffusion process and methanotrophic activity may be important reasons for the enrichment in δ13CCH4. Another oilfield in the northern Tarim Basin, the Yakela condensed gas filed, had shown similar variation in δ13CCH4 . The δ13C isotope showed an obvious enriched trend (-38.1‰ to -3.5‰) from deep strata to the near surface.
4.4. Comparison of CH4 in Flux, Soil Gas, and Absorbed Gas in Dawanqi Oilfield
Dryland soil is considered to be a net biotic sink of atmospheric CH4 because of the methanotrophic oxidation [35, 36]. However, in the tectonically active area and sedimentary basin hosting the oil and gas reservoir, microseepage could overcome the methanotrophic consumption occurring in dry soil throughout large areas, especially in the winter season [3, 13, 37, 38]. Gas seepage to the surface is strictly related to the existence of two geological features, a gas migration source and a preferential route for gas migration. The preferential routes of gas flow are zones of enhanced permeability (i.e., sand horizons within a clayey sequence) and tectonic discontinuities (i.e., faults and fracture networks). Normally, the secondary permeability caused by the fracturing and faulting induced by tectonic movements is the leading factor controlling CH4 seepage [39–41]. These channels provide minimum loss of pressure drive and loss of flow rate.
Table 5 shows the flux and CH4 in soil gas of representative test sites in Dawanqi oilfield. Compared to the absorbed gas, CH4 in soil gas has a better correspondence with CH4 fluxes, especially in faulted zones. At the edge of the fault block, the locations of anomalously high CH4 fluxes, such as the 18-7, 545-12, and 543-28-3, also had anomalously high CH4 concentrations in soil gas. Similarly, the sites of low fluxes had low CH4 concentrations 3 m below the surface. In the central zones of the fault block, most sites of anomalously high CH4 concentration in soil gas (such as 542-12, 558-8, and 545-13) exhibited low flux at the surface. Therefore, CH4 fluxes of faulted zones are generally higher than unfaulted zones.
The shallow reservoir and active tectonic structure are conducive to gas migration to the near surface. Both soil gas and absorbed gas samples showed high CH4 concentration. Figure 4 is the soil gas areal distribution of the study area in Dawanqi. This is consistent with the measurements in this study and relevant petroleum exploration data from the surrounding area . Soil gas CH4 showed high concentrations in the DW105 and DW106 regions, where the absorbed soil gas CH4 measurements were also anomalously high. In the DW109-5, DW109-7, and DW109-15 regions, soil gas samples showed high CH4 concentration, while the CH4 concentration in absorbed gas is low. Both gases showed high concentrations in the fringe of the fault block because the faults and fractures in faulted zones provided a channel for gas migration. As shown in Figure 4, high CH4 concentration also appeared in the center region of the fault block in the near surface. The sandstone in the Quaternary and the Kangcun formation was in the nonlithified to partially lithified state. The soil is characterized by enhanced permeability, shallow burial depth, and poor cementation increasing migration of gas. Gases could vertically transfer to the surface along the rock pores and fractures, showing anomalously high CH4 concentration in the central region of the fault block. On the surface, few locations in the center of the fault block (such as 555-16 and 329.9 mg m-2 d-1) had high fluxes. Most high CH4 fluxes appeared along the fault. Considering some of high flux points in faulted zones are located in the immediate vicinity of production wells, for example 18-7, there is also the possibility of methane leaks along the columns of the production wells [42, 43] but CH4 fluxes near most wells in the Dawanqi oilfield do not have obvious high concentration. It seems that faulting was the key factor for controlling the CH4 seepage in the Dawanqi area. Gas also migrated from the surface to the atmosphere in the center region of the fault block because of the high permeability and shallow reservoir in Dawanqi oilfield.
In Dawanqi oilfield of the arid northern Tarim basin of China, CH4 was prone to migrate upwards by microseepage from the shallow reservoirs to the surface through faults, fissures, and rock pore networks. The soil gas and absorbed gas samples showed high concentrations of hydrocarbons and consistency in locations of anomalies. The abnormally high concentrations of CH4 were mostly distributed along faults, which show that faults are the key factor for controlling the CH4 seepage. Microseeping gas at Dawanqi showed an enrichment in δ13CCH4. The extreme 13C enrichment can be explained by the fractionation effects of diffusional migration and methanotrophic oxidation. The enhanced permeability and shallow reservoir in Dawanqi oilfield create a favorable environment for migration by diffusion. The diffusion may also play an important role in gas migration in the Dawanqi area. The extreme 13C enrichment also indicates that the CH4 in the migration process is strongly affected by methanotrophic oxidation due to the dry climate in Dawanqi and the summer sampling season. However, the methanotrophic oxidation is not able to consume all microseeping CH4, resulting in a substantial loss of CH4 into the atmosphere. The methanotrophy rate is obviously affected by seasons. Therefore, a future study of CH4 migration and oxidation in winter is needed.
The data used to support the findings of this study are included within the supplementary information files (available here).
Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.
This work was supported by the National Natural Science Foundation of China (Grant Nos. 41872126 and 41373121). The authors would like to thank Giuseppe Etiope (Istituto Nazionale di Geofisica e Vulcanologia, Rome, Italy) and R.W. Klusman (Colorado School of Mines, USA) for his valuable comments and suggestions. We also appreciate Mr. Suo Xiaodong (Bureau of Geophysical Prospecting Inc., China National Petroleum Corporation) for offering oil exploration information.
Supplementary information file 1 had molecular and isotopic composition of soil gas samples and absorbed gas samples, and we processed the data. Supplementary information file 2 had gas concentration analyses of soil gas and absorbed gas in the background area. (Supplementary Materials)
P. Ciais, C. Sabine, G. Bala et al., “Carbon and other biogeochemical cycles,” in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of IPCC, T. F. Stocker, D. Qin, G.-K. Plattner et al., Eds., Cambridge University Press, Cambridge, UK, 2013.View at: Google Scholar
R. W. Klusman, M. E. Jakel, and M. P. LeRoy, “Does microseepage of methane and light hydrocarbons contribute to the atmospheric budget of methane and to global climate change?” Association of Petroleum Geochemical Explorationists Bulletin, vol. 11, pp. 1–55, 1998.View at: Google Scholar
M. Prather, D. Ehhalt, F. Dentener et al., “Atmospheric chemistry andgreenhouse gases,” in Climate Change 2001: The Scientific Basis, J. T. Houghton, Y. D. Ding, D. J. Griggs et al., Eds., pp. 239–287, Cambridge University Press, Cambridge, UK, 2001.View at: Google Scholar
G. Etiope, G. Papatheodorou, D. P. Christodoulou, G. Ferentinos, E. Sokos, and P. Favali, “Methane and hydrogen sulfide seepage in the northwest Peloponnesus petroliferous basin (Greece): origin and geohazard,” AAPG Bulletin, vol. 90, no. 5, pp. 701–713, 2006.View at: Publisher Site | Google Scholar
R. W. Klusman, “A geochemical perspective and assessment of leakage potential for a mature carbon dioxide–enhanced oil recovery project and as a prototype for carbon dioxide sequestration; Rangely field, Colorado,” AAPG Bulletin, vol. 87, no. 9, pp. 1485–1507, 2003.View at: Publisher Site | Google Scholar
J. H. Tang, Z. Y. Bao, W. Xiang, S. Y. Qiao, and B. Li, “On-line method for measurement of the carbon isotope ratio of atmospheric methane and its application to atmosphere of Yakela condensed gas field,” Environmental Sciences, vol. 27, pp. 14–17, 2006.View at: Google Scholar
J. H. Tang, Z. Y. Bao, W. Xiang, and Q. H. Guo, “Daily variation of natural emission of methane to the atmosphere and source identification in the Luntai fault region of the Yakela condensed oil/gas field in the Tarim Basin, Xinjiang, China,” Acta Geologica Sinica, vol. 81, no. 5, pp. 801–840, 2007.View at: Google Scholar
H. W. Kuang, S. Z. Niu, and X. L. Chen, “The diagenetic characteristics and its controlling factors of the Kangcun formation in the Dawanqi oilfield, the Tarim Basin,” Geosciences Journal, vol. 17, pp. 211–216, 2003.View at: Google Scholar
H. W. Kuang and G. C. Jin, “Reservoir characteristic and evalution of the Kangcun formation in the Dawanqi oilfield, Tarim Basin,” Journal of Geomechanics, vol. 11, pp. 81–89, 2005.View at: Google Scholar
M. J. Zhao, Y. Song, S. B. Liu, and S. F. Qin, “The diffusion influence on gas pool: Dawanqi oilfield as an example,” Natural Gas Geoscience, vol. 14, pp. 393–396, 2003.View at: Google Scholar
Chinese Petroleum Corporation, Dawanqi Surface Oil and Gas Geochemical Exploration Report, China Petrochemical Press, Kuerle, China, 2003.
X. D. Suo and D. Y. Yu, “Application of geochemical exploration in the on-going development of Dawanqi oil field, Xinjiang, China,” Natural Gas Geoscience, vol. 16, pp. 64–68, 2005.View at: Google Scholar
J. H. Tang, W. Xiang, Z. Y. Bao, S. Y. Qiao, and Q. H. Guo, “Effect of geogenic emission of methane on the atmosphere,” Geological Science and Technology Information, vol. 25, pp. 75–82, 2006.View at: Google Scholar
G. Etiope, A. Feyzullayev, A. V. Milkov, A. Waseda, K. Mizobe, and C. H. Sun, “Evidence of subsurface anaerobic biodegradation of hydrocarbons and potential secondary methanogenesis in terrestrial mud volcanoes,” Marine and Petroleum Geology, vol. 26, no. 9, pp. 1692–1703, 2009.View at: Publisher Site | Google Scholar
Q. L. Wang, J. P. Bao, Z. G. Xie, and S. G. Wang, “Geochemical characteristics and genesis of natural gas in the Dawanqi oilfield,” Journal of Yangtze University, vol. 9, pp. 28–31, 2012.View at: Google Scholar
Z. Huaiping, C. Tongjin, L. Wu, and W. Guojian, “Features of methane carbon isotope and patterns of hydrocarbon migration in Tabei area,” Oil & Gas Geology, vol. 26, pp. 450–455, 2005.View at: Google Scholar
R. W. Klusman, M. E. Leopold, and M. P. LeRoy, “Seasonal variation in methane fluxes from sedimentary basins to the atmosphere: results from chamber measurements and modeling of transport from deep sources,” Journal of Geophysical Research: Atmospheres, vol. 105, no. D20, pp. 24661–24670, 2000.View at: Publisher Site | Google Scholar
L. Leiter, “Leaky oil wells in Allegheny National Forest prove Pennsylvania needs conventional drilling rules more than ever,” 2018, https://earthworks.org/blog/leaky-oil-wells-in-allegheny-national-forest-prove-pennsylvania-needs-conventional-drilling-rules-more-than-ever/.View at: Google Scholar
University of Vermont, “New study first to predict which oil and gas wells are leaking methane,” 2018, https://phys.org/news/2018-12-oil-gas-wells-leaking-methane.html#jCp.View at: Google Scholar