This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights
Author's personal copy
Radiation Measurements 50 (2013) 116e120
Contents lists available at SciVerse ScienceDirect
Radiation Measurements journal homepage: www.elsevier.com/locate/radmeas
Radon measurements over a natural-gas contaminated aquifer D. Palacios a, *, E. Fusella b, Y. Avila a, J. Salas c, D. Teixeira a, G. Fernández d, A. Salas d, L. Sajo-Bohus a, E. Greaves a, H. Barros a, M. Bolívar e, J. Regalado e a
Universidad Simón Bolívar, P.O. Box 89000, Caracas YV 1080, Venezuela Instituto de Estudios Avanzados (IDEA), Caracas, Venezuela c Universidad Central de Venezuela, Caracas, Venezuela d Universidad del Zulia, Maracaibo, Venezuela e Servicios Geofísicos-PDVSA, Puerto la Cruz, Venezuela b
h i g h l i g h t s < High Radon/Thoron ratios were localized near the natural-gas emanations in a river. < Natural gases are ascending trough a deep geological fault. < Apparently, the radon anomaly shows the site where natural gas enters the aquifer. < Natural gas source may be related to leaks in the structure of abandoned gas wells.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 December 2011 Received in revised form 21 September 2012 Accepted 30 October 2012
Radon and thoron concentrations in soil pores in a gas production region of the Anzoategui State, Venezuela, were determined by active and passive methods. In this region, water wells are contaminated by natural gas and gas leaks exist in the nearby river. Based on soil gas Radon data surface hydrocarbon seeps were identified. Radon and thoron concentration maps show anomalously high values near the river gas leaks decreasing in the direction of water wells where natural gas is also detected. The area where the highest concentrations of 222Rn were detected seems to indicate the surface projection of the aquifer contaminated with natural gas. The Radon/Thoron ratio revealed a micro-localized anomaly, indicating the area where the gas comes from deep layers of the subsoil. The radon map determined by the passive method showed a marked positive anomaly around abandoned gas wells. The high anomalous Radon concentration localized near the trails of ascending gas bubbles at the river indicates the zone trough where natural gases are ascending with greater ease, associated with a deep geological fault, being this the main source of methane penetration into the aquifer. It is suggested that the source of the natural gas may be due to leaks at deep sites along the structure of some of the abandoned wells located at the North-East of the studied area. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Radon Thoron Soil pores Natural gas LR-115 detector Aquifer
1. Introduction According to the inhabitants of a locality in Anzoategui state, Venezuela, there have been intense natural gas leaks (mostly methane) in several sites in the nearby river for about 10 years. Moreover, during the drilling of two water wells for public consumption there was methane gas leaks and the wells had to be sealed. Isotopic analyses indicated thermogenic origin of methane. The village is located in an area where there are many, mostly
* Corresponding author. Tel./fax: þ58 2129063590. E-mail addresses:
[email protected],
[email protected], daniel210755@ yahoo.es (D. Palacios). 1350-4487/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.radmeas.2012.10.016
inactive, gas wells. The penetration of natural gas into the aquifer can occur through naturally occurring fault conduits and fracture systems that connects with deep underground gas layers or with a breakdown somewhere along the structure of a gas well. According to Kristiansson and Malmqvist (1982), nondiffusive transport of Radon can be easily explained by carrier-gas transport, so this non-reactive gas is an ideal tracer for gas transport from deeper layers to the soil surface, and its detection is one of the indicators of hydrocarbon leaks from subsoil (Li et al., 2006; Dyck and Jonason, 2000; Ghahremani, 1985). The advantage of using Radon as tracer of light hydrocarbons lies in that the latters might be involved in shallow biological reactions (Fleischer and Turner, 1984). The location of active faults can be achieved by detecting Radon over a large area. The Radon anomalies make it possible to
Author's personal copy
D. Palacios et al. / Radiation Measurements 50 (2013) 116e120
highlight more emissive zones that are either related to main faults or to secondary fractures acting as migration pathways (Landrum et al., 1989; Hus et al., 1999; Patrick et al., 2011). If a strong Radon anomaly is not associated with uranium mineralization, most likely this anomaly is due to the ease of gas migration through highly permeable deep fracture zones possibly caused by geological faults. Irrespective of the origin, the geogases tend to migrate towards the surface due to pressure and buoyancy effects creating areas of anomalous degassing. When the geogas microflow crosses the groundwater may form a stream of bubbles (Sikka and Shives, 2001). Bubble flows related to faults may occur in different geological environments and have been theoretically and experimentally recognized as a mechanism for rapid gas migration. How to characterise hydraulic connectivity of faults/fractures with a river and alluvial aquifers and how to locate these permeable structures are becoming very critical questions for groundwater risk assessment and water management in oil and gas producing areas. The aim of this study was to determine the areal distribution of Radon and Thoron concentrations in soil gas on a large region that includes the village, for the purpose of detecting the location of structures (fault zones) and suggest which is or may be the source responsible for the occurrence of natural gas in the river and aquifer. 2. Materials and methods The study was conducted in Freites municipality, Anzuategui state, Venezuela (Fig. 1). The local topography is flat with little variation in the soil properties. Data were collected on a grid composed of eight sets of rectangular rings centered on the locations of the two contaminated water wells, four abandoned gas wells suspected of gas leaks and the river location where gas was observed to emanate. A total of 210 locations, 250 m apart, covering approximately 22 km2, were analyzed for soil gas Radon. Measurements were carried out during the dry season. In the active measurement, soil gas was sampled through steel sampling tubes at a depth of 60 cm. Radon and Thoron measurements were performed with radiation monitors (AB-5 Pylon, Ottawa, Canada) coupled with 150-cm3 Lucas cells. The monitors were adapted with a simple external filtering system to eliminate
117
moisture and small soil particles. Flow meters were also employed to ensure the cells were completely full with soil gas at the end of the pumping cycle and for Thoron corrections. The pumping cycle was for 3 min followed with nine 1-min counting periods. Radon and Thoron concentrations were calculated from the recorded count rates. Radon in soil gas was also monitored for long time periods (passive method) using LR-115 type II track detectors from Dosirad Co., France, fixed at the bottom of cylindrical diffusion chambers (65 mm diameter and 75 mm height high density plastic cups). The top of the diffusion chambers were covered with a 30 mm thick polyethylene foil to keep aerosols, Radon daughters and Thoron gas out of the measuring volume. Three cups with Radon detectors inside were located at the bottom of each hole drilled at 60 cm depth. The holes were covered with the extracted soil. The time of exposure was 45 days. Detector etching was done under standard conditions (10% NaOH solution at 60 C during 120 min). After etching, films were washed with distilled water and dried in a dust free chamber. The alpha tracks recorded on the films were counted using an optical microscope coupled to a digital camera. Track images were processed by the ImageJ free Internet access software. Relative Radon concentrations were given by the average track densities since the exposure time, etching conditions and method of track reading were the same for all detectors. Radon and Thoron contour maps were created in the GSþ program (Robertson, 2008). 3. Results 3.1. Active radon measurements Fig. 2 shows the spatial distribution of soil gas Radon concentrations in the studied area. In zone A there are no oil or gas wells, active or inactive, and as far as we know it is not considered a prospective area. The Radon maximum concentration values are near the location where gas emanates from the river and decrease in the direction to the location of the contaminated water wells. Since the aquifer is unconfined, the observed behavior seems to indicate the ascending flux of natural gas dissolved in groundwater, and hence the superficial projection of the gas-contaminated aquifer and its hydraulic connection with the river emanations.
Fig. 1. Map of Venezuela showing where the study took place (left). Location of a water well, where methane gas leak occurred, respect to adjacent gas wells and river gas emanations (right).
Author's personal copy
118
D. Palacios et al. / Radiation Measurements 50 (2013) 116e120
Fig. 2. Map of soil gas
222
Rn concentrations obtained by the active method.
Apparently, the anomaly located at zone A is related to the migration of hydrocarbon gas from some source through an interconnected fracture system or a fault. Due to the river proximity it is probable that these gas flows are related to the trails of ascending gas bubbles observed at the surface of the river. Radon high concentration was also observed in zone B where there are some abandoned gas wells. Apparently, this area is spatially connected with the sites where there are the contaminated wells. However, the relationships between the levels of Radon concentrations do not seem to indicate the abandoned gas wells as the direct and near-surface contamination source of the aquifer since in the proximity of the water wells the Radon concentrations in soil gas are higher than in zone B. The intensities of the observed anomalies indicate that the site near the river gas
emanation (A) is the main conduit of gas penetration into the aquifer. However, a relation in depth of this conduit with the gas wells in zone B should not be discarded. Fig. 3 shows that Radon/Thoron ratios are smaller than 0.5, so according to Card et al. (1985) if anomalies are found they cannot be explained on the basis of Uranium mineralization. The Radon/ Thoron ratios show an anomaly only in zone A. It indicates that both isotopes have been brought to the surface by other carrier gases from a deeper source (Hus et al., 1999). Since 222Rn has a much longer half-life than 220Rn, it can be transported from greater depths before decaying. The existence of a deep fault in that location, through which gases are transported and penetrate the aquifer, could explain the observed river gas emanations due to its hydraulic connection with the aquifer. Considering that the shape
Fig. 3. Map of Radon/Thoron ratios in the studied region.
Author's personal copy
D. Palacios et al. / Radiation Measurements 50 (2013) 116e120
Fig. 4. Map of the relative
222
119
Rn concentrations in soil pores obtained by the passive method.
of Radon anomaly across faults indicates dip direction, it can be assumed the conduit through which natural gases penetrate into the aquifer to be approximately vertical with slight inclination to the South-West, due to the rounded shape of the anomaly and the nearly sharp fall of Radon concentration at its edges. 3.2. Passive radon measurements Fig. 4 shows the relative Radon concentration map in soil pores given by the track densities induced in LR-115 detectors. The same anomaly of high Radon concentrations near the river gas emanation is apparent in zone A. However, the highest Radon concentrations were found in zone B where there are a few abandoned gas wells. That area also appears to be characterized by high soil permeability and the existence of an active flow of ascending geogases carrying Radon, since the behavior of the latter cannot be explained on the basis of lithological variations. An apparent spatial connection exists with the nearest water well that showed methane gas escape. Unlike the results of active measurements, Radon levels seem to suggest that this area could be the possible source of the aquifer contamination, since soil gas Radon concentrations gradually decreased toward the water wells. Also a spatial connection is apparent between the gas wells (B) and the site in the river where emanations takes place (A). This connection was not observed with the active measurements. In contrast to the active measurements, results obtained with passive detectors are less affected by variations of meteorological conditions and reflect average concentrations during the exposure time. A possible reason for not detecting the highest concentrations near the river gas emanation could be due to the low quantity of data in that zone as several LR-115 detectors were lost. Both Radon measurement methods indicate that the occurrence of gas emanations in the river and gas escape during the drilling of water wells are controlled by a single conduit (fault or interconnected fractures in zone A) for the ascent of natural gas to the surface. Results obtained using magneto-telluric methods also show the existence of a vertical conduit connecting the zone of the river gas emanations with very deep zones in the subsoil. Additionally, the resistivity distribution in depth indicates an apparent
connection between some of the abandoned gas wells in zone B and this conduit. Although by both measuring methods the highest soil gas Radon concentrations were obtained approximately in the same areas, which can be explained on the basis of little influence of meteorological conditions in zones of active seepage hydrocarbons along faults and fractures, this study continues repeating over time to evaluate the weight of temporal variations of the Radon and leakage of gases on the distribution of soil gas Radon concentrations. The fact that a deep fault or fracture system exists near the zone where the gas emanations are observed in the river, indicates the deep origin of the transported gases that penetrate into the aquifer. The results of soil gas Radon measurements by passive and active methods do not discard the existence of a breakdown in the structure of abandoned gas wells in zone B. Accumulated gasses may have acquired enough pressure to migrate through a system of interconnected natural fractures to the geological fault or conduit that intersects the aquifer. It is improbable that the gas comes from a deep gas reservoir as in that case it would exit with higher pressure at the river site. Additionally, it is highly unlikely that in the last 10 years methane started to migrate through natural faults and fractures from a reservoir and coincidentally reach the aquifer at the same time oil and gas development started in the studied area, after having been down there for over 65 million years. Since Radon anomalies are more intense in the neighborhood of the gas emanations in the river, this suggests that if gas leaks from some of the wells in zone B, it is not taking place near the surface or is preferably ascending by the wall of the well but rather very deep. 4. Conclusions 1. Surface hydrocarbon seeps were identified based on soil gas Radon data. 2. Through the study of the Radon/Thoron ratio it is concluded that the Radon anomalies are not due to Uranium mineralization. The anomalous zone for the Radon/Thoron ratio seems to indicate a high permeability of the soil caused by the effect of a fracture system or nearly vertical deep geological fault.
Author's personal copy
120
D. Palacios et al. / Radiation Measurements 50 (2013) 116e120
2. The high anomalous Radon concentration localized near the trails of ascending gas bubbles at the river indicates the zone trough where natural gases are ascending, being this the main source of methane penetration into the aquifer. The gas emanation in the river and in the water wells are controlled by this source. 3. The zone where the highest values of 222Rn were obtained by the active method seems to indicate the superficial projection of the aquifer into which the water well was perforated. 4. It is suggested that the source of the natural gas may be due to leaks at deep sites along the structure of some of the abandoned wells indicated in the North-East of the studied area. Acknowledgments This study has been financed by Repsol Exploración Venezuela, S.A (LOCTI project GPIE 33-2079/4916) and FONACIT (Project S12001000954). References Card, J.W., Bell, K., Denham, G.M., Shah, S.R.A., 1985. Radon decay product measurements in radiometric uranium exploration: implications for petroleum exploration. The Oil and Gas Journal 83, 114e116.
Dyck, W., Jonason, J.R., 2000. In: Hale, M. (Ed.), Geochemical Remote Sensing of the Sub-surface. Ó2000 Elsevier. Chapter 11. Fleischer, R.L., Turner, L.G., 1984. Correlations of Radon and carbon isotopic measurements with petroleum and natural-gas at cement, Oklahoma. Geophysics 49 (6), 810e817. Ghahremani, D.T., 1985. Radon prospecting for hydrocarbon: potential strategy for devonian shale gas in N.E. Ohio; PhD thesis, Department of Geological Sciences, Case Western Reserve University, p. 259. Hus, R., Dehandschutter, B., Bobrov, V.A., Acopachov, N.E., 1999. Active Fault Identification Using Radon Measurements Around Lake Teletskoye (Altai, Russia). Royal Museum of Central Africa, Annual Report 1997e1998, pp. 177e 201. Kristiansson, K., Malmqvist, L., 1982. Evidence for nondiffusive transport of Rn in the ground and a new physical model for the transport. Geophysics 47, 1444e 1452. Landrum, J.H., Richers, D.M., Maxwell, L.E., Fallgatter, W., 1989. A comparative hydrocarbon soil-gas study of the Flathead Valley and Townsend Valley areas, Montana. Journal of Geochemical Exploration 34, 303e335. Li, Y.L., Yuan, G.J., Peng, D., 2006. Discussion on geochemical exploration predicts the depth of oil reservoir (Chinese). Wutan Huatan Jisuan Jushu 28 (4), 367e371. Patrick, R., Frédéric, P., Bharat, P.K., Frédéric, G., Mukunda, B., Soma, N.S., 2011. Temporal signatures of advective versus diffusive Radon transport at a geothermal zone in Central Nepal. Journal of Environmental Radioactivity 102, 88e102. Robertson, G.P., 2008. GSþ: Geostatistics for the Environmental Sciences. Gamma Design Software. Plainwell, Michigan USA. Sikka, D.B., Shives, R.B.K., 2001. Mechanisms to Explain the Formation of Geochemical Anomalies Over Oilfields AAPG Hedberg Conference “Near-surface Hydrocarbon Migration: Mechanisms and Seepage Rates” September 16e19. Vancouver, BC, Canada.
