Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://geophys.geol.msu.ru/shevnin/publ/179.pdf Äàòà èçìåíåíèÿ: Wed Oct 17 15:56:41 2012 Äàòà èíäåêñèðîâàíèÿ: Sat Feb 2 23:48:35 2013 Êîäèðîâêà:

STUDY OF PETROLEUM CONTAMINATED SITES IN MEXICO WITH RESISTIVITY AND EM METHODS
Vladimir Shevnin, Mexican Petroleum Institute, Mexico city, D.F. Omar Delgado Rodriguez, Mexican Petroleum Institute, Mexico city, D.F. Aleksandr Mousatov, Mexican Petroleum Institute, Mexico city, D.F. Hector Zegarra Martinez, Mexican Petroleum Institute, Mexico, D.F. Jesus Ochoa Valdes, Mexican Petroleum Institute, Mexico city, D.F. Albert Ryjov, Moscow State Geological Prospecting Academy, Russia.

Abstract
Hydrocarbons are among the main factors of geological medium contamination. We differentiate long-term contaminations lasting years or decades of years and short-term contaminations or single accidents. The first produces give more evident geophysical anomalies, whereas anomaly strength of the second depends on the time since the accident occurred. After 6-12 months following the accident this type of contamination gives measurably low resistivity anomalies. Our experience with contaminated sites characterization in Mexico shows that low resistivity anomalies caused by hydrocarbon contamination is possible to localize with the help of vertical electrical sounding (VES) or with electromagnetic profiling (EMP). Such contamination gives low resistivity anomaly as a result of petroleum biodegradation at shallow depth in the earth. It is more difficult to characterize the second type of contaminated sites because the anomalies are not as intensive. Short-term contamination is more abundant in the oil industry.

Introduction
Study of oil biodegradation with the help of laboratory reactors (Atekwana et al., 2001) and in the field demonstrates that the most evident changes (decreasing) of ground resistivity take place above groundwater level in the lower part of the vadose zone (Abdel-Aal et al., 2001). In field studies, such low resistivity layers were found in the lower part of the vadose zone by W. Sauck (1998, 2000), D. Werkema et al. (2002, 2003) with the help of vertical resistivity probe (VRP). During last year of investigation our group studied 8 different sites of oil contamination in Mexico. These sites are different in contamination age and scale, depth of groundwater level (GWL), environment and surface conditions and the cause of contamination. Some information about these sites is in Table 1.

167


Table 1.: List of contaminated sites studied with geoelectrical methods (VES and/or EM profiling) N Name of site Origin of Environment Surface Depth of. Cont. contamination GWL, m age, years 1 Paredon, Tabasco Oil borehole rural Open 1.5-2 35 2 3 4 5 6 7 8 Campo ­ 10, Veracruz Reynosa, Tamaulipas Km 42, Tabasco La Venta, Tabasco Altace, Mexico city Sanchez Magallanez, Tabasco Km 124, Tabasco Oil wastes rural & industrial Open utilization plant Oil refining rural Open 60% factory Pipeline accident rural Open Pipeline accident Oil products storage Pipeline accident Pipeline accident rural industrial rural rural Swamp Concrete 90% Swamp Open 1.5-2 10-15 3-4 0 4 0 1-4 30 50 1 1 15 1 1

This field and interpretational experience demonstrated possibilities (and restrictions) of VES and EMP methods and their interrelation with geochemical methods of contamination study. Preliminary study of contamination area with geoelectrical methods before application of geochemical methods has more advantages when groundwater level depth is more than 1 m, when superficial layer is hard enough to obtain soil samples, when the ground is covered with asphalt or concrete layer. In all these cases geoelectrical methods give valuable information for planning and optimizing geochemical probing. And geoelectrical methods can give more detailed maps of contamination zones than geochemical sampling. Depth position of contamination obtained with geoelectrical methods help to avoid perforation into aquifer to diminish risk of petroleum contamination of this layer during sampling process. VES method gives valuable information about aquifer and aquiclude layers, helps to calculate their filtration coefficients on known formulas (Salem, 2001) and to estimate aquifer vulnerability (Kirsch et al., 2003).

168


Geoelectrical Results and Considerations
VES results visualization. Vertical electrical sounding (VES) application on contaminated site has the purpose to characterize A contamination with depth and in plan. For that purpose a system of profiles is used. Each site is being crossed by profiles with multi electrodes measuring technology called in some papers as Electrical Resistivity Tomography (ERT) or Electrical Imaging (EI). Such profiles measurements allow two-dimensional interpretation. We used for 2D 1 B interpretation software Res2DInv (Loke and Barker, 1995, 2 3 1996). 4 This 2D interpretation has several good features that 5 made the interpretation results suitable for different 6 visualizations. Resulting model has the same layers number 7 8 for all soundings along all profiles of the site. Thicknesses of Figure 1.: Matrix of apparent resistivity all blocks in the same layer are equal. With the help of data obtained with vertical electrical additional software X2IPI f, % 0 soundings along profile on Electrical (Bobachev, 2003) it is possible to convert Imaging technology (A) and matrix of f, % 1 Res2DInv model into IPI 2D data interpretation (B). Profile 6 format model having the Profil e 5 same blocks widths in all Profil e 4 Profile 7 f, % 2 Profile 8 layers and reference points Profil e 3 Profil e 2 exactly below each Profile 1 sounding point (Figure 1). Layer 1 f, % Layer 2 After that with the help of 3 Layer 3 IPI2Win software .......... .......... (Bobachev, 1994), we .......... f, % .......... 4 create files in Surfer format .......... both as sections for each Layer N f, % Figure 2.: Presentation of VES survey profile and maps for each 5 layer. Due to these along profiles as data cube. possibilities we call VES data collection for each site as data cube (Figure 2). In some cases we f, % 6 construct not only horizontal maps, but maps for any surface inside the cube (dipping or curved). The last feature is useful when we have evident layers dip at the site (Delgado et al., 2004). f, % 7 What form of VES data visualization is better, sections or maps? Visualization in sections has less interpolation between measuring points. For maps construction we need interpolate between Figure 3.: Histograms for each profiles. But maps have less resistivity range than sections (electrical horizontal layer in data cube properties change more with depth than in plan) (Figure 3). As a result for the site Km124 (1-7) and we can reach more resolution in maps visualization that gives us histogram for all resistivity possibility to localize weaker anomalies. To increase visualization sections (0). 169
Profile M N

1 2 3 4 5 6 7 8 9

AB /2 (m)

Profile

De pt h (m)

15 5 0

10

17

10

20

50

100

200

500

1000 ,Ohm.m

10

0

10

20

50

100

200

500

1000

,Ohm.m

20

10

0

20

50

100

200

500

1000

,Ohm.m

15

10

0

20

50

100

200

500

1000

,Ohm.m

17

10

0 20 20

50

100

200

500

1000

,Ohm.m

10

0

20

50

100

200

500

1000

,Ohm .m

20

10

0

10

20

50

100

200

500

1000

,Ohm.m

17

10

0

4

7

10

20

50

100

, Oh m . m


resolution we apply statistical distribution analysis (histograms) to adjust resistivity range (Figure 3) and color scale range. Without special adjustment of resistivity range frequently it is not possible to localize oil contamination anomalies on sections and maps. Table 2.: List of initial and adjusted histogram ranges (Figure 3) Sample Rho min, Rho max, Range Adjusted collection Ohm.m Ohm.m max/min Rho min, Ohm.m All data 10 1600 35 160 Layer 1 20 350 30 17 Layer 2 65 1000 120 15 Layer 3 100 1500 200 15 Layer 4 60 1600 140 27 Layer 5 55 1000 67 18 Layer 6 38 140 38 7 Layer 7 6 80 13 13 Mean value of range 16 In case of aged contamination resistivity ratio between uncontaminated and contaminated soil is between 2 and 5. To localize anomaly of contamination we need to use maximum resolution of visual presentation and it is easier to obtain in case of maps. But both forms sections and maps help us to formulate idea of contamination distribution in space.
P aredo n

Adjusted Adjusted Rho max, range Ohm.m max/min 700 20 220 7.3 700 5.8 1300 6.5 1100 7.9 420 6.3 140 3.7 65 5 6
GWL dep th: 0-2 m
X, m

0 -4 -8 -12 Z,m0

8

16

24 6

32 10

40 16

48

56 27

64 45

72 74

80

88 121

96

104

Pro file 9
0 -5 -10 -15 -20 Z,m0

,

Ohm. m

Reyn osa

GWL dep th: 10-15 m

Contamination Indicator's Horizon 20 40 60 80 100 120 140 X, m 160 W. Sauck and his colleagues applied for Profile 6 2.7 7.4 20 55 150 400 a, Ohm.m localization of contamination with depth vertical resistivity probing (VRP) method that is a kind Figure 4.: Appearance of contaminated zones on 1140 resistivity sections.
Y, m
1120

,

Ohm.m

,O

h m .m
55

1100

33

1080 20 1060 12 1040 7.4

1020

1000

4.5

Figure 5.: Resistivity map for contamination indicator's horizon at the site Paredon. 170

980 900

X, m
920 940 960 980 1000 1020 1040 1060 1080


of resistivity log in shallow boreholes (Sauck, 1998; 4400 2000; Werkema et al., 2002). They found that aged Y, m contamination appears as low resistivity horizon a 4300 bit above ground water level. Vertical electrical sounding in multi electrode configuration and 2D interpretation can 4200 also localize this layer. In some cases it is possible to see this layer in resistivity cross-sections (Figure 4100 4) in other cases in maps. We called this horizon as contamination indicator. To characterize oil 4000 contamination in plan we build resistivity maps at the level of the contamination indicator's horizon. When we can localize this layer on resistivity 3900 sections we reorganize data to make a map of this horizon (Figures 5-6). Such map gives optimal 3800 presentation of contamination position in plan.
PM 1

1900

2000
01

2100
31

2200
PM 6

2300

2400

,ohmPM 5 6

150 90 55

m

11 9 PM 4

33
14

20 12 7.3

16

4.5 2.7

17 PM 3

1.6
20

1 0.6

3700

3

Recalculation of Soil Resistivity into Petrophysical Contaminated zones Parameters 3600 Soil resistivity depends on water content and 1900 2000 2100 2200 2300 2400 X, m its salinity, clay content, porosity and some other Figure 6.: Resistivity map for contamination factors. There are a lot of models describing dependence of soil resistivity from these factors indicator's horizon at the site Reynosa. (Archie (1942), Waxman and Smits (1968) and many others). We use petrophysical model developed by A.Ryjov (Ryjov, 1987; Ryjov, Sudoplatov, 1990). If model describes dependence of soil resistivity from some factors we have a chance to find some of these petrophysical factors from soil resistivity data. In laboratory we measure soil resistivity versus water salinity and estimate clay content, porosity and cation exchange capacity (Shevnin et al., 2004). It is possible also to estimate these three parameters for soil resistivity (estimated from VES interpretation) and constant water salinity. Using this approach we can recalculate resistivity crosssections and maps into petrophysical cross-sections and maps. For uncontaminated soil these parameters (clay content, porosity and cation exchange capacity) are close to true petrophysical parameters, estimated with traditional methods in laboratory. For contaminated soils we receive anomalous
1140
1140

0.37

Y,m
1120

Clay content, %
60 50

Y,m
1120

PM-1

CEC, g/l
55 33

1100

1100
S-1 3

S-10

S-1

1080

40

1080
PM- 4 S-1 2 S-9

20

S-2
S-3

PM- 5

1060

30

1060
S-11 S-7

AV

12

1040

20

1040

S-8 PM- 3 S-6

S-4

7.4

1020

10

1020

S-5 PM- 2

4.5

1000

0

1000
PM- 6

2.7

980 900

X, m
920 940 960 980 1000 1020 1040 1060 1080

980 900

X, m
980 1000 1020 1040 1060 1080

920

940

960

Figure 7.: Maps of clay content and cation exchange capacity for the site Paredon. 171


parameters, but these parameters help us to localize contamination. Mature contamination gives increased clay content and CEC and anomalous porosity (increased or decreased depending on clay content in soil). This apparent change of petrophysical parameters reflects real increase of superficial conductivity in contaminated zones (Abdel Aal et al., 2004). For example direct measurements of clay content at site Paredon (both geological laboratory determination and geoelectrical sampling (C) curve) shows that maximum clay content is 40%. But at the map of clay content in figure 7 contaminated areas show anomalous clay content between 40 and 60% (boundary line is 40%). Application of Electromagnetic Profiling Method for Oil Contamination Mapping Electromagnetic profiling can be also used Y, m for oil contamination mapping. We applied EM-31 , Ohm.m instrument (Geonics) for this purpose. In some cases we obtained similar information like VES (Figure 4), but 8-10 times more rapidly (Figure 8). EM has higher horizontal resolution, whereas VES has good vertical resolution. In some areas we can use both methods. In other situation (concrete cover with steel bars) EM profiling can't be applied. Sometimes VES method is difficult to apply (for example in swamp areas) and EM profiling is the only possible method of geophysical mapping. We visualized EM profiling results in the form of apparent resistivity instead of conductivity X, m maps to make these results similar to VES Figure 8.: Apparent resistivity map of the site resistivity maps. Paredon obtained on EM mapping with EM31. Because of these maps (from VES and EMP) similarity we performed recalculation of EMP resistivity maps into petrophysical maps (taking into account groundwater salinity). This operation is not correct because we use apparent resistivity instead of true resistivity, but in some cases we received good results similar to VES method (Figure 9 in comparison with Figure 7). To recalculate EMP resistivity maps into petrophysical maps we need information about water salinity (or water resistivity) that is not a problem in swamp area.
266 0
a

264 0

50 40 33 27 22 18 15 12 10

262 0

260 0

258 0

256 0

8.2 6.7

254 0

954 0

956 0

958 0

960 0

962 0

964 0

966 0

Y,m
2640

2660

CEC, g /l CEC 33

Y,m
2640

2660

Cla y con te nt, %

Clay
30 25

2620

20 2620

20 2600 12 2600 15 2580 7.4 2580 10 2560 4.5 2560

5

2540

2.7

2540

0

2520 9520

X, m
9540 9560 9580 9600 9620 9640 9660 9680

2520 9520

X, m
9540 9560 9580 9600 9620 9640 9660 9680

Figure 9.: Maps of clay content and cation exchange capacity for site Paredon estimated for apparent resistivity map (Figure 7). 172


Example of Contamination Study at the Site Km 124. Oil contamination at this site occurred one 260 Por.,% Y,m year before our study as a result of pipeline 250 24 leakage. Area of accident is situated at 25 meters 240 from small artificial lake used for fish-breeding. 22 This area has a lot of rains that sometimes 230 PE2 20 changes water level in the lake at 2.5 meters. As a 220 result groundwater level also changes and oil 18 contamination is smearing in soil at 2-3 meters 210 16 depth interval. This situation is demonstrated at 200 X, m figure 10 as a series of maps for the depth 1, 2, 3 14 360 370 380 390 400 410 420 430 440 450 and 4 meters approximately. In this case we use 260 soil porosity like contamination indicator, but Y,m Por.,% 35 other parameters (soil resistivity, clay content and 250 cation exchange capacity) show similar anomalies 240 resulted in oil contamination. 34 At figure 11 there are histograms for 4 230 PE2 parameters: resistivity, clay content, porosity and 220 cation exchange capacity. Grey intervals show 33 210 anomalous values f, % Resistivity of each parameter 200 A X, m resulted in 32 360 370 380 390 400 410 420 430 440 450 contamination. In 260 f, % Clay Por.,% our opinion clay Y,m 35 B 250 content and porosity in 240 34 f, % figure11 have Porosity C 230 33 more resolution PE2 220 between 32 f, % uncontaminated 210 CEC and contaminated D 31 zones. That is why 200 X, m 30 Figure 11.: Histograms for we used porosity 360 370 380 390 400 410 420 430 440 450 demonstrate 260 resistivity, clay content, to Por.,% contamination in Y,m porosity and CEC (Km 124). 36 250 Gray blocks - anomalous figure 10. Figure 12 240 values of each parameter. 35 demonstrate 230 physical parameters distribution with depth PE2 34 estimated from mean VES data for each layer of 220 2D interpretational model. Upper part of section 210 33 until 5 m consist of pure sand. After 5 m it is 200 sandy loam with clay content about 10%. X, m 32 Filtration coefficient was calculated using the 360 370 380 390 400 410 420 430 440 450 P ipe li ne z one bou ndar i es B or ehol e wi th o il co ntam ina tion formula published in the paper (Salem, 2001): 2.09 Figure 10.: Maps of porosity for 4 layers calculated K f = 0.6653 F , (m/day), where F=soil/water. In the paper (Custodio, Llamas, 1983) filtration from resistivity maps for the site Km 124.

1

20 10 0

2

, Ohm.m

10

20

30

50

70 10 0

20 0

50 0

10 00

20 10 0

Clay content, %

40 20 0

0

5

10

15

20

25

30

35

40

45

50

55

60

Por osity, %

20 10 0

14

16

18

20

22

24

26

28

CEC, g/l

0.01

0.02

0.05

0.1

0.2

0.5

1

3

4

173


coefficient for soil like in site Km 124 is in interval 1.7 - 8.6 m/day that is in good agreement with our determination (1.2 - 16 m/day) based on soil and water resistivity.
10 00



, Ohm.m

50 0

0 - Clay co ntent, % 2 4

A

0

B

0

C

0

D

0

E

2

2

2

2

20 0

4
10
10 0

4

4

4

20
50

6

6

6

6

8
20

8

8
Cl ay content, % Humidi ty, %

8
K fil tr. , m/day

10 10 20

Humidity, %
50 10 0

10



, Ohm.m

Z, m

10 20 50 10 0

50 0

Z, m

10

0

2

4

6

8

10

Z, m

10

10 2

10

20

50

Figure 12.: Distribution of resistivity (B), clay content (C), soil humidity (D) and filtration coefficient (E) with depth. A - Theoretical dependence of soil resistivity from clay content and humidity.

10 0 Z, m1

5

10 20

Conclusions
1. Vertical resistivity cross-sections normally have greater resistivity range than maps (electrical properties change more with depth than in plan) that makes their visualization with the purpose of contamination localization more difficult. In most cases we can't reach needed resolution without adjustment of resistivity intervals and color scale range with the help of histograms. Higher visual resolution obtained in maps gives us possibility to localize weaker anomalies. 2. Vertical electrical sounding in multi electrode configuration and 2D interpretation can localize contamination indicator's horizon at the aged contaminated sites when groundwater level (GWL) is at a depth from 1 to 15 m. This horizon is located close and slightly above GWL. For better characterization of oil contamination in plan we recommend preparing resistivity maps for the level of the contamination indicator's horizon. 3. For soil resistivity recalculation into petrophysical parameters we need to know groundwater salinity (determined from water resistivity). Using this approach we can recalculate resistivity crosssections and maps into petrophysical cross-sections and maps. For uncontaminated soil estimated petrophysical parameters (clay content, porosity and cation exchange capacity) are close to true petrophysical parameters, found with traditional methods in laboratory. For contaminated soils we receive anomalous parameters, but these parameters help us to localize contamination. The true cause of apparent change of petrophysical parameters in contaminated zones resulted in increase of superficial conductivity as was found by Abdel Aal et al. (2004). Joint usage of resistivity and petrophysical parameters helps receiving more detailed characterization of uncontaminated and contaminated soils. 4. EMP method faster acquires data and has more horizontal resolution in comparison with VES. Each method has its own advantages and disadvantages at contamination study. In some cases we can recalculate apparent resistivity values of EMP method into petrophysical parameters taking into account groundwater salinity. 5. Field study of contaminated sites includes VES or EMP methods and water resistivity measurements. Sometimes we make soil resistivity study versus water salinity in the laboratory to estimate soil characteristics: clay content, porosity and cation exchange capacity with maximum accuracy.

174


References
Abdel-Aal, G. Z., Werkema, D. D., Sauck, W. A. Jr. and Atekwana, E., 2001, Geophysical investigation of vadose zone conductivity anomalies at a former refinery site, Kalamazoo, ML, in Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems, 1-9. Archie, G. E., 1942, The Electric Resistivity Logs as an Aid in Determining some Reservoir Characteristics. SPE-AIME Transactions, 146: 54-62. Atekwana, E., Cassidy, D. P., Magnuson, C., Endres A. L., Werkema, Jr., D. D. and Sauck W. A., 2001, Changes in geoelectrical properties accompanying microbial degradation of LNAPL, in Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems, 1-10. Abdel Aal, G. Z.; Atekwana, E. A.; Slater, L. D.; Atekwana, E. A., 2004, Effects of microbial processes on electrolytic and interfacial electrical properties of unconsolidated sediments. Geophys. Res. Lett., Vol. 31, No. 12, L12505 p.1-4. Bobachev, A.A., 1994, IPI2Win software: http://geophys.geol.msu.ru/rec_labe.htm . Bobachev, A.A., 2003, X2IPI software: http://geophys.geol.msu.ru/x2ipi/x2ipi.html . Custodio, E., Llamas, M. R., 1983, Hidrologia Subterranea. Ed. Omega S.A. Barcelona. Delgado-RodrÌguez, O., Shevnin, V. Ochoa-ValdÈs, J. and Ryjov A., 2004, Geoelectrical characterization of a site with hydrocarbon contamination as a result of pipeline leakage. SEG2004, pp. 1448-1451. Kirsch, R., Sengpiel, K.-P. and Voss, W., 2003, The use of electrical conductivity mapping in the definition of an aquifer vulnerability index. Near Surface Geophysics, 2003, Vol. 1, 13-19. Loke, M.H. and Barker, R.D., 1995, Least-squares deconvolution of apparent resistivity pseudosectious. Geophysics, 60, 1682-1690. Loke, M.H. and Barker, R. D., 1996, Rapid least-squares inversion of apparent resistivity pseudosections using a quasi-Newton method. Geophysical Prospecting, 44, 131-152. Ryjov, A., 1987, The main IP peculiarities of rocks, In the book "Application of IP method for mineral deposits' research". Moscow, 1987, p. 5-23. (In Russian). Ryjov, A. A., Sudoplatov, A. D., 1990, The calculation of specific electrical conductivity for sandy clayed rocks and the usage of functional cross-plots for the decision of hydrogeological problems, In the book "Scientific and technical achievements and advanced experience in the field of geology and mineral deposits research. Moscow, pp. 27-41. (In Russian). Salem, H.S., 2001, Modelling of lithology and hydraulic conductivity of shallow sediments from resistivity measurements using Schlumberger vertical electric soundings. Energy sources. 23 (7): 599-618 Aug-Sep 2001. Sauck, W. A., 1998, A conceptual model for the geoelectrical response of LNAPL plumes in granular sediments, in Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems, pp. 805-817. Sauck, W. A., 2000, A model for the resistivity structure of LNAPL plumes and their environs in sandy sediments. J. App. Geophys., 44, 151 -165. Shevnin, V., Delgado RodrÌguez, O., Mousatov, A., Ryjov, A., 2004, Soil resistivity measurements for clay content estimation and its application for petroleum contamination study. SAGEEP-2004, Colorado Springs. p. 396-408. Waxman, M.H. and L.J.M. Smits, 1968, Electrical conductivities in oil-bearing shaly sands: Journal of the Society of Petroleum Engineering, 8, 107-122. Werkema, D. D. Jr., Atekwana, E. A., Endres, A. and Sauck, W. A., 2002, Temporal and spatial variability of high resolution in situ vertical apparent resistivity measurements at a LNAPL

175


impacted site. in Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems, 2002, 13esc4, 11 pp. Werkema, D. D., E. A. Atekwana, A. Endres, W. A. Sauck, and Cassidy, D. P., 2003, Investigating the geoelectrical response of hydrocarbon contamination undergoing biodegradation, Geophys. Res. Lett., 30(12), 1647, doi:10.1029 / 2003GL017346.

Acknowledgements
The authors consider a pleasant debt to express gratitude to Mexican Petroleum Institute, where in framework of the program PIMAS (Programa de InvestigaciÑn del Medio Ambiente y Seguridad) this study was fulfilled.

176