Chap16

Chapter 16

Applying Magnetotellurics

by

Arnie Ostrander

Arnie OstranderArnie Ostrander is an oil and gas exploration consultant specializing in the integration ofmagnetotelluric methods and surface geochemistry in frontier basin exploration and inunderdeveloped stratigraphic plays in producing basins. He earned his B.A. in geology in1974 from the University of Montana. He began his professional career with Zonge Engi-neering and Research Organization from 1975 to 1985, was with Phoenix Geoscience, Inc.from 1988 to 1991, and has been an independent consultant since 1991.

Overview • 16-3

This chapter discusses the nature and uses of magnetotellurics (MT), a method of survey-ing the subsurface from the surface. Although MT cannot provide the resolution of seismicsurveys, it is less expensive and, more importantly, can be used in places where seismic isimpractical or gives poor results.

Introduction

Overview

This chapter contains the following topics.

Topic Page

What is Magnetotellurics (MT)? 16–4

What Does an MT Survey Measure? 16–5

How Are MT Data Acquired? 16–6

Case History: Frontier Basin Analysis (Amazon Basin, Colombia) 16–8

Case History: Rugged Carbonate Terrain (Highlands of Papua New Guinea) 16–9

Case History: Precambrian Overthrust (Northwestern Colorado) 16–10

Case History: Volcanic Terrain (Columbia River Plateau) 16–11

References 16–12

In this chapter

16-4 • Applying Magnetotellurics

Magnetotellurics is an electrical geophysical technique that measures the resistivity ofthe subsurface. This is the same physical parameter that is measured in a borehole resis-tivity log.

Definition

What is Magnetotellurics (MT)?

MT differs from an inductive electric log in three major ways:

Magnetotellurics Measurements Electric Log Measurements

Made from the surface Made subsurface from inside a borehole

Depth of investigation is a function of both frequency Depth of investigation is the depth of the at which the measurement is taken and the average borehole measuring device below the surfaceresistivity of the subsurface

Respond only to changes in average bulk resistivity Respond to individual rock layers along the wall of the borehole

The figure below shows the simplified relationship between a lithologic log, an electric log,an MT sounding, and an inversion run using the MT sounding data.

How MT differsfrom electriclogs

Figure 16–1.

We can also take electric log data and run a forward MT model to produce an MT sound-ing curve.

Subsurface layers are resolved by inverse modeling of MT data acquired across a spec-trum of frequencies, as illustrated in Figure 16–1.

Subsurfacelayers resolved

The rule-of-thumb for MT resolution for depth of burial vs. layer thickness is 10:1. Forexample, to “see” a layer at a depth of 1,500 m (5,000 ft), the thickness of the layer needsto be approximately 150 m (500 ft) or more. Low-resistivity layers are more easily delin-eated than high-resistivity layers. It is difficult for MT to resolve more than three or foursubsurface layers.

MT resolution

Applying Magnetotellurics • 16-5

Two basic alternating current (AC) measurements are taken in an MT survey: a horizon-tal magnetic field (H-field) measurement and an electrical field (E-field) measurement.The E-field is always measured perpendicular to the H-field data.

What ismeasured?

What Does an MT Survey Measure?

The H-field is the “source” signal, or the primary field. It propagates across the surface ofthe earth. Because it does not travel in the subsurface, the H-field data do not provideinformation about the subsurface geology.

Very limited information about the subsurface geology can be interpreted from the verti-cal H-field if this component is measured. The vertical H-field is called the tipper.

The horizontal H-field is measured with a horizontally oriented magnetic coil. The tip-per is measured with a vertically oriented coil.

Be careful not to confuse an MT survey with a magnetic survey. An MT survey does notmeasure the magnetic properties of the subsurface rocks, as does a magnetic survey.

The H-field

The E-field is the secondary field, generated by the H-field propagating across the surface.Each time the primary H-field (an AC signal) switches polarity, a secondary E-field (cur-rent flow) is generated in the subsurface. Thus, the horizontal E-field data provides infor-mation about the subsurface geology.

This is the same physical principle as the alternator in a car. An alternating or spinningmagnetic field (H-field) sets up current flow in the wire windings in the alternator, whichin turn charges the battery. In the case of an MT survey, the “wire” is the earth.

The E-field is measuredwith a grounded dipole typi-cally 50–200 m long. All sub-surface geology informationis contained in the E-fielddata. However, without theH-field data, we cannot cal-culate resistivity.

The figure at right showsthe relationship between the E- and H-fields.

The E-field

Figure 16–2.

The resistivity calculation is a simple ratio of the primary source signal (H-field) and thesecondary current flow in the earth (E-field), with a modifier for the frequency at whichthe data were acquired:

Apparent Resistivity =

where:E = magnitude of the E-fieldH = magnitude of the H-fieldf = frequency

EH

15f

2

×

Resistivitycalculation


The data are collected using a microprocessor-controlled voltmeter. The voltmeter is infact a system of complex hardware/software devices that includes amplification, filtering,A/D conversion, stacking and averaging, and various data-enhancement algorithms.

Acquisitioninstrumentation

How Are MT Data Acquired?

There are two types of MT surveys: natural source (Vozoff, 1972) and controlled source(Goldstein and Strangway, 1975). The equipment and the operational procedures forthese two types differ considerably.

Types of surveys

The natural-source data-acquisition system typically measures four components: Ex, Ey,Hx, and Hy. The Ex component is oriented perpendicular to the Ey component. This is alsotrue for the H-field components.

The predominant low-frequency (< 1.0 Hz) signal source for natural-source data issunspot activity. The dominant high-frequency (> 1.0 Hz) source is equatorial thunder-storm activity.

Although H-field data do not provide information on the subsurface geology (when only Hx

and Hy components are measured), the vertical H-field component—if measured—pro-vides information on the surface geology.

The figure below shows a typical MT setup for a natural-source survey.

Natural-sourcesurveys

Figure 16–3.

The controlled-source system uses a high-power transmitter and motor/generator set totransmit a discrete AC waveform. This signal is transmitted into a grounded dipole typi-cally 600–1,200 m (2,000–4,000 ft) long. The transmitter is normally located 3–6 km (2–4mi) from the survey line.

Controlled-source surveys


Normally, only the Ex (parallel to the transmitter dipole) and Hy components are mea-sured.

The figure below shows a typical MT setup for a controlled-source survey.

Controlled-source surveys(continued)

Figure 16–4.

The choice of MT method depends on the survey objectives. Natural-source data are bestsuited for regional surveys where the stations are widely spaced (e.g., frontier basinanalysis). Controlled-source data are best suited for mapping structural detail where thestations lie along a continuous profile at 100–200-m (300–600-ft) spacings. The maximumdepth of exploration for the controlled-source method is 3,000–4,500 m (10,000–15,000 ft)in a typical volcanic, carbonate, or granite overthrust terrain. Natural-source data haveconsiderably deeper penetration but poorer resolution at shallower depths.

Which methodis better?

How are MT Data Acquired? continued

MT can be valuable in areas that yield poor-quality seismic data and where acquiringseismic data is very expensive. The following table indicates where to use MT and thereasons for using it.

Locations Reasons for Using MT

Carbonate terrains Poor-quality seismic data

Volcanic terrains Poor-quality seismic data

Granite overthrusts Poor-quality seismic data

Regional surveys Less expensive than seismic; generates prospects to detail with seismic

Remote areas Less expensive than seismic

Rugged terrains Less expensive than seismic

Fracture zones Excellent tool for mapping

Where to use MT


A regional exploration program to study a large unexplored area in the Colombian Ama-zon basin was conducted by Amoco Production Company in 1987 and 1988 (Burgett et al.,1992). This study area was very large [approximately 300,000 km2 (115,000 mi2)] andremote with dense jungle cover, rugged terrain, and limited road access.

The first phase of the program consisted of 31,700 km (19,700 mi) of airborne gravity andmagnetics. The large-scale structures delineated in these surveys were then furtherinvestigated by MT. The MT survey was feasible with a light helicopter because the crewwas small and equipment was light and compact. Data were collected from 43 sites, witha typical spacing of 10–20 km (6–12 mi).

Introduction

Case History: Frontier Basin Analysis (Amazon Basin, Colombia)

The MT data clearly delineated a thick sedimentary section with internal units that couldbe correlated from site to site. Three resistivity “packages” were observed:• 40–100 ohm-m (sedimentary)• 150–250 ohm-m (sedimentary)• >1000 ohm-m (crystalline basement)

The figure below shows a simulated cross section in the Amazon basin based on MT data.

Survey results

Figure 16–5. Drafted from data in Burgett et al., 1992.

Encouraged by the evidence from the MT survey, Amoco decided to shoot a small seismicprogram and drill a shallow stratigraphic test. This program was positioned on the edgeof a subbasin defined in the MT data. There generally was good agreement between theMT data, the seismic data, and the borehole geology.

The airborne gravity and magnetic data, followed by the surface MT survey, provided avery cost-effective means of regional basin definition and led directly to a well-positionedseismic survey and well site.

Post-MT program


The Papuan thrust belt is both an expensive and difficult area in which to acquire seismicdata. The area is typified by rugged mountainous terrain, dense equatorial jungle, andthick, heavily karstified limestone. The karstified limestone in some areas is also overlainby heterogeneous volcanics. The few coherent seismic reflectors are lacking in characterand continuity, and the data in general are extremely noisy.

The sedimentary section in this area, however, is an excellent MT target (Billings andThomas, 1990). This sequence observed in MT data is a simple three-layer package. Theupper layer is the high-resistivity Darai Limestone, the middle layer is low-resistivityLeru Formation clastics, and the third layer is high-resistivity basement rocks. Therefore,the MT data provide a subsurface map of the base of the Darai and the top of the base-ment. The addition of an upper high-resistivity volcanic layer in some areas usually doesnot complicate this interpretation, except that it may not be possible to differentiate thebase of the volcanics from the top of the Darai.

Introduction

Case History: Rugged Carbonate Terrain (Highlands of Papua New Guinea)

More than 2,500 MT sites have been acquired in Papua New Guinea by numerous compa-nies involved in exploration in the region (Mills, personal communication, 1994). BPExploration (Hoversten, 1992) acquired MT data over both the Angore anticline and theHides anticline. The interpreted models from these two data sets provide depth estimatesof the base of the Darai Limestone to within 10% of the measured depth in the Angore 1well. In both cases, the seismic data aided the interpretation.

The figure below shows the 2-D MT model beneath the Angore-1 well and the base of theDarai Limestone as observed in the well.

Survey results

Figure 16–6. Drafted from data in Hoversten, 1992.


MT can be used in an overthrust environment to delineate conductive sediments beneatha resistive thrust plate. It is often difficult to acquire good-quality seismic data in an over-thrust area where high-velocity (high-resistivity) rocks overlie low-velocity (low-resistivi-ty) sediments.

The Precambrian overthrust in the Bear Springs area of northwestern Colorado is anexample (Mills, 1994).

Introduction

Case History: Precambrian Overthrust (Northwestern Colorado)

The MT station near the drill hole (see diagram below) shows a thin, near-surface conduc-tor on top of the resistive Precambrian thrust sheet. This is a wedge of Quaternary andTertiary sediments. Beneath the thrust, a thick conductive section of Cretaceous sedi-ments is observed.

The figure below is an 11-station MT profile across the thrust.

Survey results

Figure 16–7. Drafied from data from Mills, 1994.

These data provide the following structural details:• Thickness of Quaternary and Tertiary cover• Thickness of Precambrian thrust sheet• Thinning of Cretaceous sediments to the south• Depth to top of Paleozoic sediments• No differentiation between Paleozoic and basement

A very detailed subsurface structural map could be obtained in this area using a 3-D grid,controlled-source MT survey depicting the Precambrian/Cretaceous thrust contact andthe top of the Paleozoic section.

Structuraldetails


Seismic methods do not work well in areas covered by volcanics because of the dispersivenature of the volcanics and because of the decrease in acoustic velocity at the base of thevolcanics.

Volcanic terrain, however, is an ideal environment for MT because it is a simple, three-layer stratigraphic package: resistive basalts over conductive sediments, which in turnoverlie resistive metamorphic or granitic basement rocks.

Introduction

Case History: Volcanic Terrain (Columbia River Plateau)

The cross section below is a 13-station MT natural source survey profile. This east–westsection begins near the Idaho–Washington border and extends approximately 75 mi (120km) to the west (Mills, personal communication, 1994).

Survey results

Figure 16–8. Drafted from data from Mills, 1994.

These data provided the following structural details:• Considerable variation on the thickness of the volcanics• Considerable variation in the depth to top of basement• Basalts thin to the east• Sediments thin to east and eventually disappear• Basement resistivities are an order-of-magnitude higher on the east end of the profile

Controlled-source MT data could provide 3-D imaging of individual prospects.

Structuraldetails


Billings, A.J., and J.H. Thomas, 1990, The use and limitations of non-seismic geophysicsin the Papuan thrust belt, in C.J. Carman and Z. Carman, eds., Proceedings of the FirstPNG Petroleum Convention: Port Moresby, New Guinea, p. 51–62.

Burgett, W.A., A. Orange, and R.F. Sigal, 1992, Integration of MT, seismic, gravity, andmagnetic data for reconnaissance of the Colombian Amazon: 54th meeting, EuropeanAssociation of Exploration Geophysicists, Expanded Abstracts, p. 428–499.

Goldstein, M.A., and D.W. Strangway, 1975, Audio-frequency magnetotellurics with agrounded electrical dipole source: Geophysics, vol. 40, p. 669–683.

Hoversten, G.M., 1996, Papua New Guinea MT: looking where seismic is blind: Geophysi-cal Prospecting, vol. 44, p. 935–961.

Mills, A., 1994, Zephyr Geophysical Services, personal communication.

Vozoff, K., 1972, The magnetotelluric method in the exploration of sedimentary basins:Geophysics, vol. 37, no. 1, p. 98–141.

References

Chap16

Education

hfield data

primary hfield

horizontal hfield

hfieldthe hfield

secondary field

primary field

vertical hfield

hfield propagating