gds1101-rev - matheson area, magnetic and electromagnetic …

Report on Matheson Airborne Geophysical SurveyGeophysical Data Set 1101 - Revised

MATHESON AREA

Ontario Airborne Geophysical SurveysMagnetic and Electromagnetic DataGeophysical Data Set 1101 - Revised

Ontario Geological SurveyMinistry of Northern Development and MinesWillet Green Miller Centre933 Ramsey Lake RoadSudbury, Ontario, P3E 6B5Canada


1

TABLE OF CONTENTS

CREDITS .....................................................................................................................................................................2

DISCLAIMER .............................................................................................................................................................2

CITATION...................................................................................................................................................................2

1) INTRODUCTION..............................................................................................................................................3

2) SURVEY LOCATION AND SPECIFICATIONS ..........................................................................................4

3) AIRCRAFT, EQUIPMENT AND PERSONNEL ...........................................................................................6

4) DATA ACQUISITION ......................................................................................................................................9

5) DATA COMPILATION AND PROCESSING..............................................................................................10

6) MICROLEVELLING AND GSC LEVELLING ..........................................................................................18

7) FINAL PRODUCTS ........................................................................................................................................28

8) QUALITY ASSURANCE AND QUALITY CONTROL..............................................................................30

APPENDIX A TESTING AND CALIBRATION..............................................................................................36

APPENDIX B PROFILE ARCHIVE DEFINITION ........................................................................................43

APPENDIX C ANOMALY ARCHIVE DEFINITION.....................................................................................46

APPENDIX D KEATING CORRELATION ARCHIVE DEFINITION.........................................................48

APPENDIX E GRID ARCHIVE DEFINITION ..............................................................................................49

APPENDIX F GEOTIFF AND VECTOR ARCHIVE DEFINITION .........................................................50

APPENDIX G HALFWAVE ARCHIVE DEFINITION ..................................................................................51


2

CREDITS

This survey is part of the Operation Treasure Hunt geoscience initiative, funded by the OntarioGovernment.

List of accountabilities and responsibilities:� Andy Fyon, Senior Manager, Precambrian Geoscience Section, Ontario Geological Survey

(OGS), Ministry of Northern Development and Mines (MNDM) – accountable for theairborne geophysical survey projects, including contract management

� Stephen Reford, Vice President, Paterson, Grant & Watson Limited (PGW), Toronto,Ontario, OTH Geophysicist under contract to MNDM, responsible for the airbornegeophysical survey project management, quality assurance (QA) and quality control (QC)

� Lori Churchill, Project and Results Management Coordinator, Precambrian GeoscienceSection, Ontario Geological Survey, MNDM – manage the project-related milestoneinformation

� Zoran Madon, OTH Data Manager, Precambrian Geoscience Section, Ontario GeologicalSurvey, MNDM – manage the project-related digital and hard copy products

� Spectrem Air Limited, Johannesburg, South Africa, under sub-contract to Fugro AirborneSurveys, Calgary, Alberta - data acquisition and data compilation.

DISCLAIMER

To enable the rapid dissemination of information, this digital data has not received a technicaledit. Every possible effort has been made to ensure the accuracy of the information provided;however, the Ontario Ministry of Northern Development and Mines does not assume anyliability or responsibility for errors that may occur. Users may wish to verify criticalinformation.

CITATION

Information from this publication may be quoted if credit is given. It is recommended thatreference be made in the following form:

Ontario Geological Survey 2003. Ontario airborne geophysical surveys, magnetic andelectromagnetic data, Matheson area; Ontario Geological Survey, Geophysical Data Set 1101 –Revised.


3

1) INTRODUCTION

Recognising the value of geoscience data in reducing private sector exploration risk andinvestment attraction, the Ontario Government embarked on “Operation Treasure Hunt” (OTH).The OTH initiative comprises a three-year, $29 million program that commenced April 1, 1999. Itincorporates:

� airborne geophysics (high-resolution magnetic/electromagnetic surveys and someradiometric surveys)

� surficial geochemistry (lake sediments and indicator minerals)� bedrock map compilation� methods development (e.g. electro-geochemical modelling applied to exploration and 3-

D geological/geophysical modelling)� delivery of digital data products.

The OGS was charged with the responsibility to manage OTH. The OGS sought advice about themineral industry needs and priorities from its OGS Advisory Board – a stakeholder boardincluding representatives from the Ontario Mining Association, Ontario Prospectors Association,Prospectors and Developers Association, Aggregate Producers Association of Ontario, Chairs ofOntario University Geology Departments, Canadian Mining Industry Research Organisation andGeological Survey of Canada. The OGS Advisory Board mandated a Technical Committee toadvise the OGS on geographic areas of interest within Ontario where collection of new data wouldmake the greatest impact on reducing exploration risk.. Various criteria were assessed, including:

� commodities and deposit types sought� prospectivity of the geology� state of the local mining industry and infrastructure� existing, available data� mineral property status.

In late 1999, MNDM commenced an ambitious program of airborne magnetic and electromagneticsurveys as part of OTH geoscience initiative. The project involved four survey contractors, fivedifferent electromagnetic systems and more than 105,000 line-km of data acquisition.

The airborne survey contracts were awarded through a Request for Proposal and ContractorSelection process. The system and contractor selected for each survey area were judged on manycriteria, including the following:

� applicability of the proposed system to the local geology and potential deposit types� aircraft capabilities and safety plan� experience with similar surveys� QA/QC plan� capacity to acquire the data and prepare final products in the allotted time� price-performance.

In February 2002, PGW was retained by MNDM to microlevel and level to a common datum allOTH aeromagnetic surveys. Any OTH surveys adjacent to existing AMEM surveys weresubsequently merged to form supergrids. PGW commenced this project in March 2002.


4

2) SURVEY LOCATION AND SPECIFICATIONS

The Matheson survey area is located in north-eastern Ontario (Figure 1). It ties on to the northernboundary of block A of the Kirkland Lake survey area (Geophysical Data Set 1102). It straddlesthe Destor-Porcupine Fault Zone in the northern portion of the Archean Abitibi Subprovince,incorporating metavolcanic, metasedimentary and a variety of (ultra)mafic intrusive rocks. Glacialoverburden cover in the area is extensive, reaching 20 to 30 m thickness, incorporating significantclay cover. The commodities sought include gold, komatiite copper-nickel, intrusion-relatedcopper-nickel-platinum group elements, volcanogenic massive sulphides, and potential diamond-bearing kimberlites. The Spectrem2000 time-domain electromagnetic and magnetic system,mounted on a fixed wing platform, was selected by MNDM to conduct the survey.

Figure 1: Matheson survey area.

The airborne survey and noise specifications for the Matheson survey area were as follows:

a) traverse line spacing and direction� flight line spacing was 200 m� flight line direction was 150� azimuth from true north� maximum deviation from the nominal traverse line location could not exceed 50 m over a

distance greater than 3000 m� minimum separation between two adjacent lines could be no smaller than 150 m or larger

than 250 m


5

b) control line spacing and direction� at a regular 1500 m interval, perpendicular to the flight line direction� along each survey boundary (if not parallel to the flight line direction)� maximum deviation from the nominal control line location could not exceed 50 m over a

distance greater than 3000 m c) terrain clearance of the EM receiver bird

� nominal terrain clearance was 52 m� altitude tolerance limited to �15 m over a long chord not to exceed

3000 m, except in areas of severe topography d) aircraft speed

� nominal aircraft speed was 61 m/sec� aircraft speed tolerance limited to �6.0 m/sec, except in areas of severe topography

e) magnetic diurnal variation

� could not to exceed a maximum deviation of 2.5 nT peak-to-peak over a long chordequivalent to the control line spacing (1500 m)

f) magnetometer noise envelope

� in-flight noise envelope could not exceed 0.1 nT, for straight and level flight� base station noise envelope could not exceed 0.2 nT

g) EM receiver noise envelope� the noise envelope could not exceed:

X-coil (channels 5, 6 and 7) - �500 ppmZ-coil (channels 5, 6 and 7) - �500 ppmover a distance exceeding 3000 m


6

3) AIRCRAFT, EQUIPMENT AND PERSONNEL

Aircraft and Geophysical On-Board Equipment

Operator SPECTREM AirAircraft type Basler DC3 – TP67Registration ZS- ASNNominal aircraft speed 60 m/s

EM systemTransmitter height above ground 91 mReceiver bird height above ground 52 mTx – Rx horizontal separation 136 mTransmitter coil axis VerticalReceiver coil axes X : horizontal, parallel to flight direction

Y : horizontal, perpendicular to flight directionZ : vertical

Current waveform square waveProcessed waveform signal STEP responseBase frequency 90 HzTransmitter loop area 420 m2

RMS current 960 amperesRMS dipole moment 400 000 A.m2

Pulse width 5555 microseconds (full duty cycle)Receiver bandwidth 21.7 to 5555.6 microsecondsDigitising rate 46080 Hz / componentRecording Rate 5 HzMeasured parameters X, Y and Z-components of dB/dT, deconvolved to

B-fieldNumber of windows 8 per componentWindow distribution Pseudo-binaryWindow Number Window Centre (�s) Window Width (�s)

1 21.7 21.72 54.3 43.43 119.4 86.84 249.6 173.65 501.0 347.26 1030.8 694.47 2072.5 1388.98 4155.8 2777.8

Magnetic systemBird height above ground 72 mBird location 19 m below and 41 m behind centre of aircraftSensor Scintrex CS-2 Sensor with SPECTREM


7

Counter/Sync SystemRecording Rate 5 HzSensitivity 0.01 nTResolution 0.1 nT

Positioning systemSensor Novatel RT-20 GPS receiver with Fugro Omnistar

differential correctionsRecording Rate 5 Hz

Other sensorsRadar Altitude Honeywell HG9000 with 5 Hz sampling with 0.3 m

resolutionLaser Altitude Riegl with 5 Hz sampling with 0.03 m resolutionBarometric Pressure Rose Mount with 1 Hz samplingTemperature (OAT) PT-100 RTD with 1 Hz samplingAnalogue Chart Recorder RMS GR-33Flight Camera Sanyo 168H Time Lapse VCR Model TLS-1600P

with Sony 1/3" Hyper HAD CCD B/W camera (510x 492 picture elements)

Base Station Equipment

Magnetometer Scintrex SM-4Sensor Scintrex CS-2Recording Rate 2 HzLocation Inside Val d'Or AirportBackup Magnetometer Geometrics G856Location Behind Confortel Hotel, Val d'Or

Field Office Equipment

Computer Purpose built, rack mounted Pentium PC withswappable hard drives, 5.2 Gb magneto-opticaldrive and 15 Gb DLT tape drive.

Printer Canon Bubblejet

Field Personnel

The following personnel were on-site at various stages of the acquisition program.

Simon B. Bosch Crew Chief and logistics managerGys Oosthuizen Technician / OperatorKieran Bloom Technician / OperatorLen Trouw Aircraft Engineer


8

Peter Wilson CaptainTosh Ross PilotSteve Lynch Geophysicist

Prior to the survey the following personnel were responsible for maintenance on the survey system.Peter Boniface Aircraft EngineerTeo Hage Geophysicist, Systems EngineerEric Steele Electronic engineerPhil Klinkert Geophysicist


9

4) DATA ACQUISITION

The survey operations were based in the town of Val d'Or, Quebec, where hangar facilities toaccommodate the DC3 aircraft were available. These statistics include the survey of the Reid -Mahaffy test area, which totalled 96.8 line-km.

General statistics

Survey Dates January 28 to February 23, 2000Total line kilometres 10,902Days of production 19Production flights 30Days completely lost to weather 8

A series of calibration and test flights were performed on the system, including:- Magnetometer lag check.- GPS navigation lag and accuracy check.- Altimeter calibration.- Magnetometer heading (cloverleaf) check.

Details of these tests and their results are given in Appendix A.

The initial quality assurance of the data was carried out in the field. Routine quality controlincluded checking the analogue chart records and digital plots of the profiles. A statisticalanalysis assisted in verifying that noise levels and flight line deviations matched the contractspecifications. All the data sets were gridded as a further check.

The magnetic base station data was also reviewed daily.

The line data was sent back to Johannesburg, South Africa via file transfer protocol after eachday's flying. Further quality control and the complete processing of the data was carried out inJohannesburg.


10

5) DATA COMPILATION AND PROCESSING

Personnel

The following personnel were involved in the compilation of data and creation of the finalproducts:

Pete Holbrook Data processorMatthew Holbrook Data archivistTim Archer GeophysicistPhil Klinkert GeophysicistPeter Leggatt GeophysicistSteve Lynch Geophysicist

Base maps

Base maps of the survey area were supplied by the Ontario Ministry of Northern Developmentand Mines.

Projection description

Datum: NAD27 (NTv2)*Ellipsoid: Clarke 1866Projection: UTM (Zone 17N)Central Meridian: 81� WestFalse Northing: 0 mFalse Easting: 500,000 mScale factor: 0.9996* The geophysical data was collected using the WGS84 datum then transformed to NAD27

datum using the Geomatics Canada National Transformation (Version 2) software.

Data Processing

Most of the data was processed using the DFA Intrepid package. Some detail processing andfinal map products were generated in the Geosoft package OASIS montaj version 4.3.

Magnetic Processing (Figure 2)

The magnetic processing included:� Magnetic diurnal removal� IGRF removal� Tie-line levelling� Decorrugation� Micro-levelling


11

Base Station DataThe magnetic base station data, used to remove the diurnal variation, was de-spiked and low passfiltered before being included in the database.

Tie-line levellingTie line levelling is used to remove the diurnal variation and errors due to instrument drift, both areassumed to vary slowly over time. Tie-line levelling is an iterative process:

� Calculate the misclosures at the crossover points of the tie and traverse lines. Themisclosure is the difference between the magnetic value on the tie line and the traverse line.The misclosures are weighted by the gradient of the total field at the crossover point.

)1.0(1

gradienteWeight�

�

� The error is approximated by a piecewise polynomial as a function of time along a flightand then along a tie line.

These steps are repeated until a good fit has been obtained.

GridsBefore gridding the data is corrected for diurnal magnetic variation and the extrapolatedInternational Geomagnetic Reference Field. The lag correction is 40 m and the grid cell size 40 m.

DecorrugationThis is a grid based operation designed to reduce the residual errors that the tie-line levelling doesnot remove. These are due to inaccuracies in the crossovers, localized diurnal activity, and localaltitude variations.Elongated anomalies with the following characteristics are removed:

� 2 times the line spacing perpendicular to the line direction� 2 times the tie line spacing parallel to the line direction� small dynamic range

Micro-levellingApplies the corrections made to the grid to the profile data and thereby enhances the line data byremoving the final residual errors. The micro-levelled data are then gridded with a cell size of 40m.

Second Vertical Derivative of the Residual Magnetics

The final grid of the corrected total magnetic field values was then used as input to create thesecond vertical derivative. The second vertical derivative was calculated from this grid byconvolution.


12

StreamedMagnetometer

Data

Remove EMfield

Aircraft binarydata file

Removespikes

Extractmagnetic

data

Removediurnal

BaseMagnetometer

Remove spikes/ low pass

RemoveIGRF

Correct forlag

Grid data

De-corrugate

Write correctionsback to profiles

(microlevel)

Profile database

Tie linelevelling

Generatefinal grids

Figure2: Flowchart of the magnetic data processing

Keating Correlation Coefficients

Possible kimberlite targets have been identified from the residual magnetic intensity data, based onthe identification of roughly circular anomalies. This procedure was automated by using a knownpattern recognition technique (Keating, 1995), which consists of computing, over a movingwindow, a first-order regression between a vertical cylinder model anomaly and the griddedmagnetic data. Only the results where the absolute value of the correlation coefficient is above athreshold of 75% were retained. The results are depicted as circular symbols, scaled to reflect thecorrelation value. The most favourable targets are those that exhibit a cluster of high amplitudesolutions. Correlation coefficients with a negative value correspond to reversely magnetizedsources. It is important to be aware that other magnetic sources may correlate well with the verticalcylinder model, whereas some kimberlite pipes of irregular geometry may not.


13

The cylinder model parameters are as follows:Cylinder Diameter: 200 mCylinder Length: infiniteOverburden Thickness: 30 mMagnetic Inclination: 74.4 �NMagnetic Declination: 11.1 �WMagnetization scale factor: 100Maximum data range: 8 962.4 nTNumber of passes of smoothing filter: 0Model window size: 15Model window grid cell size: 40 m

Digital Elevation Processing

Initially, the GPS height and the radar altimeter channels were visually inspected and any spikes ordiscontinuities removed. A Naudy filter was applied to both channels. The GPS height channelwas then gridded and the resultant grid checked. Due to the nature of the GPS data, a degree ofdecorrugation was applied to the grid with the corrections then written back to the database.

The radar altimeter channel was then subtracted from the corrected GPS height channel in thedatabase and the resultant channel gridded for verification.

Electromagnetic Processing (Figure 3)

On-Aircraft ProcessingSome of the most important EM data processing was carried out on the aircraft as it acquired thedata. The first processing stage was stacking the data to 512 samples. The data was thendeconvolved to remove system response. The deconvolution processes the transmitter waveformto a square wave with a maximum of +1000 and a minimum of -1000. A square transmitterwaveform was chosen as a periodic approximation of the step response.

In the next stage of processing the data was binned into 8 channels or windows. The times for eachof the channels is given in Chapter 3. As the SPECTREM system makes its measurement whilethe transmitter is switched on, it is necessary to separate the primary (transmitted) field from the(induced) secondary field. The assumption is made that the induced field will have decayed to aminimal amount at the time the last channel is sampled. As the last channel only measured theprimary field, it can be subtracted from the other channels to separate the secondary field. Hencethere are actually 7 channels of geological information in the final data.


14

Profile dataThe profile line data have been drift corrected and micro-levelled. The drift is particularlynoticeable on the later time channels and has been applied to channels 4 to 7. This is an iterativeprocess, with the assumption that there is a constant drift on a single line. This is reasonable if thelines are short. The processing steps are:

� The channel data are clipped retaining the data in the resistive areas where the responseshould be close to zero.

� The average of the clipped data is then calculated and subtracted from the channel data.The steps are then repeated, refining the correction.

A decorrugation and micro-levelling process (see magnetic processing section for more detail) hasbeen applied to all the channels to reduce small residual errors that have not been corrected throughthe drift correction method.

Decay ConstantThe decay constant or tau values were calculated by fitting an exponential function to the EMamplitude data. The calculation used all of a component's channels with an amplitude greaterthan a noise threshold. The decay rate of the exponential is related to the conductance andconductor geometry. A slow rate of decay, reflecting a high conductance, will be represented bya high decay constant value. The unit for the decay constant is milliseconds.

Apparent ConductivityThe apparent conductivity for each channel was calculated from its channel amplitude and theaircraft height. An apparent conductivity is the conductivity of a half space that would produce anamplitude equivalent to the measured response. It is useful in providing a physically sensible unitand partially compensates for aircraft ground clearance variations. The unit for apparentconductivity is millisiemens/metre.

GridsThe data were corrected for lag before gridding. With the asymmetrical geometry of fixed wingEM systems, the ideal lag correction differs for conductors of different shape and dip. A set ofgrids with a range of lag corrections was generated to select the optimum. The final lag correctionis a compromise that produces the most coherent image. The X component has been lagged 90 mfor windows 1 to 3, 100 m for window 4, 110 m for windows 5 and 6, and 120 m for window 7.The Z component was lagged 90 m (approx. 7.5 fiducials) for all windows. The data were griddedusing an Akima spline with a cell size of 40 m. A decorrugation filter was applied to remove theherringbone effects created by geometrical asymmetry.


15

StreamedData

Stack to 512samples

Deconvolve toSquare wave

Bin to 8windows

Removeaircraft's field

Removeprimary field

Removespikes, lowpass filter

Correct forlag

Calculateapparent

conductivity

Drift correctionfor EM levels

Calculatetime

constant

Filter andremovespikes

Extractwindowed

data

Grid dataDecorrugateWrite correctionsback to profiles

(microlevel)

Aircraft binary datafile

Profile database

Figure 3: A Flowchart of EM data processing.


16

EM Anomaly Interpretation

The EM anomalies were selected and parameterized by an interpreter using proprietarySPECTREM software. The software records the physical location of the electromagneticconductors along with various parameters associated with the conductor. These parameters includean anomaly grade, the conductance, its mid-time (window 4) residual X channel amplitude, itsestimated depth below ground surface, its dip with respect to the nominal survey direction, and themagnitude of its associated magnetic anomaly. The reference model for estimating conductorparameters is a 300 m by 300 m wire loop below a conductive horizontal sheet, whichapproximates a typical volcanogenic massive sulphide target with overburden. If the actualconductor differs greatly from the model its parameters will be inaccurate, particularly in the depthestimate.

Estimated Conductor DepthCaution needs to be taken when using the depth estimates provided in the EM anomaly listings.The 300 m by 300 m wire loop model approximates a typical volcanogenic massive sulphidetarget, but bears very little resemblance to a body of appreciably different dimensions, such as atypical stratiform deposit. For this reason, depth estimates are unreliable for bodies of dimensionsvery much greater than 300 m by 300 m (estimated depths are too shallow) and very much less(estimated depths are too deep).

EM Anomaly GradingAn anomaly grading scheme has been devised to assist in prioritising which anomalies should firstbe considered for ground follow-up. This grading scheme is essentially geophysical, being acumulative assessment by the interpreter of the likelihood of a particular anomaly being aprospective mineral target. Anomaly grade takes cognisance of such features of the anomaly as itspeak shape (width and amplitude), its conductance (conductivity-thickness product) and itsmagnetic association.

SPECTREM EM anomalies are graded A, B, C or D, with grade A anomalies being the mostfavourable.

During the airborne survey, the pilots record the various types of culture and this information isstored with the AEM profile data for viewing by the interpreter. AEM anomalies caused byculture, were automatically given the lowest grading, D. Anomalies where culture was notrecorded were graded A, B, C or D, depending on the parameters as described above.

Interpretation NotesMost of the AEM anomalies in the area were due to culture. For this reason no spatial filteringwas applied to the AEM profile data before the anomalies were selected, because the interpreterwanted to resolve bedrock conductors that were situated close to cultural conductors.

The abundant power lines, transmission lines and electromagnetic radiation sources such astelecommunication transmitters in the region, resulted in a fair amount of “low frequency” noiseacross long sections of the AEM profiles in many parts of the area surveyed. This low frequencynoise was in most cases easily recognised on the profiles, but it made detection of smallamplitude AEM anomalies uncertain in the affected areas. Except for the low frequency noisecaused by cultural sources, the system noise levels were generally low (i.e. around 120 ppm peak


17

to peak for the Z5, Z6 and Z7 windows). From theoretical modeling carried out in thisenvironment, these low noise levels should easily allow detection of a 20 siemens, 300 x 300-metre vertical plate conductor at a depth of 150 metres below surface in a moderately conductiveenvironment.

Cultural features such as roads, transmission lines and railway lines recorded by the pilots andlocated within 80 metres of picked AEM anomalies, were designated as cultural on the AEMplans.

The interpreted AEM anomaly parameters of conductance and dip are expected to be fairlyreliable. The depth estimates are not reliable because they are based on a 300 x 300 metre bodywith the flight-line passing over the centre of the body. Therefore if the conductor hasdimensions smaller or larger than 300 x 300 metres and / or if the flightline did not pass over itscentre, which is probably true in the majority of cases, then the depth estimates will be incorrect.

There was a significant amount of surficial cover in the area surveyed, but the conductivity-thickness product was, in nearly all cases, less than 2 siemens, so this essentially had no effect onthe detection of bedrock conductors. Finally the area flown was fairly active magnetically. Thereare a considerable number of magnetic rock units in the region and some of them are stronglyfolded. There are also a large number of crosscutting north-south and some northeast trendingdykes in the area which possibly resulted in incorrect magnetic associations made on some of theAEM anomalies.


18

6) MICROLEVELLING AND GSC LEVELLING

Microlevelling

Microlevelling is the process of removing residual flightline noise that remains afterconventional levelling using control lines. It has become increasingly important as the resolutionof aeromagnetic surveys has improved and the requirement of interpreting subtle geophysicalanomalies has increased. The frequency-domain filtering technique known as “decorrugation”has proven inadequate in most situations, as significant geological signal might be removedalong with noise. In addition, the microlevelling correction is applied to the profile data,whereas decorrugation corrects only grids. The separation of noise from geological signal andthe correction of the profiles, are the key strengths of the PGW’s microlevelling procedure.

The PGW microlevelling technique resulted from a new application of filters used in the processof draping profile data onto a regional magnetic datum (Reford et al., 1990). It is similar to thatpublished by Minty (1991).

Microlevelling is applied in two steps. The decorrugation steps are as follows:� Grid the flightline data to a specified cell size using the minimum curvature gridding

algorithm.� Apply a decorrugation filter in the frequency-domain, using a sixth-order high pass

Butterworth filter of specified cut-off wavelength (tuned to the flightline separation),together with a directional cosine filter, so that a grid of flightline-oriented noise isgenerated.

� Extract the noise from the grid to a new profile channel.

At this stage, the noise grid may be examined toensure that the flightline noise has been isolated,and to determine what parameters will berequired to separate the true residual flightlinenoise from the high-frequency geological signalincorporated in the filtering described above.

The steps for the microlevelling procedure are asfollows:

� Apply an amplitude limit to clip or zerohigh amplitude values in the noisechannel, if desired.

� Apply a low pass non-linear filter (Naudyand Dreyer, 1968), so that only the longerwavelength flightline noise remains,forming the microlevel correction.

� Subtract the microlevel correction fromthe original data, resulting in the final,microlevelled profile channel.

In the example shown, the data are windowed

Decorrugation noise grid


19

from an airborne magnetic and electromagnetic survey flown in the Matachewan area of Ontario,over typical Archean granite-greenstone terrain (Ontario Geological Survey, 1997). Standardcorrections (e.g. diurnal, IGRF, conventional tieline levelling) were applied to the magnetic data.However, a considerable component of residual flightline noise remains, due for example toinadequate diurnal monitoring or tieline levelling difficulties.

Middle panel: red – decorrugated noise channelgreen – noise channel after zeroing high valuesblue – non-linear filtered noise channel (=microlevel channel)

Lower panel: magenta – original total magnetic fieldgrey – microlevelled total magnetic field

The resultant microlevelled channel can then be gridded for comparison with the original data.In addition, it is useful to examine the intermediate noise channels in profile and grid form, toverify that the desired separation of residual flightline noise and geological signal has occurred.


20

Total magnetic field, before microlevelling. Total magnetic field grid, after microlevelling.

Microlevelling can be applied selectively to deal with noise that varies in amplitude and/orwavelength across a survey area. It can also be applied to swaths of flightlines, where moreregional level shifts are a problem due to inadequate levelling to the control lines. This can beparticularly useful on older surveys where tieline data may no longer be available.

Microlevelling will not solve all problems of flightline noise. For example, positioning errors(e.g. poor lag correction) may result in some level shift that microlevelling will reduce.However, shorter wavelength anomalies will still remain mis-aligned. Line-to-line variations insurvey height result in anomaly amplitude variations. Again, microlevelling will reduce longwavelength level shifts, but cannot compensate for localized amplitude changes.

Decorrugation Parameters

Decorrugation requires a database of geophysical data, oriented along roughly parallel surveylines. Surveys with more than one line orientation should be separated into blocks of consistentline direction. The profile channel to be microlevelled should have had all standard corrections,and conventional tieline levelling, already applied. Only traverse lines should be selected formicrolevelling (i.e. no tie-lines).

Flight Line SpacingThe nominal flightline spacing is required to design the filter parameters. If a survey containsblocks flown at different line spacings, better results will likely be obtained if these blocks aremicrolevelled separately. If one is attempting to remove wider level shifts, across swaths oflines, then the average width of the swath should be specified instead.

Flight Line DirectionThe nominal flightline direction is required so that the directional filtering incorporated in thedecorrugation process has the correct orientation. Survey blocks flown with different linedirections should be microlevelled separately.


21

Grid Cell Size for GriddingThe cell size chosen should be small enough so that the residual flightline noise represented inthe grid of original data is well-defined on a survey line basis. Thus, a grid cell size of ¼ the linespacing or smaller is recommended. However, a cell size that is too small (i.e., less than 1/10 theline spacing) will not improve the microlevelling results, and will increase the processing timerequired.

Decorrugation cut-off wavelengthThis parameter defines the cut-off wavelength of the sixth-order, high-pass Butterworth filter,that is combined with a directional cosine filter (power of 0.5) oriented perpendicular to theflightline direction, to extract the residual flightline noise component from the grid of theoriginal data. A wavelength of four times the line spacing has typically proven to produce thebest results. Setting this wavelength too small will not give the filter enough width to isolate theeffect of each flightline. Setting it too large will extract more geological signal than necessary.

Microlevelling Parameters

Once decorrugation has been applied, it is recommended that the decorrugation grid be reviewedand compared to the original data. This is best done by shaded relief imaging. The purpose isto:

� Ensure that the parameters chosen when decorrugation was applied have properlyisolated the residual flightline noise.

� Measure the amplitudes (e.g. determine the peak-to-trough amplitude variations betweenthe survey lines) and wavelengths (in the flightline direction) of the residual flightlinenoise, from the decorrugation grid.

Amplitude limit valueThe amplitude limit defines the value estimated by the user as the maximum amplitude of theresidual flight line noise in a survey. If the absolute value of the decorrugation noise channelexceeds the specified amplitude for a given record, then it will be clipped to that value, orzeroed, depending on the mode chosen. This is one of the techniques employed to separateresidual flightline noise from geological signal. It is assumed than any responses of higheramplitude reflect geology.

The user should also consider the sources of noise for the particular survey that is beingmicrolevelled. When considering aeromagnetic data, the noise amplitudes produced by somesources (e.g. diurnal variation) are not affected by the geological signal of an area, whereas thenoise amplitudes from others (e.g. height variations) are affected by the geology, particularlywhere the magnetic gradients are strong.

If the user does not want to apply an amplitude limit, than a large value, exceeding the dynamicrange of the decorrugation noise channel, should be specified. This dynamic range can bedetermined from the channel statistics.

Amplitude Limit ModeThere are two choices for the amplitude limit mode:


22

Zero mode – This will set any value in the decorrugation noise channel, whose absolute valueexceeds the specified amplitude limit value, to zero prior to application of the non-linear filter.This is suited to areas where the responses exhibit steep gradients (e.g. magnetic survey overnear-surface igneous and metamorphic rocks). It has the effect of dividing a simple, highamplitude response into three parts: two flanks centred on a zeroed section, allowing a shorternon-linear filter wavelength to be applied, if appropriate. It also reduces the possibility of thisfilter distorting a response whose wavelength is close to the filter wavelength.

Clip Mode – This will set any value in the decorrugation noise channel, whose absolute valueexceeds the specified amplitude limit value, to the amplitude limit value (with appropriate sign)prior to application of the non-linear filter. This is suited to areas where the magnetic responsesexhibit shallow gradients (e.g. magnetic survey over sedimentary terrain). It is also appliedwhere the wavelengths of the residual flightline noise in the line direction are clearly muchgreater than those of the geological signal in the decorrugation noise grid.

Naudy Filter LengthThe Naudy non-linear low pass filter (Naudy and Dreyer, 1968) is used due to its superiorqualities for either accepting or rejecting responses beyond the specified filter length. A linearfilter, in contrast, would smear an undesirable, short wavelength response into the filtered data,rather than completely remove it. It is applied to the amplitude-limited noise channel, to removeany remaining geological signal. The filter length is set to half the length of the shortest linearnoise segments visible in the decorrugation noise grid. In most situations, the lengths of thesenoise segments will still be considerably larger than the wavelengths of geological signal. Theexception occurs where there is strong signal due to geology (e.g. magnetic dykes) that strikesubparallel to the line direction. In such cases, it is wise to choose a fairly long filter length forthe first pass of microlevelling, and then shorten the filter length for any subsequentmicrolevelling applied only to survey lines (or parts thereof) where problems remain.

Naudy Filter ToleranceThis parameter sets the amplitude below which the filter will not alter the data. Formicrolevelling, it is recommended that this value be set quite small (e.g. 0.001 nT for magneticdata) as otherwise, the filtered noise channel may contain low amplitude, high frequency chatterthat will then be introduced into the microlevelled channel when the correction is applied.

Quality ControlOnce the microlevelling process has been applied, it is instructive to study five parameters, bothin profile and gridded form: original unmicrolevelled data, decorrugated noise, amplitude-limited noise, non-linear filtered noise (i.e. microlevel correction) and microlevelled data. Thiswill allow the user to determine if separation of residual flightline noise from geological signal issatisfactory, and whether any levelling problems remain.

Shaded relief imaging of the total magnetic field and its residual component and/or 1st/2nd

vertical derivatives will verify that the residual line noise has been minimised, and that new linenoise has not been introduced. A grid of the microlevel correction will confirm that geologicalsignal has not been removed.


23

GSC Levelling

In 1989, as part of the requirements for the contract with the Ontario Geological Survey (OGS)to compile and level all existing GSC aeromagnetic data (<1989) in Ontario, PGW developed arobust method to level the magnetic data of various base levels to a common datum provided bythe Geological Survey of Canada (GSC) as 812.8 m grids. The essential theoretical aspects of thelevelling methodology were fully discussed in Gupta et al. (1989), and Reford et al. (1990). Themethod was later applied to the remainder of the GSC data across Canada and the high-resolution AMEM surveys flown by the OGS (Ontario Geological Survey, 1996).

Terminology:

Master grid – refers to the 200 metre Ontario magnetic grid compiled and levelled to the 812.8metre magnetic datum from the Geological Survey of Canada.

GSC levelling – the process of levelling profile data to a master grid, first applied to GSC data.

Intra-survey levelling or microlevelling – refer to the removal of residual line noise describedearlier in this chapter; the wavelengths of the noise removed are usually shorter than tie linespacing.

Inter-survey levelling or levelling – refer to the level adjustments applied to a block of data; theadjustments are the long wavelength (in the order of tens of kilometres) differences with respectto a common datum, in this case, the 200 metre Ontario master grid, which was derived from allpre-1989 GSC magnetic data and adjusted, in turn, by the 812.8 metre GSC Canada wide grid.


24

Ontario Master Aeromagnetic Grid (Ontario Geological Survey, 1999). The outline for the sampledataset to be levelled (Vickers) is shown.

The GSC Levelling Methodology

Several data processing procedures are assumed to be applied to the survey data prior tolevelling, such as microlevelling, IGRF calculation and removal. The final levelled data isgridded at 1/5 of the line spacing. If a survey was flown as several distinct blocks with differentflight directions, then each block is treated as an independent survey.

1.Create an upward continuation of the survey grid to 305m

Almost all recent surveys (1990 and later) to be compiled were flown at a nominal terrainclearance of 100 metres or less. The first step in the levelling method is to upward continue thesurvey grid to 305 metres, the nominal terrain clearance of the Ontario master grid. The grid cellsize for the survey grids is set at 100 metres. Since the wavelengths of level corrections will begreater than 10 to 15 kilometres, working with 100 metre or even 200 metre grids at this stagewill not affect the integrity of the levelling method. Only at the very end, when the levelcorrections are imported into the databases, will the level correction grids be regridded to 1/5 ofline spacing.

The unlevelled 100 metre grid is extended by at least 2 grid cells beyond the actual surveyboundary, so that, in the subsequent processing, all data points are covered.

2. Create a difference grid between the survey grid and the Ontario master grid


25

The difference between the upward continued survey grid and the Ontario master grid, regriddedat 100 metres, is computed. The short wavelengths represent the higher resolution of the surveygrid. The long wavelengths represent the level difference between the two grids.

Difference grid (difference between survey grid and master grid), Vickers survey.

3. Rotate difference grid so that flight line direction is parallel to grid column or row, ifnecessary.

4. Apply a first pass of a non-linear filter (Naudy and Dreyer, 1968) of wavelength on the orderof 15 to 20 kilometres along the flight line direction. Reapply the same non-linear filter acrossthe flight line direction.

5. Apply a second pass of a non-linear filter of wavelength on the order of 2000 to 5000 metresalong the flight line direction. Reapply the same non-linear filter across the flight line direction.

6. Rotate the filtered grid back to its original (true) orientation.


26

Difference grid after application on non-linear filtering, Vickers Survey.

7. Apply a low pass filter to the non-linear filtered grid

Streaks may remain in the non-linear filtered grid, mostly caused by edge effects. They must beremoved by a frequency-domain, low pass filter with the wavelengths in the order of 25kilometres.

Level correction grid, Vickers Survey.

8. Regrid to 1/5 line spacing and import level corrections into database.


27

9. Subtract the level correction channel from the unlevelled channel to obtain the level correctedchannel.

10. Make final grid using minimum curvature gridding algorithm with grid cell size at 1/5 of linespacing.

Total Magnetic field and Second Vertical Derivative Grids

For most surveys the reprocessed total field magnetic grid was calculated from the finalreprocessed profiles by a minimum curvature algorithm (Briggs, 1974). The accuracy standardfor gridding is that the grid values fit the profile data to within 1 nT for 99.98% of the profiledata points. The average gridding error is well below 0.1 nT.

Minimum curvature gridding provides the smoothest possible grid surface that also honours theprofile line data. However, sometimes this can cause narrow linear anomalies cutting acrossflight lines to appear as a series of isolated spots.

The second vertical derivative of the total magnetic field was computed to enhance small andweak near-surface anomalies and as an aid to delineate the contacts of the lithologies havingcontrasting susceptibilities. The location of contacts or boundaries is usually traced by the zerocontour of the second vertical derivative map.

An optimum second vertical derivative filter was designed using Wiener filter theory andmatched to the data (Gupta and Ramani, 1982) of individual survey areas. First, the radiallyaveraged power spectrum of the total magnetic field was computed and a white noise power waschosen by trial and error. Second, an optimum Wiener filter was designed for the radiallyaveraged power spectrum. Third, a cosine-squared function was then applied to the optimumWiener filter to remove the sharp roll-off at higher frequencies.

The radial frequency response of the optimum second vertical derivative filter is given by:

H2VD(f) = (2f*π)2*(1-exp(-x(f)))

Where x(f) is the logarithmic distance between the spectrum and the selected white noise.

Survey Specific Parameters

The following decorrugation and microlevelling parameters were used in the Matheson survey:No microlevelling was required

The following GSC levelling parameters were used in the Matheson survey:Distance to upward continue: 233 metresFirst pass non-linear filter length: 10000 metresSecond pass non-linear filter length: 2500 metresLow pass filter cut-off wavelength: 15000 metresComments: none


28

7) FINAL PRODUCTS

Map products at 1/20,000

� Residual magnetic field contours, plotted with flight path and EM anomalies on a planimetricbase.

Digital plot files at 1/50,000

� Residual magnetic field in colour with contours, presented with EM anomalies on aplanimetric base.

� Shaded colour image of the 2nd vertical derivative of the magnetics, presented with theKeating kimberlite coefficients, on a planimetric base.

� Colour EM decay constant (X-component) with contours, presented with EM anomalies on aplanimetric base.

Profile database

� Profile archive at 5 samples/sec in both Geosoft GDB and ASCII format.

EM Anomaly database

� EM Anomaly database in both Geosoft GDB and CSV format.

Kimberlite coefficient database

� Keating kimberlite coefficient database in both Geosoft GDB and CSV format.

Data grids

� Geosoft data grids, in both GRD and GXF formats, gridded from coordinates in both NAD27and NAD83 datums, of the following parameters:

� Digital Terrain Model.� Residual Magnetic Intensity.� GSC Levelled Magnetic Field.� 2nd Vertical Derivative of Magnetics.� 2nd Vertical Derivative of the GSC Levelled Magnetic Field.� EM Decay Constant (X-component).� EM Decay Constant (Z-component).� Apparent Conductivity (X-component).� Apparent Conductivity (Z-component).


29

GeoTIFF images of the entire survey block

� Colour residual magnetics on a planimetric base.� Colour shaded relief of 2nd vertical derivative of magnetics on a planimetric base.� Colour EM decay constant (X-component) on a planimetric base.

DXF vector files of the entire survey block

� Flight path.� EM anomaly locations� Keating kimberlite coefficients.� Residual magnetic field contours� EM decay constant (X-component) contours

Halfwave data files

� These are raw ASCII files, covering one flightline of data per file, delivered on separate DVDs.These files contain the 256 points of the TDEM waveform, stacked to 5 Hz sampling, for thefour components Tx, X, Y and Z.

Project report

� Provided in both WORD and PDF formats.


30

8) QUALITY ASSURANCE AND QUALITY CONTROL

Quality assurance and quality control (QA/QC) were undertaken by the survey contractor(SPECTREM Air), by PGW (as OTH Geophysicist), and by MNDM. Stringent QA/QC wasemphasized throughout the project so that the optimal geological signal was measured, archivedand presented.

Survey Contractor

Important checks were required during the data acquisition stage to ensure that the data quality waskept within the survey specifications. All data was processed to preliminary level in the field todetect quality problems as they arose. The profiles and grids were inspected for each dataset. Thefollowing list details the other data quality checks that were performed during the course of thesurvey.

Daily quality control

Navigation data

� The flight path was plotted and inspected after each day’s flying to ensure that itmatched contract specifications. Statistics were also generated to check that theflown lines did not differ excessively from the planned flight path. Aircraft speedwas calculated to check for erratic jumps in the GPS data.

� The altimeter data was checked to see that the specified ground clearance wasmaintained within safe and practicable flying limits.

Magnetometer data

� The diurnal data was checked visually every day for large deviations. The chorddeviation of the specifications was checked computationally. Lines were re-flown incases where the diurnal gradient was greater than specification or seemed to haveotherwise affected the quality of the final magnetics grid.

� The noise levels of the airborne magnetometer were checked using a high pass filterto isolate high frequency noise. As the sensor is installed in a towed bird it was notnecessary to consider aircraft compensation.

Electromagnetic data

� Flying was stopped during times of high wind or sferic activity. The turbulencecaused by high winds can lead to excessive bird motion and a deterioration in dataquality. Sferic activity obviously affects all EM systems. The aircraft operatormonitored the level of sferic interference during flight to be sure that data qualityremained within expected limits.


31

� The EM data was checked after each flight for noise and drift levels. Noise levelswere checked by viewing digital and analogue profiles.

Near-final field products

In field processing was limited to lag corrected grids of magnetics, elevation and all the EMchannels.

Complete databases were prepared for the OTH QA/QC Geophysicist at each of the qualitycontrol visits. Near-final grids, processed in the office and sent back to the field, were alsoavailable.

Quality control in the office

Review of field processing of magnetic and electromagnetic data.

Many of the quality assurance checks applied in the field were repeated in the office, particularlyfor profile noise levels. Additional checks were also made of gridded data quality. Theprocessing of the data was carried out as the survey was in progress, enabling any substandarddata to be quickly re-flown.

Various imaging enhancements were used to test the quality of the girds. The magnetic data wastested by applying a second vertical derivative filter.

Interim products

Archive files containing the raw and processed profile data, the EM anomaly database and thefinal gridded parameters were provided to MNDM for review and approval.

Creation of 1/20,000 maps

After approval of the interim data, the 1/20,000 maps of the total field magnetics with the EManomalies and the base maps were created and verified for registration, labelling, droppingweights, general surround information, etc.


32

OTH Geophysicist

The OTH Geophysicist conducted on-site inspections during data acquisition, focusing initially onthe data acquisition procedures, base station monitoring and instrument calibration. As data wascollected, it was reviewed for adherence to the survey specifications and completeness. Anyproblems encountered during data acquisition were discussed and resolved.

The QA/QC checks included the following:

Navigation Data� appropriate location of the GPS base station� flight line and control line separations were maintained, and deviations along lines were

minimized� verified synchronicity of GPS navigation and flight video� all boundary control lines were properly located� terrain clearance specifications were maintained� aircraft speed remained within the satisfactory range� area flown covered the entire specified survey area� differentially-corrected GPS data did not suffer from satellite-induced shifts or dropouts� GPS height and radar/laser altimeter data were able to produce an image-quality DEM� GPS and geophysical data acquisition systems were properly synchronized� GPS data were adequately sampled

Magnetic Data� appropriate location of the magnetic base station, and adequate sampling of the diurnal

variations� heading error and lag tests were satisfactory� magnetometer noise levels were within specifications� magnetic diurnal variations remained within specifications� magnetometer drift was minimal once diurnal and IGRF corrections had been applied� spikes and/or drop-outs were minimal to non-existent in the raw data� filtering of the profile data was minimal to non-existent� in-field levelling produced image-quality grids of total magnetic field and higher-order

products (e.g. second vertical derivative)

Time-domain Electromagnetic Data� selected receiver coil orientations, base frequency, primary field waveform and secondary

field sampling were appropriate for the local geology� raw "streaming" data were recorded and archived� data behaved consistently between channels (i.e. consistent signal decay)� noise levels were within specifications, and system noise was minimized� bird swing and orientation noise was not evident� sferics and other spikes were minimal (after editing)� cultural (60 Hz) noise was not excessive� regular tests were conducted to monitor the reference waveform and system drift, and

ensure proper zero levels� filtering of the profile data was minimal


33

� In-field processing produces image-quality images of apparent conductivity and decayconstant (tau).

The OTH Geophysicist reviewed interim and final digital and map products throughout the datacompilation phase, to ensure that noise was minimized and that the products adhered to the OTHspecifications. This typically resulted in several iterations before all digital products wereconsidered satisfactory. Considerable effort was devoted to specifying the data formats, andverifying that the data adhered to these formats.

MNDM

MNDM prepared all of the base map and map surround information required for the digital andhard copy maps. This ensured consistency and completeness for all of the OTH geophysical mapproducts. For Matheson, the base map was constructed from digital files of the 1:20,000 OBM mapseries.

MNDM worked with the OTH Geophysicist to ensure that the digital files adhered to the specifiedASCII and binary file formats, that the file names and channel names were consistent, and that allrequired data were delivered on schedule. The map products were carefully reviewed in digital andhard copy form to ensure legibility and completeness.

The SPECTREM survey aircraft flying over Kirkland Lake


34

REFERENCES

Briggs, Ian, 1974, Machine contouring using minimum curvature, Geophysics, v.39, pp.39-48.

Fairhead, J. Derek, Misener, D. J., Green, C. M., Bainbridge, G. and Reford, S.W. 1997: LargeScale Compilation of Magnetic, Gravity, Radiometric and Electromagnetic Data: The NewExploration Strategy for the 90s; Proceedings of Exploration 97, ed. A. G. Gubins, p.805-816.

Gupta, V., Paterson, N., Reford, S., Kwan, K., Hatch, D., and Macleod, I., 1989, Single masteraeromagnetic grid and magnetic colour maps for the province of Ontario: in Summary of fieldwork and other activities 1989, Ontario Geological Survey Miscellaneous Paper 146, pp.244-250.

Gupta, V. and Ramani, N., 1982, Optimum second vertical derivatives in geological mappingand mineral exploration, Geophysics, v.47, pp. 1706-1715.

Gupta, V., Rudd, J. and Reford, S., 1998, Reprocessing of thirty-two airborne electromagneticsurveys in Ontario, Canada: Experience and recommendations, 68th Annual Meeting of theSociety of Exploration Geophysicists, Extended Technical Abstracts, p.2032-2035.

Keating, P.B. 1995. A simple technique to identify magnetic anomalies due to kimberlite pipes;Exploration and Mining Geology, vol. 4, no. 2, p. 121-125.

Klinkert, P.S., Leggatt, P.B and Hage, T.B. 1997. The Spectrem airborne electromagnetic system- Latest developments and field examples, in Proceedings of Exploration 97: Fourth DecennialConference on Mineral Exploration, p. 557 – 564.

Minty, B. R. S., 1991, Simple micro-levelling for aeromagnetic data, Exploration Geophysics, v.22, pp. 591-592.

Naudy, H. and Dreyer, H., 1968, Essai de filtrage nonlinéaire appliqué aux profilesaeromagnétiques, Geophysical Prospecting, v. 16, pp.171-178.

Ontario Geological Survey, 1996, Ontario airborne magnetic and electromagnetic surveys,processed data and derived products: Archean and Proterozoic “greenstone” belts – MatachewanArea, ERLIS Data Set 1014.

Ontario Geological Survey, 1997, Ontario airborne magnetic and electromagnetic surveys,processed data and derived products: Archean and Proterozoic “greenstone” belts – Black River-Matheson Area, ERLIS Data Set 1001.

Ontario Geological Survey, 1999, Single master gravity and aeromagnetic data for Ontario,ERLIS Data Set 1036.

Palacky, G.J. and West, G.F. 1973. Quantitative interpretation of INPUT AEM measurements;Geophysics, vol. 38, p. 1145 – 1158.


35

Reford, S.W., Gupta, V.K., Paterson, N.R., Kwan, K.C.H., and Macleod, I.N., 1990, Ontariomaster aeromagnetic grid: A blueprint for detailed compilation of magnetic data on a regionalscale: in Expanded Abstracts, Society of Exploration Geophysicists, 60th Annual InternationalMeeting, San Francisco, v.1, pp.617-619.

Smith, R.S. and Annan, A.P. 1997. Advances in airborne time-domain EM technology; inProceedings of Exploration 97: Fourth Decennial Conference on Mineral Exploration, p. 497 –504.

Smith, R.S. and Annan, A.P. 1998. The use of B-field measurements in an airborne time-domainsystem, Part I: Benefits of B-field versus dB/dT data; Exploration Geophysics, vol. 29, p. 24 –29.

Smith, R.S. and Keating, P.B. 1996. The usefulness of multicomponent time-domain airborneelectromagnetic measurements; Geophysics, vol. 61, p. 74 – 81.

Wolfgram, P. and Thomson, S. 1998. The use of B-field measurements in an airborne time-domain system, Part II: Examples in conductive regimes; Exploration Geophysics, vol.29, p. 225– 229.


36

APPENDIX A TESTING AND CALIBRATION

Introduction

In order to test the reliability, repeatability and accuracy of the airborne survey systems a seriesof calibrations were flown. The tests were:

� Magnetometer heading� Magnetometer lag� EM pitch-roll-yaw� Altimeter calibration� GPS positioning

After the first QC check an additional test was carried out to investigate cultural noise that wasaffecting a large portion of the survey area.

Magnetometer heading

To test the magnetometer heading error, data was acquired with the aircraft flying in the fourcardinal directions over a central point (i.e. a clover leaf test). The greatest difference from themean of values over the central point was 1.55nT. The results are summarised in the tablebelow.

Line Direction Altitude Fid Time LaggedMag

Base Mag Residual Diffmean

21010 N 92.6 28183 57276 57732.0 57200.0 532.0 0.8521020 S 96.6 28404 57402 57733.6 57201.3 532.3 0.5521030 W 101.9 28667 57549 57734.3 57199.9 534.4 -1.5521040 E 97.5 28879 57684 57733.2 57200.5 532.7 0.15

mean 532.85


37

Magnetometer lag

The lag in the magnetometer system was checked by flying in opposing directions over a culturalmagnetic feature. The raw magnetic profile (in red) and lag corrected data (in green) are shownin the "Magnetometer Lag Correction" plot. The lag correction was 3.5 fiducials, or 0.5 seconds.Magnetometer lag is also checked by gridding and imaging the data from the survey area.

EM roll-pitch-yaw

To test the EM compensation the aircraft climbed to a level high enough to eliminate any groundeffect. It then flew roll, pitch and yaw manoeuvres that were at least as harsh as a typical surveyflight. The resultant EM data remained within the noise specification of 500 parts per million or1 part per 2 thousand. The aircraft attitude was recorded with pitch and roll meters that showdegrees of rotation on the left hand scale.


38

E M P itc h T e s t

-2

-1 .5

-1

-0 .5

0

0 .5

1

-5

0

5

1 0

1 5

2 0

2 5

3 0

x 5 x 6 x 7 z 5 z 6 z 7 P itc h R o ll

E M R o ll T e s t

-2

-1 .5

-1

-0 .5

0

0 .5

1

-5

0

5

1 0

1 5

2 0

2 5

3 0

x 5 x 6 x 7 z 5 z 6 z 7 P itc h R o ll

E M Y aw T es t

-2

-1 .5

-1

-0 .5

0

0 .5

1

-5

0

5

10

15

20

25

30

x5 x6 x7 z5 z6 z7 P itch R o ll


39

Altimeter calibration

The altimeter calibrations were performed over a lake near the survey area. The aircraft flew atheights of 200, 300, 400 and 500 feet above ground level, as indicated by the pilot’s instruments.The differentially corrected GPS data was used to calibrate the barometric altimeter but it mustbe remembered that atmospheric variation prevents an absolute correction. The relative heightsof the other altimeters can also be calibrated by the GPS. The laser altimeter is self-calibratingand was used to calibrate the radar altimeter. A full altimeter calibration was flown both beforeand after the survey. The results of the pre-survey calibration are tabulated below.

Height GPSZ Baro Z Laser Radarfeet meters min max mean min max mean min max mean min max mean

200 60.6 355 374 369 315 332 327 55 74 69 50 71 65300 90.9 374 406 398 333 361 353 74 106 98 71 106 97400 121.2 400 442 430 360 396 385 101 142 130 100 146 132500 151.5 443 472 465 399 426 418 143 173 164 146 179 170

Differences300 to 200 29 26 29 32400 to 200 61 58 61 67500 to 400 96 91 95 105

The post survey results are tabulated below

Height GPSZ Baro Z Laser Radarfeet meters min max mean min max mean min max mean Min max mean

200 60.606 368 421 382 373 417 386 56 110 71 50 109 65.5300 90.909 402 447 410 397 438 408 91 136 99 87.5 137 97400 121.21 427 493 443 420 477 437 117 183 132 115 187 132500 151.52 462 527 478 449 502 465 152 215 167 154 223 171

Differences300 to 200 28 22 28 31.5400 to 200 61 51 61 66.5500 to 400 96 79 96 106


40

GPS positioning

The accuracy of the GPS position was tested by flying directly over the Timmins airport VORtransmitter. The given position of the VOR transmitter was 472705 E 5379719 N, known to anaccuracy of about 25 m in the X and Y direction. The aircraft flew over the transmitter on twoflights to establish an average position of 472691 E 5379732 N. The flight lines are plottedbelow.

Cultural noise

In the Matheson area there was a great amount of cultural noise, arising from both passive andactive sources. Passive noise is caused by cultural conductors such as the ground loops formedby power-lines and fences. Current is induced in cultural conductors in the same way it isinduced in geological conductors. They are considered as noise as they obscure the geologicalresponse. The cultural conductors were obvious as long linear features, most common in thecentre of the area. A picture shown below, with different types of power-lines, railway lines andbuildings was taken whilst on a survey line.

Rather than just returning an induced signal, active noise sources transmit with their own power.Power lines and radio transmitters are both active noise sources. The base frequency of EM


41

systems is chosen to nullify the effect of noise generated by power-lines. Usually the powerfrom radio signals is not great enough to affect EM data.

However, in the Matheson area there were active noise sources with enough power to affect thedata. The Z component data in portions of the western half of the area was affected by a

spurious transmitted signal. The photo on theleft shows one of the transmitter towers in thesurvey area. Data collected near these towerswas often noisy.

A line and its re-flight are shown in profilebelow . This line was re-flown, days after theoriginal, to test that the noise was createdexternally. The noise is obvious in the northernend of the line. The southern end of the lineexhibits a more typical noise level. It can beseen that the noise occurs in the same part ofboth lines, indicating an external source. Thereis no correlation between the noisy signal on theoriginal line and the re-flight.


42


43

APPENDIX B PROFILE ARCHIVE DEFINITION

Survey 1101 was carried out using the time-domain Spectrem2000 electromagnetic and magneticsystem, mounted on a fixed wing platform. A transmitter base frequency of 90 Hz was used.

Data File Layout

The files for the Matheson Geophysical Survey 1101 are archived as a 3 CD set with the filecontent divided as follows:

CD – 1101a,c- Profile database in ASCII (XYZ) format- ASCII (GXF) grids- EM Anomaly database (CSV format)- Keating correlation (kimberlite) database (CSV format)- DXF files of entire survey block at 1/50,000 for:

- Flight path- EM anomalies- Keating correlation (kimberlite) anomalies- Total field magnetic contours- Decay constant (X-component) contours

- GEOTIFF images (250 dpi) of the entire survey block at 1/50,000 for:- Total field colour magnetics with base map- Colour shaded relief of 2nd vertical derivative with base map- Colour Decay constant (X-component) with base map

- Project report (Word97 and PDF formats)

CD – 1101b- Geosoft Binary (GRD) grids- EM Anomaly database (GDB format)- Keating correlation (kimberlite) database (GDB format)- DXF files of entire survey block at 1/50,000 for:

- Flight path- EM anomalies- Keating correlation (kimberlite) anomalies- Total field magnetic contours- Decay constant (X-component) contours

- GEOTIFF images (250 dpi) of the entire survey block at 1/50,000 for:- Total field colour magnetics with base map- Colour shaded relief of 2nd vertical derivative with base map- Colour Decay constant (X-component) with base map

- Project report (Word97 and PDF formats)


44

CD – 1101d- Profile database in Geosoft montaj (GDB) format- EM anomaly database (GDB format)- Keating correlation (kimberlite) database (GDB format)- Project report (WORD97 and PDF formats)

The content of the ASCII and binary file types are identical. They are provided in both forms tosuit the user’s available software. Some of the larger files have been compressed using Winzip.

Coordinate Systems

The profile and electromagnetic anomaly data are provided in four coordinate systems:Universal Transverse Mercator (UTM) projection, Zone 17N, NAD27 datum, NTV2 local

datum;Universal Transverse Mercator (UTM) projection, Zone 17N, NAD83 datum, North American

local datum;Latitude/longitude coordinates, NAD27 datum, NTV2 local datum; andLatitude/longitude coordinates, NAD83 datum, North American local datum.

The gridded data are provided in the two UTM coordinate systems.

Line Numbering

The line numbering conventions for survey 1101 are as follows:Flightlines - 81020, 81030 to 85050Tielines – 89010, 89020 to 89400

Profile Data

The profile data are provided in two formats, one ASCII and one binary:thmatheson.xyz - flat ASCII filethmatheson.gdb - Geosoft OASIS montaj binary database file (no compression)

Both file types contain the same set of data channels, summarized as follows:

Channel Name Description Units

x_nad27 easting in UTM co-ordinates using NAD27 datum metersy_nad27 northing in UTM co-ordinates using NAD27 datum metersx_nad83 easting in UTM co-ordinates using NAD83 datum metersy_nad83 northing in UTM co-ordinates using NAD83 datum meterslon_nad27 longitude using NAD27 datum decimal-degreeslat_nad27 latitude using NAD27 datum decimal-degreeslon_nad83 longitude using NAD83 datum decimal-degreeslat_nad83 latitude using NAD83 datum decimal-degreesgps_z_asl_nad27 GPS Z (NAD27 datum) meters above sea levelgps_z_asl_nad83 GPS Z (NAD83 datum) meters above sea levelradar_raw raw radar altimeter meters above terrainradar_final corrected radar altimeter meters above terrainlaser_final corrected laser altimeter meters above terrainbaro_raw raw barometric altimeter millibars


45

baro_final corrected barometric altimeter meters above sea leveldem digital elevation model meters above sea levelfiducial fiducialflight flight numberline_no full flightline number (flightline and part numbers)line flightline numberline_part flightline part numbertime_utc UTC time secondstime_local local time seconds after midnightdate local date YYYYMMDDheight_mag magnetometer height meters above terrainmag_base_final corrected magnetic base station data nanoteslasmag_raw raw magnetic field nanoteslasmag_edit edited magnetic field nanoteslasmag_diurn diurnally-corrected magnetic field nanoteslasigrf local IGRF field nanoteslasmag_igrf IGRF-corrected magnetic field nanoteslas mag_lev levelled magnetic field nanoteslasmag_final micro-levelled magnetic field nanoteslasmag_gsclev GSC levelled magnetic field nanoteslasheight_em electromagnetic receiver height meters above terrainem_x_raw raw (stacked) step response, X-component, on-time (7 channels)

parts per two thousand of the primary fieldem_z_raw raw (stacked) step response, Z-component, on-time (7 channels)

parts per two thousand of the primary fieldem_x_final filtered step response, X-component, on-time (7 channels)

parts per two thousand of the primary fieldem_z_final filtered step response, Z-component, on-time (7 channels)

parts per two thousand of the primary fieldtau_x decay constant (tau) for X-component microsecondstau_z decay constant (tau) for Z-component microsecondsconductivity_x apparent conductivity for X-component millisiemens per metreconductivity_z apparent conductivity for Z-component millisiemens per metre

In thmatheson.xyz, the electromagnetic channel data are provided in individual channels with numerical indices (e.g.em_x_final[0] to em_x_final[6]). In thmatheson.gdb, the electromagnetic channel data are provided in arraychannels with seven elements.


46

APPENDIX C ANOMALY ARCHIVE DEFINITION

Electromagnetic Anomaly Data

The electromagnetic anomaly data are provided in two formats, one ASCII and one binary:thmaanomaly.csv – ASCII comma-delimited format (Microsoft Excel file)thmaanomaly.gdb – Geosoft OASIS montaj binary database file



x_nad27 easting in UTM co-ordinates using NAD27 datum metersy_nad27 northing in UTM co-ordinates using NAD27 datum metersx_nad83 easting in UTM co-ordinates using NAD83 datum metersy_nad83 northing in UTM co-ordinates using NAD83 datum meterslon_nad27 longitude using NAD27 datum decimal-degreeslat_nad27 latitude using NAD27 datum decimal-degreeslon_nad83 longitude using NAD83 datum decimal-degreeslat_nad83 latitude using NAD83 datum decimal-degreesdem digital elevation model meters above sea levelfiducial fiducialflight flight numberline_number full flightline number (flightline number + flightline part number)line flightline numberline_part flightline part numbertime_utc UTC time seconds after midnighttime_local local time seconds after midnightdate local date YYYYMMDDem_x_final filtered step response, X-component, on-time (7 channels)

parts per two thousand of the primary fieldem_z_final filtered step response, Z-component, on-time (7 channels)

parts per two thousand of the primary fieldtau_x decay constant (tau) for X-component microsecondstau_z decay constant (tau) for Z-component microsecondsconductivity_x apparent conductivity for X-component millisiemens per metreconductivity_z apparent conductivity for Z-component millisiemens per metreheight_em electromagnetic receiver height meters above terrainanomaly_no nth anomaly along the survey lineanomaly_id unique anomaly identifieranomaly_type_letter anomaly classification letteranomaly_type_number anomaly classification numberanomaly_grade_letter grade A (best) to D (worst)anomaly_grade_number grade 1 (best) to 4 (worst)conductance conductance of dipping plate model siemensdepth depth of dipping plate model metersdip dip of dipping plate model degreesdip_direction dip direction of dipping plate model degreesheading direction of flight degreessurvey_number survey number

In thmaanomaly.csv, the electromagnetic channel data are provided in individual channels with numerical indices(e.g. em_x_final[0] to em_x_final[6]). In thmaanomaly.gdb, the electromagnetic channel data are provided in arraychannels with seven elements.


47

The unique anomaly identifier (anomaly_id) is a ten digit integer in the format 1LLLLLLAAAwhere 'LLLLLL' holds the line number (and leading zeroes pad short line numbers to six digits).The 'AAA' represents the numeric anomaly identifier (anomaly_no) for that line padded withleading zeroes to three digits. For example, 1000101007 represents the seventh anomaly on Line101. When combined with the survey number (survey_no), the anomaly identifier provides anelectromagnetic anomaly number unique to all surveys archived by the Ontario GeologicalSurvey.

The codes for anomaly_type_letter and anomaly_type_number are as follows:N 1 anomaly due to geological source? 2 anomaly probably due to geological source (possibly cultural)P 3 anomaly probably due to cultural source (possibly geological)C 4 anomaly due to cultural source


48

APPENDIX D KEATING CORRELATION ARCHIVE DEFINITION

Kimberlite Pipe Correlation Coefficients

The Keating kimberlite pipe correlation coefficient data are provided in two formats, one ASCIIand one binary:thmakc.csv – ASCII comma-delimited formatthmakc.gdb – Geosoft OASIS Montaj binary database file



gps_x_final differentially corrected GPS X (NAD83 datum) decimal-degreesgps_y_final differentially corrected GPS Y (NAD83 datum) decimal-degreesx_nad27 GPS X in UTM co-ordinates using NAD27 datum metresy_nad27 GPS Y in UTM co-ordinates using NAD27 datum metresx_nad83 GPS X in UTM co-ordinates using NAD83 datum metresy_nad83 GPS Y in UTM co-ordinates using NAD83 datum metreslon_nad27 longitude using NAD27 datum decimal-degreeslat_nad27 latitude using NAD27 datum decimal-degreescorr_coeff correlation coefficient percent x 10pos_coeff positive correlation coefficient percentneg_coeff negative correlation coefficient percentnorm_error standard error normalized to amplitude percentamplitude peak-to-peak anomaly amplitude within window nanoteslas


49

APPENDIX E GRID ARCHIVE DEFINITION

Gridded Data

The gridded data are provided in two formats, one ASCII and one binary:*.gxf - ASCII Grid eXchange Format (revision 3.0)*.grd - Geosoft OASIS montaj binary grid file (no compression)*.gi - binary file that defines the coordinate system for the *.grd file

The grids are summarized as follows:

thmamag27.grd/.gxf IGRF-corrected magnetic field in nanoteslas (UTM coordinates, NAD27 datum)thmamag83.grd/.gxf IGRF-corrected magnetic field in nanoteslas (UTM coordinates, NAD83 datum)thmamaggsc27.grd/.gxf GSC levelled magnetic field in nanoteslas (UTM coordinates, NAD27 datum)thmamaggsc83.grd/.gxf GSC levelled magnetic field in nanoteslas (UTM coordinates, NAD83 datum)thma2vd27.grd/.gxf second vertical derivative of the IGRF-corrected magnetic field in nanoteslas per square-

metre (UTM coordinates, NAD27 datum)thma2vd83.grd/.gxf second vertical derivative of the IGRF-corrected magnetic field in nanoteslas per square-

metre (UTM coordinates, NAD83 datum)thma2vdgsc27.grd/.gxf second vertical derivative of GSC levelled magnetic field in nanoteslas per metre-squared

(UTM coordinates, NAD27 datum)thma2vdgsc83.grd/.gxf second vertical derivative of GSC levelled magnetic field in nanoteslas per metre-squared

(UTM coordinates, NAD83 datum)thmadem27.grd/.gxf digital elevation model in metres above sea level (UTM coordinates, NAD27 datum)thmadem83.grd/.gxf digital elevation model in metres above sea level (UTM coordinates, NAD83 datum)thmaconx27.grd/.gxf X-component apparent conductivity in siemens per metre (UTM coordinates, NAD27

datum)thmaconx83.grd/.gxf X-component apparent conductivity in siemens per metre (UTM coordinates, NAD83

datum)thmaconz27.grd/.gxf Z-component apparent conductivity in siemens per metre (UTM coordinates, NAD27

datum)thmaconz83.grd/.gxf Z-component apparent conductivity in siemens per metre (UTM coordinates, NAD83

datum)thmadcx27.grd/.gxf X-component decay constant (tau) in microseconds (UTM coordinates, NAD27 datum)thmadcx83.grd/.gxf X-component decay constant (tau) in microseconds (UTM coordinates, NAD83 datum)thmadcz27.grd/.gxf Z-component decay constant (tau) in microseconds (UTM coordinates, NAD27 datum)thmadcz83.grd/.gxf Z-component decay constant (tau) in microseconds (UTM coordinates, NAD83 datum)


50

APPENDIX F GEOTIFF AND VECTOR ARCHIVE DEFINITION

GeoTIFF Images

Geographically referenced colour images are provided in GeoTIFF format for use in GISapplications.

The images are summarized as follows:

thmamag27.tif IGRF-corrected magnetic field with planimetric base (UTM coordinates,NAD27 datum)

thma2vd27.tif shadowed second vertical derivative of the IGRF-corrected magnetic fieldwith planimetric base (UTM coordinates, NAD27 datum)

thmadcx27.tif X-coil decay constant with planimetric base (UTM coordinates, NAD27datum)

Vector Archives

Vector line work from the maps is provided in DXF ASCII format as follows:

thmaem27.dxf electromagnetic anomalies (UTM coordinates, NAD27 datum)thmakc27.dxf Keating correlation targets based on the IGRF-corrected magnetic field

grid (UTM coordinates, NAD27 datum)thmapath27.tif flight path of the survey area (UTM coordinates, NAD27 datum)thmamag27.dxf contours of the IGRF-corrected magnetic field in nanoteslas (UTM

coordinates, NAD27 datum)thmadcx27.tif contours of the X-coil decay constant in microseconds (UTM coordinates,

NAD27 datum)

The layers within the DXF files correspond to the various object types found therein and haveintuitive names.


51

APPENDIX G HALFWAVE ARCHIVE DEFINITION

Halfwave Data

The halfwave data are stored in ASCII files named line_number .stk (e.g. 82120.stk for line82120). On the DVD, these files have been zipped to save space.

Each ASCII file contains five columns of data: the fiducial and the four measuredelectromagnetic components Tx, X, Y, and Z, stored in that order. These components are:Tx - the amplitude of the primary (transmitted) fieldX - the amplitude of the secondary field as seen by the X coilY - the amplitude of the secondary field as seen by the Y coilZ - the amplitude of the secondary field as seen by the Z coil

These components are stored in digitiser units, where 1 digitiser unit is approximately 0.477microvolts.

Each fiducial contains a 256-point waveform (i.e. fiducial value repeats 256 times). Each fiducialrepresents a 0.2 second sample. This is an eighteen-fold stack from the original sampling rate of 90Hz to the halfwave sampling rate of 5 Hz.

The windowing of the halfwave data to eight channels, after transformation to the step response,is defined as follows:

Window 1: sample 2 1 sample wideWindow 2: samples 3 to 4 2 samples wideWindow 3: samples 5 to 8 4 samples wideWindow 4: samples 9 to 16 8 samples wideWindow 5: samples 17 to 32 16 samples wideWindow 6: samples 33 to 64 32 samples wideWindow 7: samples 65 to 128 64 samples wideWindow 8: samples 129 to 256 128 samples wide

Window 8 is then used to normalize the step response for windows 1 to 7, as a means ofremoving the transmitter-receiver coupling changes of the primary field measured at the receiver.

gds1101-rev - matheson area, magnetic and electromagnetic …

Documents