Geological correlation of magnetic susceptibility and profiles from Nordre Strømfjord, southern West Greenland

The Palaeoproterozoic Nagssugtoqidian orogen is dominated by reworked Archaean gneisses with minor Palaeoproterozoic intrusive and supracrustal rocks. The Nagssugtoqidian orogen (Fig. 1) was the focus of regional geological investigations by the Geological Survey of Denmark and Greenland (GEUS) in 2001 (van Gool et al. 2002, this volume). In conjunction with this project, geophysical studies in the inner part of Nordre Stromfjord, Kuup Akua and Ussuit were undertaken as part of the Survey’s mineral resource assessment programme in central West Greenland. The studies include geophysical modelling of airborne magnetic data, follow-up studies of aeromagnetic anomalies by magnetic ground surveying, and geostatistical treatment and integration of different geological, geophysical and geochemical data. The aim is to obtain an interpretation of the region in terms of both regional geological features and modelling of local features of relevance for the mineral resource assessment. This paper presents an account of the field work and some of the new data. The work was carried out from a rubber dinghy in the fjords and from helicopter-supported inland camps. In situ measurements of the magnetic susceptibility of rocks and magnetic ground profiles were carried out during a period of 25 days in June and July 2001. In total 133 localities were visited from three camps. The data collected will be used together with magnetic properties and density of rock samples determined in the laboratory for geophysical modelling of the area. The petrophysical data will constrain the geophysical and geological interpretations and thus provide a higher degree of confidence in the models. Magnetic susceptibility

The Palaeoproterozoic Nagssugtoqidian orogen is dominated by reworked Archaean gneisses with minor Palaeoproterozoic intrusive and supracrustal rocks.The Nagssugtoqidian orogen (Fig. 1) was the focus of regional geological investigations by the Geological Survey of Denmark and Greenland (GEUS) in 2001 (van Gool et al. 2002, this volume).In conjunction with this project, geophysical studies in the inner part of Nordre Strømfjord, Kuup Akua and Ussuit were undertaken as part of the Survey's mineral resource assessment programme in central West Greenland.The studies include geophysical modelling of airborne magnetic data, follow-up studies of aeromagnetic anomalies by magnetic ground surveying, and geostatistical treatment and integration of different geological, geophysical and geochemical data.The aim is to obtain an interpretation of the region in terms of both regional geological features and modelling of local features of relevance for the mineral resource assessment.This paper presents an account of the field work and some of the new data.
The work was carried out from a rubber dinghy in the fjords and from helicopter-supported inland camps.In situ measurements of the magnetic susceptibility of rocks and magnetic ground profiles were carried out during a period of 25 days in June and July 2001.In total 133 localities were visited from three camps.
The data collected will be used together with magnetic properties and density of rock samples determined in the laboratory for geophysical modelling of the area.The petrophysical data will constrain the geophysical and geological interpretations and thus provide a higher degree of confidence in the models.

Magnetic susceptibility
Magnetisation is defined as the magnetic moment per unit volume.The total magnetisation of a rock is the vector sum of the remanent magnetic moment that exists irrespective of any ambient external magnetic field, and the induced magnetic moment that exists because of the presence of the external magnetic field.The strength and direction of the induced magnetic moment is proportional to the strength and direction of the external magnetic field.The proportionality factor is termed the magnetic susceptibility (denoted with the symbol χ and assumed to be a scalar quantity).In the following sections all quantities are referred to the SI system (Système International) in which the magnetic susceptibility becomes dimensionless.
The magnetic susceptibility of rocks was measured with a hand-held magnetic susceptibility meter (Fig. 2).To obtain estimates of the remanent magnetic component and more precise results of the magnetic susceptibility it is also necessary to investigate rock samples in the laboratory.The in situ measurements presented in this paper were obtained during the field work; the results of laboratory investigations currently being carried out at the petrophysical laboratory at the Geological Survey of Finland are not yet available.
The amount and distribution of the magnetic minerals in a rock determine the magnetic response measured along a profile.The content of magnetite (Fe 3 O 4 ) and its solid solution ulvöspinel (Fe 2 TiO 4 ) is the dominating factor in crustal rocks (Blakely & Connard 1989).The magnetic susceptibility of gneiss is normally between 0.1 x 10 -3 SI and 25 x 10 -3 SI (Telford et al. 1998).

Data acquisition and processing
The magnetic susceptibility meter used in the field was a Geo Instrument GMS-2 (Fig. 2).Depending on the homogeneity of the rocks and the size of the outcrop, ten to forty readings were taken at each locality to ensure a proper statistical treatment of the measurements.Outcrops were selected so as to provide the most representative measurements of the rock on unweathered, smooth surfaces.In total 3444 readings were taken at the 133 localities.In the statistical treat- ment the measurements were grouped according to locality, rock type and geological province.In cases of very heterogeneous rocks, the relative proportions of the rock types present were estimated and data weighted accordingly.Magnetic susceptibility measurements were also made as a secondary task by two other field teams in the western part of Nordre Strømfjord, at Attu, in the Ikamiut area and at Lersletten, but these data are not included in this presentation.Two long magnetic profiles were made (Fig. 3) with measurements of both the total field and the vertical gradient using a magnetic gradiometer (Geometrics G-858); another magnetometer (Geometrics 856) was used as base magnetometer.The sampling distance along the profiles was approximately 1 m.As an example, the magnetic total field intensity from the south-Fig.2. The hand-held magnetic susceptibility meter is small and easy to use.Measurements are taken first with the meter at the rock surface, followed by a reference reading with the meter held up in the air.The photograph shows the first step of the measurements on typical gneiss lithologies in the central part of Ussuit fjord.ernmost two kilometres of the profile undertaken from the eastern inland camp south of Ussuit is discussed in a later section.

Regional geology
Reworked Archaean gneisses with minor Palaeoproterozoic supracrustal and intrusive rocks dominate the Nagssugtoqidian orogen.The region studied in this paper lies in the eastern part of the central Nagssugtoqidian orogen (CNO; Marker et al. 1995).The CNO is bounded by the Nordre Strømfjord shear zone to the north and the Ikertôq thrust zone to the south (Fig. 1).
Subregion 1.The rocks of the northern CNO flat belt are dominated by Archaean orthogneisses with a grey to white colour and variably developed banding; major open upright antiformal structures are characteristic.The gneisses are intercalated with narrow belts of Palaeoproterozoic supracrustal rocks, spatially associated with the calc-alkaline Arfersiorfik intrusive suite.The supracrustal rocks are often strongly foliated and migmatised with several leucosome phases, and comprise mafic amphibolite bodies and layers, ultramafic bodies, pelitic schists, marble and calc-silicate rocks, and fine-to medium-grained quartz-rich paragneisses with biotite and garnet.
Subregion 2. The Nordre Isortoq steep belt separates the northern CNO flat belt and the southern CNO.The steep belt is a zone of steeply dipping and isoclinally folded orthogneiss and paragneiss, and is dominated by an up to five kilometres wide belt of supracrustal rocks.The supracrustal rocks comprise mainly pelitic and psammitic paragneisses, with lesser amounts of mafic to ultramafic bodies and layers, amphibolites and calc-silicate rocks.The gneisses are very variable in appearance, and range from felsic migmatitic gneiss types to more pelitic and mafic types.The rocks are in granulite facies.
The study region is cross-cut by several NE-SWtrending faults of unknown age.The rocks to the west of Kuup Akua and north of the northern border of the Nordre Isortoq steep belt, including the Ussuit area, are all in amphibolite facies.The rocks to the east of Kuup Akua and further north are in granulite facies.

Regional aeromagnetic data
The aeromagnetic anomaly data for the study region resulting from project Aeromag 1999 (Rasmussen & van Gool 2000) were obtained by subtraction of the International Geomagnetic Reference Field (IGRF) from the measured data, and correlate well with the surface geology of the region (Figs 1, 3).
The subdivision of the CNO and the boundary features are clearly reflected in the aeromagnetic anomaly data.The Nordre Strømfjord shear zone stands out as a sharp discontinuous ENE-WSW lineament.Magnetic domains can also be recognised coinciding with the three geological subregions.
Subregion 1.The northern CNO flat belt is characterised by elongated and curved, short wavelength anomalies reflecting the folded nature of this domain.These anomalies are superimposed on a regional magnetic field level of around zero.
Subregion 2. The Nordre Isortoq steep belt stands out as an ENE-WSW-trending regional magnetic low with superimposed low amplitude, elongated, short wavelength anomalies.Based on the magnetic data alone, it may be argued that the northern border of the steep belt should be placed more northerly than that depicted in Fig. 3, for which only the central part has so far been confirmed by mapping.The low magnetic anomaly is partly due to the presence of supracrustal rocks.Uniform low magnetic response of supracrustal rocks is confirmed from many other regions of the world (Card & Poulsen 1998).
Subregion 3. The southern CNO has a high magnetic regional level and is characterised by closely spaced short wavelength anomalies with steep horizontal gradients.Several fold structures can be recognised in the magnetic anomaly patterns.A NNW-SSE-trending low magnetic feature cross-cuts the eastern part of the southern CNO.The anomaly is weak in the steep and flat belt regions.
The border between the CNO and the southern Nagssugtoqidian orogen (SNO; Fig. 1; van Gool et al. 2002, this volume) stands out as a very sharp and large gradient, which can be correlated with the Ikertôq thrust zone.

Magnetic susceptibilities of different rock types
The magnetic susceptibility measurements presented here show that the different rock types exhibit a wide range of susceptibility values within the same formation, and even on the same outcrop.The measurements for the main rock types of the studied area are given in Fig. 4 and Table 1.
The susceptibility in SI units for the entire data set ranges from 0.0 to 91.41 x 10 -3 SI.The rock types examined include orthogneisses, paragneisses, a variety of supracrustal rocks, and intrusives related to the Arfersiorfik quartz diorite.The highest values correspond to orthogneisses, whereas some marbles and gneisses are virtually non-magnetic.

Gneiss
The susceptibility distribution for all types of gneisses is shown in Fig. 4A.In total, 2372 measurements were made on 77 gneiss localities.The variability of the gneisses in the field is reflected in very variable magnet-ic susceptibilities ranging from 0.0 to 68.18 x 10 -3 SI, with a geometric mean value of about 1.21 x 10 -3 SI.The negative skewness (Table 1) of the measurements in Fig. 4A shows an asymmetric tail extending towards lower values.This may reflect that the generally low measured susceptibility values have a too low mean, perhaps due to near-surface weathering of the rocks.
In general, the metamorphic facies is reflected in the susceptibility values, with high values for granulite facies gneisses and lower values for amphibolite facies gneisses.This is probably caused by the formation of magnetite under granulite facies metamorphism (Clark 1997).Moreover, the gneiss type is clearly reflected in the susceptibility values, with low values for paragneisses and higher values for orthogneisses.Visible magnetite was often observed in migmatites, which possibly indicates formation of magnetite during migmatisation, and is reflected in the high susceptibility values.

Rock type
The skewness characterises the degree of asymmetry of a distribution around its mean.The geometric mean is the mean of all the obtained susceptibility values larger than zero for the given rock type.

D C
Fig. 5. Magnetic total-field intensity map with shaded relief.The response observed from the airborne and ground magnetic survey profiles south of Ussuit (see Fig. 1).The NNW-SSE-trending white line C-D shows the location of the ground profile.The N-S-trending white line is the airborne magnetic profile.The white lines also define the zero level for the magnetic total field intensity data shown as a black curve.
The grey circles show locations where selected susceptibility values were obtained in the field.
have a reducing effect, which hinders the formation of magnetite.As was the case for the amphibolites, the negative tail of the distribution possibly reflects weathered rocks.

Marble
Marbles from five localities are very similar, all with very low susceptibility values (Fig. 4E) due to the high content of non-magnetic calc-silicate minerals.The highest values for marble were obtained at one locality where the marble contained thin intercalated mafic mica schist bands and was penetrated by pegmatite veins.In general, the magnetic susceptibility is almost negligible, and the marble lithologies can thus be considered as forming non-magnetic units.

Intrusive rocks: Arfersiorfik quartz diorite
Susceptibility values were taken at six outcrops of the Arfersiorfik quartz diorite, and fall into two groups (Fig. 4F).The first group has high values ranging from 0.69 to 6.88 x 10 -3 SI, while the second group has lower values between 0.01 and 1.10 x 10 -3 SI.Field observations indicate that the quartz diorite varies in appearance from dark to light coloured types, due to varying amounts of mafic components, quartz content and grain size, which may explain the variance of the susceptibility values.

Magnetic profile data
The NNW-SSE profile measured from the eastern inland camp south of Ussuit is perpendicular to the southern border of the Nordre Isortoq steep belt (line C to D in Figs 3, 5).The profile runs from the low magnetic zone of the steep belt into a more irregular high magnetic anomaly zone.The profile was laid out as a straight line, with start and end points together with every 100 m interval determined by use of the Global Positioning System (GPS).
The central part of this ground profile crosses a small positive anomaly.One of the aims was to compare the details obtained from the ground measurements with a profile from the Aeromag 1999 survey (Rasmussen & van Gool 2000).The airborne magnetic profile was flown at an altitude of 300 m, runs N-S and intersects the ground profile (Fig. 5).The difference in content of short wavelength anomalies in the two survey types (Fig. 5) clearly illustrates the attenuation with increased distance to the sources, which has significant implications for the amount of detail that can be acquired from the airborne data.However, a clear correlation with the observed surface geology is confirmed by the ground profile.
The sharp positive anomalies observed in the central part of the ground profile correlate with a 100-150 m wide zone containing ultramafic rocks, whereas the lower magnetic anomalies reflect gneiss lithologies.The locations of the lowest anomalies can be related to calc-silicate horizons observed in the field.Based on these observations it can be concluded that the small positive anomaly in the aeromagnetic data originates from the presence of ultramafic rocks.
Combined forward modelling and inversion undertaken with tabular bodies as the principal model is shown in Fig. 6 for both the ground and airborne profile.The free parameters in the final inversion are location, size, thickness and magnetic properties.Some initial modelling with the dip angle as free parameter indicates that a steep northward dip of the bodies gave the best data-fit.In the final inversion the dip angles for all bodies were identical, except one body for which it was necessary to deviate slightly from the common angle in order to obtain a proper data-fit.The relatively thin alternating bodies of rocks with different magnetic properties, necessary in the modelling, reflect the banded nature of the geology in the study region.
To test the agreement of the field susceptibility measurements with the values obtained by the modelling, the modelling was undertaken without any constraints on the magnetic properties, but with the assumption that the direction of magnetisation was aligned along the present direction of the geomagnetic field.Thus the modelling does not distinguish between a remanent magnetisation in the direction of the geomagnetic field and the induced magnetic component.The magnetic susceptibility values for the bodies in the modelling range from 0 to 92 x 10 -3 , with a mean around 30 x 10 -3 .This is one order of magnitude higher than the geometric mean values of the measured susceptibility values, but within the range obtained from the measured values.An explanation to this discrepancy may be that the remanent magnetic component contributes considerably to magnetisation; however, this has not been confirmed by laboratory measurements on rock sample from the Survey's archive.
Although modelling of potential field data is known to be highly ambiguous, the model presented above includes features that are expected to be common to all models that are realistic representations of the geology.More detailed modelling and further study including measurements of the magnetic properties are warranted.

Conclusions and further work
The aeromagnetic data reflect the regional geology well.Further work will involve interpretation through processing and modelling.
The ongoing construction of a large database of magnetic susceptibilities and other petrophysical parameters, coupled with observations on rock types and structures, will help to elucidate the correlation between the geology and magnetic responses, and is a prerequisite for realistic geological interpretations of the aeromagnetic surveys from the area.
The field measurements show that the magnetic susceptibility is variable within the same rock type, and even on individual outcrops there are considerable variations.Gneiss and schist lithologies in particular have very variable susceptibilities, probably reflecting the variable nature of the lithologies, e.g.pelitic to psammitic.Ultramafic rocks and amphibolites, and to a lesser extent some intrusives of the Arfersiorfik quartz diorite suite, show relatively high magnetic susceptibilities within a narrow range.Marble is essentially a non-magnetic rock type.All susceptibility values obtained from the different lithologies are within the typical range for such rock types (Clark & Emerson 1991;Shive et al. 1992;Clark 1997;Telford et al. 1998), and are in agreement with values obtained in previous investigations (Thorning 1986).The variable susceptibility values of the rock types reflect the different nature of the rocks and their different geological histories, e.g.metamorphism, hydrothermal alteration, bulk composition, etc.More work will be necessary to analyse the susceptibility values in relation to these factors.The discrep-Fig.6. A: The airborne profile data (A-B, the dashed white line in Fig. 3) and the resulting model from the modelling with the projection of the ground profile (line C-D in Figs 3, 5).B: The ground profile data and the resulting model from the modelling.Measured magnetic total field intensity data are shown in black, and the response of the models in red.Green-coloured bodies are the magnetic bodies used to model the measured data.The grey shaded regions in the model correspond to magnetic reference level; i.e. zero magnetic susceptibility.ancy between the susceptibility values obtained in the field and those indicated by modelling will also have to be investigated further.The ground magnetic profile carried out during the field season illustrates well the significant difference in resolution of the geological details that are possible from different survey types, at the same time confirming the correlation of geology and airborne anomalies.The investigations will continue in the 2002 field season, when ground geophysical surveys will be undertaken in connection with lineament studies and the study of a mineralised horizon in amphibolite at the fjord Inuarullikkat (Stendal et al. 2002, this volume).The database of the magnetic susceptibility of rocks will be supplemented with new measurements and with laboratory determinations of petrophysical properties when these become available.

Fig. 3 .
Fig. 3. Magnetic total-field intensity map with shaded relief.The tectonic boundaries and subregions of the central Nagssugtoqidian orogen stand out as distinct lineaments and zones in the magnetics.Shading is with illumination from the north-north-west.Black triangles mark the position of the three field camps in the study area.Dotted lines mark the boundaries within the central Nagssugtoqidian orogen.Closely spaced dots indicate boundaries mapped in the field, and wider-spaced dots extrapolations based on the aeromagnetic data.The airborne and ground magnetic profiles are shown with dashed (A-B) and full white lines (C-D), respectively.The part of the ground profile used for modelling is shown as the grey part of the line C to D (see also Fig. 5).

Fig. 4 .
Fig. 4. Magnetic susceptibility distribution in per cent for different rock types.