DEVELOPING GEOPHYSICAL SIGNATURES TO CONSTRAIN GEOLOGIC MAPPING

DEVELOPING GEOPHYSICAL SIGNATURES TO CONSTRAIN GEOLOGIC MAPPING

by Jeff Wynn, Sue Karl, Bruce Smith, and Anne McCafferty US Geological Survey

INTRODUCTION Greens Creek, one of the largest producing volcanogenic massive sulfide (Ba-Pb-Zn-Ag) deposit in southeast Alaska, is found on north-central Admiralty Island in southeastern Alaska (Newberry and Brew, 1997; see Figure 1). It and other known VMS prospects in the region are found only in Triassic Hyd Group volcanic rocks. Rocks underlying Admiralty, as well as Kupreanof, Zarembo, and Etolin, Island to the south are part of the Alexander Terrane, have been inferred to represent an island arc depositional environment and tectonic setting that was biogeographically isolated from North America during the Silurian, and were accreted to North America in the Cretaceous (Berg and others, 1978; Karl and others, 1999; 2000). Proterozoic and Paleozoic rocks of the Alexander Terrane are unconformably overlain by Middle to Upper Triassic volcanic and sedimentary rocks of the Hyd Group.

Future success of mineral exploration in the region is dependent on accurate geologic maps. Because much of the region is covered by dense vegetation, water, and tidal mud-flats, ground and airborne geophysical information have proven to be the most effective tools for finding appropriate host rocks for more Greens Creek type VMS deposits. A large airborne geophysical survey, incorporating both electromagnetic (EM) and magnetic sensors, was flown over Kupreanof, Zarembo, and Etolin Islands during May of Figure 1. Index map showing Zarembo Island. 1

1997 (Burns and Liss, 1997). This survey was funded by the village of Wrangell, the US Bureau of Land Management, and the State of Alaska, with a goal of strengthening the regional economy. In addition to the airborne survey, ground magnetic and VLF-EM resistivity profiles were acquired, and were closely coordinated with detailed geological mapping. These profiles were acquired over key areas where geology could be unequivocally mapped, in order to develop geophysical signatures of key rock units, in particular the lithologically similar Cannery Formation, Hyd Group, and Seymour Canal formation argillaceous rocks. This work was done as part of a much larger regional-scale mineral resources assessment extending from Admiralty Island to Prince of Wales Island and covering most of southeast Alaska (See Figure 1).

The results of this effort are a set of geophysical signatures for key geologic units in the region. These geophysical signatures have been used to interpret the airborne geophysical survey in areas where it is difficult or impossible to examine rocks directly due to water, tidal flat, or vegetation cover. The method was effective for some geologic units, but not all. Complications arise in part because electromagnetic methods do not work over water or tidal flats, and also because metamorphism to greenschist/amphibolite facies in the region overprint some of the rocks. Furthermore, some of the original mapped geologic units are age-correlated “catch-all” units, incorporating both sedimentary and volcanic components. These generalized geologic units, the metamorphic overprint issue, and the geophysical limitations discussed earlier make defining geophysical signatures more difficult.

Despite these complications, we have developed a set of geophysical signatures and predictive geophysical models (see McCafferty and others, 1999) for select geologic units and mineralized areas. Through a closely-coordinated geologic-geophysical iterative process, older regional geologic maps have been dramatically updated. Among other things, Triassic Hyd argillites can now be separated from the Cannery formation argillites, and there are substantially larger areas now known to be underlain by Triassic rocks and therefor permissive for VMS mineralization. The Cannery formation is now known to have two distinct “geophysical subpopulations”, apparently differing only by the degree of metamorphism. This latter distinction

2

reflects a lower porosity in the more metamorphosed rocks, however magnetite content is apparently little changed by the metamorphic process.

As part of the ground follow-up to the airborne geophysical survey, a series of geologic/geophysical profiles were acquired on Kupreanof, Woewodsky, and Zarembo Islands during May of 1998. A description of the extended project’s findings is beyond the scope of this paper, so we will focus on just Zarembo Island, a large island in the center of the study area, to demonstrate Figure 2. Geo-profiles on Zarembo Is. the methodology we used for developing geophysical signatures and how the signatures were used to constrain geologic mapping.

The locations for the geophysical profiles were chosen to cross key rock units, structures, and mineral occurrences, in order to identify positions and orientations of geologic contacts, to verify airborne geophysical survey contacts, and to define ground-based geophysical signatures of units to help interpret the airborne geophysical data. The profiles on Zarembo Island are shown in figure 2; a key profile discussed later, designated “Road 6594-6590 SE Zarembo” is shown in red.

THE AIRBORNE EM AND MAGNETIC DATA

Figure 3 shows the airborne magnetic data for Zarembo Island. The major magnetic highs coincide Figure 3. The magnetic map of Zarembo Island. 3

with mapped Cretaceous intrusives, but a major intrusive in the northeast of the island has no apparent magnetic anomaly. Figure 4 shows the 7200 Hz airborne EM results converted to resistivity values. The blue colors around the edge of the island are the extremely low resistivities due to the surrounding seawater. The intrusive bodies on the island generally manifest high resistivity (i.e., low porosity). Note the relatively low resistivity of the through-cutting valley that bisects the island from the east to the northwest side. This may be a major suture or perhaps a wide, through-going fracture zone channeling relatively more water than the surrounding rocks.

Figure 4. The resistivity map of Zarembo Island.

Figure 5 shows Euler deconvolution solutions (see Blakely, 1996) posted against shaded-relief topography. Euler deconvolution solutions are calculated wherever there is a gradient or change in the magnetic field intensity. Consequently, a solution plot also generally shows the locations of the underlying lithologic contacts. The color spectrum and increasing circle

Figure 5. Structure inferred from Euler deconvolution of the magnetic grid of Zarembo Island. Note arrows showing dips.

4

diameters together indicate depth-to-magnetic-source, based on solutions calculated at the gradient points in the magnetic data. The arrangement of the solutions, while providing depth-tosource information, can also be used to infer lithologic transitions, faults, and other structural information. In some cases, on-lapping solutions can be used to even infer dip to a contact or fault. A close examination of the solutions in the Road 6594-6590 SE Zarembo area in the southeastern quadrant of Zarembo island shows consistent south or southwestern dips at the gradient points (note arrows in figure 5). These dips are in the 30o to 50o range; better precision than this would require modeling.

THE GROUND SURVEY - INSTRUMENTATION AND METHODOLOGY

In our ground follow-up geophysical profiling effort, we used a proton-precession magnetometer, a magnetic susceptibility meter, and a VLF-EM resistivity unit. The magnetometer (with a 1 nTesla sensitivity setting) was used in the profiles to map magnetic mineral content, such as magnetite and pyrrhotite, and to distinguish magnetic signature characteristics (including spatial coherence and noise). We can often use these different signature characteristics to help us distinguish between buried lithologies of otherwise similar magnetite content. The depth of investigation of the magnetic method is limited only by the profile length; short-wavelength anomalies are caused by shallower sources than the broader, long-wavelength anomalies, and we can see more high-frequency (e.g., shallower source) anomalies using the ground magnetic data than the airborne magnetic data. Computer modeling or Euler deconvolution is required to get depths-to-source. In the survey area, we generally encountered several magnetic transitions on each profile (each would usually connote a lithologic boundary).

We used a magnetic susceptibility meter (with a sensitivity around 2 x 10-5 SI units) whenever we encountered a distinct geologic unit that we could add to our susceptibility database. We also used it whenever we encountered an unusual (sometimes negative) magnetic anomaly that might have represented a reversely-polarized rock unit. The magnetic susceptibility data were useful for adding a third dimension to the geophysical signatures used for classifying

5

individual geologic units. They have been useful for subsequent modeling of individual magnetic anomalies in follow-up work.

Table 1 lists accumulated results of susceptibility measurements for geologic units in areas of good exposure that our team geologists were confident had been mapped correctly. The results suggest that Hyd argillite can be distinguished from Hyd volcanic rocks using magnetic data, but there may be some overlap. Cannery Formation rocks have uniformly low magnetic susceptibility, somewhere in the lower end of the Hyd argillite range, however we noted two distinct populations of Cannery when we examined the resistivity data. One of these units (PMc1 on figure 10) observed to the north on Kupreanof Island is substantially more resistive, suggesting more silica and less carbon than the other population (PMc-2), observed on Zarembo Island and the nearby Lindenberg Peninsula. This likely is due to the higher degree of metamorphism to the north and west of Zarembo Island. Unweathered gabbro is generally very magnetic, as are the ubiquitous Tertiary dikes (Qtd) found throughout the area.

Table 1. Geologic Units vs. magnetic susceptibility. Unit

Susceptibility, x 10-5 SI

Rock Unit Description

units Trhv

Hyd volcanic rocks

30 - 80

Trha

Hyd argillite

5 - 30

PMc1

Cannery - cherty argillite, silicic turbidites

10 - 20

PMc2

Cannery - argillite, turbidites

5 - 20

KJss

Seymour Canal greywackes

20 - 100

Qtd

Tertiary/Quaternary dikes

up to 5,000

Tgbk

Gabbro

~2000

The VLF-EM system we used takes advantage of military submarine communications transmissions to map ground impedance (see, for instance, McNeill and Labson, 1991) which the instrument translates into an apparent resistivity value. Experimentally we have learned that in 6

coastal Alaska we must be more than 125 meters from the high-tide mark to avoid effects of seawater on the resistivity measurements. The surprisingly short distance here undoubtedly is due to the substantial rainfall in southeast Alaska, reducing the saltwater incursion that would normally be expected in tidal zones.

We generally used a 25 - 100 meter spacing for the magnetometer stations (shorter spacings were used to better define particular anomalies), with a 50 - 500 meter station-spacing for the VLF-EM resistivity system. We used coarser spacing for the resistivity measurements because the VLF-EM readings took a substantially longer time to make, and in coastline measurements they had to be made at least 125 meters inland from the high-tide mark where the magnetometer survey was generally conducted. This generally entailed substantial bush-bashing, and couldn’t be carried out with the same speed as a magnetometer survey. When unusual anomalies were encountered, however, the nominal VLF spacing of 200 - 500 meters was reduced to better define the target.

Detailed geologic mapping was conducted simultaneously with each geophysical profile. We constructed geologic cross-sections using the ground geophysical profile data plus lithologic and structural data recorded by the geologist along the same traverses.

DISCUSSION OF A TYPICAL PROFILE SOUTHEAST ZAREMBO ISLAND:

The scope of this

Figure 6. Airborne, ground geophysics, and geology, Zarembo Is. 7

paper precludes detailed examination of all the ground profiles; these however can be found in Wynn and others, 2000. For this reason we will consider just one representative profile across the southeastern quadrant of Zarembo Island (figure 6). In the following text, where map units such as “Trhv” are encountered, the “Tr” stands for the Triassic geologic epoch. Trh represents the Triassic Hyd Group undivided; Trhv is the Triassic Hyd volcanics, and Trha is the Triassic Hyd argillite. Trhsv contains sedimentary and volcanic rocks, and PMc represents chert, shale, and sandstone of the Cannery formation. We were able to divide PMc into two distinctive subunits (PMc-1 and PMc-2) using geophysical signature information (see following text). KJss refers to Seymour Canal Formation, which consists dominantly of volcaniclastic sandstones in this area. For more detailed descriptions of these and other units, see Karl and others (1999; 2000).

The “Road 6594-6590 SE Zarembo” profile was designed to cross several major lithologic transitions in southeast Zarembo Island. The upper panel shows the airborne data, and the middle panel shows the ground geophysical profiles together. The match between ground and airborne profiles is not exact because they didn’t follow exactly the same path or even direction (the airborne profile was assembled from a number of different flight-lines). The bottom panel shows the interpreted geologic cross-section based on the geophysical profiles and coincident rock sample collection. There are several narrow magnetic highs at station 500, station 6500 (which may be a dipolar feature), and at 8000; these are probably Tertiary mafic dikes. Station numbers are distance in meters from the beginning of the particular profile. A negativelypolarized magnetic anomaly at stations 900-950 is distinctive and unusual (it probably represents a Tertiary dike) and a susceptibility reading was taken here. The junction of roads 6594 and 6590 takes place near a bridge at station 4300. Flat, relatively noise-free magnetic data also suggests a distinctive lithology and relatively few Tertiary dikes from stations 9700 to 10,400; fine-grained turbidites of Seymour Canal Formation (KJss) were seen to outcrop at the north edge of this zone around station 10,400. There are consistently high resistivities throughout most of the profile.

8

The starting point of this profile is a quarry of alkali-feldspar granite (Tmaz in the geologic map of figure 10; the GPS waypoint is SEZ-0, and this granite continues to station 2800). In these rocks, clinopyroxene is altered to chlorite, and there are the usual, ubiquitous mafic dikes. At station 500 there is a mafic dike at least 7 m thick. This is probably the source of the magnetic anomaly seen in the profile here. Somewhere between stations 2800 to 4000 there is an intrusive contact; the resistivity and magnetic data suggest that it may be around station 3500, and they also suggest that the contact dips roughly northeast. At station 4000 there is a quarry (GPS waypoint SEZ-3) containing hornfelsed cherty rock with calc-silicate layers (garnet, epidote, diopside, albite). This is very siliceous and thin-bedded Cannery Formation (PMc-2) dipping northeast; the resistivity data support this northeastern dip also. The siliceous rocks are hornfelsed to biotite, quartz and pyrite; they are also relatively nonmagnetic and quite resistive. Station 6400 is still hornfelsed Cannery Formation. North of the profile around station 6600 there is a dike of (Tmaz) granite that may be the cause of the sharp dipolar magnetic anomaly seen here. The contact is not exposed, but somewhere near here (judging from the magnetic data around station 7200) there is a depositional contact of Triassic volcaniclastic rocks (Trhsv) lying on the more siliceous Cannery (PMc-2) sediments. At station 7900 there is thinly laminated chloritic metatuff, Trhsv, with felsic Tertiary dikes present. The very strong magnetic high at station 8200 could be a dike of alkalic-feldspar granite (probably Tmaz, but unfortunately we have no susceptibility measurements of the Tmaz unit here to be certain, and this could as easily be caused by a mafic dike).

Somewhere around station 9600 the magnetic and resistivity data suggest there lies the contact of the Seymour Canal Kjss graywacke turbidites over green Triassic volcaniclastic rocks (Trhsv); the resistivity data again suggest the dip of this contact is probably northeast. At station 10,500 the profile crosses complexly-folded graywacke and argillite turbidites (KJss). Another apparently different lithology with a similar magnetic high but a lower resistivity is seen between stations 10,900 and 11,100. It appears to be a discrete, buried, different lithologic unit up to 400 meters wide. The only thing like this that we have seen on Zarembo Island is a felsic porphyry (QTd) observed in a roadcut near Baht Harbor in the west-northwest corner of the island. A final, geophysically-distinct lithology lies between station 11,200 (GPS waypoint SEZ-9) and the end of 9

the profile at GPS waypoint SEZ-10; Seymour Canal Formation rocks (KJss) also outcrop beneath a bridge at station 12,000. As we note the apparent dips in both resistivity and magnetic data, we see that the geophysical profiling can not only identify buried lithologic contacts, but in some cases provide useful structural information about them.

DISCUSSION: GEOPHYSICAL SIGNATURES OF THE GEOLOGIC UNITS IN THE STUDY AREA.

An examination of the magnetic and resistivity data in the study area suggests that some units can be readily discriminated from others, but also that some units - especially those catchall units defined primarily according to age - are too lithologically varied, resulting in widelyvarying physical properties being manifested within a single map unit.

Constraints: It is not realistic to compare absolute magnetic values from profile to profile as is normally done with airborne magnetic data - the ground profiles are acquired in different regions on different days. Within a given magnetic profile, however, it is possible to use both magnetic susceptibility and the magnetic variability within a given geologic map unit to help characterize that unit. In our case we use a simplified, three-level classification scheme of profile magnetic data based on relative frequency-content. For instance, in Table 1 we see that Hyd volcanic rocks (Trhv) are consistently more magnetic than Hyd argillite (Trha), and Hyd sedimentary-volcanic rocks (Trhsv) lie somewhere in between. However, we can also observe that there is more magnetic variability within the Ktif intrusive than in other units, including PMc1 (the more magnetic of the two populations of Cannery Formation containing volcanic rocks), that have otherwise similar magnetic susceptibilities.

10

We have assembled the physical properties of the more important units in the study area into a more graphical form in figures 7 and 8. Figure 7 shows magnetic susceptibility (and its ranges) for 11 key geologic map units. In this figure it is easy to see that Hyd volcanics have much lower magnetic susceptibilty than the Ktif intrusive or the TKd diorite. However, if we used magnetic susceptibility alone, we would not be able to distinguish between Triassic Hyd volcanics (Trhv) and other units, such as the Seymour Canal greywackes (KJss). Note that the two “geophysical”

Figure 8. Resistivity vs. geology, entire study area. Note also the magnetic variability, a measure of the frequency-content of the mag.

populations of Cannery Formation can also be distinguished or characterized geologically: PMc-1 represents sandstones, chert, and volcanics, while PMc-2 is consistently carbonaceous chert and shales. Figure 8 shows that they can generally be distinguished by their contrasting resistivity alone.

It would difficult to separate the two Cannery populations from the Hyd argillite solely on the basis of susceptibility measurements. Using resistivity, however, the Zarembo Cannery population (PMc-1) is more siliceous and much more resistive than Hyd argillite, so we can thus

Figure 7. Magnetic susceptibility vs. geology, entire study area. 11

distinguish them - at least on Zarembo Island. Gabbro and the ubiquitous Tertiary dikes (QTd) are strongly magnetic; so strongly, in fact, that even relatively narrow dikes can be seen clearly in the airborne magnetic data (see McCafferty and others, 2000). Seymour Canal sedimentary rocks (KJss) appear in both susceptibility measurements and profile data to be moderately variable magnetically, which thus provides us with another characteristic signature feature of these rocks.

We have chosen to also classify some units (where we have enough information), as magnetically “Flat”, “Variable”, or “Noisy” (Figure 8). The “Flat” characterization implies a high degree of homogeneity, at least insofar as magnetic mineral content goes. A “Noisy” classification implies unusually high magnetic heterogeneity within the geologic unit; another way to describe this would be to say the magnetic profile data have a high frequency content. This may be due to a number of things, including mineralized rocks with localized magnetite or pyrrhotite in veins, or significant remanent magnetization. However, a more likely explanation would be local alteration or a characteristic lithologic variability within the geologic map unit. The high variability may also be due to serpentinized shear zones (often causing an intense, narrow magnetic high), or it may reflect localized altered shear zones (typically causing a broad local magnetic low). The “Variable” classification falls somewhere in between these two extremes, and as used here is a somewhat subjective call. We have tried in this classification scheme to distinguish inherent magnetic variability from the magnetic “spikes” (generally singlestation anomalies) caused by the highly magnetic Tertiary dikes found throughout the study area (note: the field crew made a strenuous effort to avoid the ubiquitous culverts and steel logging cables during the profile measurements).

Resistivity data, on the other hand, can be considered and treated as absolute numbers, and we have shown these data in figure 8 as ranges of resistivity within a given geologic map unit. To a small extent these ranges may be subjective, since localized faulting or the unseen proximity of a culvert can lead to resistivity lows, and certain igneous dikes, that we may have not seen, can cause local resistivity highs. It is important to keep in mind that resistivity measurements in the study area, because of almost constant rainfall, are in almost all cases roughly correlatable with porosity, not ionic content of pore-fluids. For this reason, high silica content in the cherty PMc-1 12

(Cannery) unit will generally give very high resistivities, whereas the more sooty (carbonaceous) population (PMc-2) is more conductive. Massive sulfides such as those observed elsewhere in the study area will typically give rise to unusually low resistivites. Where we see occasional localized low resistivity values in the PMc-1 Cannery unit (the lower end of the resistivity range in figure 8), it may indicate local shear zones or carbonaceous layers..

Discrimination between units and predictive geophysical models: Having stated these limitations, we can see from figures 7 and 8 that using ground geophysical profiles most geologic units can be discriminated from others -- if sufficient data are available. We can also see a wide range of resistivities for the Trhsv, PMc1, and Ktif units, suggesting that these geologic units are not narrowly inclusive, but in fact contain a range of rocks with substantially different porosities. Despite similar wide ranges of resistivities, we can look at the profiles and still use them to distinguish Trhsv from PMc1 because the former has substantially variable magnetic values, whereas the PMc1 unit is magnetically inert and homogeneous - “flat” (the limited susceptibility data support this). The Ktif intrusive unit seems superficially like the Trhsv volcaniclastics in the geophysical data, but the high magnetic variability within the Ktif intrusive is striking - probably due to local alteration, magnetic dikes, or inclusions of nearby country rock. With a long enough profile length in Trhsv units, we were able to use geophysical data to distinguish between the two units relatively easily. Although it may be difficult to compare between profiles, contacts within profiles are relatively obvious.

It appears difficult, but perhaps not impossible, to discriminate between Trhv (which should be more resistive, since it’s mostly volcanic rocks with little argillite) and Trha (which should have lower resistivity, being mostly argillite), using the geophysical profiles. Both have low resistivity, both are flat magnetically. Fortunately both are target units for our project, but this problem demonstrates the limitations of this method.

13

Figure 9 shows a predictive geophysical model for the airborne survey for Triassic Hyd argillite (the colored band displayed from northwest to southeast). Note that this band is on trend with other known Triassic Hyd rocks to the northwest on Kupreanof Island. This model takes into account the range of geophysical properties observed in both mag and EM data, and was developed in close coordination with our team geologists.

Finally, figure 10 shows the revised geologic map for Zarembo Island derived from the geologic sampling and the geophysical profiling described above. This map is substantially different from the older reconnaissance geologic map... in large part because it utilizes the airborne geophysical data in the form of the predictive geophysical models like the one shown in figure 9. This map includes approximately 30% more Triassic rock than the older reconnaissance map, dramatically improving the chances for finding another Greens Creek - type VMS deposit.

Figure 9. The geophysical model for Triassic argillite. 14

Summary: Most geologic units broadly distributed throughout the study area can be separated from each other using ground geophysical profile data - if there is enough profile data available and test sites can be unambiguously identified as being one unit or another. There was relatively little Mzg and Kdi sampled by our survey program, so these units must be considered still poorly-characterized, at least in a geophysical sense. With additional data, they may prove sufficiently distinctive to Figure 10. The resulting revised geologic map of Zarembo Island. Only the intrusives and the Triassic rocks are colored for clarity. permit us to separate them using geophysical profile data.

We have the theoretical advantage of using three effective physical properties that we can use to distinguish rocks on the ground: magnetic susceptibility, magnetic variability or spatial 15

frequency content, and resistivity. In practical terms, however, we cannot describe all rock units by all three geophysical signatures, so a 3-D matrix is not shown here. These limitations stem from the limited number of outcrops that we have confident age-dates for, as well as limited exposures in general. The ground profiles nevertheless proved very useful to fine-tune our interpretations of the airborne survey.

The geophysical profiles also allow us to characterize lithologic transitions under large areas of rain forest, muskeg, and mud-flat cover, because in many cases we can assign distinctive geophysical signatures to specific geologic units. Defining lithologic transitions is relatively straightforward, based on magnetic field strength within a given profile, frequencycontent of the magnetic profile data (referred to in figure 8 as “Magnetic Variability”), and different resistivities. However, it is not always possible to say that each particular geologic unit has a distinctive geophysical signature over the entire study area. This occasional uncertainty is due to several factors, some physical and some geological. Most geologic units were originally distinguished or named in large part by their age (often indirectly inferred). A single unit may therefor include a wide range of facies or phases, with substantially different amounts of silica (which will increase the resistivity), graphite (which will decrease the resistivity), and even substantial variation in magnetite content (which will strongly affect the magnetic survey data, both its amplitude and frequency-content). In some cases, strong localized hydrothermal alteration may remove or convert the magnetite to hematite, locally reducing the magnetic susceptibility of the rock, and the associated argillization will often dramatically reduce the resistivity (and increase the conductivity) at the same time due to increased clay-content. Serpentinization in mafic rocks can often give rise to widely-varying magnetic susceptibilities over a short profile length. Further, in areas where low-porosity (and therefor high resistivity) rocks have been sheared up by tectonic events and faulting, we will often observe a much lower resistivity in the local sheared area due to increased weathering and water-filling in and around the fractures, along with a concomitant change in magnetite content (which is usually but not always lower under these circumstances). Increasing degrees of metamorphism within a rock unit caused by different depths of burial can often lead to a lower porosity and higher resistivity for a rock in one area than for the 16

same rock unit in another area. Recognizing this difference means recognizing subtle differences in metamorphic grade. The resistivities for igneous units are generally high except where there has been shearing or alteration; lower resistivities should be expected in sedimentary rocks with greater porosity, but as stated above also in areas where there has been argillic alteration. Silicaflooding during metamorphic events may substantially increase the resistivity of a sedimentary rock (such as cherty PMc1, though its generally higher resistivity may have more to do with lower graphite content in this case).

CONCLUSIONS:

Geophysical profiling on Zarembo Island and elsewhere in the region has shown that some VMS deposits may be conductive or not conductive; apparently this depends upon both the percentage of sulfide present as well as the inter-connectivity of individual sulfide grains (which is to some extent related to the percent of total sulfides present), alteration of the host rocks, and the conductivity of the host (hanging wall and foot wall) rocks. The lesson here is that EM methods will probably not identify disseminated sulfide deposits in SE Alaska. In these cases the more labor-intensive induced polarization method, which is more sensitive to surface area than to grain interconnectivity or volume, may be necessary to map and characterize some sulfide deposits in the study area, but this is a locally specific, not a regional geophysical tool.

The measured total magnetic field is generally higher in mafic igneous rocks and lower in most sedimentary rocks (Van Blaricom, 1992). There are many cases in the world, however, where felsic igneous rocks host substantial magnetite, as is the case with the notably magnetic Tertiary granites on Zarembo and Etolin Islands (Karl and others, 1999; 2000). We have noted that most mafic dikes in the study area have a distinctive high-mag, high-resistivity geophysical signature. We have also noted, however, that the Ktif intrusive is less magnetic than Tmaz even though the Ktif is more mafic (color index