gold environments and favourability in the nuuk area of southern west greenland

GOLD ENVIRONMENTSAND FAVOURABILITY

in the Nuuk area ofsouthern West Greenland

No. 9 - February 2007

GO_09.qxp 23/02/07 12:48 Page 1

So far the interplay between the mag-matic, tectonic and metamorphic evo-lutions and their mineralising eventshas been the main resource evaluation.

During the last couple of years,focus on the primary geological envi-ronment hosting the mineral depositshas started to improve the evalua-tions. Multidisciplinary, spatial, quan-titative analysis of geo-data is usedfor this purpose. The construction ofquantitative, favourability maps for aspecific element, such as gold, is aresult of this.

The new approach has been usedin the Nuuk region in West Greenland,where the statistical, spatial analysishas been used to evaluate the goldpotential. The analysis also providesvaluable input to the interpretation ofthe mineralisation and the primary geo-logical environments.

Geological environments andmineralisation in the Nuuk region

The Nuuk region, located within the Archa-ean North Atlantic craton (the Nain cratonin Canada), comprises several differenttectono-stratigraphic terranes, which hadindependent, geological evolution until theywere tectonically assembled at c. 2.7 Ga.The terranes host several supracrustal belts;the oldest is the Isua greenstone belt withan age of c. 3.8 Ga and younger belts areMesoarchaean. The different belts and areasare described according to their dominatingenvironment.

The investigations of the environmentsrequire the combination of detailed geo-logical mapping, geochemistry, petrographywork, isotope work and geochronologicalstudies.

Island-arc environmentRecently volcaniclastic rocks of andesiticcomposition were discovered in the region.The belt forms major parts of the Meso-archaean, amphibolite facies supracrustalbelt around Qussuk bay, on central andeastern Bjrneen and on central Stor

(3071 Ma). Meta-volcanic rocks have beensubject to intense synvolcanic hydrothermalalteration associated with gold-coppermineralisation (grab samples with up to 2g/t Au). Later hydrothermal events mayalso have affected the occurrences of goldin certain areas.

The Ivisaartoq supracrustal belt, with amaximum age of 3075 Ma, is interpretedas a forearc supra-subduction environ-ment. This belt underwent at least twostages of calc-silicate, metasomatic alter-ation between 3075 and 2963 Ma attrib-

2

The Qingaaq gold prospect at Stor located within a supracrustal sequence. The prospect is currentlybeing drilled and evaluated by NunaMinerals A/S. The lake in the lower left corner is 470 m abovesea level. The top in the background is the mountain Qingaaq (1616 m. a.s.l.), and the top in theforeground is Little Qingaaq (1070 m.a.s.l). Photo courtesy: NunaMinerals A/S.

Volcaniclastic, meta-andesitic rock fromBjrneen with fiamme textures.

GE

OL

OG

Y A

ND

OR

E

9 /

20

07

GOLD ENVIRONMENTS ANDFAVOURABILITYin the Nuuk area of southern West Greenland

GO_09.qxp 23/02/07 12:48 Page 2

6430'N

65N

52W 51W

Isukasia

Fiskef

jord

50W

InlandIce

DavisStrait

Godt

hbs

fjord

Godthbsfjord

Nuuk

Ameralik

Qooqqut

Stor

Bjrneen

Ivisaartoq

25 km

Qussu

k

QA

Isua Greenstonebelt

S

Nordlandet

Sermitsiaq

Sillissianguit

North Atlantic craton

Nagssugtoqidianorogen

KetilidianKetilidianorogenorogen

Ketilidianorogen

Nuukregion

Gneiss - infracrustal

Major faults and shear zones

Supracrustal environments:

Unknown

Island Arc

Ocean Floor

Magmatic

Groups of gold showings:

Isua gold groupBjrneen gold groupStor gold groupUngroup showings

uted to seafloor hydrothermal alteration.The Ivisaartoq belt hosts volcanic massivesulphide deposits (semi-massive to massivepyrite-pyrrhotite) and disseminated tung-sten in altered komatiites (channel sampleswith 0.44% WO3 over 2.5 m). Showingswith up to 3.5 g/t Au are recorded withinthe Ivisaartoq belt.

Ocean-floor environmentThe 3.8 Ga Isua greenstone belt representsan ocean-floor environment. The dominat-ing rock types in this belt are pillow-struc-tured tholeiitic and high Mg-basaltic rockswith intercalated, extensive bands of chertand banded iron formation, turbidites aswell as garnet-mica schist representingmetamorphosed sediments. Gold occur-rences have been found in several settingswithin the Isua greenstone belt.

In silicate facies iron formation (up to 1.2 g/t Au)

In fuchsite-stained quartz-rich rocks (up to 3.1 g/t Au over 3 m)

In shear-zone hosted, carbonate-altered ultrabasic rocks (grab sampleswith up to 100 g/t Au).

A hydrothermal episode affected a c.3.8 Ga intrusive tonalite sheet in the west-ern part of the Isua greenstone belt. Thebelt contains up to 14% combined Pb andZn with up to 6.8 g/t Au and 180 g/t Ag.

Well-preserved, volcanic successionsinterpreted as environments dominated byocean-floor settings are described fromseveral other areas in the Nuuk region. Bi-modal, volcanic succession in greenschistfacies is described from the Nunatak 1390in the Tasiuarsuaq terrane. Voluminouslylayered amphibolites and well-preservedpillow lava at Bjrneen, associated withhomogenous, metagabbroic amphiboliteare also interpreted as a primary oceanfloor environment. A small VMS showingis located in the south-western part of thebelt at Bjrneen. This showing containsup to 3.55% Zn, 1.33% Pb, 33 g/t Agand 0.153 g/t Au.

Magmatic environmentThe Nordlandet and Fiskefjord areas aredominated by mafic to ultramafic sequencesincluding dunite, peridotite, pyroxenite,norite and amphibolites hosted by tona-litic and trondhjemitic gneisses. In manycases relict magmatic layering is preserved.

This area comprises probably both primarymagmatic environments and relicts ofgreenstone belts. The mafic to ultramaficcomplexes south of Fiskefjord show en-hanced values of Au and PGE. The SeqiOlivine Mine operated by Minelco AB islocated just north of central Fiskefjord.The deposit contains at least 100 milliontonnes of high-quality olivine and is host-ed within a large dunite body.

The supracrustal belt at central Stor isof a complex character and much debated.It seems that this belt may contain charac-teristics of both continental and oceanicvolcanic and sedimentary rocks. The areais furthermore affected by intense foldingand major shearing, making straight inter-pretations even more difficult. The area mayrepresent an accretion environment alongboundaries of different terranes (amalga-mated around 2.7 Ga). Central Stor hoststhe Qingaaq Gold prospect which is inves-tigated and drilled by NunaMinerals A/S.Several gold bearing structures, 436 mwide, have been found at Qingaaq withgrades from 0.5 to 6 g/t Au and locally24 m wide zones with grades up to 20g/t Au.

3

G O L D E N V I R O N M E N T S A N D F A V O U R A B I L I T Y

Simplified schematic geo-logical map of the Nuukregion with general indi-cation of different geo-logical environments.

Outlines of the rock unitsare from digital versionsof 1:100 000 and 1:2500 000 scale geologicalmaps published by theSurvey. Q: Qingaaq; A: Aappalaartoq; S: Seqi Olivine mine.

The statistically definedgroups of gold showingsin the region are alsoindicated (a gold show-ing is defined as a localitywhere in situ rock sam-ples yielded 1 g/t Au).

GO_09.qxp 23/02/07 12:48 Page 3

Mineral potential - addressingmultivariable datasets

Visual inspection and comparison of e.g.lithological, topographical, geochemicaland geophysical maps are necessary waysof interpreting geological features acrosspoorly exposed or little explored areas. Inmineral exploration the maps and data areused visually to identify anomalies believedto reflect the existence of mineral occur-rences. However, if the regional data arewidely spaced, and if the occurrences areof limited size or poorly exposed, the ano-malies may be too subtle to be recognisedby a simple visual approach. In such cases,statistical data analyses may help identifydeviations from the background variationsand trends in the data. Furthermore, visu-ally based correlation between distributionsis often limited to a few parameters andwith the wealth of to-days digital data,multivariable relationships are easily over-looked or not possible to deduce visually.

The principles of the statisticalmethodologyAll spatial datasets are converted info aform, where they can be compared andsubjected to statistical analysis. Each data-set is presented as a regular grid using acommon grid cell size denoted a pixel. In

the case of gold potential in a region, pix-els with and without gold showings areregistered. To construct a gold favourabili-ty map includes analysing the characteris-tic data signatures of the pixels, with andwithout gold showings, in multiple data-sets and then integrating the data signa-ture to calculate the gold showing proba-

bility for each pixel. Besides enabling thequantification of the gold favourability ina certain area, the statistical approach pro-vides quantitative signatures of specificgeological features, e.g. the gold show-ings, which can be used and integrated ingeological models and interpretations.

4


Deformed pillow lava fromsouthern part of Bjrneen.

Rusty chromite-magnetite banding from magmatic layered norite in the Fiskefjord area.

GE

OL

OG

Y A

ND

OR

E

9 /

20

07

GO_09.qxp 23/02/07 12:48 Page 4

Data

No.

of s

how

ings

ShowingsNo showings

Empirical distribution functions are calculated for different datasets one for'showings', one for 'no showings'

? ?? ? ? ? ? ? ? ? ? ? ?? ?? ? ? ? ? ? ? ? ? ?

? ?? ? ? ? ? ? ? ? ? ? ?? ? ? ? ? ? ? ? ? ? ? ?

? ?? ? ? ? ? ? ? ? ? ?? ?? ? ? ? ? ? ? ? ? ? ?

?

?

?

Quantified signaturesin different datasets

A search for similar signaturesis carried out.The degree of similarities isused to construct mineral potential maps

Mineral potential map

highlow

favourability for showings

Geoscientific data sets are pixelated

2

1

4

3

AuAu

Au

5


Illustration of the statistical approach:

1: Regional datasets and gold showings (Au) are referred to acommon grid with fixed cell size; a pixel.

2: Empirical distribution functions for pixels with and withoutknown gold showings are used to establish the signatures ofthe showings in each dataset.

3: Likelihood ratio functions for all datasets are calculated onthe basis of the two previous functions, expressing the likelinessof a showing to be present. Groupings of gold showings arebased on constructed mineral potential maps for each showing.Pixel by pixel, these maps outline the predicted favourability fordata signatures similar to the selected showing, i.e. the poten-tial for a similar showing to be present.

4: Finally, mineral potential maps are calculated for each identi-fied group of showings. These maps are based on the signa-tures found to be indicative for each of the different groups.

GO_09.qxp 23/02/07 12:48 Page 5

6GE

OL

OG

Y A

ND

OR

E

9 /

20

07

Rust zone in massive sulphide occurrence at Ivisaartoq.


Gold favourability map for theNuuk region

Data analysedSixty-nine different geoscientific datasets including location of known gold show-ings, data from geological maps, streamsediment geochemistry, aeromagnetic data,aeroradiometric data and data from linea-ment analysis have so far been included inthe statistical analysis. All datasets are pix-elated. A gold showing is in the currentstudy defined as a locality where a rockgrab sample has yielded >1 g/t Au.

Grouping of gold showingsInstead of treating all known gold show-ings as one group, it is statistically ana-lysed whether the showings should be splitinto groups according to common signa-tures in the datasets. For the Nuuk region,this analysis resulted in the identificationof at three main groups of gold showings:the Stor, the Bjrneen and the Isuagroup.

Resulting gold favourability mapBy using the identified common signaturefor gold showings, it is possible to estimatethe probability for similar data signatureswithin all pixels. It is not surprising that

areas immediately surrounding known goldshowings are predicted as favourable, butmore interestingly, a number of favourableareas are also predicted outside areas withknown gold showings.

Gold bearing rustzone in the Qussuk area.

GO_09.qxp 23/02/07 12:48 Page 6

7G O L D E N V I R O N M E N T S A N D F A V O U R A B I L I T Y

Fiskefjo

rd

InlandIce

DavisStrait

Godt

hbs

fjord

Godthbsfjord

Ameralik

Qooqqut

Fiskefjo

rd

InlandIce

DavisStrait

Godt

hbs

fjord

Godthbsfjord

Ameralik

Qooqqut

Nuukregion

Q

A

Q

A

6430'N

65N

52W 51W

Isukasia

Q

A

Q

A

50W

Serm

itsiaq

Nuuk

Stor

Bjrneen

Ivisaartoq

25 km

6430'N

65N

52W 51W

Isukasia

Q

A

Q

A

50W

Serm

itsiaq

Nuuk

Stor

Bjrneen

Ivisaartoq

25 km

Top 685 km2 most gold favourable area (5% of entire study area)

Top 137 km2 most gold favourable area (1% of entire study area)

GneissIce

AmphiboliteMetasedimentUltramafic rock

Fault and shear zonesLate Archaean granites

Stor showingsBjrneen showingsIsua showings

Ungrouped showings

Isua gold group

The top 137 km2 and 685 km 2 most favourable areas for:

Bjrneen gold group

Stor gold group

Simplified geological maps with the toprated 137 km2 and 685 km2 (respectively1% and 5% of the analysed area) outlinedas most gold favourable area for the threegroups of gold showings (Isua, Bjrneenand Stor groups). Outlines of the rockunits are from digital versions of 1:100 000and 1:2 500 000 scale geological mapspublished by GEUS.

GO_09.qxp 23/02/07 12:48 Page 7

Some of the most notable areas out-lined as most favourable for gold are eastof the fjord Qooqqut, areas at Stor, areasat Sermitsiaq, areas south and west ofIsuakasia and areas in outer and inner partof Fiskefjord. Especially in the latter area,hardly any prospecting for gold has beendone, but the area comprises bothsupracrustal rocks and structures that aresimilar to those in areas with known goldshowings. Field work in the inner part ofFiskefjord recently revealed a strong alter-ation of the country rock in that area;mainly bleaching, silicification and epidoti-zation. Rocks from the area yield elevatedgold content with up to 0.377 g/t Au inamphibolite-hosted quartz veins.

The spatial distributionThe spatial distribution of all outlined favour-able areas suggests the existence of a mainregional tract favourable for gold and goldmineralising environments. Most of the fa-vourable areas for the different groups ofknown gold showings are situated within a


8

GE

OL

OG

Y A

ND

OR

E

9 /

20

07

Vertical gradient of total magnetic intensity field [nT/m]

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6

Nor

mal

ised

freq

uenc

y

0.0

2.0

4.0

6.0

8.0

Cs [ppm]6

Nor

mal

ised

freq

uenc

y

0.0

0.8

1.6

2.4

3.2

3 4 52

Ni/Mg ratio5 15 25 35 45 55

Nor

mal

ised

freq

uenc

y

0.00

0.04

0.08

0.12

0.16

As [ppm]2 6 10 14 18 22 26 30

Nor

mal

ised

freq

uenc

y

0.0

0.2

0.4

0.6

0.8Isua grou p Bjrneen grou p Stor grou p Background

Four examples of statistically derived data signatures for the three main groups of gold showings inthe Nuuk region, displayed together with the background signature (pixels without any gold show-ings).

Rust zone with 377 ppb Au in quartz-veined amphibolite, central Fiskefjord.

GO_09.qxp 23/02/07 12:48 Page 8

NNE-trending tract from the town of Nuukto Isukasia. The tract embraces most ofthe previously known gold showings inthe region as well as geochemical goldanomalies, and the zone has previouslybeen suggested as being prospective forgold.The NNE-trending tract outlined asbeing favourable for gold is common tothe general orientation of the supracrustalbelts in this area and to the regional large-scale faults. In general, supracrustal unitsoutside the tract constitute similar propor-tions, but these are not indicated as beingwithin the most favourable areas by thepresent statistical analysis. This indicatesthat the supracrustal units within the tractdeviate from those outside.

Signatures of gold showingsThe systematic construction of data signa-tures of showings and the backgrounddraw attention to the significance of par-ameters that are not immediately or tradi-tionally regarded as indicative of gold min-eralisation. It has previously been estab-lished that the combination of elevated Asin stream sediment and gold mineralisationhas not been found outside Stor. How-ever, the possible significance of the Ni/Mgratio in relation to gold has not previouslybeen substantiated in a quantitative way.This observation is an incentive to conductfurther studies of host rock properties interms of their Ni and Mg behaviour. Therare and unique As signature of the Stor

group poses a problem in the search forareas holding a potential for new show-ings. Since this signature exists on Storsolely, this area will be outlined as favour-able only for the Stor group. Consequent-ly, the As data are omitted in the calcula-tion of pixels with a potential for Stor-type showings. It has before been arguedthat Cs may be considered a pathfinderelement for gold mineralisation associatedwith granite-related, hydrothermal alter-ation. The signature for Cs in stream sedi-ment confirms that the gold at Stor andBjrneen is indeed located in a Cs-richenvironment, but also suggests that themineralisation at Isua has a different nature.

The last example illustrates the indicatedsignificance of the vertical gradient of thetotal magnetic field intensity (denotedVGTMI). This signature for the Stor andBjrneen groups (0.2 to 0.1 nT/m) is iden-tical to the background signature, and isconsequently regarded as non-indicative forthese groups. The Isua group has a moredispersed signature (0.5 to 0.8 nT/m), withan indicative positive and negative VGTMIthan the other groups. This probably reflectslocal-scale changes in lithology. Other data-sets, which yield a characteristic signature ofone or more of the groups of gold show-ings, are the Au, Cs, Rb, La, Th, and U con-centrations in stream sediment geochemisty.


9

Empirical simplified model of geological environments and mineral occurrences present in the Nuuk region.

Basalticmagmatism

Calc-alkalinevolcanism

GraniteTTG

Midoceanicrift

Ivisaartoq(~3.075 Ga)HydrothermalAu, W

Seqi(~3.03 Ga)MagmaticOlivine

Shield

Ultramafic-mafic intrusions

Isuakasia(~3.8 Ga) BIFAu, Fe

QussukBjrneen(~3.07 Ga)VMS, hydrothermalAu, Cu, Zn, Pb

Bi-modalvolcanism

Islandarc

Ultramaficintrusion

Sillissisanguit(~3.03 Ga)MagmaticNi, Cu, PGE

QingaaqAappalaartoq(~2.85 Ga)Hydrothermal, Au

Fault-/shearzonesuture zone

Granite intrusions

Ultramafic-mafic intrusions

Continetal margin

Back-arcbasin

Godthaabsfjord(~2.63 Ga)Pegmatite Tourmaline, beryl, U, Th, REE

Qrqutgranite

Visible gold in drill core from the Stor prospect. Photo: NiunaMinerals A/S.

GO_09.qxp 23/02/07 12:48 Page 9

Perspectives

The statistical approach has a potential ofidentifying and mapping other types ofmineral occurrences as well as specific tec-tonic units in poorly known areas and canbe used on local or regional scale, depend-ing on the resolution of the involved data-sets. The statistical approach has been usedto examine gold showings in a moderatelyexplored region.

The approach could also be applied toeconomic deposits from well-known his-toric or active mining camps. As shown inthe case in the Nuuk area, the statisticalapproach should always be used in combi-nation with other geological approaches.

The results of a statistical analysis shouldbe cross-validated and critically balancedaccording to geological knowledge.

Concluding remarks

The detailed investigations of primary geo-logical environments of supracrustal rocksin the Nuuk region have revealed a greatdiversity in environments, which all carrygold. The results from the statistical analy-sis have been positive in the sense thatareas with high favourability for new goldoccurrences have been outlined.


10

GE

OL

OG

Y A

ND

OR

E

9 /

20

07

Hydrothermal alteration with epidotization,south of Ameralik fjord.

GO_09.qxp 23/02/07 12:48 Page 10

11

G E M S T O N E S O F G R E E N L A N D

Drill cores are lifted by helicopter duringoperation at Little Qingaaq. Photo courtesy:NunaMinerals A/S.

GO_09.qxp 23/02/07 12:48 Page 11

12

GE

OL

OG

Y A

ND

OR

E

9 /

20

07

GE

OL

OG

Y A

ND

OR

E

7 /

20

07

Key references

Appel, P.W.U., Coller, D., Coller, V., Heijlen, W.,

Moberg, E., Polat, A.,Raith, J., Schjth, F.,

Stendal, H. & Thomassen, B. (2005): Is there

a gold province in the Nuuk region? Danmarks

og Grnlands Geologiske Undersgelse Rapport

2005/27, 79 pp., 1 CD-ROM.

Eilu, P., Garofalo, P., Appel, P.W.U. & Heijlen,

W. (2006): Alteration patterns in Au-mineralised

zones of Stor, Nuuk region West Greenland.

Danmarks og Grnlands Geologiske Undersgelse

Rapport 2006/30, 73 pp.

Garde, A.A. (in press): A mid-Archaean island

arc complex in the eastern Akia terrane, Godt-

hbsfjord, southern West Greenland. Journal of

Geological Society.

Hollis, J.A., van Gool, J.A.M., Steenfelt, A. &

Garde, A.A. (2005): Greenstone belts in the

central Godthbsfjord region, southern West

Greenland. Geological Survey of Denmark and

Greenland Bulletin 7, 6568.

Hollis, J.A., Schmid, S., Stendal, H., van Gool,

J.A.M. & Weng, W.L. (2006): Progress report on

the 2005 field work: geological mapping, regional

hydrothermal alteration and tectonic sections.


Rapport 2006/7, 171 pp.

Juul-Petersen,A., Frei, R. Appel, P.W.U.,

Persson,M. & Konnerup_Madsen, J. (in press):

A sher zone related greenstone belt hosted gold

mineralization in the Archean of West Greenland.

A petrographic and combined Pb-Pb and Rb-Sr

geochronological study. Ore Geology Reviews

(2007).

Stendal, H., Secher, K., Nielsen, B.M., Schn-

wandt, H.K. & Thorning, L. (2005): Greenland

geological environments and mineral resources.


Rapport 2005/8, 207 pp.

Stensgaard, B.M., Chung, C., Rasmussen, T.M.

& Stendal, H. (in press): Assessment of mineral

potential using cross-validation techniques and

statistical analysis: A case study from the

Paleoproterozoic of West Greenland. Economic

Geology, 101.

Steensgaard, B.M., Steenfelt, A. & Rasmussen,

T.M. (2006): Gold potential of the Nuuk region

based on multi-parameter spatial modelling.

Progress 2005. Danmarks og Grnlands Geo-

logiske Undersgelse Rapport 2006/27, 207 pp.

Polat, A., Appel, P.W.U., Frei, R., Pan, Y., Dilek, Y.,

Ordes-Caldern, J.C., Fryer, B., Hollis, J.A.

& Raith, J.G. (2007): Field and geochemical

characteristics of the Mesoarchean (~3075 Ma)

Ivisaartoq greenstone belt, southern West

Greenland: Evidence for seafloor hydrothermal

alteration in supra-subduction oceanic crust.

Gondwana Research 11, 6991.

Front cover photographThe mountain Aappalaartoq (1440 m)

named after the Greenlandic word for

red mountain in the background.

Several gold-bearing structures have

been located within the supracrustal

package making up the mountain. Photo

courtesy: NunaMinerals A/S.

Bureau of Minerals and Petroleum(BMP)

Government of GreenlandP.O. Box 930

DK-3900 NuukGreenland

Tel: (+299) 34 68 00Fax.: (+299) 32 43 02

E-mail: [email protected]: www.bmp.gl

Geological Survey of Denmark and Greenland (GEUS)

ster Voldgade 10DK-1350 Copenhagen K

Denmark

Tel: (+45) 38 14 20 00Fax.: (+45) 38 14 20 50E-mail: [email protected]: www.geus.dk

AuthorB.M. Stensgaard & H. Stendal, GEUS

EditorKarsten Secher, GEUS

Graphic ProductionCarsten E. Thuesen, GEUS

PhotographsGEUS unless otherwise stated

PrintedJanuary 2007 GEUS

PrintersSchultz Grafisk

ISSN1602-818x

Gold-mineralised quartz veins hosted in amphibolite at the Qingaaq gold prospect, central Stor.

GO_09.qxp 23/02/07 12:48 Page 12

gold environments and favourability in the nuuk area of southern west greenland

Documents