Dunn Colin E. Biogeochemistry in Mineral Exploration

Подождите немного. Документ загружается.

Results of the Athabasca survey showed that within an area of 10,000 km

all

trees sampled yiel ded concentrations of at least 10 ppm U. Inside this vast area, the

50 ppm U contour encompassed 3000 km

, and the 100 ppm contour extended over

an area of 1000 km

, reaching a maximum concentration of 2270 ppm U at Rabbit

Lake (Dunn, 1982). In hindsight, this unique occurrence of a massive biogeochemical

uranium province near the eastern margin of the Athabasca Sandstone could have

been identiﬁed by sampling just 1 tree per 1000 km

. There is further discussion of

this U anomaly in Chapter 11.

The extraordinary size of the Wollaston U anomaly appears to be rare, although a

regional Mo ‘province’ approaching this size has been identiﬁed in Russia (Kovalevsky,

pers. comm., 1983). Subsequently, a large Mo anomaly deﬁned by lodgepole pine bark

(Pinus contorta) has been established around the Endako molybdenum mine in central

British Columbia. The contour of 20 ppm Mo in bark ash (approximately ten times

background) extends for over 5000 km

(Dunn, 1999). In the Endako area, although

dust from the mining operations contributes to the high levels encountered, historical

records indicate similar levels in plants collected from the mine site in the 1960s during

the early days of mining operations (Warren et al., 1953; Warren and Delavault, 1965;

Warren, pers. comm., ca. 1988).

In addition to identifying major geochemical provinces, low-density surveys can be

undertaken to outline large geochemical systems and significant mineral ‘camps’. As

part of a regional survey of about 20,000 km

of Nova Scotia, focussing primarily on

balsam ﬁr (Abies balsamea), at every ﬁfth ﬁr sample site the outer bark from red

spruce (Picea rubens) was collected at a sample density of just 1 tree per 50 km

within

a 5000 km

area of eastern Nova Scotia (Dunn, 1989). For this survey, the network of

roads and forestry trails was used for gaining access, and the trees sampled were each

at least 50 m from any roadside or other sign of disturbance. From simply plotting

multiples of the median values of the element concentrations, the distribution patterns

of Au, As, Sb and Se each outlined the main gold camps (Fig. 5-1) of which the north-

easterly trend from Goldenville to Forest Hill was the most notable.

Additional examples of low-density biogeochemical surveys in Canada include

publications from Man itoba (Fedikow et al., 1997a,b, 1998, 1999, 2000, 2002), in

central British Columbia (Dunn et al., 1994c, 1996b; Dunn and Hastings, 1998, 1999,

2000), in Newfoundland (Dunn et al., 1995b) and in Alberta (Seneshen et al., 2005).

In Australia, Cohen et al. (1998, 1999) conducted two reconnaissance surveys that

included the analysis of foliage from several common plant species. The larger of

these surveys encompassed 14,000 km

of north-east ern New South Wales and, since

it included a comparison with stream sediments, sample sites were close to streams

(Cohen et al., 1999). Among their conclusions, they found that genera capable of

accumulating high concentrations of trace elements in the vicinity of mineralization

included Co and REE in Eucalyptus,AsinCallistemon (bottle brush) and Au in

Casuarina (River She-oak).

This last-mentioned survey brings up another point for consideration – the pre-

ferred topographic location of sample stations. In some areas, there is sufﬁcient

132

Survey Design and Comparisons with Other Sample Media

Fig. 5-1. Low-density reconnaissance survey (1 per 50 km

over approximately 5000 km

)of

eastern Nova Scotia, Canada, showing concentrations of Au and As in ash of red spruce outer

bark (Picea rubens). Similar patterns were shown by Sb and Se (Dunn, 1989). Analysis by

INAA.

133Biogeochemistry in Mineral Exploration

density of streams to permit sample collection from near the base of a slope draining

into a stream valley. In areas with moderately thick surﬁcial cover, it is better to

collect at this break in slope rather than from valley ﬂoors, be cause in the stream’s

ﬂoodplain plants are likely to be rooted in exotic detritus brought down by streams in

times of high water ﬂow. Consequently, the biogeochemical signature of samples

from a ﬂoodplain may provide a subtle signature that reﬂects an allochthonous

component of material rather than the preferred signature of a nearby source that is

more likely to be obtained from a break-in-slope sample location.

Moderate density (1 site per 1– 10 km

)

In the previous section, reference was made to 20,000 km

covered by regional

biogeochemical surveys in Nova Scotia. Surveys were conducted in early summer on

ﬁve occasions betw een 1987 and 1995 at a sample density of approximately 1 site per

8km

using balsam ﬁr twigs (Abies balsamea), because this proved to be the most

ubiquitous tree (e.g., Dunn et al., 1989, 1992a,b, 1994a,b, 1996a,b; Dunn and Balma,

1997). At most of the ﬁr sample sites, there were also spruce present (red spruce and/

or black spruce) and so outer bark of spruce was collected at the same time and

additional datasets were published for comparison (Dunn et al., 1992a,b). At that

time, the optimal analytical procedures involved reduction of tissues to ash by con-

trolled ignition to 475

C with analysis by INAA, and by ICP-ES (aqua regia di-

gestion) for those elements not readily determined by INAA.

The Nova Scotia regional biogeochemical surveys generated exploration data that

deﬁned known and previously unknown areas of metal enrichments. As an example

of the signatures obtained, Fig. 5-2 shows U and W in balsam ﬁr twigs (concen-

trations in ash) from a survey of the south-western part of the Province conducted in

the spring of 1991 (Dunn et al., 1994a). The area of anomalous concentrations

encompasses the East Kemptville tin/tungsten mine that was in production from

1985 to 1992, and several satellite zones of mineralization that were not exploited.

Other elements in the ﬁr twigs that exhibited similar patterns of enrichments included

As, Be, Cs, Li, Rb, Sn and Ta. Although some of this signature was probably related

to airborne dust it was noteworthy that other sites, remote from known mineral-

ization, yielded some anomalous values of similar magnitude.

The fastest way to establish biogeochemical patterns over large tracts of land is by

helicopter. Although the cost of helicopter rental is high, the speed of collecting

samples is sufﬁciently rapid that the over all economics are favourable. At the time of

writing, the entire cost per sample to run a survey in remote areas of Canada is

around US$120. This is about two-thirds to three-quarters the cost of a regional

helicopter-supported lake sediment or stream sediment survey, and includes all

components of a survey from planning, collection, analysis to a ﬁnal report . Sample

densities are typically in the range of 1 site per square kilometre to 1 per 4 km

Examples of results from helicopter-borne surveys are given in Chapter 11.

134

Survey Design and Comparisons with Other Sample Media

Fig. 5-2. Regional survey of south-western Nova Scotia, Canada, using twigs of balsam ﬁr

(Abies balsamea). Uranium and tungsten shown as ppm in ash, determined by INAA. Location

of disused East Kemptville Sn/W mine centred in area of anomalous metal enrichments (Dunn

et al., 1994a).

135Biogeochemistry in Mineral Exploration

A survey designed to locate kimberlites beneath a Quaternary cover in central

Alberta involved the collection of white spruce (Picea glauca) tops at a sample

spacing of 2 km (i.e., 1 site per 4 km

). An added component to this study was the

collection of an additional ﬁve samples over known kimberlites, in order to establish

their biogeochemical signatures (Seneshen et al., 2005 ). The resulting sample grid

pattern and the locations of the principal kimberlites known at the time of the survey

are shown in Fig. 5-3.

In retrospect, the basic sample spacing of 2 km proved too coarse for clearly

deﬁning the locations of kimberlites, and 1-km spacing would have been preferable.

Some results of this survey are discussed in Chapter 11.

Semi-detailed surveys (1 site per 0.25– 1 km

)

Surveys at this scale can be undertaken by helicopter in difﬁcult terrain, or ground

traverses – either on foot, by truck or four-wheel drive vehicles in open arid terrain,

SAMPLE SITES

White Spruce

Top Stems

metres

K14

TQ155

Kimberlite:

[5 additional

samples over

each known

kimberlite]

0 2000 4000

Fig. 5-3. Sample pattern established for collection of tree tops (2 km grid) to assist in locating

concealed kimberlites in central Alberta (Seneshen et al., 2005).

136 Survey Design and Comparisons with Other Sample Media

by all-terrain vehicles (‘quads’) on rough forest trails, or by snow-mobiles along trails

during the long cold winter months of the northern continental masses. Biogeo-

chemical sampling may be the only feasible geoch emical sampling method in north-

ern climates during the half of the year that experiences snow-co vered terrain and

frozen ground.

In the rugged Cordillera del Co

ndor region of the western Amazon, close to the

border between Peru and Ecuador, a survey programme included the collection of

foliage from 366 tree ferns (Chonta con Espinas – Cyathea spp.) on an even grid of

sample stations spaced at intervals of 500 m (i.e., 1 site per 0.25 km

). Figure 5-4

shows plots of Au and As with the 98th percentile of the dataset selected as the

maximum for plotting, such that values above this value are all shown as the same

intensity of shading. The former artisanal Au workings at Chinapintza (epithermal

Au/Ag) are clearly evident, and a second area (Anomaly A) farther south also

shows a Au/As relationship. A strong mercury anomaly was associated with the

0.5

1.5

2.5

3.5

4.5

GOLD in Tree Fern Foliage

Cordillera del Condor

Au ppb

Max. 16.8

Ecuador

Peru

Chinapintza

Anomaly A

2 km

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.1

As ppm

Max. 2.9

ARSENIC in Tree Fern Foliage

Cordillera del Condor

Ecuador

Peru

Chinapintza

Anomaly A

2 km

Fig. 5-4. Western Amazon – Au and As in dry foliage of tree fern. Data courtesy of An-

gloGold-Ashanti. Determinations by ICP-MS on aqua regia digestion of dry tissues.

137Biogeochemistry in Mineral Exploration

Chinapintza Au, and Ag anomalies were peripheral to Anomaly A, indicating zoning

of the mineralization. Details of this survey are provided in Chapter 11.

In northern Saskatchewan the small but high-grade Rottenstone Ni-PGM deposit

was mined out between 1965 and 1968, but there are indications of additional de-

posits in the area. There is no all-weather road access and the heavily forested terrain

has a veneer of glacial deposits. It is cut by streams and rivers and is dotted with

many lakes such that geochemical surveys are difﬁcult and very time-consuming to

undertake.

To provide a rapid and broad reconnaissance geochemical survey of a 130 km

area that had been staked by Uravan Minerals Inc., about 800 black spruce (Picea

mariana) tops were collected from a helicopter during a ﬁve-day period in September

1999. Samples were collected at 500-m spacing, with alternating offset lines to pro-

vide minimum distance between sample points in accord with an idealized grid shown

in Fig. 5-5.

In practice, this sampling pattern was easy to achieve, and digressions from the

designated sample locations were mostly only a few metres in order to locate an

appropriate tree or to avoid a lake. Many of the larger lakes had islands on which

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

100

101

102

103

104

105

106

107

111

Fig. 5-5. Idealized offset grid sampling pattern to provide minimum distances between sam-

ple points.

138 Survey Design and Comparisons with Other Sample Media

trees were available for sampling, and the extensive drainage network required that

sample sites only be shifted a short distance from the desired grid. The sample

interval was reduced to 250 m over a 48 km

area of primary interest around Rot-

tenstone Lake that is outlined in the centre of Fig. 5-6. The results of this survey are

discussed in Chapter 11.

Detailed surveys (25– 200 m spacing)

In desert areas, such as the south-western United States, detailed surveys using

widespread species (e.g., sagebrush [Artemisia tridentata]) can be undertaken by

sampling on a predetermined grid pa ttern. However, surveys in the dense temperate

and boreal fores ts of North America and the densely vegetated tropical areas of the

world are often controlled by logistics – access by road, foot, boat, ﬂoat-plane or

helicopter. In these areas, a sample spacing of about 200 m has proven effective and

feasible, even though sampling on a regular grid pattern may be precluded by the

presence of streams, bogs, or perhaps the absence of the desired species. In Pre-

cambrian Shield terrain of many parts of the world there is commonly a preferred

Kenyon

Taylor C

Lake

Rottenstone

Lake

502000

504000 506000 508000 510000 512000 514000 516000 518000

520000

6236000

6238000

6240000

6242000

6244000

6246000

6248000

6250000

6252000

Albert

Lake

Fig. 5-6. Sample pattern achieved for the collection of black spruce tops around Rottenstone

Lake, Saskatchewan.

139Biogeochemistry in Mineral Exploration

trend to metamorphic grade and to structure, such that an effective compromise is to

sample across strike at 200 m intervals along lines spaced 1 km apart. Extensive

surveys at this sample spacing in northern Saskatc hewan have used bark of black

spruce (Picea mariana) and twigs of mountain alder (Alnus crispa). Among several

zones of anomalous Au concentrations that wer e identiﬁed, one was in an area

dominated by alder, and follow up work using alder at closer sample spacing (mostly

a 100 m 100 m grid) identiﬁed the spatial relationship among Au, Mo and Cs

(Fig. 5-7) in alder twig ash (most recent three years of growth). These three elements

Fig. 5-7. Metals in ashed twigs of alder (Alnus crispa) – ash yield approximately 2%: (a)

Location of survey area, (b) Gold, (c) Molybdenum and (d) Cesium. La Ronge gold belt of

northern Saskatchewan, Canada (Dunn et al., 1990). Analysis by INAA.

140 Survey Design and Comparisons with Other Sample Media

in plant tissues quite commonly exhibit a spatial relationship to gold mineralization.

Subsequent drilling intersected gold mineralization, but to date only sub-economic

grades have been found.

In an area with known or suspected mineralization, a simple orientation survey

can be co nducted to determine if a biogeochemical response is obtained in common

plants. A single line is inadvisable, since the validity of any anomaly that arises may

be doubtful. By sampling along two lines oriented normal to the presumed strike of

mineralization, a positive signature on both lines would provide an indication of the

strike of the mineralized body. It is an expeditious procedure to collect along one

line going away from a base camp and a second returning to it. If time permits,

of course, greater substantiation of anomalous trends can be obtained from a

four-line sampling program. Figure 5-8 shows examples of iodine anomalies in cedar

foliage (western redcedar) along two lines that traversed a suspected zone of breccia-

hosted Cu porphyry at the Boundary Zone, Mount Polley, in central British

Columbia. The trend of the mineralization can be determined from the locations of

the anomalies.

From these proﬁ les, a sketch map could be plotted outlining the predicted lo-

cation and trend of mineralization (Fig. 5-9 ).

The plot shown in Fig. 5-9 shows sample spacing at 100 m, closing down to 50 m

then 25 m over the zone of interest. Similarly, quartz-hosted vein-type gold may

require sample spacing at 50 m or as close as 25 m. To sample more closely than this

may be impractical or not possible because of the generally lower density of plant

distributions, although on occasion, according to local geological conditions, spacing

at 10 m may be warranted. In a situation where a narrow quartz vein is known to have

irregularly spaced zones of mineralization along strike, sampling may be spaced at

100 m or 200 m along a vein. However, sampling normal to the strike of the vein may

be reduced locally to 10 m in order to delineate mineralization. Figure 5-10 shows an

idealized situation of an effective sampling pattern over a gold-bearing quartz vein. In

this situation sampling over the vein at 10 m intervals has provided a four-point

I ppm I ppm

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0 500 1000 1500

Western Traverse

South North

Approx. Location of

mineralization

metres

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0 500 1000 1500

Eastern traverse

South North

Approx. Location of

mineralization

metres

Fig. 5-8. Iodine in dry tissue of western redcedar (Thuja plicata) at the Boundary Zone,

Mount Polley, British Columbia. Analysis by HR-ICP-MS after a warm water leach (Dunn et

al., 2006a,b).

141Biogeochemistry in Mineral Exploration