Characteristics of Bacillus cereus Spore Counts in Soils with Underlying Gold Mineralization in Northwestern Sichuan, China: Results from a Pilot Study

INTRODUCTION

The concept of using microorganisms as a tool for mineral exploration was proposed almost 50 years ago (Slavnina, 1957). Two decades later, Watterson et al.(1985, 1983a, b) reported on a relationship between spore counts of Bacillus spp. (almost exclusively B. cereus) and heavy metal anomalies in soils over ore deposits. These findings suggested the possibility of using B. cereus as an index to underlying metal mineralization. Field studies were carried out in several locations of different climate conditions (Parduhn, 1991) and results were promising, especially in semiarid areas. Further application research on the bioexploration method was carried out in several countries including Belgium, Burkina Faso, France, Mexico and Sudan (Melchior et al., 1996, 1994; Melchior, 1993; Moureau et al., 1991; Neybergh et al., 1991, 1989). Based on the previous literature, it is hypothesized in this study that there is a common spatial association between B. cereus spore count and soil gold content in spite of the differences in the climate, geological setting and mineralizations. In China, research of microbial bioexploration is still in its infancy (Wang et al., 1999; Xie et al., 1997; Li et al., 1994). The purpose of this study is to screen soil samples at three mine sites located on the same tectonic crustal block (the Aba crustal block) in an effort to seek association between B. cereus spore counts and gold content as a prelude to developing a bioexploration index. This method has been tested in the gold deposits of Axi, Manaoke and Dongbeizhai, northwestern Sichuan Province of China (Fig. 1).

Figure  1.  Sketch of the study area.

DownLoad: Full-Size Img PowerPoint

PHYSIOGRAPHICAL AND GEOLOGICAL SETTING

The sampling region is a typical canyon with elevations ranging from 1 800 to 4 300 m, located in the northeast of the Qinghai-Tibet plateau.A continental temperate altiplano climate prevailsin the region with an annual mean rainfall of 600-700 mm.The annual av-erage temperature increases gradually from Axi (0-3℃) to Manaoke (3-6℃) and to Dongbeizhai (5.7℃).Specifically, Axi has a sub-humid sub-frigid altiplano climate, while Dongbeizhai and Manaoke have a sub-humid temperate altiplano climate.The vegetation changes from mountainous boskage meadow in Axi to mountainous taiga in Dongbeizhai.

Slightly alkaline drab soil and brown soil developed in Dongbeizhai, while Axi is characterized by subacidity-neutral sub-alpine meadow soil (Zhang et al., 1997). The altitude of Manaoke is over 1 000 m higher than that of Dongbeizhai, which leads to the differences in climate and vegetation. The soil in the Manaoke gold deposit bears a mixed character of the soils developed in Dongbeizhai and Axi. Thus, the soil in Manaoke can be regarded as the transition of the two typical agro types, the drab/brown soil and sub-alpine meadow soil.

Geotectonically, the study area (the Aba crustal block) is in the triangle region bordered by the Yangtze platform, Yushu-Jinshajiang fold belt, and southern Qinling fold belt (Fig. 2). The crustal block has been characterized by active metamorphism, volcanism, magmatism and metallogenesis. Most of the discovered gold deposits developed in large-scale, regional deep-fracture zones along the boundary of the block in northwestern Sichuan. The host strata for these gold deposits vary from Cambrian, Silurian and Devonian to Triassic. However, 80 % of the gold deposits occurred in the Triassic, especially in the turbidite of Middle to Late Triassic (Yang et al., 1999).

Figure  2.  Map of the Aba crustal block (modified after Zhang, 1999).

1. deep fault zone; 2. fault zone; 3. pre-Triassic; 4. Triassic; 5. gold deposit.

DownLoad: Full-Size Img PowerPoint

The Manaoke gold deposit is hosted in the turbiditic sandstones and slates of Late Triassic.Gold mineralization occurs in strongly altered structural brecciasand hydrothermal quartz veins along an NW-SE trending reverse fault fracture system, dipping NW, paralleling to the strike of the stratigraphic bedding.The quartz veins and structural breccias are characterized by abundance of Au-bearing sulfides including pyrite, arsenopyrite and realgar.The individual lode extends several hundred meters in length and the thickness varies from several to tens of centimeters.The gangue minerals are quartz and calcite.The hydrothermal alteration is defined by silicification, carbonatization, pyritization and weathering by limonitization.For the detailed geology of the mineral deposit, refer to Lü et al. (1999), Wang and Fu (1999), Wang et al. (1999) and Zhang (1999).

The potential ore-bearing structures of the Axi gold deposit consist of crushed fracture systems in the exomorphic zone of Yanshanian diorite where skarn and metasomatic mineralization occur (Liao et al., 1999). The presence of intrusive rocks distinguishes the Axi gold deposit from the other two studied herein. Au mineralization is spatially related with the lithological contact between sedimentary rocks and dioritic intrusives. Gold is mainly hosted in pyrite, and the gangue minerals are quartz and calcite.

The Dongbeizhai disseminated gold deposit is one of the largest deposits in northwestern Sichuan. Gold is mostly found in pyrite, which is the most abundant sulfide; however, native gold, arsenic sulfides and antimonite also occur. The primary geochemical signature is defined by an assemblage of Au-As-S-Sb-Hg. Au-bearing sulfide lodes are closely associated with carbonaceous slates and phyllites in the Xinduqiao Formation of the Upper Triassic (Yin et al., 1999; Zhou et al., 1995; Zheng, 1989) and are generally confined to the clastic footwall block of a north-south trending fault (within a distance of 0-60 m from the fault), which is parallel to the strike of the stratigraphic bedding. The mineralization zone extends 4 km in length while the width of the Au-bearing sulfide veins can reach up to 4 m. Quartz and calcite are the gangue minerals. Hydrothermal alteration is characterized by strong silicification, pyritization, carbonatization and weak argillization.

EXPERIMENTAL METHOD

Soil samples from five transects perpendicular to the strike of the orebodies were collected for microbiological and elemental analyses at 5 m intervals and at every 2.5 m when close to the zones of gold mineralization. All samples were collected in July.

Samples were taken at the depth of 10-25 cm in the horizon B of soil profiles to avoid surface contamination. Scoop used was ethanol sterilized and samples were kept in sterilized plastic bags to prevent the loss of moisture during the transportation. After air-drying in a super clean bench, which was sterilized with UV light, the samples (200-300 g each) were divided into two parts. One aliquot was sieved to -40 mesh for microbiological analysis. The other was sieved to -200 mesh for the quantitative analyses of the contents for Au, Ag, Cu, Pb and Zn. The methodology for microbiological analysis was similar to that employed by Watterson (1985). Samples were heated to 90 ℃ for 60 s to inactivate all vegetative cells and cooled in tap water before aliquots of diluted samples were pour plated on egg yolk agar (EYA). The plates were incubated at 30 ℃ for 15 h. Thus spores that resisted the heat pre-treatment germinated and grew as colonies on the EYA plates. B. cereus colonies are characterized by the presence of an opaque to translucent halo because of the hydrolysis of lecithin (added as egg yolk to the medium). In the initial phase of this study, bacteria from suspected B. cereus colonies were isolated and characterized by physiological and biochemical tests to ensure proper identification based on colonial morphology (Xie et al., 1997). The detection limit was 10 CFU (colony forming units) per gram air-dried soil for bacterial analysis.

To evaluate the background of B. cereus, soil samples from no-mining areas were also analyzed. The data were pooled for each area and the threshold was defined based on the frequency histogram.

Atomic absorption spectrometry (AAS) was used for the elemental analysis of Au, Ag, Cu, Pb and Zn in the soil samples. The Au and Ag analyses were carried out by graphite furnace AAS and Cu, Pb and Zn by flame-AAS. Au was analyzed after the sample was digested in an HCl/HNO₃ mixture and the interfering organic compounds were removed by extraction with methylisobutyl ketone. Aqua regia was used to dissolve Ag and a hot HCl-HNO₃-HF digestion was used for Cu, Pb and Zn. The detection limits for Au and Ag are 0.5 ng/g soil and 0.01 μg/g soil, respectively, and 0.5 μg/g soil for Cu, Pb and Zn.

RESULTS

Manaoke Gold Deposit

Two parallel soil transects, designated as M1 and M2, perpendicular to the strike of the orebodies, were sampled in the Manaoke deposit.

M1 transect

In total, 23 soil samples from the B horizon were collected along the M1 geological transect. As shown in Fig. 3, five orebodies (designated as O₁, O₂, O₃, O₄ and O₅) have been revealed by the 205 Geological Survey Team of Sichuan Province through routine geochemical and geophysical methods on the M1 transect. Two (O₂ and O₃) of these orebodies were evaluated to be high-grade. Corresponding to these orebodies, four sites of elevated gold content (G₁, G₂, G₃ and G₄) were found in overlying soils of the transect. Only four gold anomalies were identified because O₄ and O₅ could not be differentiated on the basis of the soil gold analysis.

Figure  3.  B. cereus spore counts and metal contents in soils overlying the M1 geological transect at the Manaoke deposit. The geological transect is after the 205 Geological Survey Team of Sichuan Province.

1. anomaly of spore counts; 2. anomaly of soil gold contents; 3. calcareous siltstone; 4. sericite slate; 5. auriferous orebody.

DownLoad: Full-Size Img PowerPoint

The counts of B. cereus spores varied greatly on this profile and were characterized by a clustering of high and low counts (Fig. 3). A threshold value of 240 CFU/g air-dried soil based on frequency histogram of B. cereus spore counts was designated to differentiate background value and anomalous counts. The highest spore counts were two orders of magnitude higher than the background value. Only three distinct anomalies of B. cereus spore counts (B₁, B₂ and B₃) were found in the overlying soils. Low spore counts occurred not only in the soil with low gold content (northeast of the transect) but also in soils with high gold contents overlying the orebodies (at the sites of 8-13 m and 42-47 m on the transect). The three spore anomalies (B₁, B₂ and B₃) were associated with the three high soil gold anomalies (G₁, G₂, G₃), which corresponded to the underlying medium- to high-grade orebodies (O₁, O₂, O₃). No anomalous spore counts were detected in the soils associated with the low-grade orebodies O₄ and O₅.

The highest spore anomalies B₂ and B₃ flanked the highest soil gold anomaly G₃, which corresponded to one of the two high-grade orebodies (O₃). Spore anomaly B₁ flanked the soil gold anomaly G₂, which corresponded to the other high-grade orebody (O₂). B₁ may also be associated with G₁ which corresponded to the orebody of O₁, but further data southwest of the transect are not available. Thus the three spore anomalies flanked the soil gold anomalies with the highest grades of underlying gold mineralization. The soils between these spore anomalies were characterized by low spore counts. Consequently, spore count anomalies were clustered in the soils corresponding to the top and bottom walls of the underlying orebodies, whereas low soil spore counts exactly overlaid the orebodies. Figure 3 also shows multi metal analyses for the soil samples, but there was no apparent association between the spore counts and the concentrations of metals.

M2 transect

17 soil samples from the B horizon were collected along the M2 geological transect for both bacterial and gold analyses. Only one ore body was clearly revealed by the 205 Geological Survey Team of Sichuan Province through routine geoexploration methods on the profile. Low spore counts occurred in soils where the orebody (O₁) outcropped and at the southwest of the transect (Fig. 4). High spore counts (B₁ and B₂) were detected in soils flanking the subsurface orebody (O₁). The gold content in soils was higher in the southwest transect as compared to the northeast. Only one of the high values corresponded to the subsurface orebody (Fig. 4).

Figure  4.  B. cereus spore counts and gold content in soils overlying the M2 geological transect of the Manaoke gold deposit. The geological transect is after the 205 Geological Survey Team of Sichuan Province.

1. auriferous orebody; 2. anomaly of spore counts; 3. feldspathic quartzose sandstone.

DownLoad: Full-Size Img PowerPoint

Axi Gold Deposit

The results from two soil transects in the Axi deposit are shown in Fig. 5. High spore counts of B. cereus (B₁ and B₂; B₃ and B₄) were detected in soils flanking the orebodies of O₁ and O₂. The highest count was 13 000 CFU/g soil, while low counts of spores were found at the northeast of the profiles and in soils, which overlaid the orebodies. The soil gold content showed similar trends with the spore counts on the transects, although the corresponding peaks were slightly shifted from each other. The spore distributions in the two profiles are similar to that seen in the profiles of the M1 and M2 in the Manaoke deposit.

Figure  5.  B. cereus spore counts and gold contents in soils overlying the two geological transects of the Axi gold deposit. B₁-B₄. the anomalies of spore counts; O₁-O₂. auriferous orebodies.

DownLoad: Full-Size Img PowerPoint

Dongbeizhai Gold Deposit

Seven orebodies were discovered on the geological transect by the Geological Team of Northwestern Sichuan through routine geoexploration methods. The variation of B. cereus spore counts along the transect is shown in Fig. 6. The highest spore count was 6 500 CFU/g soil, which is two orders of magnitude higher than the counts in most samples of the profile. The two spore anomalies (B₁ and B₂) corresponded to the outermost edges of the cluster of orebodies. However, B₂ partially overlapping an underlying orebody may indicate the combined results from the adjacent underlying orebodies. The other counts between B₁ and B₂ on the transect could not resolve individual orebodies because they were so thin and in very close proximity to each other. The gold contentin soil was not analyzed for this profile.

Figure  6.  Variation of B. cereus spores in soils overlying a geological transect of the Dongbeizhai gold deposit. The transect is after the Geological Team of Northwestern Sichuan.

1. Xinduqiao Formation of the Upper Triassic; 2. Xuebaoding Formation of Lower Carboniferous; 3. limestone; 4. cataclastic slate; 5. orebody; 6. Quaternary; 7. mine adit; 8. anomaly of spore counts.

DownLoad: Full-Size Img PowerPoint

Signal Intensity of Spore Counts Caused by Gold Mineralization

As with most geochemical exploration methods, it is useful to employ the background value and abnormal value in this study to discriminate the low counts away from and overlaying auriferous orebodies, and the relatively high spore counts flanking subsurface mineralization. The results (Table 1) show the average background spore counts of the Axi deposit was 312 CFU/g soil with a range of < 10 to 800 CFU/g soil. The background spore count of the Dongbeizhai deposit varied from < 10 to 40 CFU/g soil. The background spore counts of the Manaoke deposit varied from < 10 to 240 CFU/g soil, averaging at 53 CFU/g soil. The data may suggest some association between the background spore counts and soil types (Table 1). The meadow soil (Axi) had the highest average background values and the drab soil (Dongbeizhai) had the lowest. The transitional soil in Manaoke had an intermediate average background. The threshold to discern gold mineralization varies from one deposit to another. For example, soils with spore counts of 400 CFU/g soil can be regarded as anomalies in Dongbeizhai but not in Axi because of the different thresholds.

Table  1.  Signal intensity revealed by B. cereus spore counts

 | Show Table

DownLoad: CSV

DISCUSSION

Relatively low spore counts were found in soils, which directly overlaid the gold containing orebodies. One reason for this association may be the high gold content in these soils. In a study of the El Real mineralization, Mexico, Melchior et al. (1994) reported that B. cereus spore counts peaked at locations where the auriferous quartz veins did not reach the surface. Low spore counts were found in soils where the orebodies outcropped, suggesting that metal toxicity may have suppressed B. cereus spores. However, the total metal contents analyzed except gold showed no apparent association with the spore counts in soils in the present work. Parduhn (1991) also reported low spore counts in soils where the orebody outcropped the surface at the Mesquite deposit, southern California. Insufficient information was given about the soil types in her study however, and therefore it might not be comparable to this present study.

In the M2 profile the spore counts decreased with increasing gold content in soils. Generally, the spore counts increased upslope, while soil gold content increased downslope. The increase in gold content may have resulted from the effect of creep.

Although the distribution trend of B. cereus spore counts in the Axi gold deposit is similar to that of the Manaoke deposit (low spore counts overlying outcropped areas of orebodies with spore anomalies at both sides), the relationship between the subsurface orebodies and the soil gold contents is somewhat different from each other. On the M1 profile in the Manaoke deposit, soil gold anomalies just overlay the underlying orebodies. However, in the Axi deposit, the soil gold anomalies flanked the subsurface orebodies. Thus the soil gold anomalies had a pattern of distribution comparable to that of spore counts in Axi. Further research is needed to examine the distribution patterns of soil gold contents and B. cereus spore counts at different gold deposits.

In general, the data of all soil transects show that the spore counts of B. cereus can indicate the soil gold anomaly and the underlying gold mineralization. The spore counts close to the underlying gold mineralized areas were two to three orders of magnitude higher as compared to the background. A spatial shift was apparent between the anomalies of B. cereus spore counts and the corresponding soil Au content.

These data suggest that the distribution pattern of spore counts can outline the subsurface mineralization, whereby orebodies underlay low spore counts with two spore anomalies. This kind of indication is particularly apparent in the two profiles of the Manaoke deposit and two soil profiles in the Axi deposit. This pattern was present in deposits characterized by different soil types (drab/brown soil for Dongbeizhai, meadow soil in Axi, and the transitional soil type in Maonaoke).

The signal intensity of B. cereus spore method in response to the underlying mineralization has yet to be elucidated. High counts associated with underlying gold mineralization were found in the Axi deposit that also had a higher background spore level. The signal to noise ratio (abnormal counts to background counts) seemed to be the highest at the Dongbeizhai gold deposit that had the lowest background of spore counts. The background of B. cereus spore counts is related to complex environmental factors such as the soil type, temperature, rainfall and humidity whilst the distribution pattern of spore count anomaly is mainly associated with the subsurface mineralization, and the anomaly intensity is affected by both the underlying mineralization and background.

No spore anomalies were found in the soils overlying the low-grade orebodies. It is conceivable that low-grade orebodies will be missed with this microbial geoexploration technique. In general, the technique presented in this paper is best suited for field application in combination with routine, conventional geoexploration methods.

ACKNOWLEDGMENT

We thank the Geological Team of Northwestern Sichuan and the 205 Geological Survey Team of Sichuan Province for their help with the fieldwork and providing the geological profile data. The participation of Zhang Jun, Wang Zhiping, Zhou Xuewu, Zhou Qiaowei, Xiong Wei and Zhang Xiaojun is gratefully acknowledged. We thank Olli H. Tuovinen for helpful comments on the manuscript. The work was supported by the Ministry of Land and Resources (No. 2000441 and 95-02-002-03), the Ministry of Education (Outstanding Youth University Teachers Fund), and the National Natural Science Foundation (No. 40202011) of China.