*4.2.5 Multivariate analysis*

Multivariate analysis has been carried out on the significant metal contents from samples of tailings (Brunita), tailings + colluvial sediments (San Quintín), and tailings + watercourse sediments + bedrock (San Cristóbal and Las Moreras) (**Figure 4**). Statistical data processing was carried out using Minitab® 16 software. The multivariate analysis was based on clustering (group average linkage dendrograms, Euclidean distance) of the set of samples and significant trace elements (Ag, As, Cd, Cu, Pb, Sb, Sn, and Zn). The dendrogram of the metals and As in the Brunita tailing samples shows the metallic signature of the district ores: Ag-Pb-Cd-Zn, Cu, and As-Sb-Sn, with As being mainly related to Sb (tetrahedritetenanntite mineral group). Ag-Pb-Cd-Zn signature is clearly defined due to the mineral source. In the case of San Quintín, the dendrogram from tailings and colluvial sediments reflects again the metallic signature of the district (Pb-Ag-Sb, Cu, and Zn-Cd to a certain extent [13]), with As mainly related to Sb (bournonite and boulangerite) and Pb-Ag (galena). Some samples display a strong affinity to the Ag-Pb-Sb-As association, whereas other samples display Cd-Zn affinity. The same

#### **Figure 3.**

*Total gaseous mercury (TGM) seasonal distribution in the San Quintín area: (a) summer and (b) winter values. Modified from Martín-Crespo et al. [6].*

#### **Figure 4.**

*Dendrograms (distance Euclidean) of metals: (a) Brunita tailings; (b) San Quintín tailings and colluvial; (c) Mazarrón tailings; and (d) Mazarrón watercourse sediments and bedrocks.*

metallic signature has been obtained from tailings and colluvial sediments, reflecting the same origin for both kinds of samples. The external origin of Hg is reflected by the highest obtained distances. The same occurs for the samples from Mazarrón district, reflecting the metallic signature of the district (Pb-Ag-Sb, Zn-Cd, and Cu to a certain extent [14]) with As being mainly related to the Sb (stibnite and sulfosalts) and Pb and Ag (galena). Some slight differences are displayed in the samples from sediments and bedrocks, particularly the larger range of Cu content.

All dendrograms presented in **Figure 4** are in good agreement with field data, and mineralogical and geochemical features of tailings and watercourse deposits (**Tables 2–4** respectively). In summary, the metallic signature of the three districts is clearly defined in the samples from tailings and affected sediments.

#### **4.3 Complementary techniques**

#### *4.3.1 Electrical resistivity tomography (ERT)*

Additional information about the characteristics of the mine tailing deposits can be obtained from electrical resistivity tomography (ERT) data. The major pieces of information that this method provides are related to both the thickness of the deposits and the occurrence of AMD (both inside the mine pond as flowing out through the dyke or the base). **Figure 5** shows several examples of the type of information derived from the application of ERT to different mine ponds, resulting in a valuable tool that completes the information derived from mineralogical and geochemical techniques.

As a general rule, the materials that constitute the mine pond infilling are characterized by a medium to fine texture and high-water content. Moreover, due to oxidization of sulfide minerals, the pH of the water stored in the mine pond infilling is frequently acidic (pH < 5) in character. Opposite to this, the host rock where the mine pond is placed should be the host rock of the mineralization, typically metamorphic and/or igneous rocks of coarse texture and extremely low water content. Thus, a high resistivity contrast between the mine pond infilling (low to very low resistivity values) and the host rock (medium to high resistivity values)

**101**

**Figure 5.**

*mine deposits and the presence or not of acidic water.*

*Geoenvironmental Characterization of Sulfide Mine Tailings*

exists, allowing an accurate characterization of the boundary between both rock types and providing good estimations of the thickness of the mine pond deposits. Different mine pond thickness values and bottom geometries (dashed white lines) are imaged in **Figure 5**. Monte Romero and San Quintin mine ponds show simple bowl-shaped geometries with a thickness of ~3 and 10 m, respectively (confirmed with data from a borehole in the case of San Quintin mine pond 1), whereas Mina Concepcion and Brunite mine ponds exhibit a stepped bottom geometry with variable thickness (~6–10, and ~5–12 m, respectively). Where different rock units are present below the mine pond, instead of a homogeneous lithology, an estimation of the thickness of the different units can also be obtained. This is the case for San Quintin mine ponds, where a ~10 m thick sedimentary unit of colluvial deposits overlies the metasediments that constitute the regional basement. As mentioned before, a low pH value for the water contained in the mine pond deposits is also frequent, resulting in lower resistivity values in comparison with water with circumneutral pH. Therefore, the occurrence of acidic water inside a mine pond is revealed by extremely low resistivity values, normally lower than 1 ohm m. This is the case for Monte Romero, Mina Concepcion, and Brunita mine ponds where areas

*ERT profiles obtained at four different mine sites. Each profile provides information about the thickness of the* 

*DOI: http://dx.doi.org/10.5772/intechopen.84795*

*Geoenvironmental Characterization of Sulfide Mine Tailings DOI: http://dx.doi.org/10.5772/intechopen.84795*

**Figure 5.**

*Applied Geochemistry with Case Studies on Geological Formations, Exploration Techniques…*

metallic signature has been obtained from tailings and colluvial sediments, reflecting the same origin for both kinds of samples. The external origin of Hg is reflected by the highest obtained distances. The same occurs for the samples from Mazarrón district, reflecting the metallic signature of the district (Pb-Ag-Sb, Zn-Cd, and Cu to a certain extent [14]) with As being mainly related to the Sb (stibnite and sulfosalts) and Pb and Ag (galena). Some slight differences are displayed in the samples

*Dendrograms (distance Euclidean) of metals: (a) Brunita tailings; (b) San Quintín tailings and colluvial;* 

*(c) Mazarrón tailings; and (d) Mazarrón watercourse sediments and bedrocks.*

All dendrograms presented in **Figure 4** are in good agreement with field data, and mineralogical and geochemical features of tailings and watercourse deposits (**Tables 2–4** respectively). In summary, the metallic signature of the three districts

Additional information about the characteristics of the mine tailing deposits can be obtained from electrical resistivity tomography (ERT) data. The major pieces of information that this method provides are related to both the thickness of the deposits and the occurrence of AMD (both inside the mine pond as flowing out through the dyke or the base). **Figure 5** shows several examples of the type of information derived from the application of ERT to different mine ponds, resulting in a valuable tool that completes the information derived from mineralogical and geochemical techniques. As a general rule, the materials that constitute the mine pond infilling are characterized by a medium to fine texture and high-water content. Moreover, due to oxidization of sulfide minerals, the pH of the water stored in the mine pond infilling is frequently acidic (pH < 5) in character. Opposite to this, the host rock where the mine pond is placed should be the host rock of the mineralization, typically metamorphic and/or igneous rocks of coarse texture and extremely low water content. Thus, a high resistivity contrast between the mine pond infilling (low to very low resistivity values) and the host rock (medium to high resistivity values)

from sediments and bedrocks, particularly the larger range of Cu content.

is clearly defined in the samples from tailings and affected sediments.

**4.3 Complementary techniques**

**Figure 4.**

*4.3.1 Electrical resistivity tomography (ERT)*

**100**

*ERT profiles obtained at four different mine sites. Each profile provides information about the thickness of the mine deposits and the presence or not of acidic water.*

exists, allowing an accurate characterization of the boundary between both rock types and providing good estimations of the thickness of the mine pond deposits. Different mine pond thickness values and bottom geometries (dashed white lines) are imaged in **Figure 5**. Monte Romero and San Quintin mine ponds show simple bowl-shaped geometries with a thickness of ~3 and 10 m, respectively (confirmed with data from a borehole in the case of San Quintin mine pond 1), whereas Mina Concepcion and Brunite mine ponds exhibit a stepped bottom geometry with variable thickness (~6–10, and ~5–12 m, respectively). Where different rock units are present below the mine pond, instead of a homogeneous lithology, an estimation of the thickness of the different units can also be obtained. This is the case for San Quintin mine ponds, where a ~10 m thick sedimentary unit of colluvial deposits overlies the metasediments that constitute the regional basement. As mentioned before, a low pH value for the water contained in the mine pond deposits is also frequent, resulting in lower resistivity values in comparison with water with circumneutral pH. Therefore, the occurrence of acidic water inside a mine pond is revealed by extremely low resistivity values, normally lower than 1 ohm m. This is the case for Monte Romero, Mina Concepcion, and Brunita mine ponds where areas of <1 ohm m inside the mine pond correspond to the presence of water with pH ranging from 2 to 3 (see **Table 5**). On the other hand, the higher (>5 ohm m, and mainly >10 ohm m) resistivity values of the infilling of San Quintin mine ponds are associated with circumneutral pH (**Table 5**).

Finally, the strong resistivity contrast between the acidic water and the host rock results to be very useful to detect if AMD is flowing through the bottom of the mine pond. Where the sealing of the mine pond is correct, the host rock shows homogenous high resistivity values along the whole boundary with the pond infilling, such as the case of Monte Romero and San Quintin mine ponds. However, where AMD flows through the host rock, discrete areas of resistivity values much lower than the ones associated with the host rock are imaged, revealing the occurrence and sense of flow of the AMD. The latter is nicely imaged in both the cases of Mina Concepcion mine pond, where AMD flows from the inner central part of the pond toward the northern edge (confirmed during the field inspection of the dyke that exhibits AMD trough it), and Brunita mine pond, where AMD flows toward the east through the host rock.
