Rachold: Major. Trace and Rare Earth Element Geochemistry
217
Factor analysis of the SP M data
Factor analyses of river water chemical data, such as principle component analysis, has been
applied to identify the mixing of different sources in numerous studies (c.f. Christophersen and
Hooper, 1992). In the present study, principle component analysis (orthogonal transformation)
of the chemical data set was used in order to: (I) reduce the large number of variables to a small
number of factors, (2) identify the general interrelationship between major and trace elements,
and (3) distinguish between SPM of individual rivers. The concentrations of 31 elements (AI,
Ti, Fe, Mn, Mg, Ca, Na, K, Ba, Bi, Ce, Co, Cs, Ga, Hf, La, Li, Mo. Nb, Ni, Rb, Sb, Sr,
Ta, Th, TI, U, V, W, Y, and Zr) in 64 samples were included in the calculation. As, Cd, Cu,
Pb, Sn, and Zn were omitted from the computation due to contamination effects.
The general variations in the chemical composition of the SPM can be explained by two
factors accounting for 71.4 % of the total variance. Figure 10 presents the factor loadings of
these two factors. The factor scores are illustrated in Figure II. Factor I (59.6 % of the
variance) clearly separates elements predominant in mafic rocks (high negative factor loadings
for Ca, Co, Fe, Mg, Ni, Ti, and V) from elements that are more abundant in acidic rocks (high
positive factor loadings for Ba, Bi, Ce, Cs, Hf, K, La, Li, Mo, Na, Nb, Rb, Sb, Ta, Th, TI,
U, W, Y, and Zr). This factor discriminates the Khatanga SPM from the Lena and Yana SPM.
While the geochemistry of Khatanga SPM is closely related to the basaltic rocks of the Siberian
Trap, the composition of Lena and Yana SPM is similar to average shale. except for some
elements. As a consequence, high negative factor scores can be observed in Khatanga SPM and
positive factor scores in Lena and Yana SPM (Figure 11). As has been discussed above, the
REE pattern of Vilyuy (L3) SPM indicates the influence of Siberian Trap basalts. This is also
seen in the factor score diagram where the Vilyuy plots between the Lena and Khatanga (Figure
11). In the Popigay (K3) SPM, the dominance of the Siberian Trap is less pronounced than in
the other Khatanga tributaries.The Popigay consequently plots very close to the Vilyuy in the
factor score diagram (Figure II).
Factor 2 (11.8 % of the variance) is characterized by high positive factor loadings for Cs, Hf,
Sb, Ti, and Zr and a negative Mn factor loading (Figure 10). This factor may be related to two
different processes: (1) The negative factor loading for Mn and the high positive factor loadings
for Hf, Ti, and Zr suggest that differences in grain-size distribution are responsible for this
factor. Due to the dominance of fine-grained material, which is characterized by high amounts
of Mn oxihydroxides, Lena SPM exhibits high Mn concentrations. Yana SPM, on the other
hand, displays higher amounts of coarse-grained material resulting in high Hf, Ti, and Zr
contents. (2) High positive loadings for Cs and Sb are related to granitic intrusions in the Yana
catchment (see above). As a result of both processes, factor 2 clearly differentiates between
Lena and Yana SPM. It can be concluded that factor analysis serves as a helpful tool to identify
the interrelationship of chemical elements in river SPM and to separate between SPM
transported by individual rivers.
Summary and conclusions
In order to unravel the sediment transport of East Siberian rivers draining to the Laptev Sea,
grain-size distribution, mineralogy, and inorganic chemical composition of SPM from the Lena,
Yana, and Khatanga rivers and their major tributaries were analyzed. Based on these data the
following conclusions can be drawn:
I The average chemical composition of Lena SPM compares well with average shale, except
for Mn-enrichment in the fine-grained SPM and Cd and Zn, which seem to be associated
with organic material. This finding indicates that the average composition of all lithologies in
the Lena catchment is similar to the upper continental crust. High feldspar contents
217
Factor analysis of the SP M data
Factor analyses of river water chemical data, such as principle component analysis, has been
applied to identify the mixing of different sources in numerous studies (c.f. Christophersen and
Hooper, 1992). In the present study, principle component analysis (orthogonal transformation)
of the chemical data set was used in order to: (I) reduce the large number of variables to a small
number of factors, (2) identify the general interrelationship between major and trace elements,
and (3) distinguish between SPM of individual rivers. The concentrations of 31 elements (AI,
Ti, Fe, Mn, Mg, Ca, Na, K, Ba, Bi, Ce, Co, Cs, Ga, Hf, La, Li, Mo. Nb, Ni, Rb, Sb, Sr,
Ta, Th, TI, U, V, W, Y, and Zr) in 64 samples were included in the calculation. As, Cd, Cu,
Pb, Sn, and Zn were omitted from the computation due to contamination effects.
The general variations in the chemical composition of the SPM can be explained by two
factors accounting for 71.4 % of the total variance. Figure 10 presents the factor loadings of
these two factors. The factor scores are illustrated in Figure II. Factor I (59.6 % of the
variance) clearly separates elements predominant in mafic rocks (high negative factor loadings
for Ca, Co, Fe, Mg, Ni, Ti, and V) from elements that are more abundant in acidic rocks (high
positive factor loadings for Ba, Bi, Ce, Cs, Hf, K, La, Li, Mo, Na, Nb, Rb, Sb, Ta, Th, TI,
U, W, Y, and Zr). This factor discriminates the Khatanga SPM from the Lena and Yana SPM.
While the geochemistry of Khatanga SPM is closely related to the basaltic rocks of the Siberian
Trap, the composition of Lena and Yana SPM is similar to average shale. except for some
elements. As a consequence, high negative factor scores can be observed in Khatanga SPM and
positive factor scores in Lena and Yana SPM (Figure 11). As has been discussed above, the
REE pattern of Vilyuy (L3) SPM indicates the influence of Siberian Trap basalts. This is also
seen in the factor score diagram where the Vilyuy plots between the Lena and Khatanga (Figure
11). In the Popigay (K3) SPM, the dominance of the Siberian Trap is less pronounced than in
the other Khatanga tributaries.The Popigay consequently plots very close to the Vilyuy in the
factor score diagram (Figure II).
Factor 2 (11.8 % of the variance) is characterized by high positive factor loadings for Cs, Hf,
Sb, Ti, and Zr and a negative Mn factor loading (Figure 10). This factor may be related to two
different processes: (1) The negative factor loading for Mn and the high positive factor loadings
for Hf, Ti, and Zr suggest that differences in grain-size distribution are responsible for this
factor. Due to the dominance of fine-grained material, which is characterized by high amounts
of Mn oxihydroxides, Lena SPM exhibits high Mn concentrations. Yana SPM, on the other
hand, displays higher amounts of coarse-grained material resulting in high Hf, Ti, and Zr
contents. (2) High positive loadings for Cs and Sb are related to granitic intrusions in the Yana
catchment (see above). As a result of both processes, factor 2 clearly differentiates between
Lena and Yana SPM. It can be concluded that factor analysis serves as a helpful tool to identify
the interrelationship of chemical elements in river SPM and to separate between SPM
transported by individual rivers.
Summary and conclusions
In order to unravel the sediment transport of East Siberian rivers draining to the Laptev Sea,
grain-size distribution, mineralogy, and inorganic chemical composition of SPM from the Lena,
Yana, and Khatanga rivers and their major tributaries were analyzed. Based on these data the
following conclusions can be drawn:
I The average chemical composition of Lena SPM compares well with average shale, except
for Mn-enrichment in the fine-grained SPM and Cd and Zn, which seem to be associated
with organic material. This finding indicates that the average composition of all lithologies in
the Lena catchment is similar to the upper continental crust. High feldspar contents