Tectonic discrimination of basalts with classification trees

Traditionally, geochemical classification of basaltic rocks of unknown tectonic affinity has been performed by discrimination diagrams. Although easy to use, this method is fairly inaccurate because it only uses bi- or trivariate data. Furthermore, many popular discrimination diagrams are statistically not very rigorous because the decision boundaries are drawn by eye, and they ignore closure, thus violating the rules of compositional data analysis. Classification trees approximate the data space by a stepwise constant function, and are a more rigorous and potentially more effective way to determine tectonic affinity. Trees allow the simultaneous use of an unlimited number of geochemical features, while still permitting visualization by an easy-to-use, two-dimensional graph. Two classification trees are presented for the discrimination of basalts of mid-ocean ridge, ocean island, and island arc affinities. The first tree uses 51 major, minor, and trace elements and isotopic ratios and should be used for the classification of fresh basalt samples. A second tree only uses high field strength element analyses and isotopic ratios, and can also be used for basalts that have undergone alteration. The probability of successful classification is 89% for the first and 84% for the second tree, as determined by 10-fold cross-validation. Even though the trees presented in this paper use many geochemical features, it is not a problem if some of these are missing in the unknown sample. Classification trees solve this problem with surrogate variables, which give more or less the same decision as the primary variables. The advantages of the classification tree approach over discrimination diagrams are illustrated by a comparative test on a sample dataset of known tectonic affinities. Although arguably better than discrimination diagrams, classification trees are not perfect, and the limitations of the method are illustrated on a published dataset of basalts from the Pindos Basin (Greece).     

Geochemical fingerprinting of oceanic basalts with applications to ophiolite classification and the search for Archean oceanic crust

Two geochemical proxies are particularly important for the identification and classification of oceanic basalts: the Th–Nb proxy for crustal input and hence for demonstrating an oceanic, non-subduction setting; and the Ti–Yb proxy for melting depth and hence for indicating mantle temperature and thickness of the conductive lithosphere. For the Th–Nb proxy, a Th/Yb–Nb/Yb projection demonstrates that almost all oceanic basalts lie within a diagonal MORB–OIB array with a principal axis of dispersion along the array. However, basalts erupted at continental margins and in subduction zones are commonly displaced above the MORB–OIB array and/or belong to suites with principal dispersion axes which are oblique to the array. Modelling of magma–crust interaction quantifies the sensitivity of the Th–Nb proxy to process and to magma and crustal compositions. For the Ti–Yb proxy, the equivalent Ti/Yb–Nb/Yb projection features a discriminant boundary between low Ti/Yb MORB and high Ti/Yb OIB that runs almost parallel to the Nb/Yb axis, reflecting the fact that OIB originate by melting beneath thicker lithosphere and hence by less melting and with residual garnet. In the case of volcanic-rifted margins and oceanic plume–ridge interactions (PRI), where hot mantle flows toward progressively thinner lithosphere (often becoming more depleted in the process), basalts follow diagonal trends from the OIB to the MORB field. Modelling of mantle melting quantifies the sensitivity of the Ti–Nb proxy to mantle potential temperature and lithospheric thickness and hence defines the petrogenetic basis by which magmas plot in the OIB or MORB fields. Oceanic plateau basalts lie mostly in the centre of the MORB part of that field, reflecting a high degree of melting of fertile mantle. Application of the proxies to some examples of MORB ophiolites helps them to be further classified as C (contaminated)-MORB, N (normal)-MORB, E (enriched)-MORB and P (plume)-MORB ophiolites, which may add a useful dimension to ophiolite classification. In the Archean, the hotter magmas, higher crustal geotherms and higher Th contents of contaminants all result in widespread crustal input that is easy to detect geochemically with the Th–Nb proxy. Application of this proxy to Archean greenstones demonstrates that almost all exhibit a crustal component even when reputedly oceanic. This indicates, either that some interpretations need to be re-examined or that intra-oceanic crustal input is important in the Archean making the proxy less effective in distinguishing oceanic from continental settings. The Ti–Yb proxy is not effective for fingerprinting Archean settings because higher mantle potential temperatures mean that lithospheric thickness is no longer the critical variable in determining the presence or absence of residual garnet.     

Density calculations for silicate liquids: Reply to a Critical Comment by Ghiorso and Carmichael

The analysis of the liquid silicate density model recently proposed in Bottingaet al. (1982) by Ghiorso and Carmichael (1984) is shown to be based on a combination of unwarranted mathematical assumptions, refusal to recognize experimental and theoretical evidence for the non-linear effect of composition on liquid silicate density, and a totally unrealistic view of the accuracy with which the thermal expansion of silicate liquids can be measured. As a consequence, none of the general or specific points raised by Ghiorso and Carmichael are relevant to the issue of which of the existing calculation models (Bottinga and Weill, 1970; Nelson and Carmichael, 1979; Moet al., 1982; or Bottingaet al., 1982, 1983) should be used. As stated in Bottinga, Richet and Weill (1983), there is a problem in using a combination of the molar volume parameters from the first three of these models because they are not mutually independent. However, the set of partial molar volumes and thermal expansion constants given in Bottingaet al. (1982, 1983) are internally consistent and mutually compatible. We remain firmly of the opinion that our latest model is an improvement over previous attempts because it conforms to a much wider set of observations, it incorporates a larger set of melt components, it calculates density and thermal expansion more accurately, and it points the way to one possible method of accommodating a non-linear phenomenon into a nonlinear model.     

对“燕山早期花岗岩基印支期侵位的岩浆动力学证据及构造意义: 基于南岭8个岩体侵位年龄计算结果”的质疑

     基于构造学、岩石学和矿床学的地质事实,本文认为南岭地区燕山早期同造山花岗岩的岩浆形成、侵位和结晶在一 个很短的时间段内完成,其时间可以用锆石U-Pb年龄代表。因此, 章邦桐等（2014） 一文的结论应该是不成立的。     

A reappraisal of the high-Ti and low-Ti classification of basalts and petrogenetic linkage between basalts and mafic–ultramafic intrusions in the Emeishan Large Igneous Province, SW China

Based on published data, we reappraise the classification of high-Ti and low-Ti basalt from the Emeishan large igneous province (ELIP) and the correlations between basalts and mafic–ultramafic intrusions. Because of the lack of clear spatial and temporal variations of different types of basalts, we suggest that the basalts in the ELIP cannot be classified into high-Ti and low-Ti groups, by TiO₂ contents and/or Ti/Y ratios. The distinctive characteristics of these high-Ti and low-Ti lavas probably result largely from the different fractionating assemblages. Whether or not fractional crystallization of the Fe–Ti oxides occurred probably is the key factor that controls the Ti abundances and Ti/Y ratios in the residual melts, e.g., lavas, although the nature of the mantle sources, variable degrees of partial melting of mantle and crustal contamination also influence the geochemical signatures of the lavas. Therefore, neither Ti abundance nor Ti/Y ratios in basalts can reflect the nature of their mantle source. Moreover, the different types of mafic–ultramafic intrusions in the ELIP cannot simply be attributed to be genetically related special types of basalts, either high-Ti or low-Ti basalts. It is likely that they are merely the cumulus phases, i.e. chamber or conduit of the basaltic lavas. Hence, caution should be exercised in the use of high-Ti or low-Ti basalts as prospecting vectors for ore deposits in the region. Potential implications are proposed that both the Fe–V–Ti oxide and Cu–Ni–(PGE) sulfide mineralization in the ELIP intrusions is largely due to the variable differentiation and crustal contamination during magmatic processes.