Search

We are entering the era of big data, and big data achievements have benefited ordinary people. However, big data application in geochemistry research has not been really appreciated. Big data is a kind of method and a pattern of thought. It is different from traditional scientific research methods and thinking in that it starts from the data and adopts full data mode. This paper introduces basalt, andesite and continental marginal arc basalt tectonic environment discrimination diagrams made by the author and his collaborators using all the global rock geochemical data available. We have discussed the following scientific problems of petrology and geochemistry using big data research: 1. we found many better illustrations in the study of discriminant diagrams, mainly relying on the relationship among the major elements, transition elements, and metal elements. What the points between the relationships are? Why did they work? These new questions need to be studied by the geochemists. 2. Data mining found that mid-ocean ridge is extremely lack of intermediate-acidic rocks. Does it mean that the upper mantle is an acute shortage of water? 3. This study found that the Miocene is the most developed epoch of magmatic activities. The Miocene appeared many important geological events in the world. Is there a connection between them? 4. The Miocene adakite most developed in the world, according to adakite exposed. There may be a giant Eurasian plateau from the Tibetan plateau to the Carpathians. 5. According to the spatial and temporal distribution of Cenozoic picrite, we put forward how to realize a global hotspot issues, etc. The paper also put forward some suggestions of further research and emphasized that science has entered the era of big data. In the era of big data, scientific classification standards have changed: any discipline which can be expressed in data is called science, and which cannot be expressed in data is called non-science. Whether it can be expressed by data is the watershed of science and non-science. In the era of big data, the geology encountered an unprecedented crisis. According to our forecast, in the foreseeable future, geophysics will be far ahead of geology. Space science will boom, and geochemistry will lead the way for a long time in the field of geology. Keywords: Big data, Petrology, Geochemistry, Basalt, Andesite, Discrimination diagram

This study evaluated the status of deteriorating water quality around the Vembanad Lake of Kerala State in India. A total of 40 water samples from lake (L, n = 5), river (R, n = 6), open well (W, n = 26), and bore well (BW, n = 9) and were analysed for major ions as per the APHA methods. Different concentrations of Ca (10–190 mg/L), Mg (2–220 mg/L), Na (10–216 mg/L), K (2–50 mg/L), HCO3 (20–400 mg/L), CO3 (10–50 mg/L), Cl (46.5–950 mg/L), NO3 (0.62–19.75 mg/L), SO4 (20.8–423 mg/L) and PO4 (1.63–16.18 mg/L) represent the three different water types: Ca HCO3, Na Cl and mixed type of Ca Na HCO3. Strong connection between the host rock and groundwater quality indicated substantial water-rock interaction. Factor analysis helped to understand the associations of chemical parameters and their sources. TDS, TH, calcium, magnesium, sodium, potassium and chloride exceeded the BIS (2010) based on desirable drinking water and irrigational quality standards in SE, W, and NW directions due to man-made bunds, industrial wastes, agricultural wastes, household sewages, tourism, lime shell, and sand mining. All point and non-point sources of pollution adversely affected water in and around the Vembanad Lake system and caused by water quality deterioration.

Heavy metals (Fe, Mn, Zn, Pb, Cu, Cr, Ni, and Cd) in core sediments from Qi’ao Island, South China were subjected to distribution determination, risk assessment, and source identification. The concentrations (in microgram per gram) of the metals decreased in the following order: Fe (27275.8 - 50893.60) > Mn (372.0 - 2063.6) > Zn (104.8 - 314.9) > Pb (44.1 - 84.8) > Cu (32.5 - 102.2) > Cr (19.6 - 467.6) > Ni (13.0 - 62.7) > Cd (1.9 - 37.8). The geoaccumulation index (I g e o ) and contamination factor ( C f ) indicated that the sediments were heavily contaminated with Cd and moderately to very highly contaminated with Cu, Cr, Ni, Pb, and Zn. Single metal risk factor (E f i ) showed that Cu, Ni, Cr, and Zn posed low risks ( 320%). In terms of risk index (RI), all study sites were considered to have a high ecological risk (RI>600). As regards pollution load index (PLI) classification, the majority of the sites were classified as “polluted” with a mean PLI of 2.8. Correlation and PCA analyses showed that the metal sources were well correlated with one another, and these sources were geogenic and industrial (agrochemicals and electroplating) wastewater discharge, as well as domestic sewage activities in the PRDE and surrounding areas. Overall, our findings showed that Qi’ao Island was moderately to highly contaminated with heavy metals.

The distribution of elements in soil is affected by various natural processes, and therefore, it is essential that exploration geochemists are able to recognize whether an elemental concentration is related to mineralization or other processes. This research aimed to recognize the elemental distribution and its controlling factors around the Zhaojikou Pb–Zn ore deposit, SE China, via fractal and compositional data analysis. Firstly, histograms over the raw data and log-ratio transformations were used to check whether the concentration data followed a normal distribution or not. The concentration distribution maps of elements were then generated via spretrum–area fractal plot to show the spatial relationships. Afterward, the inter-relationships among different elements were studied through robust principal component analysis based on compositional data theory. The factors controlling the distribution of elements were discussed and geochemical indices were proposed for future exploration. Finally, a comprehensive anomaly map was generated to show important prospecting areas. Results showed how the interpretation of distributions of elements was enhanced due to the application of data log-ratio transformation in univariate and multivariate analysis compared to the use of raw data. The histograms based on isometric log-ratio performed best among the three log-ratio transformations. The concentration distribution maps showed three anomalous areas at the surface. It clearly allowed identifying the relationships between the elements and their possible source patterns. Area I with anomalous assemblage of Pb–Zn–Ag–Cu–Cd corresponded to the current known ore deposits, while Area II with anomalous assemblage of Pb–Zn–Ag–Cu–Cd may indicate the occurrence of new potential base metal (Pb–Zn) ore bodies, and Area III with anomalous assemblage of W–As–Sb may indicate the occurrence of new potential Au mineralization/ore occurences. The robust biplot of PCA shows that the distributions of elements of interest were generally controlled by three geological processes: Pb–Zn ore-forming process, Au mineralization and weathering. Lead, Zn, Ag, Cu, and Cd, were affected mainly by Pb–Zn ore-forming processes; W, Sb, and As were affected mainly by Au mineralization, whereas Al2O3, K2O, Rb, Ba, Sr, Na2O, CaO, MgO, and SiO2 were affected to a greater extent by weathering processes. Sulphur, a mineralizing agent, was controlled by the ore-forming processes, being affected greatly by weathering. Finally, we proposed Pb > 160 ppm, Zn > 120 ppm, Ag > 193 ppb, Cu > 50 ppm, Cd > 210 ppb, W > 5 ppm, As > 25 ppm, Sb > 3.5 ppm, and S > 120 ppm as geochemical indices based on concentration-area fractal model analysis for future surficial geochemical prospecting in the Zhaojikou District.