Trace elemental and Nd-Sr-Pb isotopic compositional variation in 37 lava flows of the Mandla lobe and their chemical relation to the western Deccan stratigraphic succession, India

Abstract

The Mandla lobe is a 900 m thick lava pile that forms a 29,400 km2 northeastern extension of the Deccan Traps. Earlier, combined field, petrographic, and major element studies have shown that this lobe comprises 37 lava flows. Using a combination of trace elements (Ba, Ti, Zr, Rb, Sr) and Nb/Zr values, we group the flows into six chemical types (A–F) that are separated stratigraphically. Combined trace element and Nd-Pb-Sr isotopic data, document the presence of lavas resembling those of the Poladpur Formation and less abundantly, the Ambenali Formation of the southwestern Deccan are in conformity with the earlier reconnaissance work. In addition, our data reveal several flows similar to those of the Mahabaleshwar Formation, the type sections of which are located ~ 900 km to the southwest. Based on the isotopic data the superposition of Mahabaleshwar-like flows over flows with Ambenali- and Poladpur-like characteristics is in the same stratigraphic order seen in the southwestern Deccan type section. However, from the stratigraphy indicated by the Discriminant Function Analysis (DFA) results and the serious discrepancy between the DFA and isotopic data, it seems that few Mandla lobe flows are different and not in the same stratigraphic order as in the southwestern part of the province. To some extent the differences may be explained by faulting along four large post-Deccan normal faults near Nagapahar, Kundam, Deori, and Dindori areas across which offsets of ~150 m have been measured. This post-emplacement faulting accounts for the presence of several chemically Mahabaleshwar-like lavas at the base of the ~900 m thick Mandla lobe pile, at a lower elevation than a thick sequence dominated by chemically Poladpur-like flows. However, presence of common signature lavas (similar to that in the northeastern Deccan) cannot be ruled out in this area. They are similar to Poladpur-type lavas both chemically and isotopically. They appear in different formations and erupted at different times other than Poladpur Formation. Close similarities in petrogenetic processes between the two regions are indicated, although it is not clear whether any of the Mandla lobe lavas are far-traveled counterparts of flows cropping out in the southwestern Deccan, or whether some magma migrated laterally in dike systems over great distances. Feeder dykes have not been found in the study area except for Chakhla-Delakhari Intrusive Complex (CDIC) in Satpura region that shows major and trace elemental similarities with the Seoni lavas, although, long distance transport of magma is yet to be proved. The Poladpur-like Mandla lobe flows appear to be different flows from those of the Poldapur Formation in the southwest, as they are somewhat different in isotopic (higher 206Pb/204Pb) composition. They also differ from any known flows in the other southwestern formations, but are broadly similar to flows found in sections across the northern Deccan west of the Mandla lobe.