“Hidden” boninitic magmatism in the Zagros ophiolites. A Comparison with the Albanide-Hellenide Ophiolites.

Ophiolites surfacing along the Main Zagros Thrust Zone (MZTZ) mark the suture zone between the Arabian and Sanandaj-Sirjan continental blocks. They represent a portion of the southern Neo-Tethyan oceanic lithosphere, which originally existed between the Arabian (to the south) and Eurasian (to the north) continental margins. Several authors suggested that an intra-oceanic supra-subduction zone setting (SSZ) developed in this ocean during the Late Cretaceous. SSZ ophiolites from the Albanide-Hellenide belt are characterized by the widespread occurrence of boninitic basalts. In contrast, in the MZTZ ophiolites, these rocks have not been found so far, with the exception of one sample from the Baft area. Nonetheless, the volumetrically most abundant rock-type within the MZTZ ophiolites consists of very depleted mantle harzburgites, which have chemical features that are typical for residual mantle after boninitic-type melt extraction. Therefore, though boninitic lava flows are lacking in the MZTZ ophiolites, the occurrence of boninitic magmatism at a regional scale can be envisaged. The aim of this contribution is therefore to review available data on the Kermanshah and Sarve-Abad ophiolites (SW Iran) in search for evidence for the existence of boninitic magmatism in the southern Neo-Tethys. The Kermanshah ophiolites are composed of various incomplete sequences formed in different tectonic settings. Among these, the SSZ sequences consist of variably depleted mantle harzburgites and minor depleted mantle lherzolites. Harzburgites show close analogies with similar rocks from the Albanide-Hellenide SSZ ophiolites. They display a significant depletion in incompatible elements and rare earth element (REE), coupled with a marked light (L) REE enrichment with respect to medium (M) REE. REE modeling shows that they may represent a residual mantle after 25 – 30% removal of boninitic-type melts in an intra-oceanic arc setting. The mineral chemistry of Cr-spinels also supports this conclusion. The Sarve-Abad ophiolites consist of cumulitic lherzolites bearing minor dunite and chromitite lenses in places. The crystallization order in ultramafic cumulates is: olivine ± Cr spinel + clinopyroxene ± orthopyroxene, which is typical for boninitic melts. The mineral chemistry of Cr-spinel and pyroxenes is compatible with a genesis from a boninitic-type melt. Indeed, the calculated TiO2 and Al2O3 compositions and Mg# in the parental melt that was in equilibrium with chromian spinel and olivine are consistent with boninitic-type compositions. Whole-rock geochemistry of the ultramafic cumulates is characterized by very low incompatible element content and a general enrichment in Th with respect to Ta and Nb. Chondrite-normalized REE patterns show different trends with either (La/Sm)N < 1 and (Sm/Yb)N < 1 or (La/Sm)N > 1 and (Sm/Yb)N < 1 (U-shaped pattern). Both these patterns are compatible with boninitic-type parental melts. Accordingly, REE petrogenetic modeling indicates that the Sarve-Abad ultramafic cumulates may have formed by small degrees (5-15%) of fractional crystallization from typical boninitic melts characterized by either LREE/MREE depletion or enrichment. In conclusion, several lines of evidence indicate that episodes of boninitic magmatism occurred within the southern Neo-Tethys Ocean during the Late Cretaceous. This conclusion poses however more questions that should be solved. The main one concerns the lacking of boninitic lavas or dykes in the MZTZ ophiolites. To this purpose, further investigations should be made. Particularly, structural investigation and a careful comparison with the well-known Albanide-Hellenide ophiolites may be helpful.