Taphonomy of mammalian fossil bones from the debris-flow deposits of Somosaguas-North (Middle Miocene, Madrid Basin, Spain)

Debris-flow hosted assemblages dominated by mammalian remains are very scarce in the fossil record and few examples are reported. Herein we present a detailed taphonomic study of Somosaguas-North (Madrid Basin,Spain), a Middle Miocene mammalian-dominatedsite embedded in debris-flow deposits, in order to increase our knowledge of the mode of formation of fossiliferous assemblages in this depositional context. The assemblage includes 6592 large-mammal remains belonging to at least 10 different species. Fossils are found in matrix-supported, poorly-sorted coarse arkosic sandstones and fine conglomerates, which are interpreted as the result of successive debris-flow deposits. Breakage constitutes a pervasive taphonomic alteration in the bone assemblage; however, together with unidentifiable bone fragments, we also found complete to almost complete skeletal elements Bone remains are isolated, a fact that is indicative of a time of exposure of the carcasses long enough to enable decay and complete disarticulation. There are no associations of elements belonging to the same individual, which suggests that, prior to or during debris-flow transport, bones were profusely dispersed. Based on the taphonomic evidence, we suggest that Somosaguas-North assemblage was formed by successive debris-flow transport and burial of pre-existing thanatocoenoses. Bones from those thanatocoenoses were modified by biostratinomic agents (trampling, weathering) in varying degrees depending on their time of exposure. Debris-flow transport produced further abrasion and breakage in collagen-depleted bones. Our results are congruent with an environmental context of semiarid landscapes presenting long arid periods and highly irregular hydrological seasonality. Finally, we compared the Somosaguas-North assemblage with other debris-flow hosted vertebrate assemblages. Although Somosaguas-North shares taphonomic features with some of these sites, it is not possible to define consistent and singletaphonomic patterns for debris-flow assemblages, given the varied taphonomic histories of the remains prior to and during debris-flow transport