There is considerable debate about whether or not the Siberian Traps were formed from a mantle plume. The problem is made worse because a substantial amount of the Traps is buried beneath the West Siberian Basin.

What do we know? Firstly, the volumes of erupted (and intruded) magma were prodigious, requiring a large energy source. Secondly, it is very unlikely that the basalts originated by melting beneath the Siberian craton. This old, Archaean structure is simply too thick to allow substantial melting and, even if this did occur, the eruptives would have a geochemistry consistent with very high pressure melting; mostly, they don't. Most of the magmas probably originated from beneath the lithosphere of what is now the West Siberian Basin and the Yenesei-Khatanga Trough.

Thirdly, there appears to be no evidence of uplift recorded in the sedimentary rocks at Noril'sk and on the adjacent craton. Czamanske et al. (1998) have eloquently argued that this rules out involvement of a mantle plume, and their case is further presented here. Rather than a plume, they argue for edge-driven convection, using the ideas of King and Anderson (1995, 1998). Tanton and Hager (2000) argued that lithospheric foundering may have triggered the formation of the Traps.

However, Saunders et al. (2005) have suggested that there is evidence of pre-basaltic uplift, but it is recorded in the sequences in the West Siberian Basin. In fact, it is the absence of much Permian sedimentation in the WSB that has led to our conclusion that the area was uplifted, perhaps by a mantle plume. Furthermore, Saunders et al. (2005) argue that the Mesozoic subsidence of the WSB was initially delayed by thermal support from the mantle.

The compositions of the basalts at Noril'sk, Putorana and in the WSB are consisent with a shallowing of the depth of melting as the province was emplaced. The early basalts at Noril'sk indicate slightly deeper melting, and have compositions similar to basalts from Hawaii (albeit the former were contaminated during their ascent through the lithosphere). The main (and most voluminous) sequences of basalts involved either shallow melting and/or very extensive melting, consistent with a hot source that was allowed to decompress to shallow levels, perhaps due to crustal extension in the WSB. The rapid ascent and decompression of the plume mantle may have been assisted by delamination of the lithosphere, as discussed by Elkins-Tanton (2005).

The arguments for and against a mantle plume are finely balanced. The compositions of the early Noril'sk suites indicate a hot mantle source and, together with the uplift in the WSB, favour a plume origin for the Traps. However, the distribution and timing of the uplift needs to be better constrained. If it was a plume, where did the plume go? There is tantalising evidence that it migrated (in a relative sense) beneath the Arctic regions, and currently resides beneath Iceland (e.g., Lawver and Muller, 1994; Lawver et al., 2002).