Constructing tephrochronologies covering periods of landscape instability

Tephrochronology is an established technique for dating and correlating sedimentary sequences, but it can be challenging to deploy when tephra layers are mixed, i.e., when they contain tephra from different sources. Mixing of tephra deposits is likely a ubiquitous process in regions with high levels of volcanic activity and dynamic landscapes, but remains understudied: to our knowledge, no one has systematically analysed the frequency and degree of mixing in terrestrial sedimentary sequences, nor assessed the reliability of a mixed, visible tephra layer as a chronostratigraphic horizon. We addressed this knowledge gap by quantifying tephra mixing in sedimentary sections encompassing the Landnám tephra layer (LTL), a key chronostratigraphic horizon in Iceland. We logged nine stratigraphic profiles in three sites associated with the earliest settlement of Iceland and analysed the major element chemistry of 44 tephra layers (957 shards in total). We used a statistical technique (clustering) to identify distinct geochemical populations within each layer and made novel use of Shannon’s diversity (entropy) index ( H ˊ) to estimate the degree of mixing. Our analysis showed that 43% of the tephra layers were mixed to some degree and around a quarter exhibited marked geochemical heterogeneity. We attributed the mixing to human activities associated with Norse colonisation of Iceland, which created a mixture of unstable surfaces (where tephra could be remobilised) and stable surfaces (where tephra layer formation was possible). We propose a model for the formation of mixed tephra layers. Our work has implications for the dating of the earliest phases of human colonisation in Iceland and the use of tephrochronology in areas — both proximal and distal to tephra sources — where conditions promote surface instability.