New article in Quaternary Science Reviews

by Malin E. Kylander et al.

 

Storm chasing: Tracking Holocene storminess in southern Sweden using mineral proxies from inland and coastal peat bogs

Malin E. Kylander, Antonio Martínez-Cortizas, Jenny K. Sjöström, Jenny Gåling, Richard Gyllencreutz, Richard Bindler, Helena Alexanderson, Frederik Schenk, Benedict T.I. Reinardy, Benjamin M.P. Chandler, Kerry Gallagher, 2022. Storm chasing: Tracking Holocene storminess in southern Sweden using mineral proxies from inland and coastal peat bogs. Quaternary Science Reviews: 299, 107854, ISSN 0277-3791, https://doi.org/10.1016/j.quascirev.2022.107854.

https://www.sciencedirect.com/science/article/pii/S0277379122004851

Abstract
Severe extratropical winter storms are a recurrent feature of the European climate and cause widespread socioeconomic losses. Due to insufficient long-term data, it remains unclear whether storminess has shown a notable response to changes in external forcing over the past millennia, which impacts our ability to project future storminess in a changing climate. Reconstructing past storm variability is essential to improving our understanding of storms on these longer, missing timescales. Peat sequences from coastal ombrotrophic bogs are increasingly used for this purpose, where greater quantities of coarser grained beach sand are deposited by strong winds during storm events. Moving inland however, storm intensity decreases, as does sand availability, muting potential paleostorm signals in bogs. We circumvent these issues by taking the innovative approach of using mid-infrared (MIR) spectral data, supported by elemental information, from the inorganic fraction of Store Mosse Dune South (SMDS), a 5000-year-old sequence from a large peatland located in southern Sweden. We infer past changes in mineral composition and thereby, the grain size of the deposited material. The record is dominated by quartz, whose coarse nature was confirmed through analyses of potential local source sediments. This was supported by further mineralogical and elemental proxies of atmospheric input. Comparison of SMDS with within-bog and regionally relevant records showed that there is a difference in proxy and site response to what should be similar timing in shifts in storminess over the ∼100 km transect considered. We suggest the construction of regional storm stacks, built here by applying changepoint modelling to four transect sites jointly. This modelling approach has the effect of reinforcing signals in common while reducing the influence of random noise. The resulting Southern Sweden-Storm Stack dates stormier periods to 4495–4290, 3880–3790, 2885–2855, 2300–2005, 1175–1065 and 715-425 cal yr BP. By comparing with a newly constructed Western Scotland-Storm Stack and proximal dune records, we argue that regional storm stacks allow us to better compare past storminess over wider areas, gauge storm track movements and by extension, increase our understanding of the drivers of storminess on centennial to millennial timescales.

Fig. 1. (A) Overview map of Store Mosse and surrounding areas. The map is based on the extent of peat, aeolian sand sheets, dunes and glacifluvial material as excerpts from the Swedish Quaternary map “Jordartskartan 25,000–100 000” (© Geological survey of Sweden; SGU, 2022b), a 2-m resolution digital elevation model “GSD-Höjddata grid 2+“ (© Lantmäteriet) and the peat thickness map by Eller and Brenner (1912). All thematic areas on the map are semi-transparent to show underlying hill-shaded topography, which is relatively flat in the areas of the shown sediment types. Store Mosse is located 163–168 m asl, and the map area elevation spans ∼150–300 m asl. Locations for source samples and peat sequences are marked with circles and stars, respectively. Inset are the locations of Store Mosse (red square) and the comparison sites Undarsmosse (1), Davidsmosse (2), Hyltemosse (3), Draftinge mosse (4) and Lake Avegöl (5) (see text for references). (B, upper panel) Composite GPR profile (including profiles 55, 57, 58 and 60); (B, lower panel) interpreted version of the same GPR profile with prominent reflectors annotated by black lines. The basal reflector at ∼3.5–5 m is interpreted as the base of the peat bog.