The role of sea ice for vascular plant dispersal in the Arctic

Abstract

Sea ice has been suggested to be an important factor for dispersal of vascular plants in the Arctic. To assess its role for postglacial colonization in the North Atlantic region, we compiled data on the first Late Glacial to Holocene occurrence of vascular plant species in East Greenland, Iceland, the Faroe Islands and Svalbard. For each record, we reconstructed likely past dispersal events using data on species distributions and genetics. We compared these data to sea-ice reconstructions to evaluate the potential role of sea ice in these past colonization events and finally evaluated these results using a compilation of driftwood records as an independent source of evidence that sea ice can disperse biological material. Our results show that sea ice was, in general, more prevalent along the most likely dispersal routes at times of assumed first colonization than along other possible routes. Also, driftwood is frequently dispersed in regions that have sea ice today. Thus, sea ice may act as an important dispersal agent. Melting sea ice may hamper future dispersal of Arctic plants and thereby cause more genetic differentiation. It may also limit the northwards expansion of competing boreal species, and hence favour the persistence of Arctic species.

Keywords: Arctic; climate change; driftwood; plant colonization; plant dispersal; sea ice.

Figure 1.

Boxplot of sea ice densities along ‘not likely’, ‘potential’ and ‘most likely’ plant dispersal routes based on 137 first occurrence data of 102 species, which each have six to eight potential dispersal routes, thus a total of 901 considered pathways. The sea-ice indices range from 0 (absence of sea ice) through 4 (occasional, less than 1 month per year) to 8 (dense sea ice). The middle lines show the median, boxes indicate the interquartile range and whiskers the range of the data. The sea ice was significantly denser along the ‘most likely’ dispersal routes than along routes classified as ‘not likely’ or ‘potential’ (p < 0.001).

Figure 2.

(a) The three main dendrochronologically inferred routes of driftwood in the North Atlantic region based on a compilation of sources (electronic supplementary material). Arrow thickness indicates the relative frequency of driftwood dispersal along that route, arrow colours indicate main source areas, and dots indicate sampling sites. Current sea ice from https://nsidc.org/. (b) Ocean surface currents from [8]. (c) Dominant sea-ice pathways as controlled by dominant wind and ocean currents today. It is, however, notable that short-term variability in sea ice especially linked to shifts in wind patterns and ocean gyres may result in sea-ice transport directions that diverge from the dominant pattern (e.g. [19]).