Soil seed banks from coastal subarctic ecosystems of Bird Cove, Hudson Bay

Abstract

Coastal ecosystems at high latitudes are exposed to physical disturbances (abrasion by ice, tides, waves, and wind) and to degradation due to the grazing activities of waterfowl. The frequency of disturbance and the regenerative capacity of plants determine the recovery potential for the plant community. Since there have been no previous soil seed bank studies of coastal subarctic ecosystems, we examined the seed bank in soil samples collected from a beach ridge, a drift zone, a sand bar and a salt marsh at Bird Cove, near Churchill, Manitoba, on the western shoreline of Hudson Bay. Seedling emergence from soil samples after four successive periods of stratification at 5°C totalled 2968 seedlings of 19 species. Mean emerged seedling densities for each ecosystem were (i) 786 seedlings/m2 (beach ridge), (ii) 699 seedlings/m2 (drift zone), (iii) 2875 seedlings/m2 (sand bar), and (iv) 39 204 seedlings/m2 (salt marsh). These densities are much higher than those for temperate coastal ecosystems and for most terrestrial, subarctic ecosystems. Ninety-seven percent of the 2772 identifiable emergents were of three ruderal species: Juncus bufonius, Senecio congestus and Spergularia marina. Juncus bufonius and S. marina are on the extreme northern limits of their ranges. These three common species were found from samples collected from all ecosystems except the beach ridge, which had no species in common with the other ecosystems and was dominated by Sagina nodosa and Saxifraga tricuspidata in its seed bank. There was little correlation between abundances of species in the seed bank and those in the aboveground vegetation except in the drift zone. Ninety-five percent of the seedlings emerged from the top 5 cm of the sand bar and salt marsh soils, whereas 91% within the drift were deeper than 85 cm. Seedling emergences increased with time and cold storage treatment; freshest samples (i.e., those with 35 days of cold storage) gave 6% of the ultimate emergence but this rose to 57% after two stratification episodes and then to 100% after three. This suggested that seed bank analysis using the emergence method should consider the germination requirements of each potential seed bank species.