Soil selenium in a forested seabird colony: distribution, sources, uptake by plants, and comparison with non-seabird sites

Abstract

Seabirds vector selenium (Se) into terrestrial ecosystems in Antarctica and on tropical coral islands, but factors controlling distribution within affected soils are unknown, especially in temperate regions. At a Westland petrel (Procellaria westlandica) breeding colony on mainland New Zealand, the concentration of Se in petrel guano (3.6 mg kg–1) exceeded soil parent material (0.8 mg kg–1) and in all but two soil samples (range 1.2–4.2 mg kg–1; n = 52). External Se (Se not derived from parent material) accounted for 64 ± 9% (mean ± s.d.) of soil Se. Measurements were also made at a former seabird breeding site, and at a site with no Holocene seabird breeding. Median surface-soil Se concentrations (mg kg–1) were in the order burrow soil (2.6) > adjacent forest floor (2.2) > former breeding site (1.0) > control site (0.2), with significant differences between burrow soil and (1) the former breeding site and (2) the control site. In a linear regression model, soil pH, and δ15N were the only significant predictors of external Se in colony soil. The correlations are consistent with seabird input driving both the Se supply and increased sorptive uptake in an environment acidified by seabird guano. Despite the enhanced Se in colony soil, median foliage concentrations (tree fern 0.05 mg kg–1, nikau 0.08 mg kg–1) were close to the accepted minimum for herbivore nutrition. Seabirds therefore contribute significant Se to breeding colony soils in temperate areas, but this is not necessarily transferred to plant foliage.