Is Glacial Meltwater a Secondary Source of Legacy Contaminants to Arctic Coastal Food Webs?

Abstract

Climate change-driven increases in air and sea temperatures are rapidly thawing the Arctic cryosphere with potential for remobilization and accumulation of legacy persistent organic pollutants (POPs) in adjacent coastal food webs. Here, we present concentrations of selected POPs in zooplankton (spatially and seasonally), as well as zoobenthos and sculpin (spatially) from Isfjorden, Svalbard. Herbivorous zooplankton contaminant concentrations were highest in May [e.g., ∑polychlorinated biphenyls (₈PCB); 4.43, 95% CI: 2.72-6.3 ng/g lipid weight], coinciding with the final stages of the spring phytoplankton bloom, and lowest in August (∑₈PCB; 1.6, 95% CI: 1.29-1.92 ng/g lipid weight) when zooplankton lipid content was highest, and the fjord was heavily impacted by sediment-laden terrestrial inputs. Slightly increasing concentrations of α-hexachlorocyclohexane (α-HCH) in zooplankton from June (1.18, 95% CI: 1.06-1.29 ng/g lipid weight) to August (1.57, 95% CI: 1.44-1.71 ng/g lipid weight), alongside a higher percentage of α-HCH enantiomeric fractions closer to racemic ranges, indicate that glacial meltwater is a secondary source of α-HCH to fjord zooplankton in late summer. Except for α-HCH, terrestrial inputs were generally associated with reduced POP concentrations in zooplankton, suggesting that increased glacial melt is not likely to significantly increase exposure of legacy POPs in coastal fauna.

Keywords: Svalbard; chiral pesticides; climate change; persistent organic pollutants; sculpin; stable isotopes; zoobenthos; zooplankton.

Figure 1

(a) Satellite image (Copernicus Sentinel data [August 20, 2018]) of Isfjorden where zooplankton were sampled in May, June, and August 2018 and benthic invertebrates in August 2018. The position of the ice edge in May 2018, when land-fast ice prevented sampling at the innermost stations, is indicated in black. Stars represent the city of Longyearbyen and the abandoned mining village of Pyramiden, which represent local sources of contamination. (b) Isfjorden station map showing stations where sculpin were sampled using gillnets in August 2018. (c) Map of Svalbard with the West Spitsbergen Current (WSC) depicted in red.

Figure 2

(A) POP concentrations and (B) EFs in bulk zooplankton by month for each plankton type: herbivorous zooplankton (Calanus spp., Meroplankton) and omnivorous and predator zooplankton (Macrozooplankton and Jellyplankton). Diamonds and error bars represent the bootstrapped mean and 95% confidence interval. ∑₈PCB is defined as the sum of CB-28, CB-31, CB-52, CB-101, and CB-153 (CB-118, CB-138, and CB-180 were <LOD in zooplankton). The racemic ranges (determined using laboratory standards) are indicated as dashed gray lines. POP concentrations on a wet weight basis can be found in Figure S8.

Figure 3

Partial RDA based on log-transformed concentrations of sums of PCBs, chlordane pesticides, DDTs, and α-HCH in herbivorous zooplankton with variance (20.6%) due to lipid content removal. Constraining variables: δ¹³C-Zoo, salinity, and temperature, which explain 41% of the residual variance, are shown in blue. EF of α-HCH (in black) is included as a passive vector. Each point represents one individual sample, and color represents the sampling month with blue = May, light brown = June, and dark brown = August.

Figure 4

HCB and ∑₈PCB concentrations in filter- and deposit-feeding benthic invertebrates and sculpin vs fjord sediment concentrations. Points and error bars represent the bootstrapped mean and 95% confidence intervals based on all fjord replicates. ∑₈PCB is defined as the sum of CB28/31, CB-52, CB-101, CB-118, CB-138, CB-153, and CB-180 for zoobenthos and CB-52, CB-138, CB-153, and CB-180 for sculpin.