Climate influence on mercury in Arctic seabirds

Abstract

The historic influence of interannual weather and climate variability on total mercury concentrations (THg) in the eggs of two species of Arctic seabird in the Canadian High Arctic was investigated. Time series of THg in the eggs of northern fulmars (Fulmarus glacialis) and thick-billed murres (Uria lomvia) from Prince Leopold Island span 40 years (1975–2014), making these among the longest time series available for contaminants in Arctic wildlife and uniquely suitable for evaluation of long-term climate and weather influence. We compiled a suite of weather and climate time series reflecting atmospheric (air temperature, wind speed, sea level pressure) and oceanic (sea surface temperature, sea ice cover) conditions, atmosphere-ocean transfer (snow and rain), as well as broad-scale teleconnection indices such as the Arctic Oscillation (AO) and North Atlantic Oscillation (NAO). We staggered these to the optimal time lag, then in a tiered approach of successive General Linear Models (GLMs), strategically added them to GLMs to identify possible key predictors and assess any main effects on THg concentrations. We investigated time lags of 0 to 10 years between weather/climate shifts and egg collections. For both fulmars and murres, after time lags of two to seven years, the most parsimonious models included NAO and temperature, and for murres, snowfall, while the fulmar model also included sea ice. Truncated versions of the datasets (2005–2014), reflective of typical time series length for THg in Arctic wildlife, were separately assessed and generally identified similar weather predictors and effects as the full time series, but not for NAO, indicating that longer time series are more effective at elucidating relationships with broad scale climate indices. Overall, the results suggest a significant and larger than expected effect of weather and climate on THg concentrations in Arctic seabirds.