Abstract: Aim Here, we aim to: (a) investigate the local effect of environmental and anthropogenic factors on alien plant invasion in sub-Antarctic islands; and (b) explore whether and how functional traits affect alien species dependence on anthropogenic factors in these environments. Location Possession Island, Crozet archipelago (French sub-Antarctic islands). Methods Single-species distribution models were used to explore the effect of high-resolution topoclimatic and human-related variables on the occurrence of six alien plants colonizing French sub-Antarctic islands. Furthermore, plant responses to human-related variables and the effect of those variables in interaction with plant traits were analysed by means of a multi-species distribution model. This allowed identifying functional features mediating the influence of human activities on the occurrence probability of alien plant species. Results We observed two main invasion patterns: (a) species predicted to occur close to the introduction sites, whose occurrence probability appeared to be strongly affected by anthropogenic factors; and (b) species predicted to occur nearly everywhere on Possession Island, except in areas featuring particularly harsh climatic conditions. Differences in the influence of human-related variables on the occurrence of the alien species were mostly related to their life history, plant height and residence time, with perennial and low-statured species introduced earlier appearing less dependent on human-induced dispersal and disturbance. Conclusions We conclude that both topoclimatic and anthropogenic factors affect plant invasion on sub-Antarctic islands. Specifically, species predicted to occur close to their introduction sites appear much more dependent on human presence and activity, potentially due to the lack of key functional traits allowing them to spread successfully across Possession Island under the harsh sub-Antarctic climate. Yet, particularly severe abiotic conditions are a major constraint which equally limits the occurrence of all alien plants, irrespective of their dependence on anthropogenic factors.

Abstract: Water isotope ratios of ice cores are a key source of information on past temperatures. Through fractionation within the hydrological cycle, temperature is imprinted in the water isotopic composition of snowfalls. However, this signal of climatic interest is modified after deposition when snow remains at the surface exposed to the atmosphere. Comparing time series of surface snow isotopic composition at Dome C with satellite observations of surface snow metamorphism, we found that long summer periods without precipitation favor surface snow metamorphism altering the surface snow isotopic composition. Using excess parameters (combining D,17O, and 18O fractions) allow the identification of this alteration caused by sublimation and condensation of surface hoar. The combined measurement of all three isotopic compositions could help identifying ice core sections influenced by snow metamorphism in sites with very low snow accumulation.

Abstract: Species extinction risk is accelerating due to anthropogenic climate change, making it urgent to protect vulnerable species through legal frameworks in order to facilitate conservation actions that help mitigate risk. Here, we discuss fundamental concepts for assessing climate change risks to species using the example of the emperor penguin (Aptenodytes forsteri), currently being considered for protection under the US Endangered Species Act (ESA). This species forms colonies on Antarctic sea ice, which is projected to significantly decline due to ongoing greenhouse gas (GHG) emissions. We project the dynamics of all known emperor penguin colonies under different GHG emission scenarios using a climate-dependent meta-population model including the effects of extreme climate events based on the observational satellite record of colonies. Assessments for listing species under the ESA require information about how species resiliency, redundancy and representation (3Rs) will be affected by threats within the foreseeable future. Our results show that if sea ice declines at the rate projected by climate models under current energy system trends and policies, the 3Rs would be dramatically reduced and almost all colonies would become quasi-extinct by 2100. We conclude that the species should be listed as threatened under the ESA.