Antarctica's geological arks of life

Abstract

Antarctic biogeographers and ecologists are currently engaged in a quest for the arks of life on the Antarctic continent. Over the last decade or so, the application of molecular phylogeography to the Antarctic terrestrial species pool has strengthened the case for widespread continental endemism (Stevens et al., 2006; Convey et al., 2008). These remnant species somehow persisted in Antarctica through the extreme ice cover of the Last Glacial Maximum (LGM) during the Pleistocene. The question of how and where they did this has re-invigorated Antarctic terrestrial biology. Although the Antarctic fairy shrimp, Branchinecta gaini, is at least one biological outlier to this pattern (Hawes, 2009), it is clear that a large proportion of the contemporary terrestrial biota escaped extinction as a result of refugia. While the idea of refugia is not new, it is only recently that a concerted attempt has begun to give the concept a biological and geological definition (Convey et al., 2008). This paper briefly outlines and discusses the main emerging hypotheses. The geothermal hypothesis states that some life in Antarctica survived glaciation as a result of themal refugia provided by regions of geothermal activity. The geothermal regions around the volcanoes of Mount Erebus and Mount Melbourne, for example, are known to harbour mosses and bacteria that cannot grow anywhere else in Antarctica (Skotnicki et al., 2001; Bargagli et al., 2004). Now, a new study by Fraser et al. (2014) has used continental-scale modelling to explore the hypothesis that geothermal regions may have played a more central role in the survival and persistence of endemic Antarctic biota. Perhaps the central issue with correlating biotic communities with the presence of geothermal heat is in determining the ‘sphere of influence’ of its environmental effects. Fraser et al. (2014, p. 5637) noted that, ‘Geothermal heating from volcanic activity is generally limited to within a few kilometers of the summit crater or caldera.’ However, their model (Fraser et al., 2014) extended these ‘few kilometers’ to a 100-km radius by creating a ‘buffer zone’ around each site. This 100-km sample unit is their primary model parameter. It may have inflated the apparent fit of the model to the data. Of course, the extent to which such a parameter is arbitrary may ultimately be overshadowed by the recognition of the sheer complexity of the question. For example, sample bias (diversity measures rely largely on the ad hoc history of Antarctic biological survey efforts) is a major stumbling block. In particular, the exclusion of inland volcanoes due to insufficient data makes it difficult to be certain whether correlations between geothermal sites and diversity are not just evidence of more benign conditions at marine and coastal sites. Fraser et al.'s (2014) paper starts the process of narrowing in on what is already essentially a dataset of continental proportions. The identification and delineation of a geothermal ‘sphere of influence’ at biologically relevant spatial scales will be a paramount requirement for increasing both the resolution and ground-truth of future research. Quantitative data are also needed to connect geography to the geophysics of thermal propagation and retention through Antarctic soil and rock. From a biological perspective, the most significant regions of effect might be expected to be at the peripheries of geothermal activity where amelioration of environmental conditions (as opposed to just warmth per se) might permit persistence of terrestrial biota through the abiotic extremes of glaciation. A peripheral effect would also be most consistent with the predominance of cryophilic physiologies on the continent. Another question raised by Fraser et al. (2014) is the application of species diversity as a proxy for refugia. If species richness is the product of a ‘survivor’ effect, in which refugia have accrued the greatest diversity through temporal accretion and preservation, then diversity is a transparent signature of refugia. However, this may not be a straightforward question. At the regional scale, Maritime Antarctica is considerably more species rich than Continental Antarctica – indeed, its environmental conditions have been considered to be sufficiently different from higher latitude Antarctica for it to merit the distinct biogeographical label ‘Maritime’. To their credit, this is largely recognized in Fraser et al.'s (2014) models. But at landscape scales, younger, non-refuge sites exposed after glaciation may also accrue species due to a number of factors. Coastal sites, for example, typically accrue high nutrient inputs through biotic interactions from vertebrates (Convey et al., 2008). Climate and microclimates on the coast are also milder as a result of the heat capacity of the ocean. The ice-free refugia hypothesis states that refugia were areas of Antarctica that received reduced ice cover during the LGM. These include nunataks, polar deserts, ablation valleys and other areas characterized by their exposure and relative lack of snow cover as a result of either geographical elevation and/or katabatic winds (Convey & McInnes, 2005; Nolan et al., 2006; Convey et al., 2008; Storey et al., 2010; Magalhães et al., 2012; Fraser et al., 2014). There is no doubt that they are refugia – because they are still refugia for many species (Stevens & Hogg, 2003; Convey & McInnes, 2005; Convey et al., 2008). However, the absence of coastal and other endemic species at such sites suggests that these sites were not the only refuges of terrestrial Antarctica (Convey et al., 2008). Many species on nunataks, in particular, are like biological relicts: unique and isolated remains of species and communities now all but extinct. Linking the geology of these sites with their biology is an ongoing process – but it is only by a combination of molecular and geological dating processes that a picture of sufficiently high resolution to accommodate landscape- (and perhaps ultimately microhabitat-) scale process and structure will emerge (Nolan et al., 2006; Storey et al., 2010). The question of whether biotic diversity is a geo-historical signature of habitat antiquity or an ecological signature of Holocene environmental conditions, applies equally to the ice-free refugia hypothesis. The examination of ecological succession in micro- and macrobiotic terrestrial communities may provide some useful insights. For example, a detailed multilayered examination of diversity gradients in the Transantarctic Mountains by Magalhães et al. (2012) indicated that more complex communities are associated with newer, more recently exposed soils – i.e. diversity follows ice retreat. These findings tell a complex story of soil physicochemistry in which increasing chemical stress levels (salt accumulation) at older sites mean that newer sites are more habitable (Magalhães et al., 2012). Further work on successional processes in other Antarctic geological contexts is certainly needed to examine both the specificities and generalities of community development. The small proportion of Antarctica's endemic biota that was not wiped out by the LGM probably survived via the occupation of a range of types of refugia. Where some nunataks may be museums of all but extinct species, other sites may be biological hotspots. Our understanding of the mechanisms of LGM survival on the Antarctic continent continues to develop, but many questions remain. For example, could some pockets of life have persisted through time and changing environmental conditions via a succession of temporary refugia? Certainly, investigations into the character and diversity of these refugia will continue to inform Antarctic biology and geology for many years to come. The comments of three anonymous referees greatly improved the manuscript.