Climate change and faunal dynamics in the uttermost part of the earth



Bruce D. Patterson, E-mail:


To use the ‘lessons of the Pleistocene’ to forecast the biotic effects of climate change, we must parse the effects of history and ecology in the Quaternary record. The preponderance of Northern Hemisphere studies of biotic responses to climate change provides a limited set of players and environmental circumstances with which to decouple these drivers. In this issue Lessa et al. (2010) examine population structure in 14 species of mice distributed across Patagonia and Tierra del Fuego in southern South America. In the Southern Cone, glacial ice was alpine, not polar; major habitats were (and are) oriented N–S, not E–W; and habitable land area actually increased, not decreased, at the height of the last glacial maximum (LGM). Despite these differences, there is evidence for poleward demographic expansion in 10 of the 14 species, and phylogeographic breaks in these are likewise stepped by latitude (and presumably history) rather than by biome. Nevertheless, high latitude endemism and the antiquity of these lineages point to an extended presence in the region that very likely predates the Pleistocene.

Global warming looms as one of the most pressing and urgent environmental challenges of the 21st century. To predict the effects of rapid climate change on plant and animal populations, one would be hard-pressed to identify a better parallel than the events following the last glacial maximum (LGM). Over the last century, hundreds of research groups have focused on biotic responses to this progressive warming period, especially in North America and Europe. Except for the present, the Quaternary of the Northern Hemisphere is surely the best-studied period of geological time. A common genetic pattern of the post-glacial period is evidence for demographic expansion towards the pole from ice-free refugia at low latitudes (Mila et al. 2000; Stone et al. 2002; Starkey et al. 2003), although many others can be cited (e.g. Rowe et al. 2004).

How much of this pattern is historical, dependent on the polar location of northern ice sheets and their episodic latitudinal growth and recession? How is this pattern influenced by the current distribution of life zones, which are usually zoned latitudinally, as Merriam originally envisioned them? How was it affected by the inter-continental biotic interchanges that swept across Beringia during sea-level recessions? Do commonalities between studies in the Nearctic and Palearctic reflect their shared history and ecology as subregions of Holarctica? In this issue, Enrique Lessa, Guillermo D’Elía, and Ulyses Pardiñas examine the genetic structure of rodent populations in the Southern Cone of South America, which offers an interesting and sometimes contrasting perspective on these questions (Lessa et al. 2010).

The novelty and importance of their analysis hinges on South America being the only Southern Hemisphere landmass save Antarctica that extends beyond 40°S, yet South America is separated from Antarctica’s huge polar ice cap by insulating circumpolar currents. Quaternary glaciation in South America was and is strictly an Andean affair. And while glaciers still remain at higher elevations and latitudes, they did not cover extensive lowland areas in the Southern Cone. Even at the LGM, ice sheets were more-or-less restricted to the cordillera itself (Fig. 1). The Andes have been produced by subduction of Pacific Ocean plates moving eastward beneath the South American plate, causing this mountain range to hug the western margins of the continent. With prevailing westerlies at temperate latitudes, this north-south orientation of the Andes creates temperate and sub-antarctic rainforests on the Pacific slopes but a dramatic rain shadow in Patagonia, which lessens in severity with increasing distance from the cordillera and increasing latitude. Major floristic associations in the Southern Cone are thus oriented in north-south strips, rather than the latitudinal bands they assume in most regions. The climatic distinctions between windward and leeward locations declines towards Cape Horn, where polar influences are strong enough to extend glaciers to sea level, as along the Beagle Canal.

Figure 1.

 Extent of Patagonian ice cap at LGM and 150-m contour of the continental shelf, which would have been exposed as dry land by LGM sea-level drops (modified from Fig. 9.5 of Heusser 2003). Land cover data from ESA Globcover Project (2004 data).

During glacial episodes of the Pleistocene, sea levels worldwide dropped up to 150 m owing to eustatic and isostatic changes, draining continental shelves and producing dramatic changes to some continental coastlines. Although the Pacific margins of South America hardly changed at all, lowered sea-levels during the LGM exposed an Atlantic shelf area exceeding the maximum area covered by the Patagonian ice cap (Heusser 2003). The area of non-glaciated terrestrial habitats in Patagonia may actually have grown during the LGM (Fig. 1).

To probe the effects of these landscape and climate changes, Lessa and colleagues sampled an 801-bp fragment of cyt-b from a diverse array of cricetid rodents (Fig. 2), including all but a few of the species present. Like many small mammals, these mice tend to be stenotopic, with narrow environmental tolerances and restricted geographic distributions, making them excellent markers for biogeographic studies. The study’s sampling scheme, totalling 318 mice, covered all major areas of Patagonia and Tierra del Fuego, including the Argentine monte, Patagonian steppe and grasslands, and Valdivian and Magellanic forests.

Figure 2.

Euneomys sp. from the Somuncura plateau in Patagonia. The genus Euneomys is endemic to the Southern Cone and apparently arose there; it cannot be assigned to any of the nine tribes (groupings of genera) that encompass sigmodontine diversity (D’Elía 2003; D’Elía et al. 2007). Lessa et al. (2010) show that the genus exhibits phylogeographic breaks that document its extended history in the Patagonian–Fuegian region (photo by Ulyses Pardiñas).

Their results show that most species in this region are represented by a single lineage showing evidence for demographic expansion, as expected under a model of post-glacial expansion from a refugium. However, several models suggest that demographic expansion in all but one of the species took place well before the LGM. Glacial episodes of the Pleistocene certainly relocated fauna and flora in Patagonia, as they did in Amazonia and Beringia, but they didn’t trigger the diversification that early refugial models attributed to them—major splits in clades in both North and South America predate their isolation during the LGM (see also Klicka & Zink 1999; Rull 2008). Interestingly, cricetid species showing significant subdivision in Patagonia are split north-to-south rather than east-to-west (the principal orientation of plant formations) suggesting the predominance of historical signal over ecological determinism.

Four species of mice showed multiple, well-differentiated phylogeographic units in the Patagonian–Fueguian region. One of these is Euneomys (Fig. 2), a open-country endemic to the Southern Cone whose phylogenetic affinities to the other 82 genera of living sigmodontines are remote enough to be unclear (D’Elía et al. 2007). The other three all belong to the Abrotrichini, a tribe of five genera found in southern South America and probably autochthonous there. It is curious that these basal divergences in Sigmodontinae should exist (or persist) in Patagonia and Tierra del Fuego, because these mice colonized South America from Central America in the vanguard of the Great American Biotic Interchange only 5–8 Ma (Reig 1981; Steppan et al. 2004). That the farthest corner of the continent should support such early divergences testifies to the region’s distinctive evolutionary history. For example, despite its continental connections, the Valdivian Forests of Chile and Argentina support a rodent and marsupial fauna whose endemism rivals that of insular New Guinea (Patterson 1992).

The Lessa et al. paper typifies a new era in biological studies in South America, which have taken predictable but quantum leaps forward. Seventeenth and 18th century knowledge of this region’s mammal fauna was limited to the anecdotes of resident naturalists (often Jesuit missionaries), including Azara, Molina, Gay, and Poeppig. The early 19th century witnessed extensive, systematic surveys by European scientists, the most notable being the famous voyages of Charles Darwin and Alcide d’Orbigny. A century later, the first systematic monographs appeared, often written by North Americans or Europeans because of the resource-intensive nature of systematic research; however, these foundations enabled more focused biological investigations that increasingly were conducted by South American scientists (Patterson 2001). Now, multinational teams of Latin Americans are tackling general biological questions using comparative analyses and the latest conceptual models. The unique geography and history of the Southern Cone represent a unique ecological theatre where evolutionary plays may take novel, unscripted turns. It is an area that will doubtless illuminate many other general questions.