Geoscience Reference
In-Depth Information
the last 10 Ma. The desiccation evident in the transition from semi-arid to hyper-arid
conditions probably arose when the central Andes had attained an elevation of at
least 2 km during the middle Miocene, thereby preventing moist air masses from the
summer monsoon from reaching the Atacama.
To sum up, the Cenozoic uplift of the Andes took place in stages, with long
intervals of stability punctuated by shorter intervals of relatively rapid uplift. Major
uplift during the Miocene resulted in the creation of an effective rain shadow on the
western margin of the Andes. Desiccation of the Atacama was underway by around
25 Ma, with extreme aridity setting in between 19 and 13 Ma, once the central Andes
had reached an elevation of 1,500-2,000 m. It is possible that the Atacama has been
relatively dry for the past 150 Ma. This is because it is located in the subtropical zone
of dry subsiding air and has undergone very little latitudinal tectonic displacement in
that long interval of time.
Determining the onset of Cenozoic cooling in South America is more problematic
than simply reconstructing the history of uplift and desiccation. There was a major
reorganisation of the plankton ecosystem in the Southern Ocean at the start of the
first major Antarctic glaciation in the earliest Oligocene around 33.6 Ma ago (Houben
et al.,
2013
). However, in southern Argentina, isotopic analysis of fossil mammal
teeth revealed no change in temperature across the Eocene-Oligocene transition,
prompting Kohn et al. (
2004
) to propose that the widely held concept of cooling at
this time (see
Chapter 3
) might be in error. Plant macrofossils from Patagonia certainly
show very high plant diversity in the early Eocene around 50 Ma ago (Wilf et al.,
2003
). Nevertheless, the fossil pollen from central Colombia and western Venezuela
do indicate a decline in plant diversity at the end of the Eocene and in the early
Oligocene (Jaramillo et al.,
2006
). Because plant diversity fluctuates globally with
temperature, this decline may indicate cooling, or it may indicate a change in the
area available for tropical plants to inhabit, which is a function of both temperature
and precipitation. Markgraf et al. (
1995
, p. 143) state categorically that 'by the late
Miocene (10 million years ago), cooling had set in, and by Pliocene times, climatic
cycles had become increasingly marked'. This generalisation is supported by the
detailed pollen analysis of van der Hammen and Hooghiemstra (
2000
) in Amazonia
and by that of Helmens and van der Hammen (
1994
), Hooghiemstra and Ran (
1994
),
van der Hammen and Hooghiemstra (
1995
) and Wijninga et al. (
2003
) in the High
Plain (Altiplano de Bogota) of Colombia.
21.5 Quaternary climatic fluctuations
In common with the rest of the desert world, the last 2.5 Ma of climatic history
in arid and semi-arid South America were characterised by rapid fluctuations in
temperature and precipitation linked to the glacial-interglacial cycles of the Qua-
ternary (see
Chapter 3
). In the late Pliocene, the climatic cycles were primarily the
