Climate changes associated with the Last Glacial Maximum (LGM, 21 kyr) are probably the most extreme that terrestrial vegetation, including tropical lowland ecosystems, have been forced to respond in over the past 100,000 years. The degree of tropical cooling can be reconstructed by paleoproxies that generally indicate a minimal cooling of 3°C and a maximum cooling of approximately 7°C (Guilderson et al., 1994; Stute et al., 1995; Mix et al., 1999; Behling and Negrelle, 2001; Mora and Pratt, 2001; Behling, 2002; Urrego et al., 2005). Some debate surrounds the degree of tropical decreases in glacial precipitation, primarily because precipitation patterns are strongly regional, and thus wide discrepancies in paleoprecipitation trends occur between different reconstructions. Despite this, a value of approximately 20% decrease in LGM rainfall is typically reconstructed from pollen-proxies in tropical catchments—such as the Amazonian Basin (Bush and Silman, 2004). Paleoclimate simulations of the South American monsoon during the LGM indicate an annual reduction in rainfall across Amazonia of between 25-35% relative to today (Cook and Vizy, 2006). Research also shows that glacial decreases in rainfall likely occurred in wet as opposed to dry seasonal months (Bush and Silman, 2004).

Less ambiguous with respect to paleoclimate reconstruction is the decline in atmospheric C02 that occurred at the LGM. Direct measurement of C02 gas trapped in Antarctic and Greenland polar ice provides us with a record of C02 for the past 420,000 years (Indermuhle et al., 1999; Monnin et al., 2001). Ice core studies indicate that atmospheric C02 at the LGM was on average 200 parts per million by volume (p.p.m.V) (relative to modern day values of >380p.p.m.V), indicating an over 40% reduction in atmospheric levels during glacial periods.

A combination of climate cooling, decreasing precipitation, and low atmospheric C02 surely promoted changes in equatorial vegetation form and function, but in what way and to what extent is still a matter of discussion. There are far fewer palynological sites located in tropical regions relative to those in temperate North America and Eurasia, simply due to geological history. Discovery of new coring sites with good paleoecological reconstruction potential, however, is increasing (Bush, 2002; Mayle et al., 2000).

As a result of the current shortage of tropical palynological records, dynamic global vegetation models (DGVMs) play a key role in filling knowledge gaps where no tropical pollen profiles currently exist (Harrison and Prentice, 2003; Cowling et al., 2004). Vegetation models can also be important for elucidating the underlying mechanisms of vegetation change because models are built upon fundamental principles of plant physiology, biochemistry, and ecosystem ecology. Models differ in the way they emphasize or parameterize particular processes, but they address fundamental biophysical mechanisms nonetheless.

Research effort towards better understanding past vegetation changes is essential, for if we can't explain ecological changes that we know to have occurred within past climates, then how can we place confidence on our estimations about ecological responses to future changes in climate? By outlining knowledge on how plants physiologically and biochemically respond to different abiotic stresses, we can begin to hypothesize about how lowland vegetation may have looked in the past, and how it might be influenced in the future. Equipped with this knowledge, we are able to make more informed decisions concerning aspects of tropical conservation, as well as developing mitigation strategies before a time when unwanted ecological changes may occur.

In this chapter I introduce some of the biochemical and physiological plant processes (mechanisms) that are important in terms of ecosystem-level responses to climate change, focusing on those most influenced by Pleistocene climate. Due to the ability of models to deconvolve co-varying responses, I address the independent versus interactive effects of Pleistocene precipitation, temperature, and atmospheric C02 on lowland plant ecology. Whether or not C4 plants (mostly subtropical and tropical grasslands) experienced widespread proliferation during glacials will be a topic of discussion, one that includes recent research highlighting the potential for over-prediction of C4 plant abundance due to caveats associated with stable carbon isotope analyses. The latter sections of this review will contain more speculative discussions of the stratification of tropical lowland forests based on the results of different modeling experiments, and of the possible response of tropical soil processes to Pleistocene climate change.

0 0

Post a comment