The Amazon since the Last Glacial Maximum

For this post I read a very interesting article published in 2004 called “Responses of Amazonian ecosystems to climatic and atmospheric carbon dioxide changes since the last glacial maximum” by Mayle et al., published by The Royal Society. In it, the authors do not do any original data collection, but instead choose to analyze several already existing papers and data in order to try and paint a more comprehensive picture regarding Amazonian ecosystems since the last glacial maximum. They also go one step further and use these findings to try and predict what could be in store for the near future of the Amazonian Rainforest.

As mentioned in a previous post, the authors disagree with the possibility of the rainforest having been reduced to separate islands, as “both fossil pollen data and dynamic, process-based vegetation simulations show that most of the Amazon Basin remained forested at the LGM”. As a whole, the forest expanded after the LGM, retracted in the early to mid-Holocene due to decreased precipitation, increase in temperature, and lower carbon dioxide levels in the atmosphere. Finally, in the late Holocene the forest began to expand once again, until significant human effects kicked in and we began to make the forest shrink once again.

Figure 1. (a) Map showing the location of sites discussed in the text. The shaded area shows the current distribution of humid evergreen broad-leaf forest (rainforest). The hatched area shows the Andes mountains. Lowland unshaded areas represent seasonally dry forests and savannahs. (b) Schematic broad-scale summary trends of climatic change and/or vegetation response for each site since 21 000 cal yr BP.

The most noticeable changes, in terms of vegetation replacement such as rainforest by savannah, would have taken place in ecotonal regions. Therefore, while the Amazon Rainforest as an integrated body was most likely intact, the boundaries of the forest did change. Not only between the rainforest and savannah, but also between the rainforest and Andean cloud forests.

However, what I found to be by far the most interesting proposition of this paper is that perhaps the savannah vs. forest replacement concept is flawed. Nowadays, in the east of Brazil there are Caatinga forests, classified as seasonally dry forests. So while there seems to be a consensus that the Amazon was never replaced by savannah since the LGM, parts of the forest could have become seasonally dry. The authors of this paper mention widely-separated semi-deciduous/deciduous forests with common species, namely the Caatinga forests in the east of Brazil, Chiquitano dry forest of eastern Bolivia and the Andean piedmont dry forest in south Bolivia/northwest Argentina.

The pollen records that were always assumed to indicate the presence of a Rainforest, could just as easily indicate a primarily dry forest region. However, the similarity of these signals brings forth this possibility, but also makes it very difficult to determine the case one way or another. The authors mention efforts towards this goal, but no definitive answer has been reached. At least from what I have read, this hypothesis is at the very least plausible, not to mention very interesting.

Figure 3. SDGVM model simulations forced with the UGAMP GCM. See § 2d(i). Key: c3, C3 grasses; c4, C4 grasses; ebl, evergreen broad-leaf forest; enl, evergreen needle-leaf forest; dbl, deciduous broad-leaf forest; dnl, deciduous needle-leaf forest. Amazonian vegetation distribution: (a) Pre-Industrial; (b) Mid-Holocene; (c) LGM; NPP of Amazonian vegetation: (d) Pre-Industrial; (e) Mid-Holocene; ( f ) LGM. (t C ha_1 yr_1, tonnes of carbon per hectare per year, where 1 hectare is 104 m2.)

After analyzing a wide array of data, the paper then goes on to propose what might happen to the Amazon Rainforest in the years to come. Since there is likely to be a rise in temperature and decrease in precipitation, both likely to be aggravated by deforestation, proxy data suggests there will be more frequent wide-spread fires, leading to a replacement of current vegetation to more drought and fire tolerant vegetation, typical of savannah-type ecosystems. However, the big wild card is carbon dioxide levels, since there is nothing in the past to compare them to, especially if they reach the projected levels of twice those of the mid Holocene by the year 2050. I do not have a tremendous amount of faith in the ability of past records to accurately predict what is to come, simply because I firmly believe that, given how different the world has become since the Industrial Age, there are more factors than carbon dioxide levels which make the future difficult to model based on proxy data. One example of this, which I have discussed in previous posts, is the increasing seasonality of climate in the Amazon, which is not the same as average temperature and precipitation changes over time, considering there can be extreme variability between seasons and yet no change to the observed average yearly values.

Sources: Responses of Amazonian ecosystems to climatic and atmospheric carbon dioxide changes since the last glacial maximum, Francis E. Mayle, David J. Beerling, William D. Gosling, Mark B. Bush, Phil. Trans. R. Soc. Lond. B March 29, 2004 359:499-514; doi:10.1098/rstb.2003.1434