‘Tipping points’ for the Amazon forest

The stability of the Amazon forest–climate equilibrium is being perturbed by a number of human drivers of change (e.g. deforestation, global warming, forest fires, higher CO₂ concentrations, and increased frequency of droughts and floods). Quantitative assessments for the maintenance of the tropical forest indicate that ‘tipping points’ may exist for total deforested area (>40%) and for global warming (ΔT > 3–4°C). The likelihood of exceeding a tipping point can be greatly exacerbated by increases in forest fires and droughts, but quantification of those effects is still lacking. Forest resilience can be significantly increased if CO₂ ‘fertilization’ effect is proven to be taking place for tropical forests, but it can be offset by continued increases in temperature, rainfall seasonality, and forest fires.

Introduction

The critical importance of the Amazon forest for the maintenance of tropical biodiversity (e.g. [1]), as a significant carbon pool (e.g. [2]) and for local and regional climate stability [3] has been recognized for some time. The functioning of the Amazon as an undisturbed regional entity can also be seen as providing key ecosystems services [4^••] and the most obvious one is its role as a large carbon pool and even as carbon sink, though the latter is not firmly established as yet [5]. However, anthropogenic disturbances of many varieties are all that tropical forests have been subjected to in the last half century. The tropical forests of South America are no exception and are under the increasing influence of a suite of human drivers of environmental change, mostly related to unprecedented rates of land cover change for the last 30 years or longer [6]. The main human drivers of change for the Amazon are forest clearing (clear-cut deforestation), forest degradation and fragmentation, climate change associated to global warming, increased atmospheric concentration of CO₂, forest fires, and potential increases of climate extremes (e.g. droughts). They are all interconnected in complex ways.

A scientific question that has been gaining key importance in recent years is the evaluation of the so-called critical or ‘tipping’ points of the Earth system [7], aimed at quantitatively establishing the likelihood of crossing a threshold that could cause an element of the Earth system to jump to another stable equilibrium. The climate–vegetation equilibrium in the Amazon has been identified as one such tipping point of the Earth system [7] possibly presenting bi-stability [8]: one equilibrium state is obviously the present climate–vegetation state with tropical forests covering most of the Basin and a second stable state would have tropical savannas (or other type of drought-adapted and fire-adapted vegetation) replacing forests in a large portions of the Basin. Once a tipping point is transgressed, the time scale for fully reaching the new equilibrium state can be ‘abrupt’ in comparison to natural time scales of change, but still may take several decades to a century for the establishment of the new vegetation–climate state [3, 7].

In the following sections, we will review current knowledge on the human drivers of change and the impact they are already causing or are projected to cause in the Amazon system with a view toward identifying tipping points of that system that would lead to irreversible changes in the functioning of the tropical forest. Such changes can reach an extent as to affect greatly the capacity of the Amazon forest to render ecosystems services [4••, 9, 10••, 11••, 12••], or other drivers which might be somewhat attenuated by increased resilience of the forest because of CO₂ effects [13].

In the context of this paper, ‘savannization’ has been defined as an environmental change in tropical South America that would lead to changes in the regional climate because of either land cover change [3, 10••] or global warming [11^••] in such way as increase the length of the dry season and turn the regional climate into the typical climatic envelope of savannas. It is a statement on regional climate change and not on the ecological processes of forest replacement. On the other hand, ‘secondarization’ refers to subjecting the forest to clear-cutting and then to a repeated cycle of secondary growth, clearing and fire penetration, that leads to a significantly impoverished form of secondary growth [14^••]. Secondarization can happen in the absence of savannization and vice versa. However, the most likely scenario is that both processes would be taking place simultaneously, operating at multiple spatial and temporal scales, and strongly interacting: large-scale drivers such as basin-scale deforestation and global warming might change the Amazon climate toward a savanna-like climate with longer dry seasons. Increased forest fragmentation and degradation, coupled to more frequent penetration of forest and secondary growth fires, and higher frequency of intense droughts (the latter because of global warming), on the other hand, would act locally to reduce the resilience of the forest and induce its replacement by more fire-tolerant, typical savanna species and dominated by grasses [14••, 15••]. Other drivers of change might increase resilience of the forest. Uncertainty of changes of the hydrological cycle because of global warming prevent conclusive statements on the increase of the dry season; global warming might likely increase not only the frequency of droughts but also the frequency of floods that is an acceleration of the hydrological cycle. The severe drought in 2005 over western–southwestern Amazon [16] was followed by floods only six months later [17]. Above-normal rainfall all over the Amazon in the first five months of 2009 has led to record-breaking floods of the Amazon River and some of the main tributaries [18]. Plentiful rainfall can in principle increase forest resilience. Lastly, atmospheric [CO₂] increases has been hypothesized as being contributing to a beneficial effect on plant productivity (the so-called CO₂ ‘fertilization’ effect) underpinning forest inventory measurements of net carbon uptake by the Amazon forest [19] over the last decade. If that positive effect continues into the future [13], it may counteract to some extent the negative impacts of deforestation, global warming, and enhanced fires.