Concluding post: land cover change in South America

Dear Readers! 

I hope you have enjoyed this blog as much as I have enjoyed writing it!

South America is a large and diverse continent - and for the purpose of this blog I have mostly concentrated on its northern half, especially the Amazon area and immediate surrounding. I chose this focus due to the global significance of the Amazon rainforest in its share of terrestrial biodiversity, role in the hydrological and carbon cycle and global climate circulations. 

Through this blog, I tried to highlight the current land cover of South America and the pressures that exist on its conversion or amendment for human purposes. Posts discussed different ways in which forest resilience is interrupted or reduced, followed by a detailed assessment of the wider effects of land cover change with a special focus on the most dominant mode: rainforest to pasture conversion. The effects covered were the quite obvious reduction in biodiveristy, the two-fold impact on renewable surface water formation, the complexity of climatic effects (local, regional and global) and the role of forests and deforestation in the carbon cycle.

It is important to assess the way anthropogenic land-cover change is caused, reinforces and also interacts with the effects of global climatic change. Consequences are severe for both natural ecosystem health and the human populations within the suggested area - if I would continue this blog further, I would expand on the socio-economic perspective as well.

I believe that this blog topic has helped me manifest my understanding of the complex manner in which humans are morphing the planet, and has put this knowledge into the context of the 9 planetary boundaries. As shown throughout my blogging, quite interestingly, a spatially limited phenomenon such as land cover change in South America impacts the situation of the whole planet not only within the boundary of 'land-system change' but also in others. Those covered here were mainly influencing 'biogeochemical flows', 'biosphere integrity', 'climate change' and 'atmospheric aerosol loading'. Thus my topic felt well-chosen to highlight the tight interactions of entities within the Earth system contributing to 'Global Environmental Change'. 

(source)

The inclusion of land cover change in future climate projections

Global circulation models all, but to a different extent, include land cover change scenarios when predicting future climates. Feddema et al (2005) analysed how important it is to include these for the outcome of the modelling.

The study shows that the way human will choose to alter the land cover in the future will have important effects on the real impact of climate change on the Earth. While on a global scale, land cover changes are expected to even out in their effects on global mean temperature and the hydrological cycle. The biogeochemical and biogeophysical effects of land cover change will, however, have modulating effects on the regional scale response to climate change. This supports my analysis of the effect of land cover change on climate in my previous posts. 

The characteristics of expected future land cover change is shown in the mapping of the IPCC SRES land-cover projections (Figure). We can see that significant changes are expected to occur in South America in all scenarios by 2050, mostly focused on Western Amazonia, South-Eastern Brazil and the southern tip of the continent. The most common change is the conversion to cool/warm grass land for agricultural purposes and livestock farming. The expected drying of the continent is reflected in the spreading of savannah biomes. What I find interesting is that in the B1 emission scenario, little change is expected in the central Amazon by 2050 and 2100 (see the boxed region on South America) - this may suggest that the rainforest conservation strategies implemented (both national and international along the lines of REDD) are expected to be effective and/or that the rainforest is able to withstand the adverse effects of global climate change.

Figure, (Source)

A new way to make CO2 emission estimates from deforestation more reliable - a step forward for policy implementation?

There are limited studies that exist detailing the emissions arising from a certain land cover changes. So the IPCC shows in their "land-use change" category of SRES category desciptions, that all estimates of emissions from land-use change are calculated based on the obtainable deforestation and afforestation figures. Further, these are based on estimates from the expected carbon storage of a set area of forest. 
In 2012, Harris et al produced a baseline map of carbon emissions in tropical regions, using satellite imagery. Pairing deforested regions of interest with the carbon stock before clearing (using base map) can help with the evaluation of CO2 emissions and the reaching of conservation targets.



The map above is taken from the publication. "Distribution of annual carbon emissions from gross forest cover loss between 2000 and 2005 mapped at a spatial resolution of 18.5 km."

Despite the lowering of deforestation rates (highlighted in my first posts) in South America, it is still high to the extent of emitting 5-200 Gg of Carbon per year between 2000-05. Of course the higher end of this spectrum is only realistic for a few Amazonian regions but should still highlight the importance of tackling this problem. If deforestation effects on the local biodiversity and climate are not enough to power strict management policy implementation, maybe the visualisation of CO2 emissions is!

The uncertain, but changing, potential of the terrestrial biosphere as a future carbon sink

I am aware that the studies used in this post are not directly linked to the South American continent as such but I consider it interesting to cover the topics due to the large proportion of the global terrestrial carbon sink lying within South America.

As described in my previous post, forests have and had the potential to store large amounts of carbon. In fact many studies now agree (e.g. A, B,) that carbon uptake by natural sinks has increased over time. The reason for this is a physiological vegetation response to CO2 fertilization in the atmosphere, that increases plants' primary productivity and thus increases the carbon stored per hectare of intact forest. Ballantyne et al (2012) use a global scale CO2 mass balance analysis to show the evolution of the carbon budget 1959-2010. 

(source)

While fossil fuel emissions have been pumping CO2 in the atmosphere at an alarming rate, there is a clear increase in CO2 from the atmosphere to global sinks over the same course of time. It mirrors the atmospheric concentration - it is a dynamic response, which means that "terrestrial ecosystem carbon fluxes both respond to and strongly influence the atmospheric CO2 increase and climate change". In a modelling study by Cao and Woodward (1998), we can see that we are still in a period in which the CO2 fertilization effect is strong enough to uphold a net positive relationship between emissions and sinks, however in the future as the effect becomes saturated a leveling off of potential terrestrial sinks is expected. Cox et al (2000) even suspect that post 2050 climate change will have such altering effects that terrestrial sinks will turn to sources (and this projection does not even take anthropogenic deforestation into account yet but is purely based on climate change effects!)

Now going back to land cover change directly: a carbon dioxide removal (CDR) technique of geoengineering is the afforestation of previously converted landscapes. This way, carbon is sequestrated from the atmosphere and stored in the new-grown forests. Fellow blogger Maria Christofi has reviewed the potential for this method in elevating dangerous levels of CO2 in the atmosphere at: http://geoengineeringinquiries.blogspot.co.uk/. A question that arises to me now is whether the negative feedback loop that has been described by Cox et al (2000) renders this method useless in the future? The uncertainty regarding the vegetation-climate responses of the future definitely adds to the factors needed to consider before intervening further with land-cover changes.  

Forests, in their role as massive carbon stores

Besides being an integral part to the climate system, Forests (especially thick tropical rainforests) are a large storage of organic carbon. The carbon is taken up by terrestrial “sinks” through plant growth/primary productivity and stored within their structures, as well as in the soils underneath. Looking at the world scale, out of all carbon stored in terrestrial ecosystems, about 55% is stored in tropical forests which is at a density of 242 MG C per hectare, most of which is stored in the biomass. Lands are said to become net emitters when ecological disturbances or deforestation cause a reduction in primary productivity. Deforestation or land-cover conversion to less productive ecosystem will result in the emission of CO2 to the atmosphere via forest burning (prominent in the Amazon in the land conversion to pasture), decomposition of harvested matter, fuel-wood burning and/or soil respiration (IPCC.www).

Here I want to present a map that shows in overview the carbon sinks and sources (Pg C per year) over two time periods arising from global forests - and I will focus on South America. 
(Downward facing bars indicate storage, upward facing indicate emission)

Source: click here

We can see that the South American continent plays a large role in terrestrial carbon fluxes. What I found interesting here is that in the period 2000-2007 the net flux is still a sink but at a lower value compared to the previous period 1990-1999 (-23%) despite reduced emissions from gross deforestation. And to understand this, my topic of land-cover change is highly important. Deforestation emits CO2 almost directly as described above, but it doesn't stop there: the area left as pasture (for example) starts to act as a source again as the grasses grow, but at a much lower productivity. Therefore, the fact that the South American total area of intact tropical forest is reduced over time is in itself a loss in 'potential carbon sink' and contributed to weaken the continents total "sink" characteristic.

It is estimated that about half of anthropogenic CO2 emissions are taken up by natural sinks like the oceans and terrestrial ecosystems, but this hasn't been stable through time. A widely cited paper by Schimel et al (2001) shows that terrestrial systems have only been a net sink since 1990, powered by land use changes that supported regrowth and nutrient fertilization that enhanced primary productivity. Can these changes power a further increase in carbon uptake in the future? I quickly became aware through my reading that scientific opinions and modelling studies on the future of forests as carbon sinks is highly divided and uncertain. The straight-forward conceptualisation of "more forest- more carbon stored" is not completely accurate. It surely depends on the health of the forest (relates to my posts on forest degradation and edge effects), climate-vegetation feedbacks, and ecological changes that occur within the forests in response to climate pressure (e.g. Amazon drought 2005 shows that this is NOT negligible by turning Amazon from sink to a source through the loss of biomass (Philipps et al 2009)).

A conceptual model linking causes and effects of disturbance/change in the Amazon basin

I had read the article "The Amazon basin in transition" a while back - and now that my blogging is soon coming to an end in a couple of weeks, I think it is useful to post this graphic produced by Davidson et al (2012). It shows the interactions between global climate, land use, fire, hydrology, ecology and human dimensions within the Amazon basin.

Please have a look and see how my blog posts fit to (or are controversial to!) this conceptual net of linkages: 

Burning biomass and emitting aerosols in smoke - what are the links to climate parameters?

Biomass burning, whether to fuel settlements (as discussed in post 17/11/15) or as the cause of slash-and-burn land clearing within the Amazon, elevates atmospheric aerosol levels (besides of course emitting CO2...). The map below shows the "smoke aerosol distribution (D < 2.5 μm; in μg m–2) and wind field in the BL over South America during the transect flights from Rondonia to the western Amazon" (Andeae et al, 2004).

These increased local aerosols (aerosol optical depth) within the Amazon basin in biomass-burning season, have shown to be correlated to increased cloud cover, cloud height and increased rainfall by Lin et al (2006). 

Another study, by Koren et al (2004), concluded from satellite imagery that heavy smoke over the Amazon basin reduces cumulus cloud formation by up to 37% compared to zero-smoke. 

Andeae et al (2004) has a different take at examining the "Smoking of rain clouds" - they show that cloud droplet size is reduced, delaying precipitation rain-out at lower levels. This means, that the aerosols and particulates from the biomass burning can travel longer distances in the upper cloud formations and thus spread the effect over larger areas. What I found interesting here, is that they suggest that with this delayed precipitation, cloud cover increased in height (consistent with Lin et al 2006) and tops may overshoot into the stratosphere. With the "smoke" and water vapour now becoming part of stratosphere, these very locally sourced pollutants can have an impact on large scale circulation patterns, and especially partially affect the radiative properties of the Earth. Due to the complexity of these mechanisms and factors involved, the extent of this effect cannot yet be accurately modelled, but the probability to influence the global climate circulations has been suggested to be high.

As important as this mechanism may be, Roberts et al (2003) analysed this too but concluded that the dominant factor influencing a change to the hydrological system in the Amazon remains the dominant land-use change from forest to pasture.

Thoughts: 

We can see that there are different conclusions drawn from the examination of how the smoke of burning biomass in the Amazon affects the local and even global climate parameters. One thing is for certain - the affect on cloud cover (whether it is the increased height, reduced droplet size or even reduced/increased cloud-cover all together) makes up a forcing factor that alters the natural climate. With the alterations to the natural climate induced by the patterns of land-cover change, this added forcing is another complication for the ecosystems and populations to react to.