Fire, global CO2 and the management of savannah woodlands

When considering the potential impacts of climate change on vegetation we generally focus on future increases in temperature and changes in the severity and frequency of extreme events (eg. drought).  While the effects of increasing temperature on vegetation may only be apparent over many decades or even centuries,  rising atmospheric CO2 is with us now and arguably is already altering vegetation dynamics.  Under current CO2 levels, tree and shrub seedling growth rates are possibly several times faster than they were at the time of European arrival in Australia.  This is important because the size of many tree and shrub seedlings is closely related to their ability to survive grass fires – the bigger the seedling, the more likely they can survive a fire and go on to become an large adult tree.  In higher rainfall areas, some trees and shrubs that might have reached a size where they can survive fire in 3 to 4 years could perhaps do so in 12 to 18 months.  Conditions are shifting in favour of trees over grasses and this has big implications for how we manage landscapes with fire.

There are two recent publications that have had a huge impact on how I view the Australian landscape and both have implications for how we might manage grassy woodlands now and in the future.  The first of these, Bill Gammage’s wonderfully rich book The Biggest Estate on Earth, has challenged my ideas about current and historical land management and vegetation patterns.  The second is a research paper by William Bond and Guy Midgley which details the impact of changes in our global atmosphere  on vegetation, in particular the influence of atmospheric CO2 levels on tree densities in grasslands and grassy woodland ecosystems (savannah).

Despite the considerable differences in the two publications there is a strong theme current to both.  Both see fire as central to determining many vegetation patterns, something most ecologists and land managers would agree with.  Gammage strongly argues that in 1788 vast (but specific) areas of Australia were open and grassy owing to intentional, planned fire undertaken by indigenous Australians.  With the removal of planned fire following European settlement many of these areas rapidly thickened, promoting large destructive fires and requiring clearing of regrowth.

More recently, land retirement (or removal of livestock) in open grassy woodlands often leads to rapid regeneration, and loss of the open woodland characteristics, presenting a significant challenge for conservation management of these ecosystems.  Gammage would argue these landscapes lack the fire regimes necessary to keep country open.  Previous work by Bond and Midgely has also demonstrated the role of fire in maintaining open grassy ecosystems across the globe.

The work of Bond and Midgley adds an additional element to the historical story told in The Biggest Estate on Earth.  Their central argument is that while fires are crucial to determining the relative dominance of trees in higher rainfall savannah landscapes, atmospheric CO2 concentrations influence how effective fire is.

While increasing rates of tree and shrub growth due to increased CO2 might be good in some respects, making it increasingly possible to restore tree cover rapidly to agricultural landscapes with appropriate soils and management, it does have serious implications for how successfully we can use fire to keep country open and grassy, which is essential for maintaining much of the ground layer diversity of grasslands, grassy woodlands and grassy forests.  While trees in woodlands are good for biodiversity, dense thickets may not be.

During the last glacial period (which peaked ~20,000 years ago) atmospheric CO2 levels were less than half what they are today, approximately 180ppm (it is now 396  ppm and rising).  In southern Australia not only was it much colder and drier, but lower CO2 levels were strongly limiting for plant growth.  Under these circumstances grasses dominated many landscapes, in particular drought tolerant C4 summer active grasses, such as Kangaroo Grass (Themeda triandra).  This is despite the colder summer temperatures – C4 grasses compete more effectively with cool-season (C3) grasses (such as native Poa tussocks), at low CO2 concentrations.  Tree and shrub seedling growth in contrast was much slower and relatively few fires were necessary to tip the balance in favour of open grassy ecosystems.

By the time of European arrival in Australia CO2 levels had risen to about 280 ppm and the climate was warmer and wetter.  Trees and shrubs would have been growing faster and fires would need to have been more frequent to keep country open.  Since then CO2 has risen dramatically and throughout the world tree densities in savannah woodlands have been found to be increasing, even where historic fire frequencies have been maintained.

It is almost certain that returning to pre-1788 fire regimes will not have the same effect that it did then or several thousand years before.  There is little doubt that more frequent fires will be required to maintain open grassy woodlands.  These more frequent fires would then need to be weighed up against the possible impacts on biodiversity and current land use. Other management strategies may need to be employed if we want to keep our woodlands and forests open and grassy, including physical removal, stem injection with herbicide and strategic use of livestock.

Regardless of the rate at which our climate is changing due to human actions, the effects of rising atmospheric CO2 are already here.  Navigating the interactions between CO2, vegetation change and fire is a challenge for how we manage our land now and into the future.


Bill Gammage 2011 The Biggest Estate on Earth Allen & Unwin

William Bond and Guy Midgley 2012  Philosophical Transactions of the Royal Society B 367: 601-612

This short opinion piece was written for the Far South Coast Conservation Management Network  and can be downloaded in full here:  History__Fire__Global_CO2_Dorrough_CMN_Nwslttr_24_SPRING_2012

Natural Regeneration

Natural regeneration is often used to describe the process of tree and shrub establishment in farmed landscapes.  In some contexts it is used in the positive – recruitment of trees in landscapes with declining tree cover – but in others it may be perceived as negative – woody plant thickening.

We see natural regeneration though as having a broader meaning.  For us regeneration reflects the rebuilding of the biological and cultural diversity of agricultural landscapes.  Natural does not imply an absence of human decision making or management, but rather it reflects the need to harness and understand ecological function and ecosystem processes.  This is not a conservative view of the environment, rather it is one in which understanding of ecological processes is used to rebuild and regenerate landscapes.  This is not a landscape in which people are absent, rather it is one in which they are very present, but where management and decision-making is based on regenerating biological diversity.


Broad-scale management of native vegetation in temperate grazing lands

In the long-term, neither paddock trees nor diverse, native perennial pastures can be maintained under high-input grazing systems involving phosphate fertilizer applications.  Hence, there is a clear trade-off between productivity, and the provision of ecosystem services and conservation of biodiversity. However, there is growing evidence that low-input grazing systems, that incorporate significant pasture recovery periods and lack fertilizer application, can be compatible with their maintenance and can be profitable.  To read more click on the link to our invited paper presented to the Grasslands Society of NSW .

High input livestock production

Farm production systems and vegetation management

Developing options for conserving native vegetation in agricultural landscapes not only requires knowledge of the relationships between management practices and biota, but also knowledge of the farm production system itself and the role of native vegetation within that.   To read more see our book chapter in “Temperate Woodland Conservation and Management” or read a draft of these ideas by downloading the pdf file below.

Chapter 17 (draft)