Part 2 – The suspects

What’s killing the trees? In this week’s exciting installment, we examine the evidence implicating three suspects in the Manna Gum murder – for the story so far, please read part 1.

Suspect 1 – Agricultural practices 

In Australia, several episodes of insect related Eucalyptus dieback have been attributed to agricultural practices, which lead to degradation and simplification of ecosystems (Landsberg & Wylie 1988). This is known as ‘rural dieback’, and may be the result of one or more of the following:

  • Isolation and exposure of individual trees
  • Loss of habitat for insect predators or parasites
  • Soil degradation and compaction
  • Fertilisation and pasture improvement
  • Competition (pasture, weeds, mistletoe etc.)
  • Pollution or herbicides

A previous dieback ‘crime’ thought to have been committed by agricultural practices was the New England dieback that hit the headlines in the 1980s. Christmas beetles are a common native insect in the woodlands of eastern Australia, which rely on grass roots as larvae and then feed on eucalyptus leaves when the adults emerge. When clearing for agriculture leaves isolated trees in a paddock, the improved pastures can support huge numbers of larvae which then emerge en masse and completely defoliate the few remaining trees.

The Monaro region is well known as sheep grazing country, and has been partly cleared and used for this purpose for well over a century. However there are also some areas such as travelling stock reserves which are relatively undisturbed which can be used for comparison. I selected a number of sites, and measured the severity of dieback as well as the ‘structural complexity index’ – this includes a range of indicators such as hollow bearing trees, understorey density, and fallen logs which are related to biodiversity and resilience (McElhinny et al. 2006).

Agriculture comparison

The picture on the left shows a fairly typical paddock with regular grazing on a combination of exotic and native pasture. It is structurally simplified – fallen logs have been removed, there is no understorey and low species diversity. The picture on the right was taken just across the road at a travelling stock reserve, which in contrast has much higher structural complexity with a diverse and healthy native understorey. We would expect that the reserve should be more resilient to dieback, yet the severity is the same.

Although agricultural practices are present at the scene of the crime, these observations would suggest that we are looking for another culprit which affects grazing land and reserves equally.

Suspect 2 – Fire exclusion 

Fire is an important part of most Australian ecosystems, which have evolved to cope with regular burning. But since European settlement fire has been suppressed from many landscapes, and this has had a range of effects including changed soil conditions, nutrient imbalances and proliferation of woody shrubs (Jurskis & Turner 2002). These conditions may be stressful to trees, making them more susceptible to insect attack.

Although the Monaro region experiences frequent wildfires in summer, particularly in grassland, fire has been excluded from much of the forest and woodland areas. In 2003, a huge bushfire in the Snowy Mountains burnt several small pockets on the edges of the dieback affected area, providing an opportunity to observe the effects of burning. I again compared the structural complexity and dieback severity in burnt and unburnt areas.

Fire comparison

The two sites above are on the same property, only a few hundred metres apart, but the one on the left was burnt in the 2003 bushfire. The fire stimulated thick regrowth of acacias and eucalypts, but had no effect on the severity of the dieback. Of course, a single fire would not have the same effect as regular burning, but this does show that using fire as a management strategy for dieback would be unlikely to make a difference.

So fire exclusion is also off the hook, and that leaves one final suspect…

Suspect 3 – Climate change and extreme weather

Climate change and extreme weather have recently been implicated in an increasing number of dieback episodes around the world (Allen et al. 2010). Climate change is already occurring in eastern Australia, with warming temperatures, increasingly variable rainfall, and more extreme events such as droughts, floods and heatwaves. Insects are also extremely sensitive to changes in climate due to their short life span and huge reproductive potential.

The Monaro region has quite a harsh climate which is very different to the surrounding areas. It sits in a rain shadow caused by the Snowy Mountains, so it has very low but highly variable rainfall, and also experiences extremes of temperature. The affected area sits right in this rain shadow, and the worst affected area also has the lowest rainfall.

The onset of the Monaro dieback coincided with one of the worst droughts on record, during which the average rainfall dropped from around 550mm to 450mm per year. There has also been a longer-term change in the distribution of rainfall, with more rainfall in summer and a large drop in autumn rainfall that has shifted the dry season earlier in the year.

Figure 3 rainfalldistribution

Manna Gums normally grow in much wetter areas, and their distribution is strongly related to rainfall. The harsh climate of the Monaro was already on the edge of their range, and the recent changes may have pushed them beyond a critical threshold.

Next week I will discuss the conclusion to the Manna Gum mystery, and some possible strategies for management and conservation.


Allen, C. et al., 2010. A global overview of drought and heat-induced mortality reveals emerging climate change risk for forests. Forest Ecology and Management, Volume 259, pp. 660-684.

Jurskis, V. & Turner, J., 2002. Eucalypt dieback in eastern Australia: a simple model. Australian Forestry , 65(2), pp. 87-98.

Landsberg, J. & Wylie, F., 1988. Dieback of rural trees in Australia. GeoJournal, 17(2), pp. 231-237.

McElhinny, C., Gibbons, P. & Brack, C., 2006. An objective and quantitative methodology for constructing an index of stand structural complexity. Forest Ecology and Management , Volume 235, pp. 54-71.