Ageing the mallee: a history in burnt trees

mallee-spinifex 3
Coppicing Mallee trees and Spinifex grass in Mungo National Park. Original photo: Visit Mungo.

Trees grow short in the mallee. Little rain and poor soils stunt their growth. Over thousands of square kilometers, in semi-arid Victoria, South Australia and New South Wales, many mallee eucalypts reach just 6 to 7 meters high. The short trees burn well, especially when growing above Spinifex and, after wet years, above flammable Spear-grasses. In 1974, fires burnt over 650,000 ha of mallee in south-west NSW.

Large fires kill most of the mallee tree trunks. The trees survive by resprouting from the lignotuber (or ‘mallee stump’), and the new coppice stems grow up to form the iconic, many-trunked trees. Like an axe with new handles, each tree (or more precisely, each lignotuber) is old, but the trunks and branches are younger, and date from the most recent fire.

About ten years ago, a team led by Mike Clarke and Andrew Bennett (from La Trobe and Deakin University) turned the coppicing habit of mallee trees to their advantage. They wanted to work out the fire history of the mallee, to discover how often different areas burnt, and to find out the age of every stand of trees. The boundaries of older fires are poorly known, so they headed to the field to gather new evidence.

A formula for fire

The team measured over 40,000 stems of mallee trees. Photo: XXX.
The team measured over 40,000 stems of mallee trees. Photo courtesy of the Mallee Fire Team.

First the team first visited lots of sites where they knew the date of the most recent fire. There they measured the diameter of thousands of post-fire coppice trunks. Because they knew when each area burnt, they knew how old the trunks were. As you’d imagine, trunks in older patches were bigger than those in recently burnt areas.

Back in the lab they ran the numbers. They developed a formula that predicted trunk ages and fire dates based on the diameter measurements. ‘Awesome’ they cried. Armed with their nifty formula, they visited areas where fire dates were unknown, and measured thousands more coppicing trunks.

Back to the lab for more number crunching. They typed in the new measurements, sipped their chai lattes, and gazed at the stats package as it spat out the estimated stand ages and fire dates for all of the unknown areas. Bingo.

Their method isn’t perfect; the formula under-estimates the real age of old fires and old trees to some extent. Nevertheless, the team created a practical way to work out when every patch of tree mallee last burnt and the approximate age of every stand of mallee eucalypts – across thousands of square kilometers.

Counting the scars

Their simple technique helped re-write the fire history of the mallee. Before their field study, the team mapped recent fires from the fire scars seen on satellite images. Satellite mapping gives accurate information on all fires after 1972 – the year of the first Landsat image – but nothing before then. Many older fires are known, but their boundaries are uncertain or imprecise.

Recent fire scars in Wyperfeld National Park, Victoria. Source: Google Earth.
Recent fire scars in Wyperfeld National Park, Victoria. Source: Google Earth.

The team’s map of Landsat fire scars showed that 40% of tree mallee burnt between 1972 and 2007. The bulk of the mallee – a sizeable 60% – sits in the category ‘old mallee’, burnt more than 35 years ago. ‘Old mallee’ means ‘older than Landsat’, nothing more. Nevertheless, in the absence of better information, the satellite fire map created the impression that much of the mallee was ecologically old and perhaps in need of another burn. In Mike Clarke’s words, based on the information available to them:

the managers of one reserve in New South Wales were really keen to carry out a landscape mosaic burn to reduce what they perceived as an over-supply of old-growth Mallee … that is [mallee] greater than 35 years of age.

The world before Landsat

How old is this pre-Landsat mallee? Were the ‘old’ stands last burnt 40, 80 or over 100 years ago? To answer this question, the team divided the big chunk of old mallee into age classes using the fire dates they calculated from their tree measurements.

They found that the pre-Landsat mallee contained a huge diversity of age classes, ranging from 35 to over 160 years old. Remember that their formula under-estimated the real age of old stands. This means that the 160 year old stands could be 200 years old or more.

Importantly, they discovered that really old stands of mallee, over 100 years old, are really rare. Most pre-Landsat mallee is Gen-X and Baby Boomer mallee, 35 to 70 years old. Only 5% of mallee sites were over a hundred years old. There is no surplus of ancient mallee.

Dead stems do talk

But wait. There’s more. Burnt mallee trees have live and dead trunks. The live trunks came up after the most recent fire. The dead trunks came up after an earlier fire and were killed by the most recent fire. By measuring the live trunks, the team worked out when each stand was last burnt, as described above.

new coppice
The size of dead stems indicates the length of the period between recent fires. Original photo: Plants in Action.

Dead stems tell the stories of earlier fires. By measuring lots of dead trunks, the team could estimate the date of the second last fire. The dead stems disclosed the time between the two most recent fires (the ‘inter-fire interval’) and gave a glimpse into the ‘invisible mosaic’, the hidden pattern of past fire regimes.

A new fire ecology blossoms

If we don’t know the age of a stand of trees, a patch of bush, we don’t know much at all. We can’t predict how quickly (or slowly) habitats will change, hollows will form, litter and fuel will accumulate. We can’t say how or when the fires we light will alter ecosystems in the future. Accurate fire histories underpin future ecologies.

Mike Clarke and Andrew Bennett’s enthusiastic team unlocked the mysteries of ageing in the mallee. In future blogs I’ll describe more of their results. Their many papers have been published in the best ecological journals. But they all stem from one simple field technique. Everything sprouts from a burnt mallee stem. From little things, big things indeed do grow.


Thanks to Dr Angie Haslem and Dr Simon Watson from the La Trobe and Deakin University Mallee Fire and Biodiversity Team for fact-checking this week’s post.

Further reading

Avitabile SC, Callister KE, Kelly LT, Haslem A, Fraser L, Nimmo DG, Watson SJ, Kenny SA, Taylor RS, Spence-Bailey LM, Bennett AF & Clarke MF (2013) Systematic fire mapping is critical for fire ecology, planning and management: A case study in the semi-arid Murray Mallee, south-eastern Australia. Landscape and Urban Planning 117, 81-91.

Clarke MF, Avitabile SC, Brown L, Callister KE, Haslem A, Holland GJ, Kelly LT, Kenny SA, Nimmo DG, Spence-Bailey LM, Taylor RS, Watson SJ & Bennett AF (2010) Ageing mallee eucalypt vegetation after fire: insights for successional trajectories in semi-arid mallee ecosystems. Australian Journal of Botany 58, 363-372.

Related posts

18 thoughts

    1. Many thanks Anthony, it’s easy to write blogs like this when vibrant researchers like the mallee team do so much fantastic new work. Ecology and conservation science have a great future in their hands. Cheers Ian

  1. Great stuff as usual Ian! I still have vivid memories of your stories of fire scars and age classes of Callitris and Eucalypts at Mt Pilot during my studies. I was looking at some local fire scars yesterday (on large Turpentine and Stringybarks) and I remember you discussing that the scar would be on the opposite side of the tree to the fire direction – am I correct? I am fascinated still by fire ecology and this method of discovering fire history in Mallee Eucs provides even more detail to our understanding of ecosystem function. Cheers, Steve.

    1. Hi Steven, thanks for your reply. Yes, when a fire burns a tree trunk (especially if the fire is running up a hill), the flames will often flare up on the side of the tree away from the fire front, causing a scar on that side. Often more litter accumulates behind a tree on the uphill side which can make this more obvious too. It’s great to hear you enjoyed the Callitris field trip too. Best wishes Ian

  2. Thanks for sharing Ian. This is very important research. I find it interesting that mallees are among the oldest trees on the planet, suggesting great ecological stability, yet the ecosystems associated with the mallee fire ecology are very dynamic. I just love Eucalypts!

    1. Hi Jan, I’m hoping to write another couple of posts about fires in the mallee, and the next one will highlight lots of the ecological changes that occur between fires, which are really quite fascinating. The mallee is an amazing place, and eucs are amazingly tough trees too. Thanks for writing in, I’m glad you enjoyed the post. Best wishes Ian

  3. Before Landsat there are many series of aerial photos – I recall finding them very useful in doing field work in the Victorian Mallee in the seventies – why not use them to help refine the age/size model?

    1. Hi John, great suggestion. I didn’t work on the project so can only answer on the authors behalf. I would imagine that variable time spans between air photo runs, incomplete photo runs and missing photos would have reduced the usefulness of the air photos compared to the satellite image work. Their project covered large areas of the mallee in SA, NSW and Vic, and it would be hard to find many complete air photo runs that covered the entire region. Nevertheless, old air photos have been used in other projects to map old fires and would be a useful way to confirm potential fire dates and to identify fire boundaries back to the 1940s. The team managed to do pretty well to project past fire dates back to the mid-late 1800s using their stem size analysis, so they would still have had to extend the fire history back a long way from the first air photos. Thanks for writing in. Best wishes Ian

  4. Ian, great blog, and some very important research. I would imagine some implications, particularly in Victoria, of the over burning of these Mallee areas to reach fuel reduction burning targets. Obviously fire is integral to mallee community but do we know if this over-burning, and therefore shortening of the inter-fire interval, going to forever alter these mallee communities by changing their composition and structure?

    1. Helllo Mark, great question and big issue. Professor Mike Clarke, who headed the project, described frequent burning in the mallee as a ‘major threat’ to species that require long-unburnt stands in this video:
      Fast-forward to 16:20 minutes into the video for the relevant section. You can picture how habitat structure will change if you look at the animated charts in the next blog: Habitat features that take a long time to develop, especially hollows, will be most affected by short inter-fire intervals. Best wishes Ian

  5. Ian, a comment on Mark’s comment: from my observations walking in parks and reserves in NE NSW and SE Queensland, there is far more vegetation change occurring as the result of no burning than as the result of frequent burning. In coastal vegetation, heaths are now woodlands and woodlands are now shrublands as fire dependent species die don’t regenerate and then die. Rainforests are marching into sclerophyll forests, again die to lack of fire. This is not to say that there is a “correct” vegetation for each site. But the rate of change on many sites due to lack of fire is an issue for conservation managers concerned about biodiversity. Probably a bigger issue than too frequent burning. See also “Fire exclusion and the changing landscape of Queensland’s Wet Tropics bioregion” in Aust Forestry vol 77 no1 (2014).

    1. Hello John, thanks very much for your comment. Mark’s comment above is consistent with the take home messages from the papers from the mallee fire research project, so the situation in mallee vegetation in SE Australia is, for whatever reason, different from that in other ecosystems. I’ve been thinking of writing a blog on fire ecology in the Bunya Mountains in Queensland based on the work by Rod Fensham and colleagues. This would touch on some of the issues you mention above. I hadn’t seen the new paper you mentioned, so will chase it up and study it too, and perhaps include it in that story. Thanks again, best wishes Ian

  6. … thanks Ian, fab blog – is there anything on the age of the lignotubers? … there seem to be extravagant claims for mallee ages based on ‘the expanding universe theory’ of mallee ring circumference but do you know who’s had a go at ageing say Euca socialis lignotubers? all the best, Stephen

    1. Hello Stephen, long, long time no see! Great question. I didn’t chase up recent papers on how old Mallee lignotubers can get when I wrote this post. However I wrote a note to myself to follow up the topic in a future blog post. I haven’t made any headway on it yet but will hopefully get to it in the not too distant future. Stay tuned. Thanks again and best wishes Ian

What Do You Think? Please Leave a Reply.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s