Biodiversity publications archive

Biodiversity and Fire: The effects and effectiveness of fire management

Proceedings of the conference held 8-9 October 1994, Footscray, Melbourne
Biodiversity Series, Paper No. 8

Biodiversity Unit
Department of the Environment, Sport and Territories, 1996

Discussion: Papers 4-10

Malcolm Gill

Q: Was the choice of examples to give us extremes of the fire frequency spectrum or was there a message there that you wanted to reinforce?

Malcolm Gill: Yes, if you wanted to interpret some of the observations I gave you in terms of management, you would have an entirely different view of what to do if you were concerned with the conservation of pencil pine, for example – would you burn it at all, if rabbits were present – if rabbits were not present you might have quite a different view, although their tolerance to cool burns is unknown (assuming this was practical which it may well not be). With the mistletoe example, it depends again on your objectives, and having objectives clear for any parcel of land is absolutely essential to any consideration as to what management is appropriate. I think the lack of this has been a disappointing feature of the debate after the Sydney fires. It's tended to be any sort of solution that seemed good at the time, rather than what are the management objectives for this particular piece of land? – be it private, local government, state or federal – and what's appropriate to achieve those objectives? So yes, there's a variety of circumstances illustrated in those slides.

Q: We have been studying a peat fire in Melaleuca squarrosa swamps at Anglesea, and found that all the plant propagules were shed and consumed by the on-going peat fire, which went on for weeks. In that environment seedling recruitment was very low and were eaten by wallabies. Could that have been part of the problem with the fires in the pencil pines?

Malcolm Gill: Yes, I suppose so. This question of root suckering is really an intriguing one as well – let alone seeds. Because the plant only fruits every five or six years and doesn't hold the seed, then it regenerates by suckers which have only been observed in boggy areas – or it must re-invade from elsewhere. In the latter case the shape and size of the fire becomes important. There are a lot of unknown things about the regeneration and I am delighted that you are looking at the effects of peat fires near Anglesea. There is no careful study of effects of peat fires in Australia.

Q: Are you are able to say if pencil pines have been reduced considerably in area in Tasmania due to fires and rabbits?

Malcolm Gill: Yes

Q: Was that fire started by bushwalkers? – because many of them are in Tasmania. If so, the implications of that for the protection of these communities should be noted.

Malcolm Gill: At that time it was attributed to stockmen ... because high plains grazing was acceptable then in Tasmania.

Comment from audience: That fire was started by graziers.

Q: Some of the literature suggests that mallee and Triodia encourages fire, but from what you said, it is disadvantaged by fire.

Malcolm Gill: The Triodia and the mallee eucalypts provide fuel for fire, but fire adversely affects the viability of the shoot system of the plants. The theory was that the evolution of the mallee encouraged fire to its own advantage. That theory is sometimes called the "Mutch" hypothesis. A chap called Bob Mutch in Montana, USA published a well known article in a journal called Ecology some 20 years ago. It aroused much controversy and argument over the idea. His original paper was quite difficult to decipher; its been put in a better way as the "fire facilitation hypothesis" but still hasn't been proven; but these sorts of diagrams (referring to one of Malcolm's slides) I think start to unravel some of the aspects to it. I think the classic paper commenting on Bob Mutch's paper was entitled; "The Fire Facilitation Hypothesis – Mutch ado about Nothing".

Q: When you were talking about wet sedgeland and the need to maintain it by burning it, what sort of interval would you recommend, if any? The reason I'm asking is that we did some work in wet sedgeland in Tasmania, in a ground parrot area. We studied invertebrates in different ages of wet sedgeland and our data indicated that 20 years would be a reasonable interval.

Malcolm Gill: My comments are based on Charles Meredith's work and Dave McFarland's work in Cooloola. No fires are necessary for ground parrots if you have abundant sedgeland supplying seed and shelter (in fact ground parrots would be disadvantaged soon after fire). If you do have shrubs, then the shrub development varies with the environment. Charles Meredith was telling me that even western Victoria is different to eastern Victoria and both are certainly different to Queensland. It is the rate of shrub development that determines how often you burn if you want ground parrots. I accept that invertebrates might give you a different story. But taking a recipe from say East Gippsland and simply transposing it is not acceptable.

Brian Lord

Q: When making modelling differences between two different situations, one might forget factors that are not immediately apparent. These are the problems with modelling that I see, that it's very difficult to take into account all the factors that would be important, particularly if you are looking at very long time periods. I think it's critical to leave open the possibility of investigating factors that are not properly understood but are hidden by their temporal variability.

Brian Lord: There's a couple of problems in that. You are dead right, the bigger you get, the more states you put in, the bigger the matrix box gets. The matrix increases, not by the same number but according to the power of the number of states. The other point you make is, I suppose a challenge, and all modellers like a challenge, and that is "do we wait until all the components are there, or do we get on with it?"

I hope I have demonstrated that we have all of the components for workable modelling. As an analogy, your first attempt at building a motor car may have five wheels but at least we have something that can be out on the road. But let us not claim that models are the answer – they are not – modelling is simply a decision support system. For instance no-one can tell us what happens to all the little plants – but do we wait 100 years for long-term data or do we start modelling and say "lets not go down that path – it doesn't look safe. Lets go down these paths that look safer". Modelling is not in itself an answer, but it is a help along the way.

Q: I think you said at the outset the transition probabilities were fixed. I'm wondering if in fact that's true in reality as opposed to in a model?

Brian Lord: In the time available I wasn't able to go into the delights of non-stationary markov matrices! When we put the numbers into that box, they become fixed numbers and by driving the model with fixed numbers you end up with basically steady state systems. We know that that's not the case, so we can use equations instead of fixed numbers, equations that, given an element of time being put in – every time you change the clock, so to speak – the number is going to be different. So within the semi-Markov process there is an embedded markov chain. Equations enable you to have a non-stationary embedded markov chain within your semi-markov model. The equations take into account environmental factors that change with time e.g. fuel accumulation, weather, ignition probability.

Q: How does your model take into account episodic events like the 1 in 100 year drought or thunderstorm etc.

Brian Lord: Yes, that the difficult bit! And again in the non-stationary part they could be built in, although I don't know who will do it! It will take quite a modeller! In fact while that challenge is on, there's still a challenge for someone to come up with an equation to explain all of that graph I showed you with the asymptotic curve (of fuel accumulation); they've got the curved bit, they've got the flat bit, but they haven't got one that does both. Those are things that are still to be refined, but we do have enough to build us a model to get us going.

Q: When will it be finished?

Brian Lord: Like most of the people that are building it I will be dead. There are components that are going to go on. I would hope that this is the sort of thing that managers will hook into and say; "I need to make decisions based on information I don't really have – is there a model I can follow?". Managers still follow models – models are not about mathematics – they are about what's in your brain, your experience and what guidance that has given you for decisions for the future. Modelling is the way to go, its just a matter of degree. There are now such advances in technology; for instance computer networks including "Fire Net". I'm not quite sure of my ground here but I think you could plug into Fire Net and get every piece I've been talking about out, and build your own semi-markov. Those are the sorts of things that are available. One of the suggestions about Fire Net is that if you have a management decision to make, go into Fire Net, get some answers out and build your own sequence of ideas and then you'll be able to make decisions. Models are decision support systems for people who have to make decisions, and managers of conservation reserves are the ones that really do need to know more than we already do know at this stage.

David Cheal

Q: You showed a slide of some woody plant invasions, were you going to allude to something a little later in your talk and relate that to what you had been saying about nutrients and fire?

David Cheal: The woody plant invasion presents problems; if nutrients are really low and deficient in heathland, how can these woody plants invade heathlands? They will actually increase the total nutrient capital of a standing piece of vegetation. If heathland plants are so efficient at grabbing nutrients, how come a heathland invaded with coast tea tree has more in it than an uninvaded heath stand? Woody plant invasions are our perception, it may just be succession because of changed ecological circumstances. I'm not saying whether it is good or bad – that's a decision we make. We should be asking whether we want it, rather than the pointless question: is it natural. Whether it is a woody plant invasion or succession to a dense shrubland is your point of view, but the situation is the same.

Q: What frequency are the repeat burns that lead to a decrease in nutrients?

David Cheal: The repeat burns were as frequent as every two years. It was in the fire break at Wilsons Promontory, after 3 burns in that particular case. The area is 'blasted' – a sacrifice area if you like – the gain is better control of the fire regime elsewhere in Wilsons Prom.

Q: In the sites that you burned such as those you sampled at Wyperfeld where the species richness had increased what has happened to the cover values of those species?

David Cheal: I only estimated, rather than measured cover values for individual species. No demonstrable change was detected in overall cover, or for individual species. We would need a very large change in cover values to pick it up with estimation. Unfortunately I didn't even have time to collect density data; what we have is species presence/absence and an estimation of their cover.

Q: I appreciate your comment that we need a lot more research. But whenever we talk to conservation mangers and say that we don't know enough, most conservation managers seem to clam up and say "we won't do anything" (whilst most fire officers would probably say the opposite!). In the interim we need some approach as to what we need to do in the next 10-20 years of getting. It strikes me that the only possible approach is to in different sites try to do different things, just so we don't foreclose our options.

David Cheal: I totally agree. I wouldn't say to land managers "don't do anything". The decision not to do anything is in itself still a management decision. We are making decisions in the interim – let's make them on the basis of best available knowledge and recognise that they are imperfect decisions that will need to be re-assessed, say, ten years later, and adjust our proposals. I remember, when in the National Parks Service writing recommendations for fire regimes for certain mallee National Parks and then six years later recommending that the interfire period should be increased for one park. But the local regional manager preferred to go by the old published report rather than my new ideas. I then came to appreciate that the strength of the written word over-rode the same person saying something later on!

So we are making decisions; we should make decisions on best available knowledge; we should recognise they are imperfect and need readjustment in time.

Q: What you might call the back(?) regenerators you found in the mallee-heath; are those propagules there all that time, or are they coming in, and what's causing them to regenerate?

David Cheal: What we are calling an increase in species richness, may just be an increase in total density. As you appreciate every quadrat you do in a heathland doesn't have every species that's within your horizon. So when I pick up, say, Astroloma in a quadrat and I didn't have it before, its total density might have gone form 2/ha to 20/ha. In other words the propagules are still through the whole of the vegetation stand, but it's now become a bit more common and so its frequency increases. So the propagules have always been there; the species have always been there, but at much lower frequency; they are now establishing.

On the explanation for the change in species frequencies – unpredictable gaps were produced in that community by severe frosts (-11°C for three nights) which killed at least one of the dominant plant species, Banksia ornata. The gaps formed were filled by seedlings from species which do not require fire to germinate: fleshy fruited epacrids, Callitrix, Baeckia, Astroloma, ... that's why it was not Banksia, Casuarina or other species requiring fire.

I believe that these unusual one off environmental perturbations – very rare in time and also those very rare in space – are very important to maintain total habitat heterogeneity in a broader landscape. It also gives me another concern about extrapolating site specific studies to huge landscapes.

Q. I was wondering if slashing instead of frequent burning would be better for firebreaks, from the point of view of nutrients? Perhaps doing it in early autumn and allowing it to grow back and then maybe burning later, at the normal time?

David Cheal: This is about sacrifices and trade-offs you have to make in reserve management. I believe that it is valuable to have sacrificed marginal strips in a reserve to give you freedom with management of the interior.

There are advantages and disadvantages to slashed firebreaks: I learnt from Charles Meredith that slashed breaks are actually of some importance on Wilsons Prom. as they can provide a seed source for ground parrots. Slashed breaks are also prime sites for invasion of Kunzea and Leptospermum.
So there are down-sides and up-sides to every management action. It's the intelligent and thoughtful manager that has to try and balance
those out. The repeat burning, yes, can have a major impact on nutrients, however the soils at Wilsons Promontory actually have quite a high nutrient capital anyway, even though they're heathlands.

Mark Neyland

Q: There's nothing in the understorey and little on the ground – so why do they need to burn it all the time?

Mark Neyland: It's still policy of what is now Forestry Tasmania – the new corporation. The policy, set up in 1970, is that all the E. sieberi forests should be burnt every seven years. They mapped out the entire range of E. sieberi forests, drew it up into catchment boundaries and set it alight with incendiaries from helicopters and drip torches. With such a simple fuel load in it, its such a manageable fuel reduction exercise, they can usually get it right; and they did that for about 20 years. Current economic constraints are slowing Forestry Tasmania down and they are moving out to a more strategic approach such as burning around pine plantations. But, as written, it's still policy.

Q: You showed a slide, and commented that you thought this was likely a result of the prescribed burning regime. Have you any basis for thinking that?

Mark Neyland: Only my own observations. If you went back three slides, to where there was a gully and treeferns, it's a south-eastern slope and quite a moist site. It has all the species in it to suggest that, if you leave it alone there would be a broad leafed closed understorey. Now I can't actually say this, because all of the E. sieberi areas have been burnt, I can only infer that given the presence of broad leaved understorey species on some sites, they would look different given a long-term absence of fire. And that's what I hope to prove with these transects over the next 50 years – we can watch the change. We are asking the Commission to ensure they are not burnt. And hopefully in the longer-term we'll come up with something concrete.

Q: You had some slides of the Gahnia (sedges), and also mentioned the Casuarina thickets, and then you showed other pictures where the understorey was almost absent. Can you relate the types you showed us to particular fire regimes, or are they the result of other aspects of the environment?

Mark Neyland: I can't relate it to fire regimes, because to our knowledge that is the same throughout. The only thing I can point to is that maybe on the east coast you have a nutrient input as a result of the on-shore drift with the sea-breeze and all the coastal forests just have a more luxuriant understorey ... this is rapidly lost as you go inland. One butterfly species is restricted to within a mile of the coast.

Q: Do you know if there's any differences from the aboriginal fire regimes?

Mark Neyland: I don't think the aborigines were in there. To work in the Methinda (?) forests I had to buy carpet layer's knee-pads to take my fuel samples – because it is murderous stuff! Very few mammal scats. If I was an aboriginal with bare feet – why go into the inland forests when the resources on the coast are so much higher? Aboriginal burning regimes are therefore not relevant, and unlikely to have affected the inland forests much.

Q: I was wondering, if you are proposing to monitor over such a long period of time - what other attributes are you looking at? Are you taking climatic measurements as well? Are there met. stations you can relate to?

Mark Neyland: There would only be the local met. stations and within that range there are a couple of forestry trials with met. stations and water flow meters. All we recorded on those sites was the vegetation, but as we've surveyed them in, we know where they are for the future.

Andrew McMahon

Q: What is the consequence of leaving those acacia invasion areas alone? Is that a succession sequence going on?

Andrew McMahon: We don't believe it's succession – well it's a successional sequence that's been helped along – the dispersal agents for these things have just literally gone through the roof. There may be natural processes like fire regimes involved; but it is the post-European factors that have really facilitated propagule dispersal of many species. Things like dispersal by car tyres, birds such as starlings; there has been an explosion in mid-distance dispersal. So we do not think this is a natural phenomenon.

Q: Were areas on the heath more regularly burnt – prior to 30 years ago or so? Why is this a recent phenomenon, rather than happening two hundred years ago?

Andrew McMahon: We don't know – but we think increased dispersion is the primary thing causing this invasion, because it's so recent. It's conceivable that in the past, either before or after pre-European occupation, that if there were two fires in quick succession, say twice in three years, just in the normal run of things – and you may only need to get that every 50 years or so – it would make the acacia retreat as it would knock the seedlings off. The heath could easily carry a fire at that interval. So there may have been an advancing and contraction of this species over time, which could have been controlled by the odd event of two fires in quick succession. But more recently it seems to have been dispersal – almost like the spot-fire phenomenon – they're just spotting all over the place, and as the front gets bigger, it's just barrelling through the bush with starlings just dispersing tens of thousands of seeds all over the place.

Q: Considering the scale at which the coastal heath of the area you were studying is being progressively destroyed, what opportunity do you see to recover the heath using fire – or do you think management-wise, it's virtually impossible?

Andrew McMahon: No. I'm optimistic. I think you have to control the recruitment of it, but I think burning at the right time would minimise that. There's no getting away that there would have to be a follow up. Fire is the primary agent that I see as being the only thing that could possibly be used at such a scale. But there may need to be a follow-up; either another fire, or a weeding campaign to control the recruitment.

Q: Don't you worry about loss of the other species from the heathland – the ones that you found were missing from the quadrats you did in invaded heath?

Andrew McMahon: A major change in the dominant plants of any vegetation – that's something we should be concerned about. And that means that this acacia should be controlled before that occurs – before you lose those species. I think you really have ten or so years to get most of the heath back, or maybe 15. But after that, you are going to suffer appreciable losses.

Q: A comment and a question. Firstly, I would like to reiterate the difficulty of getting ecological burns. I asked for an ecological burn in one of my study areas in Wyperfeld National Park. It was over a year between proposal and approval to burn approximately 20 hectares, including having to make submissions to the Director, and to the Minister. Whereas tens of thousands of hectares are burnt under the justification of fuel reduction burning. I cannot understand this discrepancy in attitude.

Secondly, what you have come up against is the perennial problem with undesirable woody plant establishment in heathland. They nearly always seem to be obligate seed regenerators, which suggests that applying a fire regime is the way to control. And most recommendations focus on two fires in quick succession. The problem then is that obligate seed regenerators that you want there are also knocked out. Do you have any thoughts as to how one can overcome that conundrum?

Andrew McMahon: Unfortunately no! I think we are facing that at Wilsons Promontory at the moment. Because the Kunzia must be flowering now from the original fire. I think its a cost we are going to have to live with – a cost of the management practice. Unless you are going to get in there and rip them out by hand, and in some cases it is a ludicrous suggestion – you would have to mobilise the whole of Victoria! So perhaps successive fires are the only answer and we might have to lose those plants.

Q: You referred to the Acacia sophorae seed as ant dispersed and then as bird dispersed. I would say that it has all the characteristics of a bird dispersed seed in that it is held in the canopy and not shed - its held in the pod. If it's any consolation, the seed of Acacia sophorae has great potential as a bush tucker seed; it is very acceptable, edible-wise. Encourage people to harvest seed from that area!

Andrew McMahon: Yes. [General laughter]

Q: When you looked at the area you slashed as a fire break – did the wattle regenerate the same as it did in the burnt area – could slashing be another management tool instead of burning it?

Andrew McMahon: There was regeneration in the slashed area from seed. We looked at that 2 years after – so we are not sure – rather than one year after for the burnt area. We would assume that you would get more regeneration from the fire just because there must be a percentage of the acacia seed that is hard-seeded and therefore fire-stimulated – I think that's the case. But there was certainly substantial recruitment in the slashed area as well – but perhaps not as much as the burnt area.

Q: I was interested in your suggestion that you didn't think Acacia sophorae had a seed-bank. If the seeds are taken by ants then perhaps they are just too deep in the soil?

Andrew McMahon: Yes, that's true, good point. From some of the work that's been done on allied species, like Acacia cyclops which is the coastal dune species in Western Australia, and is a hideous weed in South Africa – they have completely different biology in terms of their seed dynamics. They have massive seed production, but the amount that gets translated into recruitment is small. But the mechanisms between the massive production and the small recruitment is yet to be disentangled. They have a huge arrow on them that ants love, so we assume that's been harvested and perhaps carted down the profile beyond where it can germinate. But there also may be a big percentage that are quite soft seeded, and so don't persist.

Q: I was wondering whether there might be another aspect superimposed on fire regimes and dispersal, and that is past grazing. I am mindful, that in other near coastal areas like Wilsons Prom. in particular, also had a prior history of stock grazing – and I think your area has been similar. After the removal of grazing and associated burning, you had these problems of invasive woody plants. I'm wondering if grazing is involved in part ?

Andrew McMahon: Quite possibly – I'm not quite sure in the case of Acacia sophorae but I know that where Kunzia ericoides is getting out of hand at Healseville Sanctuary the grazing is strongly implicated. Just by selective grazing – the Kunzia is unpalatable, it's pretty weedy in its biology, the cattle make regeneration spots for it, up it comes and its had a free for all. In that case we think grazing has had a major role in the spread of that species and it could well be implicated here.

Q: When you talk to managers about these sorts of issues, the thing they always come back with is – is this not just part of a succession? Is it a natural process that will eventually move on and establish a different sort of ecosystem in those areas. You personally answered that earlier, but what do you see as the future of these acacia thickets down in the south-west – in ecological terms?

Andrew McMahon: It's been mentioned before, but you get a bit sick of arguing about it. It's a bit like arguing about the pre-European condition – was it fire climax, climatic climax? – and so on. But as David Cheal mentioned, the ultimate decision is "what do you want"? Do you want the species-rich heathland? Do you want the heath-rat Psuedomys shortridgii? Do you want Melbloom's spider orchid? Do you want one of the hot fired diversity units in southern Victoria or do you want Acacia sophorae? And I think its down to that, and that's really important. We judge flora and fauna on its merits now. If we are maintaining biodiversity then it's clear-cut. Whether, as Ian suggested, this may have happened in the past, well that's fine, but we see things in such a short time frame. We won't be around to see the acacia turn into something else that matches the heath in bio diversity. So I think it's a matter of adding up the biodiversity arguments and saying "well, what do you really want – what do we manage for?".

Kevin Tolhurst

Q: Was there a difference between spring and autumn fires other than post-fire weather conditions?

Kevin Tolhurst: Surprisingly not. The fire intensity and flame heights and rate of spread (ROS) were pretty similar. The main differences were, as you said, the weather conditions after the fire – and the stage of development of the plants and animals had some impact. Also it was quite noticeable that the surface soil temperatures just under the litter layer were higher in the autumn fires. But there really wasn't any measurable differences in fire behaviour and intensities between those sites.

Q: But what about the soil conditions – wet or dry?

Kevin Tolhurst: Yes, that was a major difference there. The average soil dryness index for the spring fires was less than 30mm equivalent in all our spring areas and it was between 50 and 80 for the autumn burn areas. So soil conditions were certainly drier in autumn, although the litter bed, the surface moisture was pretty similar. But that's one of the reasons we had the higher soil temperature after the autumn fire, because the soil was drier, so the bottom of the litter layer was a bit drier too.

Q: What sort of fuel loads were you burning and what did you end up with after the fires?

Kevin Tolhurst: The litter fuel before burning was about 12 t/ha and was reduced to 4 t/ha after. This is not including about 3 t/ha of fragmented decomposing litter and humus, but this was the same before and after the fire anyway.

Q: What about the size of the sites that you burnt in relation to what generally happens in practice, and what about the implications of that for the immigration of vertebrates and invertebrates back in from outside of the burnt site? Could this have biased your data?

Kevin Tolhurst: Good question. In Wombat generally fuel reduction burns are between 50 and 200 ha; occasionally a 600 ha burn. In terms of movements of invertebrates and mammals – it's pretty hard to tag an invertebrate, so I can't speak so well for them – but I assume the surface active invertebrates, particularly larvae don't move very far. We tagged small mammals and the animals that we found in the site after the fire were the same ones as were there before the fire. There was very little movement across the borders of those areas. Therefore 15 ha is large enough and that is part of the basis of why we chose that size. However that is not the case for birds as they are mobile and it is difficult to differentiate between residents and those just moving through – there's far more movement in and out of those 15 ha sites. For invertebrates and small mammals, 15 ha experimental burns were ample.

Q: Could things like beetles be flying in and out to some extent?

Kevin Tolhurst: Yes; well that particular group of beetles that come into the site are ones that mass migrate, so they move around. That's why you had a similar effect (in other sites?) – because you take measurements at the same time in the long-unburnt sites and the burnt sites, you would hope that the same thing will be happening on adjacent sites.

Q: My question is about the size of sites you were burning and I am interested in bird numbers and the effect of spring burns. Where in fact are birds likely to be able to breed, if the understorey is removed?

Kevin Tolhurst: We looked at birds although I didn't have time to present the results today. For birds, the number of species was little affected by burning. Some species changed – some new ones moved in and others moved out, but probably 80-90% of the birds remained the same. The number of birds seen also remained about the same. Because its only 15 ha, the breeding success within the burnt areas is not going to have great bearing on that overall result. If you were looking at a 2000 ha area, then it might be a bit different. We couldn't get five treatment areas any larger than that – it would just become too difficult to manage. Not a fantastic study from a bird's point of view. That's the nature of birds – they can move around ... and so it provided opportunities for some species and disadvantages for others.

Q: It seems to me that it might be that – if you extrapolate those species lost from your results to several thousand hectares of fuel reduction burns ...

Kevin Tolhurst: We are looking at species rather than individuals. If we had a particular bird species that was restricted to a 1000 ha block then you would be concerned about burning all of that block in one go ... but that is not usually the case.

Q: I was going to ask about the invertebrates – and you almost answered my question – you said no invertebrate species was lost, and I was wondering how you would actually get a measurement for that?

Kevin Tolhurst: From about 80,000 specimens. We've gone through and analysed them all. We measured for a one to two year period before any burning was undertaken and have measured burnt and unburnt sites simultaneously all the way through. The work for the beetles will be published for two fires and another paper for three fires in Australian Forestry soon. One of the things that is difficult about invertebrates is it's such a big group, so variable between seasons and from site to site, the longer we measure them – for instance, even after eight years we keep coming up with new species that occur only occasionally. There are so many species that we find only once or twice, its a difficult group to work with – one of the reasons there hasn't been so much work done on them.

Q: What was the reason for your study – the motivation for it – and does it relate to forestry practices in the Creswick area?

A. It doesn't relate to forestry practices in the area. After the last Monash symposium in 1983 on fuel reduction burning it was apparent that not a lot of progress had been made into the fire ecology issues. The Forests Commission then decided to make a commitment to look at ecological effects of fuel reduction and so concern about the possible effects of fuel reduction burning was the main reason for setting up the program and this has been continued through.

Q: But why is there such a lot of fuel reduction burning in that forest?

Kevin Tolhurst: I wouldn't say that there was any more than anywhere else. It just happens to be a convenient site to set up experiments. It's burnt about the same – in terms of proportion of the area. Is that what you're thinking is the case?

Q: It just seems to be hit really very hard with the fire regimes.

Kevin Tolhurst: Ah! This is experimental stuff – I'm trying to burn it as often as possible to have maximum impact!