We get a lot of questions about Fire in the KNP and the surrounding Game Management Areas and as we are now quite well into the dry season and a lot of early burning has been done we thought that we could talk a little bit about fires. We will see more of them before the first rains – of that I have no doubt. I have an interest in fire in general. In my previous life I spent 15 years as a forest manager and I have spent far, far too many sleepless nights and hours fighting late season fires and talking to guests (who are often a bit shocked by fires) here in the Kafue to not be interested in the topic – especially how it fits into the broader ecology/landscape here as well as it’s role in shaping the landscape as we shall discuss.

Fire in General

It is generally known that Southern Africa is the ‘cradle of mankind’ and fire has been a part of man’s influence on the African landscape for thousands and thousands of years. But despite it’s close role with man it remains a highly contentious topic. So many guests that come through here express horror or shock at fire. It is inevitable I suppose. For our modern populations ‘wild fires’ (through whatever source of ignition and whatever time of year) are associated with destruction of life and property. The terrible news images from the US, Australia, Southern Europe are on our screens every year. Of course the reality is that like so many things the real root of the problem is not the fire but us humans. We are building houses, villages and towns in areas where there were no settlements before. We have tamed the landscape, planting and protecting trees and vegetation and settlements from fires. In places the vision of fire as the enemy led to many years of fighting it like a war – extinguishing it wherever it occurred, allowing fuel loads to build up year upon year and then resulting in massive destruction when fires took hold. The United States Forest Service and attitudes that have changed over the last years is what springs to mind here.  The USFS for decades excluded fires and fought them regardless of when, where and how the fires had started.  Only over the last couple of decades has this attitude changed and the use of prescribed fires started.  Not all fires are equal. Seasonality, wind, frequency of previous fires, air temperature, dew point all play major components in how a fire will burn and what destruction or indeed benefit it will bring. It is not a case of all fire being bad and no fire being good. Like everything to do with ecosystems (and man’s place and influence in them) it is far more complex than that.  An interesting National Geographic article examining the historical USFS attitude to fire (and associated problems caused by this) can be found here: 

National Geographic Fire Article

Dry season fires here in the Kafue are almost all started by man. We do not have significant electrical storms of any significance until the onset of the rains therefore the only source of ignition is man made until generally October or so. Such fires are known as anthropogenic. The early Portuguese explorers, who rounded the Cape of Good Hope in the fifteenth century, referred to the interior of South Africa in their ships logs as “Terra dos fumos” – the land of smoke and fire. What makes Africa distinct from other areas or continents is how long man and fire have been interacting in the landscape here. Estimates mention figures like the earliest evidence of human use of fire being 1.5 million years ago. We have some history here together in Africa us humans and fire! The distinct and defined wet and dry seasons here create perfect conditions for fires. Tropical savannahs around the world have been moulded by fire and here is no different. So while a huge wilderness area like the Kafue is often perceived as ‘untouched’ by man, the reality is very different. Anthropogenic fires (i.e. started by man) have been shaping the landscape here for a couple of thousand years since the iron age. And this is a difficult concept to grasp at first. When you come around a bend in the road and are confronted by a fire the natural human first reaction is not to think “ah, fire – that has been part of the shaping of the Kafue landscape for a couple of thousand years!”. The first natural reaction is that it is out of place, dangerous and destructive. If it is in October then it possibly is. If it is mid-June then the general reality is quite different as we shall see.

Early burning a protective black line around Kaingu property

Fire and the vegetation

Here in the Kafue we are very much in the Miombo belt. Miombo is the vernacular name given to the savanna woodlands found here which are dominated by a few species of tree. It is woodland in the true sense – not fully closed canopies and interspersed with grassy areas. Miombo is said to cover about 2.7 million square kilometres of South and central Africa. Miombo woodland is dominated by trees of the genera Brachystegia, Julbernardia and Isoberlinia. Being not a closed canopy forest but rather a woodland, the landscape is also rich in other plants in the under-storey (particularly grasses). The larger mammal inhabitants of this landscape are dominated by elephants, buffalo and the larger antelope such as Lichtenstein’s hartebeest and the magnificent sable antelope. However the soils in miombo woodland are poor and this is reflected in nutrient poor grasses. The miombo woodland areas does not support the abundance of animals that richer grasslands or woodland types do. This factor together with the inevitable tsetse fly is why cattle grazing has not dominated the ecosystems here. Miombo trees have evolved deep tap roots to cope with the lack of nutrients and the long dry season. In fact a large, mature miombo inhabitant tree may have a tap root 5-6m deep. In the dry season the senescent grass provides the fuel for fire which burns through the forest floor – again all of these fires are anthropogenic in the dry season.

So why did man burn the African savanna and why still? There are many reasons, some ancient and some new. The main reasons being hunting, creating fresh growth for grazers (domestic and wild) and to clear areas around habitation to reduce wildlife threats and allow cropping. These reasons still exist of course but can be added to by factors such as protective burns around habitations, to increase visibility (tourism), more incidental causes such as honey collection (which for sure are also ‘ancient’) and of course quite modern factors such as a cigarette flicked from a vehicle. Early burning (which here can be described as from the end of the rains until mid July, depending on how long the rains persisted) for the above reasons have always been preferable; the earlier the grass can be burned to remove wildlife threats and promote grazing or to allow better hunting (for example) the better. In our case we do early burns to protect property and to stimulate fresh growth thereby encouraging grazers and open sight lines for tourist game viewing. Again, the earlier the better. We also burn areas such as the Kaingu game viewing loops in the National Park so that if late fires come through they do not cause havoc – an early burned area will stop a late season fire due to the absence of fuel load. We would rather have fresh grass (and grazers) in July through to October than absolutely nothing but scorched earth in October after a late season fire.

These early fires burn relatively quickly and cool in contrast to fires later (say October) when vegetation is at it’s driest. It is estimated that Miombo in general burns with a frequency of every 2-4 years. It is further acknowledged that this frequency is probably underestimated and is in fact more. Permanent sample plots and studies of fire and miombo interaction have been carried out for 36+ years now. Without going into very dry descriptions and tables of basal area (after all this is for general interest!) these studies can be summarised as follows:

  1. Early season fires in Miombo woodland have little or no impact at all on diversity, basal area or surface litter.

  2. Late burning however suppressed woodland development and had the opposite effect of early fires (with the exception of nine species that proved to be considered as fire resistant.  It should be noted however that these changes obviously take time.  It is very rare for a late season fire here to ‘crown’ and actually burn standing trees in number.  

  3. Complete fire protection in Miombo results in the formation of a closed canopy forest, reducing the diversity of under story species (and of course the grazers that utilise them) end eventually results in the change from miombo woodland to dry evergreen forest – something that we are not keen to promote.  

For all those reasons it has, for many years, been the policy of foresters and ecologists to promote forest development and conserve biodiversity through early burning. It could be argued that the Kafue National Park (and surrounding GMAs are perhaps burning more frequently than the above estimate of 2-4 years, but research has shown that this frequency anyway may well be a general underestimate. In the absence of enough manpower, equipment and firebreaks (as well as an integrated fire management plan and recording) this is inevitable and the key point (to repeat) is that early burning has no impact on diversity and forest development.

Early season burning in Miombo woodland.  You can see how the slow, cooler fire is basically only consuming grass.  

Of course as we have seen from point two above, late season fires can have the opposite effect. This effect is further compounded if there has been no early burning for several seasons and fuel loads have been built up. Then the late fire will be extremely destructive as well as landscape changing, and not to mention effect on other inhabitants of the ecosystem as we shall discuss later.

But of course the Kafue is not purely Miombo woodland. The woodland is interspersed with grassy dambos and larger open areas. So what effect does fire have on pure grassland environments? There has been plentiful amounts of research on this – mostly originating in South and East Africa. As we have discussed earlier, fire is an intrinsic part of maintaining savanna grassland, where burning allows grass to regenerate but shrubs and trees not to, or to be more coppiced, scrubby and not to form real woodland or forests. Again, this shows up the influence of man on what are perceived to be ‘pristine’ environments but in reality have been shaped for centuries if not millenia by the powerful tool that is fire.

Controlling early season protective burns.  The use of drip torches and backpack extinguishers allows fine control. 

Studies of grassland fires have shown that whether it is a late season or early season fire there is little effect on recovery in the next growing season. However experience and common sense dictates that obviously early season fires are better in that fresh growth will still occur after an early fire, the fire will be easier to control, less dangerous and less destructive to other living organisms. We see this year in year out in the Kafue: a destructive late season fire results in a charred ‘moonscape’, but the next season, come February it is like the fire never happened. So in effect grassland fires can be said to be like the Miombo in that early burning is better, but probably from the point of view of the inhabitants (Including man) and aesthetics rather than the actual grassland itself.  But before leaving the topic of fire and flora it is also worth pointing out that late season fires can in certain circumstances actually be of benefit and be used as a management tool.  For example in areas where fires have been excluded for many years thicket development can be checked and reversed by a controlled, hot late season fire.  Interestingly this is very visible around camps here in the Kafue that have been around for many years and where (obviously) fire has been excluded.  Eliminating this thicket development is possible with late season burning, but obviously there are serious safety implications with late fires in the direct vicinity of people and property.  A real fire management challenge.  

Fire and the fauna

But what about the inhabitants? If early season fires are not really damaging to the wider landscape and the vegetation, what about the fauna and what impact does it have there? Studies here are not as plentiful as the impact of fire on vegetation, probably as the concept of non-consumptive tourism based around wildlife (and indeed what we know as ‘conservation’) is relatively new, whereas say managing grassland for increasing grazing potential is far older. Now I cannot claim to have done degree level reading on the subject, but a cursory examination of available and pertinent literature throws up some interesting descriptions and facts. Again, it is quite clear that cooler, earlier fires are generally better. But overall, like the vegetation, it would appear that fire (especially early ones) has little effect.  Again when one considers how long fire has been part of the landscape here this should not really come as a surprise.  

The influence of fire on the animal kingdom has been broken down into two distinct areas of influence. Firstly comes “First Order” fire effects which are the obvious ones – animals physically burned and injured by fire. “Second Order” is the subsequent influence of fire affecting habitat and ecosystems and that in turn affecting the inhabitants – in other words a more downstream affect from fire.

As a broad indication of how the animal kingdom has evolved along with fire here are a few observed examples (not confined to Africa alone) of how animals respond to first order fire threats:

  • Studies showing that juvenile tree frogs when played sounds of fire moved to take cover in fire resistant sites – i.e. very wet grass that cannot burn.

  • Birds flying just ahead of the fire line and finally moving into thickets that will not burn.

  • Lizards burrowing into the soil and climbing trees to avoid fire.

  • Voles finding shelter underground or moving onto bare ground to avoid fire.

  • Snakes retreating underground.

  • Adaptations to take advantage of fire: certain beetles can sense fire up to 5km away and then move into the burned areas to lay eggs in trees weakened by fire. Indeed over 30 sepecies of beetle so far are know to do this.

  • Woodpeckers thriving in burned areas because of them preying on the above beetles and their offspring.

  • It is to be remembered that 90% of terrestrial arthropods spend a large portion of their lives burrowed in the soil or the soil litter.

Again, our subjective ‘gut feeling’ tells us that a lot of these examples ring true. Anyone that has been here when the cicadas are in the trees in the late dry season would realise that even late season fires are clearly not damaging the cicada population that have yet to emerge from the soil. 

What has been clear from research is that early patchy burning (certainly not all vegetation burns in early fires no matter how much drip torch fuel is expended!) is the primary goal. Patch burns and burning of different areas on subsequent years allows refuges where (particularly insects and lower order mammal) populations can shelter and then breed and repopulate burned areas extremely quickly. Like say grassland regrowth, the repopulation can be very rapid and by the following year the effect of the early season fire can be totally negated or even seen.

Clearly more mobile fauna are less likely to be first order casualties of fire. Birds, quick moving mammals etc are able to escape early fires. Inevitably there will be casualties, very new infants spring to mind. Without wanting to sound callous, this is clearly part of the natural order, and with early fires the potential to find areas that will not burn, already burned areas or to simply flee the fire area is high. Slower moving larger animals of course are a different case. But again, fire is not always a recipe for doom: a study indicated that 30% of large Florida Box Turtles showed carapace damage from fire. Although clearly this was not mortal damage. My own experience, on a far lesser scale, was of seeing a huge leopard tortoise that I was horrified to see in an area where we had just done some early burning. On close examination it was totally unharmed and proceeded on after the fire had passed (it did hiss in quite an annoyed fashion when examined for burn injuries!).

I feel that this write up is starting to get lengthy and possibly digress from the overall goal. The overall goal (to repeat) is to try and allow visitors an insight into why early season fires are not necessarily bad, and to why they occur and what the effect is. I am conscious that birds have not featured heavily in my descriptions so far, so in our penultimate conclussion I will cut things short and simply past the final conclusion of a paper from a student of the Percy FitzPatrick institute of African Ornithology:

Managers of savanna ecosystems can take comfort from these findings. There is clearly little reason to be concerned over the medium-term effects of fires on bird communities, even for the most severe fires. Some savanna species may be dependent on fire-created habitats for breeding. However, these species tend to be highly mobile, enabling them to locate suitable breeding habitat irrespective of the fire management strategy employed, other than complete fire suppression”


Once again the obvious conclusion would the that (taking all that has been said into account) the goal of total fire suppression would be a goal to be avoided at all costs. This holds true be it for the vegetation (and the grazers, and in turn predators that depend on them), the insect life and indeed the birdlife. There will be (for sure) first order casualties, but like all complex systems this has to be balanced and evaluated in the context of the greater good. It would appear that early burning (which inevitably allows un-burned areas) is an overwhelming positive force that has existed for hundreds (if not thousands) of years and is now an intrinsic part of what shapes the landscape – certainly no more and no less than the rains, the lack of rains, the global weather patterns, and ultimately (and certainly the most profound) the influence of us humans. The overall goal it would seem is that early burning must be done with a degree of science and planning. At the moment in the Kafue the lack of fire breaks, man power and equipment means that the early burning is usually quite random in nature. The goal should be that fire management becomes a far more proactive and integrated part of the management of the landscape. However in the absence of such infrastructure and management it would seem that a strategy of fairly random early burning is not a strategy to be ashamed off and is better then the alternative.  

Perhaps the most appropriate final words should go to Phillips, J.F.V, writing in 1965 who said that (fire in sub-saharan Africa is): “a bad master but a good servant”.

Sign up to receive our monthly newsletter on what's happening at Kaingu Lodge.

We don’t spam! Read our privacy policy for more info.

Sign up to receive our monthly newsletter on what's happening at Kaingu Lodge.

We don’t spam! Read our privacy policy for more info.