Hot News Impacts

Kakadu – a climate change impacts hotspot

When ecologists, policy makers, or the public, think about the visceral impacts of climate change on Australia’s natural systems, World Heritage listed Kakadu National Park (KNP), located in the seasonal tropics of the Northern Territory, is high on the at-risk list. But looking deeper into the human-driven processes now threatening KNP, there is actually a synergy of interrelated problems requiring simultaneous management – a situation common to most biomes threatened with global warming (Brook et al. 2008).

The big issues for KNP are changed fire regimes (impacting savanna and rain forest communities), rising sea levels (affecting the floodplain wetlands), and a suite of invasive weed and feral animal species, operating across all three major ecosystems. All three threats have a climate change component, although for fire and ferals, not wholly.

The savannas, which by area make up the largest part of KNP, are at first glance apparently intact. There has been relatively little clearance of the woody component (dominated by Eucalyptus tetrodonta and E. miniata); indeed analysis of historical aerial photography has documented vegetation thickening linked to elevated atmospheric CO2, which favours the growth of woody C3 species (Banfai & Bowman 2005). However, an emphasis by Park managers on avoiding hot late dry season fires, has meant that a large proportion of KNP is burnt during the dry season, with a return time of 1 to 5 years (Williams et al. 1999).

The impact of regular early season burning on the Park’s biota and on the structuring of understory vegetation, is a topic of ongoing debate and research. Nevertheless, some long-term studies have explicitly linked high fire frequencies to species declines (Pardon et al. 2003; Andersen et al. 2005). Climate change, via increased temperatures or shifts in the timing and intensity of monsoonal rainfall, will likely enhance future fire risk (Parry et al. 2007).

The KNP wetlands support a rich and spectacular biota, including vast flocks of magpie geese (Anseranas semipalmata), which congregate in millions to feed on Eleocharis chestnuts growing on the floodplains of the Alligator Rivers system. These wetlands formed ~6,000 years ago after sea level stabilisation, following a post-glacial rise of 120 metres. Ironically, additional sea level change associated with anthropogenic global warming threatens their future viability (Brook & Whitehead 2006).

About 20cm sea level rise occurred during the 20th century. At least double that amount – and potentially >1m due to accelerated polar ice sheet melt – is predicted by 2100. Rising sea levels, in combination with intense tropical storm surges, increases the regularity and severity with which saline flows penetrate the low-lying freshwater wetlands. At the mouth of the Mary River, to the west of KNP, extensive earthen barrages have already been built in an attempt to alleviate the damage caused by salt water intrusion.

A complex network of low-lying natural drainage channels, enlarged or cross-connected by movement of feral animals such as Asian water buffalo (Bubaus bubalis) and pigs (Sus scrofa), means that even a few tens of centimetres of additional sea level rise may be sufficient to degrade or eliminate a large fraction of the floodplain communities (Traill et al. 2010). What remain will be isolated patches of freshwater wetlands within a mire of brackish swamps and saltwater mangroves.

Beyond their impact in facilitating saline intrusion of the wetlands, feral ungulates help spread weed species such as introduced pasture grasses and Mimosa (Bradshaw et al. 2007), which compete with native plants for space and nutrients. Climate change will also cause shifts in the relative ability of invasives to compete with indigenous species, especially if natives are also under stress from herbivore grazing, changing habitat quality and altered fire regimes (Rossiter et al. 2003; Brook 2008).

KNP inevitably faces a tangled web of mutually amplifying processes associated with global change. Given the global failure to achieve significant carbon emissions mitigation to date, it seems that ‘adaptation’ and ‘building resilience’ will be the buzz words for KNP conservation managers for many years to come.

By Barry Brook

Barry Brook is an ARC Laureate Fellow and Chair of Environmental Sustainability at the University of Tasmania. He researches global change, ecology and energy.

12 replies on “Kakadu – a climate change impacts hotspot”

An issue is whether the perceived optimum fire frequency has been conditioned by millenia of indigenous occupation or is that purely caused by lightning strike. I’d guess maintaining a fleet of water bombing aircraft is not cost effective, assuming the idea is to reduce the extent of burns. Saltwater incursion will affect not only the NT tropics but places like the lower Murray River in SA. At least temperate zones don’t have cane toads, yet.

Note that Ranger uranium mine was excised from Kakadu by legislation. Sen. Ludlam wants it returned to the bush. These changes to the savannah appear gradual at least until a severe El Nino. However an irreversible coral bleaching event say off Queensland will have an immediate economic impact.


Large feral animals like buffalo,camels and horses can be controlled by aerial shooting regardless of the usual bleating from the bleeding hearts.This has been done sucessfully for horses in the Central Highlands of QLD.

NSW has a R Licence system which allows shooting of ferals in state forests under strict control.I believe SA has a scheme where hunters are used in NPs to try and control the goat scourge which is particularly bad in the Flinders Ranges.Maybe NT should look at this.

Trying to defend against sea level rise is a total waste of resources.The Kakadu wetlands are going to go under salt water.Get used to it.

It has always mystified me as to why some land managers seem to regard the Aboriginal fire regime as some sort of gold standard for Australian bush.The Aborigines burnt country for their own hunter/gatherer needs.What we have now is the result of thousands of years of that regime.Is that the best we can do?

Feral “weeds” are extremely hard to control let alone eliminate.Buffel grass is a big problem in the Territory where it was introduced as a catle fodder but has now resulted in extremely hot burns which are changing the native vegetation mix.

Weeds and climate change are not reversible.Better we concentrate on what we can do.


podargus: For a certified and unashamed bleeding heart like me who
cares about animals and ecosystems, feral animals raise serious dilemmas. My default position is that people who have, by their food choices, driven the vast bulk of extinctions and plant introductions in Australia


need to establish a little credibility before I believe that their calls for helicopter
shooting of anything are worth taking seriously. I remember being in a cattle
property on the cooper many years ago with an avid pig hater and killer … in front of us was about 40 meters of trampled “rooted” river bank which
caused a flood of anti-pig abuse … on the other side of the river were large
mobs of cattle and trampled river bank as far as the eye could see … not
a word … not a murmur … but that pig was going to die for its sins.

One of the really interesting things about KNP is that it provides a great
example of where informed opinion was dead wrong when subjected to
rigorous scrutiny. The monsoon rainforests aren’t just thickening but
expanding in area … despite the pigs, the buffalos and the fires:

Kakadu will change with climate change. But some of the big dilemmas will
be whether we sit and watch, or go in with guns and chemicals blazing. Tough


Great article.

Very good points too, Geoff Russell. I agree, if we continue to severely overgraze much of Australia, feral population control will probably have fairly small benefits overall. Kakadu is a good example of a remnant “island”, surrounded by a lot of land hostile to native species, due to human use – as is the case with much of Australia’s remnant ecosystems. There’s little room for dispersal, little resilience from external pressures, continued species invasion from outside the park, decreasing edge effects etc.

That said, feral population control (i.e. animal culling, weed control) can be beneficial in the short term, by keeping internal threats to biodiversity in Kakadu at bay – if done properly. Corey Bradshaw wrote a good article on some computer models he (and others) worked on specifically to improve efficiency of feral animal control in Kakadu, which he won a Scopus Young Researcher of the Year Award for.


Tom, yes, I was a co-author on Corey’s article that you linked to, and indeed, I wrote the initial code for that model, and worked with Corey to develop all the scenarios. We later handed it over to Clive McMahon to complete (who ended up packaging this into a paper and hence is first author). I’ve done plenty of computer-assisted killing in my time…


Sorry, Geoff, I probably should have been more clear. I was mostly agreeing with this statement: “people … have, by their food choices, driven the vast bulk of extinctions and plant introductions in Australia.”

I have no real ethical issues with animal culling as long as it’s done properly, and a real benefit is actually derived.


There is a large swag of volunteers keen to go into national parks and shoot ferals. Due to some sort of cultural elitism their aspiration is too often rejected rather than embraced. Shooting ferals ought to be a national pass time. We just need the gun phobics to grow up a little.


I’ll read the McMahon paper, but have already found various stumbling blocks … like is 13-23 million a credible estimate for feral pigs in Australia?
About half as many to almost as many as there are cattle? CSIRO want to charge
me $25 to read the paper that makes this
estimate, but my gut response is ROFL. How can this many pigs only
manage the estimated $100 million in agricultural losses? They
obviously aren’t trying very hard … pretty wussie pigs.


Sorry, Barry. I totally overlooked the fact that you were one of the authors of that paper. Credit where credit is due!

Geoff, I couldn’t access the full paper either, but the abstract (I found here) starts off stating “The number of feral pigs (Sus scrofa) in Australia has been variously guessed as 0.5-1.5 million (McKnight 1976), 8.7-10.1 million (Flynn 1980), 3-6 million (Tisdell 1982) and 8-12 million (Cutler 1989).”

So the upper limit of the Cutler estimate was 12 million, while the Hone estimate was 13 million minimum. These papers were released in two successive years, and that’s a large difference, so it’d be interesting to compare methodologies. I’m sure Barry will have something more useful to say about it than I, though.


Tom, yes a got the abstract and noticed how dodgy it looked. Also found
an interesting statement
here, namely
that “Illegal hunting and release of pigs for sport is a major issue. Illegal release is the major cause of feral pig movement across the state. DNA testing has shown feral pigs have been introduced to lower south west from more than 150kms away. DNA testing and tracking devices to understand their potential migratory range has revealed that will naturally travel only about 20kms. ”

Having to release pigs to hunt is also not consistent with such a large estimate. Of course, releasing your own local pigs to hunt would probably be deemed climate friendly by hunters, because it cuts down their “food miles” :) … assuming they eat the pigs.

But back to KNP. I’ve emailed Lochran Traill to try and get a copy of the
paper Barry cited … I’m interested in the relative contribution of hunting and
habitat fragmentation to its predicted quasi extinction of magpie geese.


I’m a co-author on the Traill et al 2010 paper too Geoff, so I’ll send you a copy. Lochran was my PhD student and then postdoc.

The Australia-wide pig numbers are just guesswork, based on extrapolating plausible densities across a range of ‘strata’ habitats. For our Kakadu modelling, we used densities of pigs that were realistic for the park, stratified into a range of habitats (e.g. lowland savanna woodland, escarpment, wetlands etc.) giving total K of about 100K pigs in Kakadu.


TerjP: Australia never comes even close to meeting the kangaroo quota, and
that’s paid work. There are some people who love to spend the occasional
weekend killing things, but feral animal control is far too much like hard
work for most tastes and ground based shooting is about as efficient at
controlling feral animals as hunting elephants with toothpicks … which is why
it is rarely employed … except as a PR tool by hunting groups :)


Leave a Reply (Markdown is enabled)

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

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s