Practical guidelines for modelling post-entry spread in invasion ecology

My first open-access journal article has just been published!  It is part of a special issue in the new journal 'Neobiota':

Parry HR, Sadler RJ, Kriticos DJ (2013) Practical guidelines for modelling post-entry spread in invasion ecology. In: Kriticos DJ, Venette RC (Eds) Advancing risk assessment models to address climate change, economics and uncertainty. NeoBiota 18: 41–66. doi: 10.3897/neobiota.18.4305

If you follow the link above it takes you to the paper on the journal's website, and from there you can download the pdf.  This paper has evolved over the past couple of years out of discussion with the co-authors and other members of the International Pest Risk Mapping Workgroup, so quite a bit of discussion and thought went into it over time.  We try to give something of a 'roadmap' for modellers faced with a wide range of pest risk management questions relating to post-entry spread in invasion ecology, taking into account considerations such as spatial and temporal scales but also constraints such as, inevitably, time and money.  In the end I hope we have come up with something that is useful!

The following diagram (figure 4 in the paper) is a good summary of some of the paper's concepts, but requires a fair bit of explanation that can be found in the paper, along with a lot more exciting stuff, like dealing with uncertainty!

Flow diagram to illustrate the modelling process with concepts from the paper (please click on image to enlarge).

Presentation at the 2013 Australian Summer Grains Conference

I recently presented some of the 'hot off the press' findings from our Grains Research and Development Corporation project 'Pest Suppressive Landscapes'.  The talk was given at the Australian Summer Grains Conference 2013 and uploaded to their website as part of the proceedings.  This is a fantastic way to share the work that we are doing and they have made a great job of the recording. 

Hazel Parry
Pest suppressive landscapes: Understanding the role of habitat on Rutherglen Bug populations

Plant invasions and extinction debts

I am rather behind on posting reviews to this blog, apologies!  We continue to review papers for F1000 and I will continue to do so with Mark Lonsdale even though his role is now changing at CSIRO (he has stepped down as chief of Ecosystem Sciences after serving a long and successful 'chiefdom'!) . 

At the end of April we highlighted the following article, particularly because it questions the timeframe over which we consider 'invasions' to play out: they are probably much longer than many of us have been thinking!

Gilbert B and Levine JM, Plant invasions and extinction debts.  Proc Natl Acad Sci U S A 2013, 110(5):1744-9 

Our review:  F1000Prime.com/718000610#eval793474985  

The idea that invasive species are a distinct threat to biodiversity has been disputed in recent times. One of the lines of evidence adduced has been the paucity of examples of invasives driving native plant species to extinction [1]. This paper uses a combination of metapopulation modelling and field experiments to show that, for a suite of Californian native grasses, extinction due to competition from invasive grasses is a real possibility – it just takes time.

The authors demonstrate experimentally that the native grass populations retreat to refugia of ever smaller size and less favourable conditions, while the habitat between these refugia becomes less hospitable for seed production and establishment through competition with the invasives. Metapopulation models show that the populations become increasingly vulnerable to local extinction with less likelihood of recolonisation because of the distance from other viable seed sources.

Their modelling suggests that these extinctions may take hundreds of years to play out. While this is slow by comparison with direct habitat destruction and the likely impact of climate change, extinction is a long-term consequence of present, profound, and insidious changes to ecosystem processes.

 

References

1.  Don't judge species on their origins. Davis MA, Chew MK, Hobbs RJ, Lugo AE, ..., Ehrenfeld JG, Grime JP, Mascaro J, Briggs JC. Nature 2011 Jun 9; 474(7350):153-4

The march of the Cane toads

Despite the name of my blog, its not often I write a post about Cane toads!  But recently a study came out that uses modelling to propose an interesting solution to halt the progress of the Cane Toad across North-western Australia.

Identifying optimal barriers to halt the invasion of cane toads.Tingley R, Phillips B, Letnic M, Brown G, Shine R, Baird S. J Appl Ecol. 2013 Feb; 50(1):129-37

Our Review: Lonsdale M and Parry H, 2013.
F1000Prime.com/717979948#eval793471458
Cane toads! Is there any way of stopping their march across northern Australia? This paper uses spatial modeling to suggest an economically and technically feasible solution. The model shows not only the potential importance of man-made waterholes to the invasion process, but suggests a control strategy that targets just these key points in the landscape.

Although this approach doesn't take into account landscape heterogeneity in the dispersal kernel, it does acknowledge that dispersal is not a uniform process, in this case driven by a series of waterbodies as foci of the invasion front. The authors feel their estimates may be conservative due to their treatment of rainfall variability, but it could also be argued that they may underestimate spread from vehicle movement (particularly with the expansion of mining in the region), and the model assumes colonization is very rapid following the presence of two toads at a site. Both of these issues simply highlight the complexity of the simulation of invasion spread. The stochastic approach taken here, that allows for an exploration of the most likely scenarios in relation to a potential management solution, is a good one that could easily be extended both to incorporate more complexity and to simulate other biological invasions.

 Image from the paper (figure 1): "The study area in Western Australia. The black arrow in the lower left-hand corner shows the location of the De Gray River, which was used as an endpoint in all simulations. The dark grey region below the black arrow shows the extent of the Pilbara region. In the inset, only waterbodies that lie within the predicted distribution of toads (grey shading) are depicted."

The 100 worst

The Invasive Species Specialist Group (ISSG) is a global network of scientific and policy experts on invasive species.  They have developed a method of scoring invasive species to come up with a list of the 100 worst.  This was recently reduced to 99, as Rinderpest virus has been eradicated ('from the wild').  That leaves a spot remaining!  So, how to fill it?  Well, this is open for public debate on their website to find the new 100th villain.  Their criteria? Firstly, they developed a shortlist of 50, from:
-  Species with a high number of appearances on current 'hit lists', both international and national
-  Emerging threats: potential major impacts or rapid spread rates of species not currently problematic.

From these, they assigned a score of (i) spread - the number of continents already invaded, and (ii) impact: either ecological, economic and/or health.  Then, they eliminated those whose genus is already represented in the top 100, or those whose value can also be 'positive' (economic, ecological or health).  This leaves just 10 candidates...

I think my vote would probably go to the triffid-like Giant Hogweed Heracleum mantegazzianum, although the aquatic plants on the list also sound particularly threatening. 

However, it was a surprise to find the beautiful Lionfish Pterois volitans on there - I have often admired this in large aquariums although I was aware it is not one you would easily keep at home, being rather poisonous!  If this piscivorous fish becomes more widespread it's impact on populations of reef fish is anticipated to be enormous - as if there weren't enough threats to coral reefs already with climate change and coral bleaching.

Radio Interview: Pest Suppressive Landscapes

My first radio interview!  This is a summary of where we are at so far with our project on 'Pest Suppressive Landscapes'.  Particularly the relationship between insect pests and predators with non-crop vegetation in the landscape, and how this might be managed to better promote predators.

Please click here to access the mp3 of the interview.

 Image shows two of the 7km radius landscapes (circles) that we studied in Queensland.  We have four more landscapes, two in New South Wales and two in Western Australia.