Friday 25 November 2016

The value of open access

Back in 2012 I published a couple of book chapters on a similar topic: 'large scale agent-based simulations'.  Both were published by Springer. The first was for the Encyclopedia of Complex Systems, and as such set out some definitions and what I thought to be some insightful guidelines on approaching the challenges of large agent-based models. And by 'large' I generally mean models with large numbers of agents and or significant complexity that basically grind standard PCs to a halt...
The other chapter was for a book 'Agent-based models of geographical systems' edited by a good friend and colleague (and now recently professor) Alison Heppenstall at Leeds, with Mike Bithell (Cambridge) as co-author. That went into more depth with examples, especially around the why and how you might parallelise agents vs spatial decomposition across processors when taking a parallel computing approach.

Recently I was invited to submit an updated version of the former (Encyclopedia) chapter to Springer.  So, will I accept?

Now a little further on in my career I don't immediately leap at such opportunities as I did in the past, and I am forced to weigh up the benefits against the ever growing backlog of half written papers I should have published and the list of topics I would like to start work on. So, instead, I ask myself some questions, which are something like this:

- is this going to move my career further towards where I want to be heading?
- Will this be a significant contribution to my field?
- Is this book likely to be well read and cited?

In this case, the answers were:
- well, quite a few methodological papers came out of my PhD, but today I think I need to publish a lot more on the scientific understanding gained from my models as that is where the true value lies in 'modelling' for my career now.
- I think there are some significant advances made in this area since that book chapter, most notably Repast HPC which is designed specifically as a library to facilitate agent based modelling in parallel. I think this still focuses on parallelising agents rather than space, though it does have spatial modelling capabilities so its well worth exploring... However, on the whole I feel my 'guidelines' and suggested approaches still hold up well today, and revising may not add a great deal.
- I actually had no idea how well cited that Encyclopedia chapter was. Turns out, not very: 6 citations according to Google Scholar! Very dissapointing, particularly compared to the contemporary chapter I wrote with Dr Bithell that has been cited a much more respectable 28 times. One factor, likely a big one, is the accessibility. Now I have looked at these again on Google Scholar I realise the latter can easily be downloaded from there, but not the former!

Overall, I think I will decline unless it really will be a quick job to revise. But thinking through my response has been useful, particularly to consider making sure all my publications are as easily accessible as possible so they get read and cited!

If you are interested, here are the two book chapters  to download:

Saturday 17 September 2016

Introducing our symposium at ICE 2016!

What do you get if you cross two field ecologists with a spatial simulation modeller...?

 our symposium at ICE!

For some time now I have been thinking about the relationships between field ecology and modelling and how one can relate to the other more effectively particularly when it comes to relate movement ecology to population distributions (one thing that got me thinking about that was a 2012 paper by Oliver Restif et al. in Ecology Letters 'Model-Guided Fieldwork'... ).  At the same time, a key aspect of the spatial ecology work we do in the Pest Suppressive Landscapes team at CSIRO is consider how small scale observations and processes relate to larger scale phenomena, particularly of insect behaviour in agricultural landscapes.

And so, Nancy Schellhorn, Cate Paull and myself bring you our International Congress of Entomology (ICE) symposium for 2016:

Arthropod movement in agro-ecosystems: linking individual behaviours and population patterns across spatio-temporal scales. Just what does emerge?

With only a week to go before we depart Brisbane, Australia for Orlando, Florida, USA we are getting very excited about the conference, our symposium and our fantastic set of speakers!  So I thought I'd give you a run-down of what our symposium is all about here on my blog.

First, the official overview:
Speakers will consider how landscape-level patterns (e.g. population distributions, migration events) emerge from proximate movement behaviours (e.g. fitness seeking, oviposition behaviour, habitat selection, mating and predation), with application to arthropod pest management in agricultural landscapes. Both theoreticians and empiricists are invited to this session, with the aim to foster cross-disciplinary discussion on how we might better combine models and empirical data to reveal emergent patterns from movement ecology in agro-ecosystems at multiple spatio-temporal scales. We ask researches to draw on case studies from their work where specific emergent properties were identified, how the link was established between individual behaviours and emergent patterns, and at what scale, and to comment on what the results mean for area-wide management of arthropods in agro-ecosystems. 

 and the program (NB its on Tuesday afternoon) - please click to enlarge:

and to get you even more excited (so that we hopefully see you there!) here is the lowdown on each of our fabulous international speakers... (for full abstracts, please see the ICE website here):

First up is symposium co-organizer, CSIRO science leader and ICE Ecology and Population Dynamics Section Convener Nancy Schellhorn from Brisbane, Australia, who will be speaking about resource exploitation: the role of movement, oviposition, and the landscape context.  Nancy will bring together results from multiple studies of different insects in both urban and rural environments to make an interesting observation that the further mobile arthropods have to move to encounter a resource, the more likely they will exploit the resource once found.

Second, we have Assistant Professor of Sustainable Agriculture Megan O'Rourke from Virginia Tech, who is presenting a study she has conducted with Katja Poveda of Cornell Univ.: Linking herbivore dispersal and population dynamics in complex landscapes.  Megan and Katja make the observation that while there are dozens of studies relating land-use patterns to insect populations, there is relatively little clarity about whether landscape diversification could be used to suppress insect pests in agricultural systems.  They aim to gain some clarity by presenting  a conceptual framework for how land-use patterns can affect herbivore populations directly by mediating their ability to disperse in the landscape,supported by examples from the literature and an analysis of landscape-herbivore data.

Next up is Assistant Professor of Entomology David Crowder from Washington State University, talking about factors affecting movement of arthropods that transmit plant pathogens and implications for pathogen spread.  The work of his lab shows that movement of insect vectors at local and landscape scales is influenced by community composition, in turn influencing the dynamics of infectious insect-vectored pathogens. The community ecology of infectious disease has been grossly understudied, and their work suggests a need for more research into how the composition of ecological communities impacts vector movement and resulting pathogen transmission in agro-ecosystems.

Fourth, we introduce Mattias Jonsson a researcher at the Swedish University of Agricultural Sciences in Uppsala.  He presents work conducted with a number of colleagues that seeks to synthesize poorly understood mechanistic linkages between land use, biodiversity and ecosystem service provision in agricultural landscapes.  In particular, he presents a model to map biological control services of aphids across cereal fields in a Swedish agricultural region with varying landscape complexity.

Next is Joseph Spencer of Illinois Natural History Survey, University of Illinois, presenting work conducted with Sarah Hughson.  In their abstract they state that 'we must pay attention to what insects actually do in the field'! When observations and empirical data contradict the conventional wisdom, we should pay attention to the stories that the insects tell... drawing on 20 years studying the rotation-resistant (and now Bt-resistant) western corn rootworm in Illinois, Joseph tells us of the challenges local movement, mating, and dispersal poses for resistance management plans.

Before the break, we introduce PhD student from Rothamsted Research (UK) Aislinn Pearson and her team of supervisors.   Aislinn will be presenting her work investigating the effect that pathogen load has on insect movement behaviours: hosts may be unaffected by infection, they might evade infection by escaping contaminated habitats or infected insects may die during migration. The pathogen too may be affected by the migratory effort of its host, yet few studies have attempted to classify the relationship in this way.  She is getting some interesting results from her flight mill studies of fall armyworm, including that males but not females changing their flight behaviour in response to sublethal viral infection.

After the break is a second talk from Rothamsted Research, by research scientist Christopher Jones and his colleagues from the Insect Migration and Spatial Ecology group.  Chris takes us right down to the molecular level to gain a better understanding of large scale movement: connecting migration of pests to the genes and and biochemical pathways that drive this phenomenon, which are poorly understood.  Chris focuses on the flight propensity of cotton bollworm, where targeted sequencing and SNP analyses have revealed genetic variation in candidate genes that could contribute to key adaptations that drive long-range movement in H. armigera.  

Continuing the cotton bollworm theme, the second of the symposium co-organizers, CSIRO researcher Cate Paull from Brisbane, Australia, will tell us about her work with colleagues Andy Hulthen and Nancy Schellhorn on the landscape ecology of this pest in Australia.  Cate has been seeking to understand  key spatio-temporal drivers explaining their abundance across a region and over multiple seasons.  A key result of these studies has been to uncover that significant variation in H. armigera and H. punctigera moths caught in Bt cotton was explained by increasing proportions of Bt cotton at a landscape scale, illuminating male moth behaviour in agricultural landscapes with implications for resistance management.

Then its me, Hazel Parry, research scientist at CSIRO, Brisbane, Australia.  Also continuing the cotton bollworm theme, but this time with simulation modelling.  Obtaining empirical evidence on the efficacy of 'refuges', the cornerstone of the current resistance management system, is difficult: processes operate at multiple spatio-temporal scales, complex interactions exist, Helicoverpa are highly mobile and we must consider both the dynamics of the refuge and the landscape, as well as the pest. Agent-based simulation modelling allows us to explore emergent egg distribution patterns and landscape use by Helicoverpa based on empirical studies of underlying processes, such as movement and oviposition behaviour. I present the results from my modelling, validated against data from the field... and the model has something to say about refuge efficacy in space and time in relation to insect behaviours.

In fact, from my talk onward the rest of the symposium is primarily modelling! Next up is a fellow antipodean from the University of Lincoln (New Zealand) Audrey Lustig, a PhD student supervised by Associate Professor Sue Worner.  Audrey is using modelling to understand mechanisms that control invasive insect invasion, which are poorly understood.  She presents a general modelling framework for efficient evaluation of the relative influence of species life traits, propagule pressure and spatial heterogeneity effects on biological invasion.  This uses landscape measures to establish a quantitative relationship between landscape structure and population dynamics, and the pattern of invasion by multiple species are used to infer key drivers of invasion.

Taking an honourable penultimate 'double spot' in our symposium is researcher Matthias Becher, presenting his work and that of Professor Juliet Osborne (who sadly couldn't join us), both from the Environment & Sustainability Institute at the University of Exeter, UK.  The 'extra' time will allow Matthias to cover both the modelling and field ecology aspects of their work on bees, although as the modeller in the duo I anticipate he will focus primarily on his model, BEESCOUT.  He will discuss how detection probabilities of food patches in the landscape affect colony growth and survival and address the differences between honeybees (large colonies, recruitment to specific sites via waggle dances) and bumblebees (small colonies, no sophisticated recruitment) in terms of their susceptibility to find food sources.

Then, to our final talk.  Last but not least post-doctoral research associate in Natural Resource Ecology and Management at Iowa State University Tyler Grant and colleagues (including myself - his Monarch model originates from my agent-based model of Helicoverpa).  Tyler will present the model of monarch butterfly movement and egg-laying in an Iowa landscape.  Effective conservation efforts in Iowa and the Upper Midwest of the USA will require information on how spatial arrangements of created or restored milkweed habitat influence monarch host plant seeking behavior and egg-laying.  He uses the model to identify characteristics of habitat patches predicted to have higher egg densities, which will inform future conservation efforts.

What an afternoon this is going to be!

Thursday 1 October 2015

Why does a beehive represent the ‘anthropocene’?

I was inspired by the forthcoming event at UQ's GCI 'Anthropocene Slam' and got thinking about what object I would choose to represent the anthropocene.... a beehive!

Bee hives have only really been in mass use since the industrial revolution in the mid-19th century.  Its no coincidence that the intensification/industrialization of honey production occurred at this time, along with many other things!  A beehive represents humans controlling a natural ecological system, intensifying it, and exploiting it. 

Honey production is just one aspect of honey beekeeping.  The other is crop pollination.  For our mass global intensive agricultural systems that have developed over the past century we are now dependent on honey bees: e.g. Honey bees enable the production of at least 90 commercially grown crops in North America according to the White House.  Some crops, such as almonds, are almost exclusively pollinated by honey bees, and many crops rely on honey bees for more than 90% of their pollination. 

However, like many of our natural systems and ecosystem services that are being stretched to the max by intensive human use, honey bees are under serious threat.  Diseases such as colony collapse disorder and the pest varroa mite threaten to wipe out vast numbers across the globe. The causes of the losses of bee colonies are multiple – like many of our ‘anthropocene’ wicked problems, but most can be traced back to human resource (over)exploitation.  Loss of forage habitat, loss of genetic diversity, exposure to pesticides…
There are also spillover effects of the problems in the human-managed bee populations, such as the pests and viruses of honeybees also affecting native pollinators.  

Overall, beehives have allowed us to manipulate a natural ecosystem service to intensify our agriculture.  However, this has left our food production system very vulnerable now that beehives and honeybees are experiencing serious problems at a massive scale, posing a threat to our global food security.  I think this is a reflection of so many other ways in which humans have exploited a single natural resource to enable our population to expand, but have now taken it to a point where our ability to continue to expand our population to 9 bn and counting by 2050 is threatened by our very singular reliance on that exploitation and the effect it has … (cf. coal mining, deforestation…) 

but don't stop keeping bees!  On the contrary, we need to continue to support and improve conditions for both managed and wild pollinators, making sure we have diversity and resilience in our pollination systems both locally (in our gardens) and at an industrial scale.   

Thursday 27 August 2015

Invasion Hotspots

The latest review for F1000 by Mark Lonsdale and myself, on F1000 here
(yes, I am still doing those!!  But I rarely get time to post on the blog!)

Review of:

Modelling Hotspots for Invasive Alien Plants in India. by D Adhikari, R Tiwary and SK Barik
PLoS ONE 2015; 10(7):e0134665

Risk assessment for invasive species in the past has tended to focus mainly on species’ attributes, rather than the role of the invaded ecosystem. This paper explores the concept of invasion hotspots - regions that are potentially vulnerable to invasion - using a large set of open access species distribution datasets available from the Global Biodiversity Information Facility (GBIF). A novel aspect of their approach is that they don’t simply take into account the climatic niche, they also consider human ‘influences’ known to facilitate invasion processes by delineating ‘anthropogenic biomes’. Where climatic suitability combines with vulnerable ecoregions and anthropogenic biomes, this is considered a ‘hotspot’. The authors find that biodiversity hotspots in India are especially vulnerable as invasion hotspots, an important finding - especially so, as the regional status of invasive species in India has been comparatively little studied. It also illustrates the value for ecological science of mining open access biodiversity data. It would, however, have been useful to see scale factors explored a little more; for example, many invasion hotspots the authors identify are actually in protected areas, though these are presumably within anthropogenic biomes.

Monday 24 August 2015

Exciting job opportunity for a post-doc to work with me in Brisbane, please apply!

 CSIRO Agriculture currently have an exciting opportunity for a highly motivated Postdoctoral Fellow to join the "Pest Suppressive Landscapes" team based in Brisbane, Queensland. The team seeks to address the broad question: why does landscape context matter for the control of pests and diseases? By combining empirical ecological studies with mathematical modelling, the team aims to gain knowledge that will help address the global challenges of food security and health crises at the landscape scale. In this role you will be specifically focused on the spatial simulation modelling of fruit fly in agricultural landscapes, in order to estimate release rates of factory-produced sterile insects for effective area-wide pest management.

Thursday 16 January 2014

Bees have a chip on their shoulder

Exciting developments in micro tracking devices now allows CSIRO scientists to put tracking devices on bees.  A lot of 'buzz' has surrounded this news story, and its easy to see why - the potential for this sort of technology in understanding movement and behaviour of a wide range of arthropods is very exciting, no other methods currently come close.  They will continue to shrink the devices down, so who knows - maybe we will put some on aphids sooner or later!!

Friday 3 January 2014

Cereal aphid movement: general principles and simulation modelling

Cereal aphid movement: general principles and simulation modelling Parry HR. Movement Ecology 2013, 1:14

I'm really going in for the open access thing, this is my second publication in that format!  As aphids are widely studied and modelled, I hope that having open access to this review article will help people navigate the literature much more easily. There is an enormous amount of study out there on both the biology and ecology of aphid flight as well as many examples of models that address the different phases of movement. So, I put together a review of principles, parameters and methodologies to construct mechanistic simulation models of cereal aphid flight: including uplift, transportation, deposition and subsequent spread.  I've chosen to publish in the new journal 'Movement Ecology' as I think it will be an important scientific journal in a rapidly growing field going into the future, and I'm truly honoured to be one of the first papers published there. 

Many thanks to Andy Jensen, Eagle, Idaho, U.S.A. for the image of an alate R. padi shown above, that has been used as the thumbnail for the paper on the journal homepage.  Andy takes great aphid photos and kindly allowed me to use this image, thank you.