National parks on land and in the ocean are dying a death of a thousand cuts, in the form of bullets, hooks, hotels, logging concessions and grazing licences. It’s been an extraordinary last few months, with various governments in eastern states proposing new uses for these critically important areas.

Australia’s first “National Park”, established in 1879, was akin to a glorified country club. Now called the “Royal National Park” on the outskirts of Sydney, it was created as a recreational escape for Sydney-siders, with ornamental plantations, a zoo, race courses, artillery ranges, livestock paddocks, deer farms, logging leases and mines.

Australians since realised that national parks should focus on protecting the species and natural landscapes they contain. However, we are now in danger of regressing to the misguided ideals of the 19th Century.

Parks under attack

In Victoria, new rules will allow developers to build hotels and other ventures in national parks. In New South Wales, legislation has been introduced to allow recreational shooting in national parks, and there is pressure to log these areas too.

Is harvesting trees for timber really appropriate management of national parks?

Late last year, NSW announced a new trial to re-instate grazing in the new Millewa National Park and other reserves, following Victoria’s unsuccessful attempt to allow grazing in the Alpine National Park. And just this week, the Queensland government passed new laws that allow graziers to feed their stock in national parks during droughts.

It’s not just the land that’s under assault. NSW recently lifted bans on shore-based recreational fishing in most state marine sanctuaries. Coastal marine parks in Australia are mostly young, small (particularly the sanctuary zones), and poorly resourced. But they are vital for regulating human activities and making coastal ecosystems resilient to pollution, invasive species, resource extraction and climate change.

The picture is grim and set to get worse. In Queensland and Victoria, land clearing laws (outside of national parks) are being “relaxed”, with at least two major impacts. First, it will place an even higher value on our reserves, as more land is cleared and further degraded. Second, it will decrease connectivity between remaining patches of native vegetation, further threatening species that require large, connected habitats.

Set against a background of rapidly changing climates (and associated changes in storm and fire frequency, droughts and floods), many imperilled species will face range contractions at best, and full extinction at worst.

Parks already doing too little

Orange-bellied parrots – despite their breeding habitat being protected, they are still in decline. But they don’t need any more impacts, inside or outside national parks.

Why should all this matter?

It’s widely acknowledged that our current reserve system and efforts to conserve our native biodiversity are eminently praiseworthy, but hopelessly inadequate. Indeed, prolonged government failure to implement existing environmental laws and draw up plans for threatened species has recently resulted in court action.

The problem isn’t limited to Australia.

Biodiversity in many protected areas around the world is declining due to encroaching threats from surrounding areas. There is no free lunch; as parks suffer, their biodiversity suffers too.

Management interventions such as feral animal control, fire management and at times, grazing management, can be useful tools to achieve conservation goals in some circumstances. However, these need to be based on the best available ecological knowledge and practice and be aimed at conserving biodiversity. This is not the motivation for any of the recent changes.

Exploitation of our parks, without scientific evidence for positive biodiversity outcomes, will hasten losses. These areas need to be in the best shape possible to cope with the intensifying pressures imposed by a disrupted climate and likely increases in the frequency of species invasions.

Australia’s rich biodiversity is one of the few things our country has that is truly, globally unique. It is worth billions of dollars to our economy, and provides crucial natural services that are not easily replaced. Beyond their value to plants and animals, our national parks, wild places, and nature in general, are “good for us“.

Desperate times call for smarter measures

To illustrate how we, and the governments representing us, are failing to make use of the best available science to aid park management, and the consequences this has, we draw attention to two issues: grazing and pest animal control.

Nobody questions the stress graziers face when their stock begins to starve as drought intensifies in parts of southern and eastern Australia. But simply opening the fences to national parks is a dangerous precedent that provides, at best, a Band-Aid solution to a recurring problem.

Moving stock from pastures to parks increases the risk of spreading weeds and further [degrading natural habitats for birds and mammals, as well as sensitive water resources on which we and our livestock depend.

The current trend in drought-relief programs is helping farmers prepare for droughts. We no longer rely on emergency measures to cope with droughts that are expected and recurrent. Opening parks to grazing does not fit this model.

Australia needs to get smarter. We should do more to encourage flexibility in our agricultural and aquaculture systems. Why don’t we produce animals better suited to the unique Australian conditions?

Making better use of Australian species could also help us deal with the pest animals, overabundant herbivores (goats, camels, buffalo, deer and kangaroos) and introduced predators (cats and foxes) overrunning our parks. We have been using bullets and poison for a long time, with little evidence for an overall gain. In some cases, this approach has generated new problems.

In many regions, the best available weapon to control pest animals is the dingo. Abundant research now demonstrates that dingoes strongly limit goat, kangaroo and fox populations. Dingoes are an unrelenting and ultimately free service.

Dingoes therefore provide the perfect example of how we can start making better use of our native species to protect biodiversity more broadly, build more resilient landscapes and shift our approach from the reactive, ineffective, costly and interventionist approaches we often see at present, to more proactive, longer-term, integrated and effective conservation and management solutions.

Our parks are the last vestiges of Australian nature – a final refuge for our irreplaceable biodiversity and ecosystems. A return to the outdated views of the 19th century – when parks were little more than playgrounds for city dwellers to escape the urban _malaise_ – would run counter to everything that Australians have learnt about environmental conservation in the last 150 years.

Your participation in this research will provide important information on the interplay between technical (science-based) information and value judgements in decision making for conservation. This will help me identify factors that can help or hinder effective incorporation of evidence into decision making. The findings will (hopefully) contribute to the knowledge on how to improve communication and knowledge transfer between science and policy/practice. This may in turn, lead to better biodiversity outcomes in practice.

This research will be of greatest interest to scientific experts, policy or decision makers, managers, or practitioners working in conservation, environment or related fields. This list is not exhaustive. If you are interested in this topic and would like to take the questionnaire, your participation is greatly appreciated.

If you are interested in learning more about this study, please visit the information page. Links to the human ethics information and consent documents are available at this page.

Rest assured, your participation is anonymous, and your responses are confidential and protected by data storage security measures. You must be over 18 to participate.

Follow this link to take the questionnaire.

Do it…FOR SCIENCE!
Via Tumblr

A few years ago I did some contract work surveying nesting populations of river birds on the Waimakariri River in New Zealand.  We walked along the river bank identifying and counting birds as part of an annual census to look at populations trends over time.  We also spent a lot of time leaping in and out of a moving jetboat onto tiny patches of gravel in the middle of the river.  That led to some heart-stopping moments!

One of the more abundant species we saw was the black-fronted tern (Sterna albostriatus).  These guys nest in colonies and tend to mob anyone or anything that gets too close, including well-meaning scientists!  Which makes the counting process somewhat difficult – “Have I already counted the one wheeling around the top of my head or is that a different one?”! Black-fronted terns only breed in the eastern regions of  New Zealand’s South Island and are usually found nesting along braided rivers.  Nests are very basic, with the eggs laid among the river gravels.  Black-fronted terns feed on freshwater fish, invertebrates and worms.  Unfortunately their numbers seem to be declining, mostly likely due to predation by introduced mammals and loss of habitat.

For example, I was trying to combine data from two different regions into one file, where one region had 72 columns of data and another region had 75 columns. If you try and do this using rbind, you get an error but going through and identifying non-matching columns manually would be quite tedious and error-prone.

Here’s an example of the function in use with some imaginary data. You’ll note that Database One and Two have unequal number of columns (5 versus 6), a number of shared columns (species, latitude, longitude, database) and some unshared columns (method, data.source).

  species latitude longitude        method     database
1       y   -32.14     150.3   camera trap database.one
2       x   -32.32     153.6 live trapping database.one
3       i   -33.28     151.0   camera trap database.one
4       b   -33.38     152.5 live trapping database.one
5       e   -31.30     149.7   camera trap database.one
6       w   -34.15     151.1 live trapping database.one

      database species latitude longitude data.source accuracy
1 database.two       v   -33.53     150.6   herbarium   8.5843
2 database.two       s   -32.74     150.5      museum  13.0638
3 database.two       u   -33.45     150.3   herbarium   6.1510
4 database.two       y   -33.17     151.6      museum   1.9278
5 database.two       r   -32.49     151.9   herbarium   8.9160
6 database.two       g   -32.98     150.4      museum   0.9405

rbind(database.one, database.two)

Error: numbers of columns of arguments do not match

So I created a function that can be used to combine the data from two dataframes, keeping only the columns that have the same names (I don’t care about the other ones). I’m sure there are other fancier ways of doing this but here’s how my function works.

The basics steps
1. Specify the input dataframes
2. Calculate which dataframe has the greatest number of columns
3. Identify which columns in the smaller dataframe match the columns in the larger dataframe
4. Create a vector of the column names that occur in both dataframes
5. Combine the data from both dataframes matching the listed column names using rbind
6. Return the combined data

rbind.match.columns <- function(input1, input2) {
    n.input1 <- ncol(input1)
    n.input2 <- ncol(input2)

    if (n.input2 < n.input1) {
        TF.names <- which(names(input2) %in% names(input1))
        column.names <- names(input2[, TF.names])
    } else {
        TF.names <- which(names(input1) %in% names(input2))
        column.names <- names(input1[, TF.names])
    }

    return(rbind(input1[, column.names], input2[, column.names]))
}

rbind.match.columns(database.one, database.two)

   species latitude longitude     database
1        y   -32.14     150.3 database.one
2        x   -32.32     153.6 database.one
3        i   -33.28     151.0 database.one
4        b   -33.38     152.5 database.one
5        e   -31.30     149.7 database.one
6        w   -34.15     151.1 database.one
7        v   -33.53     150.6 database.two
8        s   -32.74     150.5 database.two
9        u   -33.45     150.3 database.two
10       y   -33.17     151.6 database.two
11       r   -32.49     151.9 database.two
12       g   -32.98     150.4 database.two

Running the function gives us a new dataframe with the four shared columns and twelve records, reflecting the combined data. Awesome!

¹ A high technical and scientific term!

Bought to you by the powers of knitr & RWordpress

A rope bridge over the Hume Freeway in Victoria (left) and glider poles on the Hume Freeway in New South Wales (right)

Firstly, I should mention a bit about the cameras. We now have them installed at canopy bridges and glider poles along the Hume Freeway in Victoria and New South Wales and some have been in place for almost six years. They’re all motion-triggered, so whenever an animal moves past the sensors, we get a photo. Then we sit down and go through all the photos (or videos for the newer cameras) and record what species went across and when. I still remember seeing the first glider cross – it was so exciting! Unfortunately flicking through photos can get a tiny bit tedious, so the 1000th glider was a touch less exciting.

That’s why I was so thrilled to see our first brush-tailed phascogale crossing a canopy bridge. Granted, that very first photo only captured the animals rear end, but it was a different rear end. A stiff black-bottle brush, as opposed to the long, fluffy rudders I’d been staring at (and counting) for hours at a time. Ferocious little carnivores with unusual breeding patterns, brush-tailed phascogales are a threatened species in Victoria and often only seen in the area as roadkill. We’ve had a handful of crossings at our Longwood canopy bridge which is great news for local populations.

My first image of a brush-tailed phascogale (left) and a less camera-shy one (right)

Most of you will be well acquainted with the common brushtail and common ringtail possum. Both are regular visitors at some of our bridges. Now I can already hear groans and mutterings of “I’ve got plenty living in my roof” or “eating my roses”, but hear me out. While city possums might have plenty of houses to live in and gardens to eat, country possums out ‘in the wild’ do it tougher. And I think we can all agree that just because a species is common, doesn’t mean it deserves to be flattened under the wheels of a truck. More importantly, just because something seems common right now, doesn’t mean it can’t easily become rare in the future. It can’t hurt to act before we have a problem.

A family of common ringtail possums (left) and a common brushtail possum with an itchy chin (right)

This paper is the culmination of years of work, in particular by Prof. David Keith of UNSW, the lead author (see an earlier blog post on one of the workshops, and other papers below). In it, we describe and provide the scientific basis for the criteria for the IUCN Red List of Ecosystems, a new method for assessing the risk of loss (or collapse) or ecosystems. The Ecosystem Red List will complement the IUCN Red List of Threatened Species, and together they will provide a better picture of state and trends in the world’s biodiversity.

A photo of the Coorong, one the ecosystems assessed against the IUCN Red List of Ecosystems criteria, by Rebecca Lester: https://theconversation.com/coorong-recovery-begins-but-still-room-for-improvement-5810

So far we just have the criteria, applied to 20 illustrative case studies from around the world, but the aim is to have a complete assessment of the world’s ecosystems by 2025 – what a lot of work! Currently the Americas are underway – who will be next?

There has been a great deal of media interest in the Red List of Ecosystems. Below is a piece published on The Conversation written by David Keith, and here are some reports in the local media – great work by David and Richard Kingsford getting this out. I’ve also added a link to Jon Paul Rodriguez, who heads up the IUCN ecosystems Red List project, speaking about the Red List.

• on the ABC radio and website
• In The Age
• On NBC news
• Jon Paul Rodriguez on YouTube

Relevant publications:

Keith, D.A., Rodríguez, J.P., Rodríguez-Clark, K.M., Nicholson, E., Aapala, K., Alonso, A., Asmussen, M., Bachman, S., Bassett, A., Barrow, E.G., Benson, J.S., Bishop, M.J., Bonifacio, R., Brooks, T.M., Burgman, M.A., Comer, P., Comín, F.A., Essl, F., Faber-Langendoen, D., Fairweather, P.G., Holdaway, R.J., Jennings, M., Kingsford, R.T., Lester, R.E., Mac Nally, R., McCarthy, M.A., Moat, J., Oliveira-Miranda, M.A., Pisanu, P., Poulin, B., Regan, T.J., Riecken, U., Spalding, M.D. & Zambrano-Martínez, S. (2013) Scientific foundations for an IUCN Red List of Ecosystems. PLoS ONE, 8(5): e62111 [link]

Rodríguez, J. P., K. M. Rodríguez-Clark, D. A. Keith, E. G. Barrow, J. Benson, E. Nicholson and P. Wit (2012) IUCN Red List of Ecosystems. S.A.P.I.EN.S [Online] 5.2

Rodriguez J.P., Rodriguez-Clark K.M., Baillie J.E.M., Ash N., Benson J., Boucher T., Brown C., Burgess N., Collen B., Jennings M., Keith D.A., Nicholson E., Revenga C., Reyers B., Rouget M., Smith T., Spalding M., Taber A., Walpole M., Zager I. & Zamin T. (2011). Establishing IUCN Red List Criteria for threatened ecosystems. Conservation Biology, 25, 21-29. [link]

Nicholson E., D. A. Keith, and D. S. Wilcove (2009) Assessing the threat status of ecological communities. Conservation Biology, 23 (2): 259-274. [link]

David Keith’s article from The Conversation:

Identifying ecosystems at risk – the new IUCN Red List

By David Keith

We know quite a lot about which species around the world are most endangered. The Red List of threatened species, developed by the International Union for the Conservation of Nature (IUCN), identifies species that are most at risk of extinction.

But scientists have become increasingly concerned that the habitats of species and the ecological processes that influence the relationships between species are not adequately considered. Now IUCN has developed a similar risk assessment framework for ecosystems, such as Florida’s Everglades, Australia’s Great Barrier Reef and Scandinavia’s Boreal Forests. It lets the IUCN rank them as endangered, vulnerable or not threatened according to the risks that they face.

For the first time, we have a scientifically robust risk assessment framework, which works across the full range of terrestrial, freshwater, marine and subterranean ecosystems.

This new framework for a Red List of Ecosystems is now published in a scientific study in the Public Library of Science journal, PLoS ONE which illustrates how it would work around the world in a trial on 20 case studies.

Together, the Red Lists for species and ecosystems will provide a more comprehensive view of the status of the environment and its biodiversity than either can on its own.

The ecosystem Red List focuses on biodiversity, habitats for species, as well as their interactions and dependencies, including food webs. The species Red List focuses on individual species, some of which may go extinct even though the ecosystems in which they live continue to remain functional.

Assessing an ecosystem

This new Red List assesses each ecosystem against five criteria.

Two of the criteria relate to an ecosystem’s distribution – how rapidly it is declining and its current extent.

Another two of the criteria relate to functional characteristics of ecosystems. They measure how rapidly and how extensively the physical and biological components of an ecosystem are degrading, particularly the processes that sustain the ecosystem and its species.

The fifth criterion allows multiple threats to an ecosystem to be assessed, as well as potential synergies between them.

Among the 20 ecosystems assessed in the study, the remote mountain ecosystems of the Venezuelan Tepui are among those at least risk of collapse. These are showing little evidence of decline in distribution or function in the past or near future. At the other extreme is the Aral Sea of central Asia, which collapsed during the 1980s and 1990s.

Eight of the case studies come from Australia, and these exhibit varying levels of threat related to land clearing and its legacies, overexploitation of water resources, pollution and climate change.

The lessons from the Aral Sea assessment are sobering. Not only were a host of species lost forever as the sea became hypersaline and dried over much of its former extent, but the ecosystem collapse led to socio-economic disaster, including the closure of regional fisheries and shipping industries. Dust storms were generated from the dry sea bed and they continue to have major impacts on infant mortality and other indicators of human health.

How will this Red List help?

The Red List risk assessment methods are a vital part of the scientific infrastructure needed to support evidence-based environmental management. Other components of the infrastructure that need further development include a base of expertise for carrying out the risk assessments.

Most importantly, we need the essential long-term data needed to underpin reliable and scientifically credible assessments. Initiatives such as the Australian Government’s Long Term Ecological Research Network, which monitors changes in ecosystems and identifies the underlying causes of change, are a vital part of this infrastructure.

Within Australia, this new international standard for risk assessment also presents an opportunity to improve alignment among the various national and state processes currently in place for managing threatened ecosystems. It could strengthen the science and promote better co-ordination of conservation efforts across jurisdictions.

Over the coming decade, IUCN will carry out a global assessment of ecosystems, largely in collaboration with local experts and authorities.

As an early warning system, the Red List of Ecosystems will help governments, industries and communities avoid ecosystem collapse and the associated socio-economic impacts by informing better environmental decisions.

Red List assessments will better target the ecosystems most vulnerable to degradation and help determine which options for investment in environmental management will work best. Ultimately, better planning and management is needed to conserve our rich biodiversity and sustain ecosystem services that support our current standards of living.

David Keith receives funding from the Australian Research Council, the Australian Terrestrial Ecosystem Research Network, the Australian Government, IUCN.

This article was originally published at The Conversation.
Read the original article.

Last night Diversity and Distributions published our latest paper, in a special issue entitled ‘Risks, Decisions, and Biological Conservation’.

The paper – co-authored by Mick McCarthy, Michael Scroggie, Kirsten Parris and John Baumgartner – describes an occupancy-based approach to metapopulation viability analyses that I think will prove a valuable extension to standard occupancy modelling.

Here’s the back story. In the early 2000’s, Darryl MacKenzie struck ecological gold by developing and publishing an occupancy modelling approach that accounts for imperfect detection. Darryl’s 2003 paper (co-authored with the gurus at the Patuxent Wildlife Research Center) was particularly influential. That paper describes a modelling approach that can be used to estimate the probabilities of extinction and colonisation for a given species using multi-season occupancy data, with the bonus of accounting for imperfect detection. But perhaps most mouth-watering for ecologists was the ability to model the probabilities of extinction and colonisation as functions of site- and landscape-level covariates (things like patch area, quality and connectivity) using standard regression techniques. Doing so allows hypotheses about the drivers of extinction and colonisation to be tested while accounting for imperfect detection.

Our paper builds on this functionality. What we’ve done is effectively tack on the ability to simulate extinction and colonisation dynamics for a given species based on the models of extinction and colonisation that result from Darryl’s approach. That is, we’ve developed a means of running occupancy-based metapopulation viability analyses using these models. We provide code for doing this in the Supplementary Material (see also my Code and data page).

It’s a fairly simple process conceptually. One fits a multi-season occupancy model to their particular dataset, estimates the effects of patch area, quality etc. on the probabilities of extinction and colonisation, extracts the parameters of the extinction and colonisation models, and then simulates the extinction and colonisation dynamics (and hence changes in occupancy) for a particular metapopulation according to these parameters. All that’s required for the simulations are measures of the relevant covariates for each site, and the initial occupancy status of each site.

The cool thing is that one can explore, using these simulations, how the metapopulation will respond to particular management scenarios. You can take away particular patches to represent habitat loss, you can tweak patch characteristics to represent habitat enhancement or degradation, and you can even add in new patches to see how habitat creation affects metapopulation viability. We do just that in the paper – examining the effect of habitat loss and creation on metapopulation viability for the endangered Growling Grass Frog (my particular muse….).

Now, in case Andrew Royle or Marc Kéry are reading, I best now tell you that we in fact used their 2007 Bayesian state-space version of Darryl’s approach. In a move that will surely see them breeze through the pearly gates, Royle and Kéry not only put in the hard-yards developing and publishing the Bayesian approach, but provided code to do it too.

The nifty thing about the Bayesian version is that it provides the opportunity to propagate uncertainty in the effects of patch characteristics on extinction and colonisation through to the simulations. I’ll leave it to the paper to explain how, but what we end up with are probability distributions for the metrics of metapopulation viability that not only encompass uncertainty due to the stochasticity of the modelled dynamics, but also from uncertainty in the parameters of the extinction and colonisation models.

That’s the approach in a nutshell – I’ll blog more about it in the weeks to come, including tutorials on how to use it for assessing metapopulation viability. Like any modelling approach, there’s assumptions and caveats that must be understood too, and so I’ll devote some time to that issue as well. In the interim, please email if you’d like a copy of the paper.

In the 1990s, Robert Lambeck proposed the idea of using a set of ‘focal species’ that are the ones that are most affected by key threats. His and subsequent studies focused on selecting species to guide restoration or revegetation of native woodland. For example, to decide how connected or close together patches of bush targeted for restoration should be, you would be guided by the needs of the species that is most dispersal limited; to define the how big patches of habitat should be, you would be guided by the species that has the largest area requirements, such as the biggest home range.

The bush rat, one of our focal species (photo from the Museum of Victoria, http://museumvictoria.com.au/discoverycentre/infosheets/bush-rat/).

The focal species approach is quite intuitive, but has faced a fair amount of criticism. First, there is a lack of evidence to support the underlying principle that focal species confer protection to co-occurring species facing similar threats; results from previous studies are ambiguous.  Second, it is not clear what the objectives of management are: what sort of landscape are we aiming for? The best for each of the focal species, or for certain (undefined) aspects of biodiversity, or for all biodiversity?  The landscape that is best for all species is impossible: species have different needs – the ideal landscape for one species will not be the best for another or for all other species. These two things are of course interrelated: it is difficult to evaluate a theory without explicit goals and measures of how well the theory performs.

Along with colleagues Hugh Possingham, Karin Frank and David Lindenmayer, I sought to understand the conditions under which the focal species concept has merit for making sound conservation decisions (just published in Diversity and Distributions). Because the focal species concept is based on population processes and the persistence of species (that is, wanting to keep viable populations of the species), both the goal and the responses of the focal species and the other species they are supposed to represent should be measured in terms of population viability, such as probability of persistence over a given time frame, which gives a common currency across the species.

We used a case study that David Lindenmayer has been working on for decades near Tumut in NSW, a fragmented landscape of patches of native forest embedded in pine plantation. In a previous study, we modelled ten species of vertebrates (four bird species, five marsupials and one native rodent) representing a range of body sizes and life history strategies. We used a method for choosing a reserve system that maximizes the persistence of multiple species, where species persistence is estimated using a metapopulation model, and is a function of the amount, quality and configuration of habitat patches and the ecology of the species.

Another focal species, the bobuck (photo by Martin Cohen, from http://upclose.unimelb.edu.au/episode/30-love-life-mountain-brushtail-possum)

We related the criteria for selecting the focal species to key model parameters for dispersal capability, home range size and fecundity – linking species characteristics to ecological theory and to our decision criteria. Using these criteria, we identified a subset of three focal species from the set of ten species. The three focal species were the bush rat, which has very limited dispersal, the red browed tree creeper, which has the largest home range size of our suite of ten species, and the mountain brushtail possum or bobuck, which had the lowest reproductive rate of our ten species, which means that it has a low population growth rate and doesn’t make many little possums to disperse out into the world.

Our goal was to find the reserve solution that maximised the combined viability across the species (summing the probabilities of persistence over 100 years). We asked if the reserve system that maximised persistence over the three focal species was the same as the system that maximised viability across all ten species. That is, our measure of the performance of the focal species approach was whether the resultant decision was the same.

And what did we find? Well, the best reserve system using the three focal species was the same as the best reserve system for all ten species. The focal species approach seemed to work. Not only that, we tested all 120 combinations of three of the ten species, and the one containing the three focal species was the only one that was the same as the decision for all ten species. The reserve system that maximised the viability across the ten species and the three focal species was different to the best landscape for any one of the species – that is, it was a compromise, and there were trade-offs between the needs of different species. This is key – it is impossible to find the best landscape for any one species without compromising the viability of other species. Nor is it the best landscape for a ‘super-bad species’ that is bad a dispersing, has a huge home range size and has few babies – that is not realistic. Instead, we have a clearly defined aim: maximise the viability of the suite of species we have enough info on and care about. And we have a way of testing the focal species approach: does it give us the same answer that we would get if we used a larger but defined subset.

The Red-browed treecreeper (photo from David Cook, http://www.davidcook.com.au/wonga.htm)

We don’t claim that the best reserve system for the focal species or the set of ten species represents the needs of all other species or other components of biodiversity (such as ecosystem processes). We bounded the aspects of biodiversity we wanted our focal species to represent, and were explicit about its limits. Therefore the best ten-species decision may not be the best for biodiversity overall, but it’s the best we can make with the information we have.

We also didn’t account for landscape change and dynamics – our model is based on a snapshot. In fact, since we did the modelling analyses, the whole system has been radically altered by bushfire and drought. How the species will respond is a matter of time, though David and his team will be monitoring it, giving us valuable information about how many aspects of biodiversity respond to change, and how we can improve our management of such dynamic and unpredictable systems.

Despite these caveats, our study not only tested the focal species approach and found that it can work, but we provided a framework for testing this and other methods for selecting the subsets of biodiversity used for decision-making. The next step is to apply similar tests to these approaches over many more case studies to find generalisations, and answer key questions such as, how many focal species are needed to make robust decisions, and how sensitive are our results to uncertainty in our information and the rules for their selection? More to follow, I hope.

Key references:

Nicholson E., Lindenmayer D.B., Frank K. & Possingham H.P. (2013). Testing the focal species approach to conservation planning for species persistence. Diversity and Distributions, 19 (5-6): 530–540 [link]

Lambeck, R.J. (1997) Focal species: a multi-species umbrella for nature conservation. Conservation Biology, 11, 849-856 [link].

Note: I’m a co-author in another paper in the same issue: Ponce-Reyes, R., Nicholson, E., Baxter, P., Fuller, R.A. & Possingham, H.P. (2013) Extinction risk in cloud forest fragments under climate change and habitat loss. Diversity & Distributions, 19 (5-6): 518–529 [link].

Yesterday, a couple of colleagues and I watched a standoff between a praying mantis and a magpie lark outside our building.  Needless to say, the magpie lark eventually won but the praying mantis definitely gets bonus points for standing up to someone many times its size.

It reminded me of watching a standoff between a praying mantis and a honey bee a few years ago in my garden in Christchurch.  I was hanging out my washing when I spotted the praying mantis attempting to catch one of the many honey bees visiting the lavender bush.  It eventually grabbed one but it took some time before it was able to manoeuver it into a safe position for eating!

Edit: some of the metapopulation modelling research that I mentioned in this talk has just been published. You can get a copy via the links here:

http://mickresearch.wordpress.com/micks-publications/#GrowlerMetapop2013

•Beetle faunas in agricultural landscapes are a robust subset of the beetle communities that existed prior to land clearing.
•Habitat restoration may be ineffective without the translocation of specialised beetles and the introduction of ground layers.
•Beetle species and trophic groups are adapted to agricultural vegetation variables rather than vegetation in remnant areas.
•Beetle community composition is influenced by environmental variables such as native vegetation structure and soil type.

My most recent paper discusses results from my thesis and is titled ‘Are incentive programs working? Landowner attitudes to ecological restoration of agricultural landscapes‘, published in the latest issue of the Journal of Environmental Management. For this study I sent-out a questionnaire to 400 landholders in the Wimmera and Benalla regions, south-eastern Australia, to determine their management of and attitudes towards revegetated and remnant habitats. I received 179 responses, and found that the majority of landholders either had undertaken revegetation on their agricultural properties, or planned to revegetate in the future. Revegetation activities were most likely to occur on private land if the landholders were in a Landcare group, had an off-farm income, and if they had undertaken revegetation in the past. The paper discusses landholder management of remnant and restored areas, and what incentives could increase landholder uptake of restoration activities. If you would like either of these papers please email me at sacha.jellinek@gmail.com

My role is to help in bringing some ecological knowledge into the revegetation and restoration program. I have recently completed an expert opinion questionnaire and workshop, with the assistance of ACERA, to try and identify all of the important vegetation communities in the region, and ways we can more effectively restore them. This information will inform state and transition models and ultimately the restoration program. Other projects include undertaking flora and fauna monitoring in restored and revegetated areas, assessing the survivorship in revegetated areas, and monitoring how effective the replanting of sedges has been for reducing erosion  and allowing the regrowth of other wetland plant communities.

That is until I went to the Dryandra Woodland in Western Australia. We were on a walk that was touted for its abundance of woylie diggings and numbats.  We saw plenty of diggings but no numbats and were lamenting this fact towards the end of the walk when there was a noise in the undergrowth caught our attention.

Echidna!

Often referred to as spiny anteaters, echidna are covered in spines and coarse hair and have a long slender snout.  They tend to be solitary and are powerful diggers with strong legs and claws.  I think that echidna diggings look superficially like a turtle has been flailing about in the leaves (others may just think I’m weird!).  Echidna feed on ants, termites, worms and insect larvae which they dig out of rotting logs and anthills.  They don’t have any teeth, instead collecting prey with a long, sticky tongue that protrudes from their snout.

As I eluded to earlier, echidna are monotremes – mammals that lay eggs.  This is clearly an unusual strategy, with platypus making up the only one other group of monotremes.  I had always assumed that laying eggs meant that echidna would build nests.  But it turns out that female echidna actually lay a single leathery egg and carry it around in a rear-facing pouch similar to the pouches of marsupials.  The egg hatches after about 10 days and the puggle (the name for a baby echidna) spends about 2-3 months hanging out in the pouch.   It suckles from patches in the skin that secrete milk as monotremes don’t have nipples.  Once the puggle begins to develop spines, it gets kicked out of the pouch (fair enough!) and into a nursery burrow.  The female echidna will return every five days or so to feed the puggle until it gets weaned at about seven months.

Echidna are found throughout Australia and New Guinea, with three recognised genera.  Only one species occurs in Australia, the short-beaked echidna (Tachyglossus aculeatus).  The name echidna comes from greek mythology – Echidna was half woman, half snake and was mother of all monsters (she sounds like a charmer!).

One of the perks of being an ecologist is that you often get to go to some pretty amazing places and work with amazing animals.  And I’ve definitely had my share of these experiences.  But one of the coolest (quite literally) projects I have been involved with has been researching Adélie penguins (Pygoscelis adeliae) at Cape Bird on Ross Island, Antarctica.

It started out as one of those being in the right place at the right time sort of situations.  I had a gap in my workload as a private consultant and a colleague at Landcare Research asked me to volunteer three months of my time to chase penguins in Antarctica.  Needless to say, I immediately said yes and started on a journey that has been a fascinating experience.  I’ve just completed my third field season at Cape Bird and I’m still pinching myself – it seems unreal to be paid (sometimes good things happen if you volunteer) to spend my summers hanging out with 40,000 Adélie penguins.

The research we do is diverse and multi-faceted, with simultaneous studies occurring at three Adélie penguin colonies varying in size by two orders of magnitude.  Most of the work revolves around studying birds of a known age.  Every year we mark up to 1000 chicks with individually numbered metal flipper bands, with the first bands deployed in 1996.  This enables us to identify how old each banded bird is and to work out survival and recruitment rates.  It also means we have a pretty awesome long-term dataset.

26306 was banded as a chick in 2002.

Male Adélie penguins start to arrive at the colony in late October and establish nest sites.  Nests are built from small stones and are just far enough apart that your stones can’t be stolen by your neighbour.  The females will arrive a few weeks later and pairs will go through a process of courtship display to establish or renew pair bonds.  After mating, the females will lay 1-2 eggs and then head back out to sea, leaving the male to incubate the eggs for a few weeks.  By the time the female returns again, the males have been at the colony for 3-4 (or more) weeks and are getting fairly hungry!  But from now on in, the males and females will switch roles every 2-5 days or so, with one going out to feed while the other takes on incubation duties.  Eggs will start to hatch after ~31 days and the chicks grow pretty rapidly, gaining over 3 kg in just 6 weeks.

These two chicks are from the same nest and would have hatched within a few days of each other. Clearly somebody has been eating all the pies!

We typically arrive in early November and are straight into monitoring the activities of banded birds.  Each day we walk around a subsection of the colony looking for banded birds and checking out what they are up to.  All nests with banded birds are gpsed and marked with a numbered tag.  These marked nests are checked every five days or so to try and determine the fate of the eggs or chicks through until the end of January.  The primary causes of nest failure seem to be incompetent parenting, fighting neighbours knocking eggs or chicks from the nest, bad weather or skua predation.  Because we spend so much time out in the colony, we often observe some pretty interesting behaviours and it’s hard to not anthropomorphise (but they totally do run around and behave like little people!).

One of the more hands-on projects I’ve been involved with has been conducting the field sampling for a physiology and foraging behaviour study that looks at differences between birds of different ages and breeding experience.  We’ve been catching a subset of the known-age birds and attaching a small device on their backs that records information about their diving behaviour; where they travel to, how long and deep they dive, the water temperature and sudden changes in direction that indicate they might be foraging.  This year at the Cape Crozier colony they were even following the tagged birds around with a remotely operated glider to check out what they were feeding on.  All very technologically cool!

Attaching an accelerometer to the back of a penguin using strips of tape.

Once the device is attached, we let them go out for one foraging trip which usually last for 1-4 days and then we catch them again.  This is where my vampire-ish tendencies come in handy as we’re taking blood from the jugular vein for the physiology part of the study.  It’s a pretty quick and easy process but it does involve sticking a one inch needle in the neck of wriggling penguin to extract 5 ml of blood.

Vampires at work.

Not surprisingly, they aren’t very happy about this and try and extract chunks of flesh from person holding them.  Luckily this isn’t me!  After taking blood, we weigh and measure the chicks.  Then everybody is put back on the nest to continue playing happy families.

Weighing an Adelie penguin chick.

There’s a whole lot of other work that goes on throughout the season, including:

• counting a subset of the active nests and chicks at each colony to work out the annual productivity rate (the mean number of chicks raised per pair).
• fencing a small subcolony at each site, where the adults have to cross a weighbridge to get to and from their nests.  This tells us how long they go out to sea for and how much food they are bring back to their chicks.
• watching adults feed their chicks, and dissecting dead chicks, to work out what they are eating.
• weighing and measuring 50 randomly selected chicks once a week to look at how chick mass and condition vary within and between seasons.
• taking photos of all of the Adélie penguin colonies in the Ross Sea and counting the penguin dots to look at trends in population size over time.

Penguins crossing the weighbridge after feeding their chicks.

It’s always a very busy season and the fact that the sun never sets means it can be hard to switch out of work mode.  But it’s such a spectacular place with some stunning scenery and wildlife that I sometimes wonder how I managed to get a job like this!

You can find out more about our Adélie penguin research at Landcare Research and www.penguinscience.com, listen to some of the researchers talk about life at a penguin colony or check out some recent publications.

Related articles

Dugger, K., Ainley, D., Lyver, P., Barton, K., & Ballard, G. (2010). Survival differences and the effect of environmental instability on breeding dispersal in an Adélie penguin meta-population. Proceedings of the National Academy of Sciences, 107 (27), 12375-12380 DOI: 10.1073/pnas.1000623107

Lyver, P., MacLeod, C., Ballard, G., Karl, B., Barton, K., Adams, J., Ainley, D., & Wilson, P. (2010). Intra-seasonal variation in foraging behavior among Adélie penguins (Pygocelis adeliae) breeding at Cape Hallett, Ross Sea, Antarctica. Polar Biology, 34 (1), 49-67 DOI: 10.1007/s00300-010-0858-0

Lescroël, A., Ballard, G., Toniolo, V., Barton, K., Wilson, P., Lyver, P., & Ainley, D. (2010). Working less to gain more: when breeding quality relates to foraging efficiency. Ecology, 91 (7), 2044-2055 DOI: 10.1890/09-0766.1

Ballard, G., Dugger, K., Nur, N., & Ainley, D. (2010). Foraging strategies of Adélie penguins: adjusting body condition to cope with environmental variability. Marine Ecology Progress Series, 405, 287-302 DOI: 10.3354/meps08514

Dugger, K., Ballard, G., Ainley, D., & Barton, K. (2006). Effects of flipper bands on foraging behavior and survival of Adélie penguins (Pygoscelis adeliae). The Auk, 123 (3) DOI: 10.1642/0004-8038(2006)123[858:EOFBOF]2.0.CO;2

Ainley, D.G. (2002) The Adélie Penguin. Columbia University Press, New York.

I’ve got two primary motivations for uploading code and data from my papers. The first is that I’ve benefited on numerous occasions from the generosity of others who have shared their code with me. From supplementary material in published papers to emails with bits of code I’ve requested, I simply could not have completed some of my research without these generous offerings. I hope to return that favour here.

The second reason is that I want the data I’ve collected to be available to others, either for further dedicated analyses or for use in meta-analyses. The further I go through my career the more I realise just how precious data are. The time, effort and money required to gather field data is perhaps evidence enough. Sadly though, recent experience has taught me just how quickly data can become irreplaceable, as populations blip out of existence.

So, swing over to ‘Code and data’ and take a look. I’ll update the page as new papers come out, and work on filling in my back catalogue over the next few months.

Just one request – if you use the code or data presented on this website, please cite the relevant paper, or, if there isn’t one to cite, cite this website instead.

Enjoy!

We visited the island to get a better understanding of the ecosystem and discuss our PhD projects with the park managers there. Christmas Island’s ecosystem is under high pressure at the moment. One major problem is caused by yellow crazy ants (Anoplolepis gracilipes). These introduced ants didn’t cause any problem for decades, until in the early 1980’s a scale insect was introduced. This insect provides honeydew to the ants, which results in the ants becoming very abundant and forming supercolonies. These supercolonies can become extremely large and attack and kill the endemic red crabs (Gecarcoidea natalis) on the island. This resulted in a severe decline in the red crab population, which is catastrophic for the entire island ecosystem since red crabs are a keystone species in the forest. Their diet is based on leaf litter and seedlings and in absence of the crabs the forest develops an extremely dense understory.

In order to control this problem, Christmas Island National Park (together with Peter Green and Dennis O’Dowd from La Trobe University) set up an island wide survey to keep track of where supercolonies form.  Every other year they survey the entire island, which is a very time consuming and exhausting process.  It would be very useful if they got a better understanding of what causes supercolony formation. One of the aims of our projects is to get a better understanding of the habitat requirements and spread of supercolonies. If we know more about the habitat these supercolonies prefer and how they spread, it might be possible to do the survey in a more cost-effective way.

Our work is still in progress so I won’t show any results here but I do want to show you some pictures (taken by Darren, Yi and myself) of the island and its inhabitants.

Enjoy!

More than 60 % of the island is national park

Red crabs are endemic to the island

They are a keystone species of the tropical forest since their diet is based on leaf litter and seedlings

When the wet season starts millions of crabs are migrating from the forest to the coast to spawn

Temporary fences are placed to keep the red crabs out of the yellow crazy ants supercolonies

Unfortunately many crabs like to ‘explore’ what’s on the other side of the fence…

Park rangers show us how they do the intensive yellow crazy ants survey

The most common argument against women in science leans on ‘the neuroscience approach’ – the idea that male brains are ‘wired’ for logic and problem solving and so are inherently better at maths and science. The flip side, is that a woman will be an inferior scientist to a man because there are limits to what her brain can achieve (the poor dear). There are plenty of resources which deal with the evidence and counter-evidence in-depth (see bottom of this post for further reading). In short, it’s not true.

via http://xkcd.com/385/

The brain is the “most complicated biological structure in the known universe” (Dr. Francis Collins, Director of the newly formed BRAIN Initiative in the US). It controls our bodies and indeed, the way we perceive and understand the world around us. We know a little about how it works, but have only just begun to scratch the surface of understanding how brain physiology (e.g. size, blood flow, or electrical activity) influences thought, emotion and intellectual potential. While other parts of the body follow fairly well understood rules (e.g. the ability to lift a weight is limited by the amount, size and type of muscle fibers), we are constantly amazed by how the human brain can adapt, grow and recover. When it comes to the brain, the possibilities seem limitless.

Given this complexity, judging academic merit based on gender seems crude, ridiculous and damaging. In my humble opinion, to understand a person’s intelligence or ability to do good science, you really need to know more than whether they pee standing up or sitting down. Sorry, but I did say it was crude. And ridiculous.

It’s not enough to ignore these ideas. We need to defy them. So when I hear broad, sweeping, ill-informed statements about my mental capacity based on gender, this is my response:

You don’t know me. You don’t know my experiences or what I liked at school. You don’t know that my love of problem solving and learning about the natural world made it inevitable that I would choose science as a profession. You don’t know how I think, why I think that way, or what motivates me. So don’t ever tell me that there is a problem I am incapable of solving because of my gender.

How will you respond?

Further reading

This is by no means an exhaustive list but just a few starting points…

A great read, deconstructing ‘neurosexism’ arguments one by one Delusions of Gender by Cordelia Fine.

Some blogs which feature great posts on women in science include Soapbox Science and The Contemplative Mammoth. Edit almost forgot to include this cracking post from EEB & flow. Check out their links to other sites.

Some articles on gender and student performance in maths and science.

Scientific journals, Nature and Animal Conservation acknowledging and addressing the problem.

Think you’re free of bias? Head to Project Implicit and take a test to assess any unconscious bias. The results might surprise you.

If you saw this graph of woodland bird classification where would you put the cut off? Should only birds which are consistently categorized as woodland birds be considered or is it OK if we are less certain?

In the meantime, Barna Mia is open to the public for evening viewings of interesting nocturnal beasts, so we went along to check out some of the inhabitants.  Held in a 4ha fenced enclosure, animals are attracted to viewing stations just after dark by offerings of food which supplements their natural diet.  We sat quietly and searched for animals using red spotlights while a ranger explained the ecology of each species as we spotted it.  It was an interesting evening and we were rewarded with five new species to check off the list (not that I would do such twitcher-esque things!) .  So today you are treated to a menagerie of mammalian delights* (plus a bonus owl).

The bilby (dalgyte – Macrotis lagotis) has to be one of the weirder looking marsupials out there.  With rabbit-like ears, an anteater-ish snout and a long tail, they are definitely hard to confuse with their closest relatives, the bandicoots.  They are omnivorous and excellent burrowers, creating extensive underground tunnels.  Interestingly, a bilby’s pouch faces backwards, apparently to prevent it filling up with dirt when they are burrowing.

The quenda (southern brown bandicoot – Isoodon obesulus) is omnivorous and leaves tell-tale conical pits in the surface of the soil as it digs for insects, spiders, tubers and fungi.  They spend the day tucked up in small nests of vegetation under dense cover.  Quenda are the most common of the mammal species at Barna Mia and are found across most of southern Australia.  However, they are vulnerable to fox predation.

Woylies (brush-tailed bettong – Bettongia penicillata) were once found across 60% of Australia but are now confined to less than 1% of the mainland and a couple of offshore islands.  They are currently listed as critically endangered by the IUCN and recent declines are thought to be a combination of fox predation, changing fire regimes, habitat destruction and a possible disease outbreak.  Woylies are omnivorous but predominantly feed on fungi which they dig up from underground with their strong front claws.  The fungi are digested by bacteria in a special part of their stomach.  These bacteria release nutrients that can then be absorbed by the rest of the digestive system.  It’s thought that they would have played an important role in dispersing fungi spores across the landscape.  Woylies spend the day in dome-shaped nests, which they build from dried grasses and leaves that they carry in their prehensile tail.

Mala (rufous hare-wallaby – Lagorchestes hirsutus) are the smallest of the hare-wallabies, members of the macropod or kangaroo family.  They are nocturnal and solitary herbivores, feeding on leaves, seeds and herbs.  Extinct on the Australian mainland, mala are now confined to two island populations off the Western Australian coast and a handful of captive breeding facilities, including Barna Mia.

Boodies (burrowing bettong – Bettongia lesueur) are small nocturnal marsupial that lives in communal burrows, the only burrowing member of their suborder (Macropodiformes).  Boodies are omnivorous, feeding on seeds, fruits, flowers, roots, fungi and termites. They were once widespread throughout semi-arid Australia but were extinct on the mainland by the 1960s.  They are now confined to several offshore islands and captive facilities.    Declines have been attributed to fox and cat predation, competition with rabbits, livestock grazing and changing fire regimes.

Not part of the official tour but we also came across this owl perched on a branch, watching as we pitched our tent.  Scientific deduction (i.e. random flipping through the Field Guide to Australian Birds) suggests that it is a southern boobook (Ninox novaeseelandiae) but I could have totally made this up.  Please correct me if I’m wrong!

Well done to those of you who have made it all the way to the bottom of this post!  In lieu of an prize, I’m going to treat you to a very brief explanation of my new photo watermark.  As you should have noticed by now, my last name is Whitehead.  Which is quite apt when you consider how early the males in my family go grey!  But the whitehead (Mohoua albicilla) is also a New Zealand forest bird.  So to celebrate my tendancy towards bird nerdness, I’ve designed a new logo.  I’d love to blog about whiteheads (of the avian variety) on Wildlife Wednesday sometime but given I’ve only seen them twice, never managed an identifiable photo and currently live in Australia, you may have to wait a while!

* I was going to drag this out to five Wildlife Wednesdays but that seemed like cheating.

I spent Easter in Western Australia camping along the southern coast and out in the wheatbelt.  We spent a couple of nights in the Dryandra Woodland, a remanent patch of eucalypt forest in the wheatbelt.  Not a bad place to spend some time, particularly if you are looking for Easter bilbies (more on this next week).

I was trying to take a photo of an inquisitive brushtail possum (something that felt quite strange being a kiwi who is used to trying to kill them) when I spotted two glowing orbs in the darkness. And then they blinked.  It was a little bit creepy and it was with some trepidation that I cautiously approached them to see if I could identify the owner (it didn’t help that my partner was winding me up with made-up (I hope!) horror stories of giant Australian nocturnal spiders!).    It turned out that we were being watched by a Western spotted frog (Heleioporus albopunctatus) who sat very obligingly for some time while the ecological paparazzi snapped away.

These guys get up to 7.5 cm in length and are pretty solid as far as frogs go.  They are found throughout the wheatbelt region of Western Australia and are typically associated with swamps and temporary water bodies.  Once the autumn rains begin, males dig burrows up to 1m deep and sit in the bottom calling to attract a mate.  Mating occurs at the bottom of the burrow (out of the sight of prying eyes) and the females lay their eggs.  The burrows will fill with water after rain and tadpoles hatch out and are washed into larger water bodies where they take 2-3 months to develop into frogs.

I achieved this by investigating the increase in treed cover at sites as a measure of vegetation extent and by using two threatened bird species to represent the outcomes for threatened species and bird species richness to represent overall species diversity. I used aerial photography to determine trends in the total treed area in the presence and absence of planting initiatives and developed habitat models for the bird species and for the species richness of birds. The original model is arranged as a Bayesian network, a method which allows this new information to be linked to the original model with relative ease.

If you have information about the current extent and composition of the vegetation the completed model allows you to predict the effect of a range of management actions on the future vegetation extent, condition and bird assemblage. Hopefully there will be an article coming out in the not-too-distant future which goes into this project in greater depth but if you have any questions in the meantime (or even afterwards) don’t hesitate to send me an email.

Reference:

Rumpff, L., Duncan, D. H., Vesk, P., Keith, D., & Wintle, B. (2011). State-and-transition modelling for Adaptive Management of native woodlands. Biological Conservation, 144, 1224–1236.

Bayesian networks are made up for nodes and links. The nodes represent the variables of interest, while the links represent directional relationships between these nodes. Links run from parent nodes (independent variables) to the child nodes (dependent variables) and imply causation i.e. that changes in values at the parent node will cause changes in the outcome at the child node. The information underlying the relationships represented as links is stored in conditional probability tables which are attached to each child node (note that the number of parents is not limited to two as it is in humans). Values may be entered into these tables directly, be informed from raw data via an inbuilt Bayesian algorithm, or from probabilistic or deterministic equations that can be entered as code. In this case I will use raw data to update my network.

Example:

Recently I have been looking at houses and it occurred to me that a Bayesian Network would help me build a frame work for how different attributes of houses affect their sale price. I came up with a list of variables which I thought might affect the sale price of houses: suburb, size of the house, number of bedrooms, number of bathrooms, whether it is open plan, size of its garden, whether it has been renovated and proximity to public transport. Because I’ll be looking at an unknown number of suburbs I decided that the network would be more useful to me if I pooled the suburbs rather than keeping them separate. That way the network will give me an idea of how much a particular house should cost in the average suburb. I put all of the other variables into the network and connected them to two variables; one for asking price and one for sale price, with links (see Figure 1). As it is the network should allow me to predict what a house of known specifications is likely to list as their asking price and what they are likely to sell for.

Figure 1: Full house hunting Bayesian Network

You will notice that the nodes are all broken into discrete states eg. Size of House is broken into small, average and large. This is because Bayesian works are unable to work directly with continuous variables. The information in a Bayesian network is stored in conditional probability tables which predict the outcomes at child nodes for each possible combination of parent node states; this would not be possible with continuous data.

I had a look at the conditional probability table for and noticed that there were way too many possible combinations of parent node states to ever be able to create an informative model by updating the network with raw data. When updating a network with raw data a Bayesian counting learning algorithm is used to update the conditional probability table. Data from one house will update one cell in a conditional probability table so, to be able to have an estimate for every type of house you have to visit the same number of houses as there are cells in the conditional probability table. In this case that would mean visiting and obtaining asking price and sale data for 648 houses, which is not achievable. Of course it would also be preferable to have several replicates of each of these different house types which would blow out the number of houses you would need to visit even further.

I could use a multiple regression analysis to work out which ones were relevant but I would need a lot of data for this as well and I wouldn’t have any information to go on until I had visited all of the houses. By using a Bayesian network I can update it as I go and get some estimates of house prices in advance. So instead I cut down on the number of parent nodes which would exponentially decrease the number of houses I have to visit. I simplified the model to only include what I deemed to be the factors which are most likely to impact house price; number of bathrooms, number of bedrooms and proximity to trains (Figure 2).

Figure 2: Reduced house hunting Bayesian network

Now it is a greatly simplified model but there are only 18 combinations which means that I don’t have to visit nearly as many houses. Additionally there will be more replicates of the different types of houses which will make the model represent house prices more accurately (Figure 3).

As you can see at the moment the model assumes an equal probability of falling into any of discretisations, because it has no underlying information, when I update it with data this will change. I have so far visited 17 houses and obtained their asking prices, the houses are yet to sell so I don’t have information about their sale price yet. Now I want to enter the information from these into the network. I make an excel spreadsheet with the names of all of the parent nodes and the child nodes along the top and values for each house below them. Then I inform the network by selecting to ‘incorp’ this ‘case file’ into the network. Initially the network assumes an even distribution and uses Netica’s Bayesian algorithm to update the probability of a house being at a particular asking price. By viewing the conditional probability table I can tell how many times I have visited a house with each combination of parent node values, in this case I have visited 9 of those 18 types of house, which means that I have seen some house types more than once. For example I have seen four 3 bedroom, 2 bathroom houses near train stations, but I haven’t seen any four bedroom, one bathroom houses near trains. As such I will be fairly good at predicting the range of asking prices that you might find for houses with three bedrooms and two bathrooms near trains. By selecting these values for my parent nodes I will be able to get an idea of the probable asking price for these houses.

Figure 3: Updated network showing the probably asking price of three bedroom, 2 bathroom houses near trains.

You can see that the asking price is still very uncertain, indicating that more data needs to be collected. This information can also be gathered by examining the conditional probability table (Table 1).

Table 1: Conditional probability table displaying the probability of houses falling into each of the house price categories (on the right) given their number of bedrooms, bathrooms and their proximity to trains.

As more data is loaded into the model the asking price estimates will represent the market value for each of the different house types more accurately. As I visit more houses I will post the updates I make to the model to demonstrate this. If any of this is unclear or if you would just like some questions answered about Bayesian networks (particularly in Netica) don’t hesitate to shoot me an email.

These whio (pronounced “fee-0”) ducklings were just a few days old when I took this photo.  They were some of the lucky ones, hatching in a river valley where introduced predators were kept at low numbers due to the hard work of Department of Conservation staff (who seem to be as threatened as the species they protect).  All of eggs and ducklings just over the hill where there was no predator control got munched by hungry stoats!

I know I’ve featured whio before on Wildlife Wednesday (the inaugural post in fact) but I have a special place in my heart for these bluest of blue ducks (and a lot of photos), having spent 5 years of my life working with them.  March is Whio Awareness Month and I had great intentions of writing an enthralling post about the plight of the whio and the work that’s bring done to protect them. But it’s the end of March already and I have a week full of meetings and deadlines, so it will have to wait for another day. But luckily the good folk at the Department of Conservation and Whio Forever have been busy telling their stories.  Head over to the Whio Forever website to learn more about project – a partnership between the Department of Conservation and Genesis Energy.

• Find out more about this unique New Zealand species and where you can find them
• Read what motivates whio rangers “Andy “Captain Whio” Glaser”, Tim “Duck Boy” Allerby, Andrew “Max”Smart and Ivan Rogers to get out on the river
• Learn how to catch a whio with the Brand Family
• Read some whio poetry or sing along to the little blue duck song
• Watch a little whio tv
• Enter the great whio adventure competition
• Find out what you can do to help
• Follow @whioforever on Twitter or Facebook
• Or read some really exciting scientific papers about whio (see below)!

Related articles

WHITEHEAD, A., EDGE, K-A., SMART, A., HILL, G., & WILLANS, M. (2008). Large scale predator control improves the productivity of a rare New Zealand riverine duck. Biological Conservation, 141 (11), 2784-2794 DOI: 10.1016/j.biocon.2008.08.013

WHITEHEAD, A., ELLIOTT, G., & MCINTOSH, A. (2010). Large-scale predator control increases population viability of a rare New Zealand riverine duck. Austral Ecology, 35 (7), 722-730 DOI: 10.1111/j.1442-9993.2009.02079.x

I should also note that this is my first post to be posted to the Research Blogging blogroll.  I wonder if this will help to make whio more popular than my most popular post so far about kakapo?

A growling grass frog – will it cross the road?  (photo by Saide Cameron)

Frogs are small animals that hop or walk along the ground.  Roads are long stretches of gravel, concrete or bitumen that allow vehicles like cars, trucks and motorbikes to go from one place to another.  These vehicles are often large and heavy, and travel very fast – much faster than frogs can hop or walk.  This means that it is very difficult for a frog to cross a busy road safely, and many thousands of frogs (maybe even millions) are run over and killed on roads around the world every year.

Why did the frog (try to) cross the road?

So why do frogs try to cross roads?  To get to the other side, of course!  Frogs are busy animals and often have places to go.  Frogs that breed in ponds may spend part of their time at a pond and part of their time in another kind of habitat.  And at certain times of the year – such as the start and end of the breeding season – they need to move between the two.  This type of movement is known as migration.  Young frogs (juveniles) may also leave the pond where they were born and move to a new one.  This type of movement is called dispersal.

Roads are a problem for frogs because they can turn a regular journey into a very dangerous one.  A number of years ago, researchers in Denmark measured how quickly amphibians (frogs and newts) moved when they crossed a road (Hels & Buchwald 2001).  The fastest species in their study was the moor frog (Rana arvalis), which moved at 2 metres per minute.  The slowest was the common newt (Triturus vulgaris), which moved at only 0.5 metres per minute.  On Danish motorways, cars and other vehicles can travel at 130 kilometres per hour, which equals 2,167 metres per minute.  That is more than 1,000 times faster than moor frogs can hop, and more than 4,000 times faster than common newts can walk!

Road kill

Tove Hels and Erik Buchwald used this information to work out the chance of a frog or newt getting safely across roads with different amounts of traffic.  The news wasn’t very good.  The moor frogs had about a 60% chance of safely crossing a road carrying 100 vehicles per hour (2,400 per day).  This means that for every 100 frogs trying to cross the road, 60 would make it and 40 would be run over.  The slower-moving newts only had about a 10% chance of crossing safely, meaning that 90 out of every 100 newts trying to cross a road with this much traffic would be run over.  On busy roads with more than 15,000 vehicles per day, the speed of the different amphibians didn’t make much difference to their chance of a safe crossing – no matter how quickly they walked or hopped, they were almost certain to be run over.

What can we do?

In Europe and North America, where lots of frogs tend to move at once, tunnels under roads have been quite useful in reducing road kill.  Fences along each side of the road direct the frogs to the tunnels, where they cross underneath the road before going on their way.  At this stage, we don’t know whether these kinds of tunnels work in other places such as Australia, where frog movements are more spread out over time.

Some conservation groups (such as Froglife in the UK) help frogs and other amphibians to cross roads – this is known as patrolling.

Otherwise, if you are driving at night – particularly if the weather is warm and wet – slow down and watch out for small animals crossing!

In my last entry I explained the notion of time preference, and  the difference between prescriptive and descriptive approach to time preference. I also talked a bit about marshmallows.

Don’t care how, I want it now!! Veruca Salt in Charlie and the Chocolate Factory is a classic example of someone who values present gains much more than future gains, discounting the future at a very high rate.

So what does all this time preference stuff have to do with invasive species management? Well, invasive species have all sorts of different impacts that occur over different time frames. Managing them involves costs, that again, occur over different time frames. In our paper, we illustrated this with an example of invasive vertebrate animals in Australia (Table I below).  These species show a typical difference in the way impacts are estimated – agricultural impacts were considered for 30 years into the future, while environmental impacts were considered over a much longer time frame of 100 years into the future. In our paper, we aimed to explore how different time frames have been, and should be, dealt with in multi-criteria decision analyses for invasive species management.

Which invasive pests should we control when we have two conflicting objectives – to minimise agricultural losses AND minimise native species extinctions??

Multi-criteria decision analysis (MCDA) is a structured method of decision support for problems where the decision-maker must balance multiple conflicting objectives. For a decision-maker choosing to allocate funds between controlling foxes, goats, and cats, the budget they allocate to each species would depend on the importance they place on agricultural impacts versus environmental impacts. A multi-criteria decision analyst would leave these impacts in their natural units, and elicit from the decision-maker a weight for type of impact based on its perceived importance. The analyst could then use these elicited weights (or ‘preferences’) to compare alternatives, and find the decision with the best expected outcome for the decision-maker. MCDA is a really useful method for decisions involving environmental or social impacts that can’t easily be converted into monetary costs.

When we looked at previous applications of MCDA to invasive species management, we found that these analyses routinely ignore the timeframe over which impacts occur. Out of 20 analyses in the literature, none specified the time frame over which the impacts of the invasive species were estimated, and none discounted impacts occurring in the future¹. This is bad news, because ignoring time means assumptions are being made about the decision-maker’s preferences that may or may not be true. For example, not using any form of discounting effectively assumes the decision-maker is indifferent to impacts occurring immediately and those occurring decades or even generations into the future.

So how should we incorporate time preference into MCDAs? With the different approaches to time preference out there, how do we know which one to apply?

MCDA is based on describing the decision-maker’s preferences. It therefore seems sensible to use a descriptive approach to time preference within MCDA, eliciting time preference from the decision-maker. Fortunately, there’s a wealth of psychological literature based on eliciting time preferences, with well-established survey methods and protocols. These surveys can easily be adapted, and the results used in an MCDA. To show how it’s done, we even conducted our own survey with a small group of biosecurity experts.

Time preference is a tricky business, but when we ignore it, we’re actually making assumptions about the decision-maker’s preferences that may not be true. By showing how and why time preference should be incorporated into MCDAs, we hope our paper will pave the way for better decision making in invasive species management.

References

1. Rout, T. & Walshe, T. Accounting for time preference in management decisions: an application to invasive species. Journal of Multi-Criteria Decision Analysis (in press 2013).

This split between positive and negative posts might give the impression that there are just two types of PhD experiences. There are carefree students who cruise through their very easy PhD’s, taking holidays and sipping cocktails in a world full of puppies and sunshine. Or, there’s everyone else, struggling, suffering and regretting taking on a PhD to begin with. This is a problem because people who identify with the latter, might feel unable to take the advice of the former. For example, I’ve had a few comments from students who love the idea of a better work-life balance, but aren’t sure it could ever apply to their own situation.

To shed the notion that you either struggle or you don’t, here are my confessions:
- I am not always very good at taking my own advice
- I am not immune to PhD related freak-outs
- sometimes I really, really struggle to make any kind of progress
- and when I struggle, I can get a bit negative

Woe is me

Based on the strangely specific description above, you’ll realise I hit a bit of a rough patch recently. I don’t think there was any one reason, I was just having a bad day (or several).

At some point though, you have to stop being upset and get back to work. Otherwise you fall further behind, get even more stressed, then fall further behind and are sucked into a nasty vortex of despair (from which you may never return!) So while I was very tempted to spend another day wallowing, I scrambled out of my comfy, chocolate-lined misery pit and returned to my laptop. I looked at the problem again, clearly and without my ‘goggles of hopelessness’ on, and it wasn’t actually that bad. In two hours it was fixed and I was back to conquering my mammoth to-do list one step at a time. I felt good, if a little annoyed that I’d let myself get so upset about it in the first place.

So, as you can see, a positive outlook is a work in progress. It’s something that you learn and practice. It’s not about never being stressed or always being positive. And it’s not about having a hassle free PhD (no such thing I’m afraid). It’s about picking yourself up when things suck. Because as fun (and easy) as it is to wallow, it’s not really helpful.

Sometimes you’ll fall down. But you shouldn’t stay there.

We all feel this way sometimes…but it doesn’t last

This beautiful white flower is often called the Mount Cook lily.  However, it is actually a member of the buttercup family (Ranunculus lyalli).  Endemic to the South Island of New Zealand, it is found in the alpine zone between 700 to 1500 m above sea level.  Plants grow up to a metre in height and have waterlily–like leaves, hence the frequent misnaming.  Flowering occurs in late spring and early summer, with flowers reaching 5-8 cm in diameter, making it the world’s largest buttercup species.

Sometimes it can be hard work sticking to our own time preferences

In 1972, psychology researchers at Stanford University conducted what became known as “The Marshmallow Experiment”. Four- to six-year-old children sat in a distraction-free room with a treat of their choice in front of them: a cookie, a pretzel stick, or a marshmallow. The children were told that they could either eat their treat now, or if they could resist eating it for fifteen minutes, they would get a second treat as a reward. This length of time isn’t trivial – Jaochim de Posada points out in his TED talk about these experiments that making a four-year-old wait fifteen minutes for a marshmallow is like saying to a group of adults “I’ll bring you coffee… in two hours”. Footage of recent replications of these experiments makes for hilarious viewing as the kids struggle to deal with the excruciating temptation in front of them.

Although these experiments were aimed at studying mechanisms of delayed gratification, they are also a nice illustration of something called ‘time preference’. Time preference describes how people value (or prefer) outcomes occurring in the present compared to outcomes occurring in the future. The marshmallow experiment is a nice test of the kids’ time preference for sugary rewards. If a child chooses to wait fifteen minutes, it means they value two future marshmallows more than one present marshmallow. If they decide to eat their marshmallow immediately, this means that they value their present marshmallow more than two future marshmallows.  

Psychologists usually investigate people’s time preference by asking them what they would prefer rather than testing what they actually do (to remove the role of self-discipline). Would you prefer one marshmallow now, or two marshmallows in fifteen minutes’ time? How about three marshmallows in twenty-five minutes’ time? Through these surveys, researchers have discovered that the way people discount future outcomes can be described by a hyperbolic function. This is a pretty robust finding – it applies to both gains and losses, and to money, health, and environmental outcomes^1,2. In fact, it’s a finding not restricted to humans – other species (such as rats and pigeons) also discount the future hyperbolically³.

Economists deal with time preference differently. If someone offers you a choice between receiving $100 today or $101 in a year’s time, it would be rational to choose the $100 now – if you invest it wisely at the market rate of 3%, you could have $103 in a year and be $2 better off. This rational, money-centric approach is the basis for the exponential discount function widely used by economists⁴. However, it’s easy to prove that this isn’t the way we actually think. For example, following the previous logic it would be rational to pass up $100 today for the promise of $105 in a year’s time, but would you actually wait a whole year for an extra $5? I wouldn’t.

In other words, the exponential discount function used by economists prescribes what people should do, while the hyperbolic function used by psychologists describes what people actually do. From there on it gets complicated. When it comes to social decision-making (e.g., government investment in environmental programs), exponential discounting has been criticised for discounting future costs and benefits too severely, and thus disenfranchising future generations⁵. The hyperbolic function, which discounts the future less than an exponential function of the same discount rate, has been suggested as an alternative⁶.  However, there’s no clear consensus in the literature on which discount function or rates are appropriate for social decision-making, particularly when it involves non-market goods and services (such as biodiversity) ⁷.

That’s a quick run-down of time preference in economics and psychology. Next post I’ll explain how all this is relevant to decision-making for invasive species management, and what our paper is about.

References

1. Chapman, G. B. Temporal discounting and utility for health and money. Journal of Experimental Psychology – Learning, Memory, and Cognition 22, 771-791 (1996).
2. Hardisty, D. J. & Weber, E. U. Discounting Future Green: Money Versus the Environment. Journal of Experimental Psychology – General 138, 329-340 (2009).
3. Henderson, N. & Langford, I. Cross-disciplinary evidence for hyperbolic social discount rates. Management Science 11, 1493-1500 (1998).
4. Samuelson, P. A note on measurement of utility. The Review of Economic Studies 4, 155-161 (1937).
5. Henderson, N. & Sutherland, W. J. Two truths about discounting and their environmental consequences. Trends in Ecology & Evolution 11, 527-528 (1996).
6. Weitzman, M. On the environmental discount rate. Journal of Environmental Economics and Management 26, 200-209 (1994).
7. Frederick, S., Loewenstein, G. & O’Donoghue, T. Time discounting and time preference: a critical review. Journal of Economic Literature 40, 351-401 (2002).

I’ve spent far too long swearing at my computer today to feel inspired to write a long post (the joys of R and spatial data).   But here is a blue damselfly (Austrolestus colensonis) from my father’s garden in New Zealand to keep you entertained for another week.   Apparently they thermoregulate by changing colour – the things you learn on Wikipedia.

When I reflect on my time in the field during my PhD, I have a mixture of emotions. The overwhelming emotion is concern. I really worry about the future of this species. Already considered endangered in Victoria, so many of the sites you find Growlers are pretty awful. Many of the sites are characterised by rubbish, polluted water, overgrown weeds, encroaching development and introduced predators. This season, fire and a lack of rain also played their parts.

However, I also feel optimism when I visit the good sites. Some of the waterbodies I visited were on private property, or in fenced areas, and these were beautiful. Some sites (although far too few) were just covered in Growling Grass Frogs. This is a species that was once abundant throughout most of its distribution, so the sites with high densities always put a smile on my face. This is what they’re meant to be like. As my field work included not only locating these frogs, but also catching them, the sites with abundant Growlers were also extremely fun. It seems fitting that surprisingly, my very last night of field work was one of these sites. Not pretty or clean and definitely not one of the safest sites I’ve visited (one night last season involved police and search helicopters), this site was unremarkable last summer. This season it was completely different. We went there on a perfect night for frogging, it had rained heavily during the day, so everything was wet and the night was extremely hot and humid. There were frogs everywhere. It was a great reminder to me that things can change. Sure, some of my previous sites have disappeared, but there are still sites with Growlers and still time to make a difference for this species. Ending my very last night of field work surrounded by so many frogs was both a blessing and a curse. It feels really good to end on such a productive and fun night… but now I want to get back out there for more.

The very first Growling Grass Frog I caught for my PhD (in 2010)

My last Growling Grass Frog (in 2013)

Well, I think this guy made the right decision! Image: http://www.CartoonStock.com.

About six months ago, I moved back to Melbourne for my first postdoc (the job you do after your PhD). Researchers who move around institutions are said to be viewed more favourably by funding bodies, so I thought I’d offer my perspectives on how this may benefit or disadvantage an early-career academic in the long term, based on personal experience.

To move or not to move can be a big decision for someone coming to the end of their PhD, and I don’t think there’s enough discussion of what this decision actually means. For reference, was based at the ANU in the Conservation and Landscape Ecology (CLE) group for my PhD, and now work with the Quantitative and Applied Ecology Group (QAECO) at the University of Melbourne. These observations are not intended to be specific to the research groups I have moved to- and from: both produce great work, are well-respected, and the people associated with them well-liked. The do have distinctly different focuses though, which to some extent has influenced my experiences. So here is what I think the main pros and cons of moving are:

Con:  The first inevitable result of moving any distance is a short-term loss of productivity. Packing up your house, finding a new place to live, getting through induction processes and finding your feet sucks up a lot of time, and that’s coming from someone who only moved 650km! I can’t even imagine what it would like moving countries, with associated language barriers and admin hassles. In the bracket of early career researchers, a short-term loss of productivity can be the equivalent of one paper, which can also be the difference between being  successful or not in grant applications.

Pro: You have the opportunity to work with a bunch of new people whose research you may otherwise have only read about in passing, and you can ear-bash them about your own work. They can, in turn, introduce you to additional new people also working on relevant stuff. Depending on how proactive you are, you can create a number of new collaborations in a short space of time, which will be all the easier because you don’t have to use phone or email to communicate. Instead, you can go for coffee (which, let’s be honest, is a much more effective means of communication.)

Image: Noise to signal (Rob Cottingham)

Con: If you’re not unusually proactive, starting out at a new place can be isolating. Because people aren’t familiar with what you do and can do, you are no longer the go-to person for the things that you were before. Occasionally, I think I felt like more of a ‘postdoc’ as a third-year PhD student, than as an actual postdoc! Friends of mine who have stayed at their PhD institutions seem to have transitioned into postdoc life a bit more easily: because the senior researchers know what they are capable of, they are readily invited into discussions about co-supervising students and collaborating/helping out on papers.

Pro: If the group you move into has a different research focus, you can develop new skills much more readily – this is some of the reasoning behind why we are encouraged to move around in the first place. From my perspective, this has been a huge advantage – things that I would never have thought about trying are now possible because there are people around me from which I can learn, or at least point me in the right direction! There is a chance that if you stay put you end up doing the same thing over and over again, as you learnt from your supervisor, and they learnt from their supervisor…

Con: When you move, you have to shift your thinking to a new research culture, and navigating the way things are done and people work can be a bigger issue than you may have anticipated. I’ve occasionally found myself asking “was that not cool that I did that/said that?” What may have been accepted as gospel during your PhD can suddenly be questioned. But this is also a…

Pro: You have to shift your thinking to a new research culture. This can have the effect of reinforcing your own arguments as you have to constructive discussion, changing your thinking about the way you view the world, or at the very least help you better understand alternative perspectives. This is what we’re meant to be seeking as early career researchers: broader horizons! (even if they are a bit intimidating)

The clincher: Of course, the main thing you need to ask yourself is where you think you (and, if you have one, your partner) will be happy. You could be the most productive postdoc in the world in a great lab, but if you hate where you’re based, it’s going to have a big impact on your mental wellbeing. I wasn’t really willing to stay put in Canberra at the time I finished, so the decision was a no-brainer for me. It also helps if there is actually work in said chosen place!

So I guess the conclusion I’ve come to is that I’ve become a more well-rounded researcher for having moved, but this has probably come at the cost of some of the core metrics that I’ll be assessed against later: papers, collaborations and students. Please feel free to comment and share your own experiences below

Yes, my thesis and I took a break. I went to Vietnam for a month, doing nothing even remotely related to research (I saw a few arboreal mammals, but that was just a coincidence). My thesis stayed at home on the laptop (and six backups. It’s not paranoia. It’s caring).

Halong

If it worked for him…

I’m not saying these ideas were my best. I’ll probably look back at my scrappy little napkin notes and find most of them need more work or are just gibberish. Still, they’re avenues I wouldn’t have explored if I hadn’t been on holiday. And they might lead to new, better ideas.

Also, I feel like the difference between a career and a job is not so much about work hours or earning potential as it is about passion. I know I’ve chosen the right career path precisely because sometimes I can’t just ‘stop thinking’ about my research. And I don’t want to.

Now I hear you thinking “Really Kylie? You went on a holiday to Vietnam and all you did was think about your PhD?” No, my slightly cynical reader. I also went abseiling. That’s not my point.

I wasn’t haunted by my PhD out of stress. But taking time away from the emails, meetings and general academic clutter allowed a few latent ideas to formulate and float to the surface. My holiday gave me perspective, clarity and insight.

Another PhD Comics classic

As counter-intuitive as it may seem, I’ve found taking regular time out to be really important, both for my work output and general mental stability. And I’m not alone – see here and here for proof. There’s no need to spend big on a long, overseas holiday (let’s face it, we’re students). Just get out. Start small. A long weekend, a short weekend, an afternoon or a long bath.

When I came back I felt refreshed and ready to go again. I put a lot of that down to floating around on Halong Bay (jealous?). Disappointingly though, my thesis didn’t spare me a second thought and remained wilfully incomplete. That said, absence did make the heart grow fonder and I’m pleased to say we are getting along well.
For now…

Reflection…

ACEAS working group

With Matt West, my PhD supervisors Mick McCarthy and Sarah Converse and a bunch of other great people, I am setting up an ACEAS Working Group to look at “Decision Making for ex-situ conservation of Australian frogs”. We have secured funding from ACEAS to bring together a group of the best experts in captive breeding, amphibian ecology and decision making, plus (or including – most of them fit in at least two of those categories) end-users like managers at State, Federal and NGO level. We’ll look at the current state of captive breeding for amphibians in Australia, trying to work out what we want for the future and what we can realistically hope to achieve.

Spotted marsh frogs breeding in captivity here at the University of Melbourne (photo by A. Hamer)

To do so, we’ll collate a comprehensive dataset of captive breeding programs for amphibians worldwide – if you are aware of any of them and wish to be involved, by all means get in touch with me NOW! We should have a webpage soon and I’ll post the link.

This is very exciting – it will be a massive effort and I am very lucky to be leading this project, with so many fantastic people involved. We’ll make sure it provides useful tools for end-users, in addition to great science.

Southern Corroboree frog recovery program

With Dave Hunter in NSW, I have been doing lots of analysis to inform the captive breeding and reintroduction program for the Southern Corroboree frog. These beautiful animals have been having a hard time in the wild in the last twenty years or so and unfortunately they’re expected to go extinct pretty soon – but their captive breeding has been increasingly successful, meaning we haven’t seen the end of Corroboree frogs yet, and they stand a better chance than most other species to be successfully reintroduced.

I’ve been using population models and simple optimization to work out release strategies for different life stages (should we release cheap but fragile eggs, or robust but expensive adults?), and whether headstarting (collecting eggs, rearing them to a stage with high survival) can replace full captive breeding, and how we can maximize viability and minimize costs under different parameter scenarios.

This is the second-coolest species I am working on – I guess there must be something special between myself and yellow and black amphibians…

Toads, yellow bellies and fungi

Speaking of first love (the one you never forget), I have been working on a mark-recapture database for Yellow-bellied toads that I’ve built over the last few years. I have used Bayesian models to determine the rates of recapture, and then fed these into a beta-binomial model of disease prevalence to assess the effectiveness of chytrid surveys.

That means: if I look for frogs at different times of the year, and I get no infected individuals when I do chytrid tests, how will my confidence that there is actually no chytrid there change?

I have applied these methods to some screening I did in 2011 and 2012 to my beloved populations in the Ligurian Apennines, working with great people at Gent University in Belgium. The papers are in review but if you’re interested in the results (they’re looking pretty good), or in doing more analysis, again get in touch NOW!

Swabbing Yellow-bellied toads in Italy (photo by S. Salvidio)

A good part of the thinking on these projects has been done while riding my bike over the last few months – long hours on the saddle are great for reflecting and working out solutions with a relaxed mind. That is, if you have a good combination of padded shorts, comfortable seat and chamois crèmes. These items are rapidly becoming as important as R or JAGS to me, and when you’re doing something wrong, they provide error messages that are a lot more interpretable (but equally painful).

Updates on all these projects should be coming soon (at least, editorial decisions should…)

Together with Laura Längsfeld, I am the first author of “Constant properties of plant-frugivore networks despite fluctuations in fruit and bird communities.” , which appeared  two days ago as a preprint in Ecology. The paper is based on the findings of our theses, which we did at the Biodiversity and Climate Research Centre under the supervision of Matthias Schleuning, Eike Lena Neuschulz and Katrin Böhning-Gaese, in order to finish our undergraduate studies. In the study, we assessed the effects of landscape modification and seasonality on interactions between plants and avian frugivores*, and on the functional and interaction diversity within plant-frugivore networks.

The process from sitting in the field to final acceptance took about two and a half years. During that, there were definitely times when I did not want to see the manuscript anymore.** In hindsight, however, the effort payed off*** and it was a very insightful and rewarding experience.

You want to know more? Feel free to check out the paper:

Plein, M., Längsfeld, L., Neuschulz, E. L., Schultheiß, C., Ingmann, L., Töpfer, T., Böhning-Gaese, K. and M. Schleuning. In press. Constant properties of plant-frugivore networks despite fluctuations in fruit and bird communities in space and time. Ecology. http://dx.doi.org/10.1890/12-1213.1

* Fancy expression for “birds that eat fruits”.

** Especially, when I had to cancel a trip to Cork, Ireland so I could finish my thesis in time.

*** Since completion of my studies I can call myself a “Dipl.-Biol” now (german occupational title for someone who passed a biology degree). And, probably more important, it is the start of my scientific track record

Salomon, Y., McCarthy, M.A., Taylor, P., and Wintle, B.A. (2013). Incorporating uncertainty of management costs in sensitivity analyses of matrix population models. Conservation Biology 27: 134–144.

If managing koalas, should we target fecundity or survival rates? And if we are unsure of the effectiveness of management, should we do a bit of both? If so, how much should we spend on each?

The idea behind this paper arises from two strands. Firstly, we have the paper that Peter Baxter led on incorporating costs into sensitivity analysis. The essence of that paper was to take standard sensitivity analysis of matrix population models and incorporate costs. Previously, recommendations from sensitivity analysis typically suggested targeting the transitions to which the growth rate was most sensitive. Peter’s paper, however, simply noted that those recommendations should account for possible differences in costs. So rather than thinking about changes in population growth as a function of changes in parameters, the analysis should focus on changes in population growth as a function of how much money is spent attempting to change parameters (e.g., a manager might compare the efficiency of managing fecundity or survival). A key gap of this paper was that uncertainty in the efficiency of management was ignored.

The second strand arose from thinking about environmental management decisions as a problem of efficient allocation of limited resources, possibly where the outcome of management is uncertain. In this case, the optimal solution can be obtained using theory derived for financial investment. Elegant solutions can be obtained under particular assumptions (e.g., assuming that uncertainty is well described by normal distributions). An important insight of this paper was that attitudes to risk and the aspiration of managers drove the optimal decision. A key gap was that uncertainty might not follow a easily defined distribution, such as a normal.

This is where Yacov’s work sprang from. Does the choice of distribution matter, especially in cases when the normal distribution assumption is likely to be violated? The answer is, unfortunately, yes. The new paper shows that the optimal allocation depends on the model of uncertainty that is assumed. The paper uses two cases studies: control of over-abundant koalas and protection of olive ridley sea turtles.

I’d argue that the effectiveness of management actions is rarely measured. Uncertainty about this measured benefit of management, perhaps measured by a standard error, is rarer still. Yet even rarer is any assessment of the distribution of the uncertainty. Yet this new paper shows that assumptions about the distributional assumptions matter when aiming to find the best way to spend limited resources. However, defining appropriate models for uncertainty is likely to be difficult. I’m not sure if there is a simple answer to that problem.

A Spectrogram of music. The colour intensity is logarithmic!

I’m one of those people that loves to listen to music while working and reading. So here are a few papers I’m enjoying along with some tracks to accompany them.

A classic paper (May, 1976) deserves a classic tune. There is some challenging maths. But I have this (unrealistic but awesome) image of Lord Robert May bopping to Blue Oyster Cult and (Don’t fear) The Reaper and doing some differential equations.

I also like this paper by Olivero et al 2012 (Olivero et al., 2012), this is a good methods paper that provides some interesting insights to defining biogeographic regions. Being a methods paper you sometimes feel like you can hardly keep you head above water, but these tracks should help you swim.

A main part of my PhD is trying to describe drivers of deep-sea biodiversity at oceanic and global scales. This paper looks at the temporal variability in ocean productivity (Behrenfeld et al., 2006). One it’s an interesting paper, but two, temporal variability should be explored when looking at predictors of biodiversity. Here is a classic Thelonious Monk album which is well known for its treacherous tempo changes. If you’re not a Jazz fan, maybe pop is more your thang…

And finally,  Belanger et al (2012) predict global biogeographic structure for shallow marine environments. An interesting paper that compares existing qualitative bioregionalisation to models built on correlations between species occurrence data and environmental predictors. Not surprisingly temperature came out on top.

Until next time,

Skip.

Behrenfeld, M.J., O’Malley, R.T., Siegel, D.A., McClain, C.R., Sarmiento, J.L., Feldman, G.C., Milligan, A.J., Falkowski, P.G., Letelier, R.M. & Boss, E.S. (2006) Climate-driven trends in contemporary ocean productivity. Nature, 444, 752-755.

Belanger, C.L., Jablonski, D., Roy, K., Berke, S.K., Krug, A.Z. & Valentine, J.W. (2012) Global environmental predictors of benthic marine biogeographic structure. Proceedings of the National Academy of Sciences, 109, 14046-14051.

May, R.M. (1976) Simple mathematical models with very complicated dynamics. Nature, 261, 459-467.

Olivero, J., Márquez, A.L. & Real, R. (2012) Integrating Fuzzy Logic and Statistics to Improve the Reliable Delimitation of Biogeographic Regions and Transition Zones. Systematic Biology,

I was talking about a recent paper in Conservation Biology that describes how we used structured decision making and value of information analysis to better manage willow (Salix cinerea) invasion on the Bogong High Plains in Victoria.  The invasion was triggered by widespread bushfires in 2003 that created the perfect conditions for willow seeds to germinate in the region’s endangered sphagnum bogs and wetlands. Hundreds of thousands of seedlings established and the clean-up began.  It wasn’t entirely clear where the willow seeds had come from; there were adults on the Bogong High Plains, some of which had been planted and some that had established from these plantings, but there are also dense populations in most of the creeks and rivers that flow from the Bogong High Plains.  My research assisted Parks Victoria identify which willow populations they should target first and I’m proud that my work has helped manage this problem better.

With $1 million spent and many thousands of willows treated the clean-up is only halfway there. I estimate it will be another 5-10 years before the majority of the willows can be removed from the endangered Sphagnum bogs.  Funding to manage the willows is starting to get difficult as fire recovery funding declines and Commonwealth funding priorities change and there is a real chance that this clean-up will remain unfinished.

Left, a Sphagnum bog on the Bogong High Plains in Victoria that escaped the fire and is free of willows and right, a bog that was burnt and is full of willows prior to treatment in 2012 (photo Carmen de Rooze).

As I reflected on my experience this morning it struck me the willow story is also one of missed opportunities.  Willows were introduced to Australia shortly after European settlement.  They are a popular garden plant and have also been widely planted for erosion control in streams (an issue caused by wide-scale vegetation clearance).  Easy to propagate, fast growing with extensive root systems, willows are a great short-term revegetation solution for waterways.  Unfortunately in the longer term they are a bit too effective, their root systems extend into channels, blocking flow and causing erosion. They also spread easily downstream with new plants able to establish from small fragments of stems and branches.  These impacts were recognised soon after they were widely planted with willow removal programs in Victoria first instigated in the 1930’s.  Nevertheless, willows continued to be widely planted until the 1990’s (sometimes accompanied by advice to manage plantings carefully).

Salix cinerea, one of the most invasive willows in Australia was widely planted in the Australian Alps during revegetation programs associated with the Kiewa River (VIC) and Snowy Rover (NSW) Hydroelectric schemes, directly adjacent to mainland Australia’s rare alpine bogs and wetlands.  It’s hard not to conclude that an invasion was inevitable.   A 1994 report by Ecology Australia commissioned by the Australian Alps Liaison Committee concluded just that, predicting widespread invasion into the alpine wetlands and recommending the removal of all willows, especially Salix cinerea, from the sub-alpine and alpine zone.  Parks managers tried to respond and some willow plantings on the Bogong High Plains were removed in 2002-03 but perhaps the opportunity was already past.  The willow population and the task ahead was already substantial, it was difficult to attract funds for a problem that hadn’t yet occurred and the project went slowly. When the fires occurred in January 2003 willows were still widespread in the region and the willows took advantage of their own opportunity.

While I missed my window of opportunity this morning and we’ve missed many of our opportunities with willows on the Bogong High Plains, there are many other species waiting in the wings.  We have an opportunity to be more thoughtful as a society and as individuals about if and where we plant exotic species, especially those with a history of invasion.  I hope we can start taking them and avoid these damaging and often costly mistakes.

Now the work of Mascia and Pailler has taken a step forward as they recently launched a web tool that tracks where PADDD actions take place in the world (and here’s a link to a news piece about the launching). This is a great continuation to their paper, which more or less concluded that with the current available data  there are no ways of quantifying the global magnitude and impacts of PADDD actions.

Screenshot from PADDDtracker showing current data points from Australia. The tool allows you to view different types of PADDD actions, their causes and current status. The data can also be filtered by time points or actions taken in different types of protected areas.

Hopefully this tool will get the critical mass behind it and eventually help us to understand how to make protected areas persistent and effective. In the meanwhile it will no doubly help us to grasp the global size of the phenomena, by simply putting dots on a map and pointing out that PADDD actions are not just singular cases – they are happening on all continents and the pace of both proposed and successful PADDD actions is increasing.

Nice work guys!

The PADDDtracker

http://www.padddtracker.org/

A vibrant kākāriki Cyanoramphus novaezelandiae perches on a branch inside the predator-proof fence at Zealandia.

From the profile info on Zealandia’s website:

Zealandia is the first facility fully-dedicated to telling New Zealand’s unique conservation story. The Exhibition presents 80 million years of natural history right up to the conservation movement of the present day, with engaging interactive features and a big screen movie.

Our beautiful 225ha sanctuary valley provides routes for everyone from wheelchair-users to all-day trampers. You can choose from various fascinating guided tours and feeding talks (all free with admission), or simply roam at your leisure to find our valley’s remarkable native wildlife living free.

Display showing aerial view of the sanctuary valley.

The rugged terrain, dappled waterways and lush native regrowth create a picturesque scene, reminiscent of intact native wilderness. With a little imagination, you can get the impression of being transported to a faraway island, rather than the rather more mundane reality of having just taken a ten-minute trip from the nation’s capital.

View from the lower lake.

The island comparison is fitting, since most successful attempts at rehabilitating imperiled animal populations have been on offshore islands. The innovative technique of marooning threatened species on predator-free islands was pioneered in New Zealand, notably by early conservationist Richard Henry. Here, the ambition is to create a mainland “island” where native biodiversity can be isolated from introduced threats via a monumental 8.6 km predator-proof perimeter fence, and lending nature a helping hand via intensive native revegetation and pest eradication inside the fence.

Feathers not fur – in the absence of mammals, birds developed “mammal-like” traits.

The island land masses of New Zealand have given rise to the most peculiar wildlife. The original flora and fauna was a riot of odd, primeval, huge, massive, behemoth, gargantuan, kooky and mischievous characters that unfortunately, are mostly either extinct or hanging onto survival by a thread. A feature of the local biota that often surprises people is the absence of native land mammals, whose typical roles were substituted by birds and invertebrates. Consequently, birds evolved distinct and curious adaptations over millions of years of geographic isolation, such as flightlessness, nocturnality and a penchant for nesting on the ground. These characteristics meant that they were particularly maladapted to the novel threats that suddenly appeared with human arrival to Aotearoa/New Zealand.

Ruru/morepork Ninox novaeseelandiae, New Zealand’s only surviving native owl.

Birdland display, exhibition.

There are two sides to the Zealandia experience: the exhibition – a contemporary and dynamic museum space, and the sanctuary itself. I was impressed by the attention and creativity that has gone into curating the exhibition, which is intended to provide context for the project, and features comprehensive information-rich displays and interactive audiovisual installations. It might be a stretch to suggest that the darkened space was intended to be evocative of the nocturnal world that was once lively with the activity of New Zealand’s favourite birds, but that was the impression that it leaves.

A shag – Karuhiruhi/Pied Shag Phalacrocorax varius varius (I believe).

The actual sanctuary feels like an alternate reality in which the course of New Zealand nature took a different turn, perhaps if fewer mammalian invaders happened to take hold and humans were less inclined towards burning down the joint. Like it or not, birds tend to be endowed with special aesthetic appeal, and admittedly, it’s a great feeling to be surrounded by birds that you’ll probably never see in the wild.

Takahē Porphyrio hochstetteri, “a rare indigenous, stocky, sedentary, flightless bird”.

Takahē – “sedentary” and apparently indifferent to human approach.

I was delighted to get up close and personal with a critically endangered takahē for the first time – and you can really get close. Seeing tuatara basking in the sun in a free setting was also a first, which I loved. In 2008, the first tuatara nest on mainland New Zealand in over 200 years was discovered at Zealandia, and it definitely seems like the Karori population is doing well. New Zealand birds might seem a little drab by Australian standards, but in person, the North Island kākā, toutouwai/North Island robin and tieke/saddleback are quite endearing. For brilliant plumage, you can easily see Kākāriki, who happen to be sensationally photogenic. A night tour is definitely in order the next time I’m in Wellington, which is the only time you can see kiwi, since they are nocturnal.

Tuatara, the only surviving members of an order dating back to the dinosaurs.

The feeling that the whole Zealandia concept is the product of orchestrated human manipulation was hard to escape, even though the wildlife experience was a truly uplifting,  inspiring and worthwhile experience. It’s unfortunate that the situation is so dire for these incredible natural treasures that their persistence depends on such tremendous concentrations of human effort and ingenuity, and of course, money. For places like Zealandia to exist is reassuring of a strong societal conservation ethic and desire to ensure that people can continue to engage with biodiversity.

Ten commandments so you can publish during your PhD,an ode to Pia Lentini and The Notorious B.I.G. Over the weekend I was trying to write up my first PhD chapter as a paper, when The Notorious B.I.G got shuffled into my playlist. It’s been 20 years since Biggie released ‘Ready to die’. This got me thinking about a talk a Post Doc in QAECO lab gave on publishing during your PhD. So I decided to procrastinate from the task at hand and write an ode to Pia and Biggie Smalls. I give you the:

Ten “Crack That Paper” Commandments (play this track and recite the commandments with me!)

I been in this game for a year, doing it for me love of animals
There’s rules so you don’t publish shit, I wrote me a manual
A step by step booklet for you to get
Your game on track, not your wig pushed back

Rule nombre uno: never start the show
without, a plan ready to unfold, modify as you go.
The calendar decreed planning specially
If your times backed up, head down bum up

Number two: never not let em (supervisors) know your next move
Don’t you know em bad boys remain in silence
Take it from your highness (uh-huh)
You need to squeeze them chaps, ask them for their tricks and tips.

Number three: do tasks mu-tip-ly
If your draft is receiving feedback, don’t put your feet up
Ramp your analysis up, or even do a quick graph
Start thinkin’ about your next chapter and get a step up

Number four: know you heard this before
Never deny your timeline supply (Be realistic with your timelines ~3 Drafts per paper)

Number five: Ignore superiors comments is a liability
I don’t care if they’re a nonce, snub remarks and your paper will bounce

Number six: that first draft edit, dead it
You think a lab-head writing your manuscript, shit forget it

Seven: this rule is so underrated
Pitch your paper and ideas to a journal list updated
Methods and “Nature” don’t mix like no fingers and an itch
The wrong pitch and you’re in serious shit

Number eight: never keep no hate in you
Them cats that review your paper get knocked back too

Number nine shoulda been number one to me
If you ain’t kickin’ goals and your paper is rejected by the journal police
If reviews think your works decent by ain’t glisten
Just remember chance plays a role, there is no need to be frettin’

Number ten: a strong word called consignment
Strictly for all men, not just the best men
If you ain’t put in the hours they’ll say “hell no“
Cause they gonna want that paper as pure as the drive snow

Follow these rules you’ll have mad papers to post up
If so, all the way to the top
Slug out your paper, watch your thesis stand up
Marker read your papers, they’ll say this is too good to pass up

Hopefully Pia will provide an updated link to her of slides. But in the mean time here are some useful links to look at when thinking about publishing your work: Joern Fischer’s writing blog: http://writingajournalarticle.wordpress.com/
and Corey Bradshaw’s: http://conservationbytes.com/2012/10/22/how-to-write-a-scientific-paper/.

Cheers Skip.

And if you haven’t heard the original here it is! (Apologies to Biggie, hopefully he doesn’t roll over in his grave).

For those who aren’t so familiar with Australian snobbery, Canberra (along with our South Australian sister, (R)Adelaide) tends to get a pretty bad rap. Boring; nothing to do; weird; where are the milk bars?!

While I nod my head appreciably at the last comment (and, dare I say, groovy wine bars?!), the first few seem to come from people who a) have never been to Canberra, b) came once during primary school to visit Parliament House or c) have very poor taste.

My response to this:

• fast, flowing mountain bike trails winding through beautiful grassy woodland all of ten minutes from Canberra’s CBD;
• Two hours to the coast, two hours to the mountains;
• And then there are the after work strolls with kangaroos, kookaburras, cockatoos, wallabies, rosellas…
  

  What I did at the weekend: Canberra to Kosciuszko by bike. Photo by Milly Brent

Clearly, I am a “nature lover” (surprising, I know) and Canberra offers “nature” in droves. As well as being great for one’s physical and mental wellbeing, this also presents some great work opportunities as the field really isn’t that far away. To illustrate, I’ll briefly introduce a couple of the field sites where some of my colleagues in the Fenner School of Environment and Society work.

Mulligans Flat–Goorooyarroo Woodland Experiment

Response variables being studied in the woodland experiment: 1 Dead wood; 2 Birds; 3 Invertebrates; 4 Vegetation; 5 Reptiles; 6 Fungi; 7 Bettong reintroduction; 8 Brown Treecreeper reintroduction; 9 Kangaroos; 10 Small mammals; 11 Litter, soil and soil microbes; 12 Exclusion of feral pests. See link to the left for source info.

Located in a couple of nature reserves 15 km north of central Canberra, the Mulligans Flat–Goorooyarroo Woodland Experiment is a partnership between the Australian National University, the ACT Government and CSIRO. The aim of the project is to find ways of improving box-gum grassy woodland for biodiversity and the experiment manipulates and monitors a whole raft of factors (see figure above).

One of the many exciting aspects of this experiment is the reintroduction of the Tasmanian Bettong (Bettongia gaimardi) – it has been extinct from the mainland of Australia for 80 years.  Regarded as an ecosystem engineer, it will be interesting to learn what effects the Bettong has on the ecosystem.

Tumut Fragmentation Study

Based in the Buccleuch State Forest 100 km west of Canberra, the idea for the Tumut Fragmentation Experiment was sparked when David Lindenmayer was flying from Canberra to Melbourne. Peering out of the plane window, David saw an area of native forest that had been cleared for a radiata pine (Pinus radiata) plantation. Rather than just bulldozing the whole lot of it though, patches of native forest had been left. Representative of the original forest, these patches varied in size from half a hectare to 200 hectares thus providing a great way to study effects of forest fragmentation on biodiversity.

The Tasmanian Bettong (Bettongia gaimardi). Photo by JJ Harrison (jjharrison@facebook.com)

In all, the Conservation and Landscape Ecology group at Fenner run seven large-scale longitudinal field studies, all located in south eastern Australia. Long-term, large-scale ecological studies are pretty rare in Australia, yet provide incredibly valuable insights because many ecological processes occur at the landscape-scale, it can take a long time for ecosystems to respond to certain actions and it can also be very hard to detect ecological responses when background levels of variability are so high (just think of weather patterns versus climate change). A major impediment to establishing long-term studies is the fact that most grants last for only a few years. While there is increasing support and appreciation of long-term studies (the merits of which are nicely illustrated by the Long Term Ecological Research Network in the US), many researchers rely on passion, strong working relationships and cheap labour (i.e. their own) to maintain such research.

I have been in the Fenner School at the Australian National University for a few months now and I am just loving it. Although I am still employed by the University of Melbourne and retain strong links with the Quantitative and Applied Ecology research group, I will be based here for the duration of my grant and hopefully, fingers crossed, beyond that.

If you are ever in town, or are keen to visit, please drop me a line. We could even go to Parliament House.

Parliament House. Source: Milly Brent

But back to the paper itself. In 2004 I was a shiny new PhD student, champing at the bit to decode the population biology of the growling grass frog. I began a mark-recapture study on the species north of Melbourne, and diligently went about obtaining tissue samples from each frog I captured for a later, prospective, genetics study. Two years and 800 frogs later, the opportunity arose to get that work underway. Josh Hale had just started his own PhD on the conservation genetics of frogs in urbanising landscapes, and was keen to collaborate on the growler project. Josh’s first task was to develop a microsatellite library for the species. He emerged from the lab a year later, gasping for air and 10 kilos lighter, bearing nine new microsatellite loci for our little green friends. You can read about those in another paper, found here. Josh then set about genotyping ~200 growlers from three population clusters that I sampled during my mark-recapture work. These clusters, which I prefer to call metapopulations, were distributed along the Merri Creek at roughly 5 km intervals. Each included pools along the creek, plus neighbouring wetlands such as quarries, swamps and farm dams. Josh used his microsatellite data to assess the pattern of population subdivision in this area, and to test the effect of geographic distance and urban barriers on genetic distance.

So what did we find? In short, remarkable levels of genetic sub-division. Growlers have long been thought to be highly dispersive; vagabonds that wander the landscape in search of optimal conditions. On the contrary, our genetic work indicates that while individual frogs may be able to undertake significant journeys, the majority either can’t or don’t. Josh’s work found strong genetic sub-divisions between each of the wetland clusters I sampled, and even some sub-division within population clusters. The latter is especially interesting because the distances involved are small – 2 km max. Thus, in line with our previous occupancy and mark-recapture work (described here), Josh’s genetic work suggests that the migration rates of growlers are low and strongly distance-limited.

Another important outcome was the fact that populations separated by housing estates, industrial estates or dual-carriage roads displayed relatively high genetic distances. We can infer from this that urbanisation does indeed fragment populations of growlers; something we’ve long suspected but lacked any specific evidence of. It’s not a surprising result – imagine yourself as a little green frog attempting to cross a dual-carriage highway. But it is a vital piece of the puzzle for our understanding of the effects of urbanisation on this species, and for mitigating those effects.

Where to from here? Well, now that you mention it, our molecular work continues. Claire Keely, a PhD student in the QAEG, is working on the broader genetic structure of growlers around Melbourne, the fine-scale determinants of gene flow, and the efficacy of alternate tissue sampling techniques. You can read more about Claire’s great work here. As for Josh and I, we have one more paper to go from our collaboration. It’s a gooden too – a comparison of contemporary genetic diversity in growler populations from the Merri with a now extinct population from the adjacent Plenty River catchment. Think ancient DNA techniques and pickled frogs collected way back in the 1960s. Oh yes, there’s an exciting blog post in that. Stay tuned…

Squirrel Glider on a rope bridge over the Hume Freeway in north-east Victoria.

Wildlife can have a tough time crossing roads. Noisy, fast vehicles and wide, open gaps in habitat make it an uninviting and risky venture. This means some animals are cut off from food, shelter or loving company on the other side of the road, young have trouble dispersing to find new territories and populations might become small and genetically isolated.

On the Hume Freeway in north-east Victoria, though, specially designed structures are making life a lot easier for squirrel gliders. Rope ladders bridge the gap between trees on either side of the freeway and wooden “glider poles” in the centre median and roadsides replace missing trees, helping these small, threatened marsupials cross safely…

Read the full article here.

A squirrel glider crossing the Hume Freeway on a rope bridge. Photo by Kylie Soanes.

The first paper from the PhD thesis of Kylie Soanes has been accepted for publication in Biological Conservation. This is the research for which Kylie recently won a student award for best spoken presentation at the Ecological Society of Australia meeting.

Kylie’s research shows that squirrel gliders are able to use glider poles and rope bridges to cross a freeway that they previously didn’t cross. She has written a piece for The Conversation about her research. You can read a submitted version of the paper in Biological Conservation here.

This post was inspired by an ID mistake I made recently. Hopefully I have now learned to not do the same in the future.

Most people are very familiar with eucalypt trees. In fact, most Australians would see one every day.  They occur in almost all vegetation types across the continent in coastal, arid, tropical and alpine environments. We all have a fairly good idea of what a eucalypt looks like despite the vast array of growth forms, bark types and fruit shapes and sizes, but very few of us have a good handle on their names.  If you asked the average person on the street to tell you the name of a eucalypt, many would start by suggesting it was called Eucalyptus something-or-other, and with over 700 species they will most likely be right.

However there are also a number of trees that are commonly refered to as eucalypts, and they look like Eucalyptus for the most part, but they are not. Technically there are 7 genera that occur within the tribe Eucalypteae (Wilson et al. 2005), however I am focusing on the two most commonly observed and misidentified genera: Corymbia and Angophora. These genera are both in the same family (i.e. Myrtaceae) and tribe (i.e. Eucalypteae) as Eucalyptus  and are indeed very similar, but they have a few distinct spotting characters that can help you to tell them apart.

Angophora: This genus contains only 10 species and are often difficult to distinguish from Eucalyptus. Perhaps the most common distinguishing features are 1) the lack of an operculum (or cap) on the fruits and ribs along the fruits, and 2) the occurrence of opposite leaves on mature individuals. Since the operculum in Eucalyptus species is formed from the petals and or sepals, Angophora species are one of few eucalypts to have visible petals. Opposite leaves occur in many Eucalyptus species but typically only in the juvenile stages.

Angophora subvelutina (Rough-barked Apple) young fruits. Note the absence of an operculum, the ribs along the capsule, and the pale petals.

Corymbia: These are mostly ‘gums’ due to the bark type, and there are currently 113 species in the genus (Parra-O et al. 2009). Their primary distinguishing character is the inflorescence, which occurs as a ‘corymb’ i.e. a compound umbel, similar to the inflorescence commonly seen in the family Apiaceae. They possess an operculum on the fruits, as do Eucalyptus species. Some common species planted in suburban areas are the Red-flowering Gum, Lemon-scented Gum and the Spotted Gum.

Corymbia ficifolia (Red-flowering Gum). Note the corymb arrangement of the inflorescence. PHOTO: Bidgee (Wikimedia Commons)

References:

Parra-O C. et al. (2009). Phylogeny, major clades and infrageneric classification of Corymbia (Myrtaceae), based on nuclear ribosomal DNA and morphology. Australian Systematic Botany, 22, 384–399

Wilson P. et al. (2005). Relationships within Myrtaceae sensu lato based on a matK phylogeny. Plant Systematics and Evolution, 251: 3–19

One of the perks of doing research is attending conferences. It is great to be able to share your work with others and find out what is happening at other institutions in Australia and internationally. In 2011 I attended the International Association of Vegetation Science (IAVS) conference in Lyon, France. The theme of the 2011 symposium was ‘vegetation in and around water: patterns, processes and threats’. I presented some of my work on assessing riparian vegetation condition in Victoria.

The IAVS is an international organisation with a very broad scope of interest in vegetation science. They publish two research journals: Journal of Vegetation Science and Applied Vegetation Science. Their next (2013) annual meeting will be held in Tartu, Estonia.

Lyon, is a beautiful city and was a fantastic location for the conference. My travelling tip for Lyon: take advantage of the bike hire system, it is the perfect mode of transport for this city.

Lyon – bike hire

Lyon

In addition to attending the conference, I was also able to spend some time with some of the alpine researchers from the Université Joseph Fourier’s Laboratoire d’Ecologie Alpine near the Col du Lautaret 90km east of Grenoble. This was an amazing place, particularly at that time of the year when the alpine meadows were in flower. This is quite a contrast to the Victorian Alps of southern Australia.

Alpine meadow – Col du Lautaret, France

The French Alps

Alpine science at the Laboratoire d’Ecologie Alpine

And finally, what sort of trip to France in July would this be without some of the Tour de France… particularly in the first year the race was won by an Australian. Go Cadel!

Cadel Evans 2011 Tour de France winner

A Greybox transplant experiment

I graduated with a Masters of Philosophy at the University Melbourne earlier this year. And the first paper to come out of that work, to paraphrase Lil’ Wayne, has just “dropped like it’s hot”.

In our new paper (in early view at Basic & Applied Ecology) Pete, Mick and I illustrate the use of Bayesian informative priors to recover the inferential and predictive power of otherwise unusable pilot study data.

You can open pretty much any textbook on experimental design and one of the first things it’ll tell you is to do a pilot study. Pilot studies are standard practice and we do them so that we don’t waste resources doing an experiment only to find we had the wrong study design. The same textbooks will also tell you that if your pilot study indicates that you have to change the design of your study, forget about the data you’ve collected and get on with the new experiment. We argue that this default stance can be wrong and that otherwise unusable pilot study data can be used to construct a Bayesian prior for an analysis of a subsequent experiment.

We showed how data from a pilot study can be used in a case study on eucalypt seedling mortality during a transplant experiment conducted in Goulburn-Broken catchment in central Victoria, Australia. Using a pilot study to construct a prior prediction of mortality rate during a subsequent larger experiment, we found that including the informative prior effectively saved us thousands of dollars. Had we ignored the pilot study and included a flat prior in the final model, we would have needed to spend the extra money on monitoring hundreds more seedlings if we wanted the same amount of information as we did when including the informative prior.

Morris, W.K., Vesk, P.A., McCarthy, M.A., (early view, 2012) Profiting from pilot studies: analysing mortality using Bayesian models with informative priors Basic & Applied Ecology.

First, I should explain what my field work is all about: I want to assess the host specificity of the pollinator assemblage on the plant species Leucopogpon gnaphalioides - the Stirling Range Beard Heath. The plant is endemic to the Stirling Ranges, a mountain range in the south-west of WA approximately 70 km north of Albany. The plant is only found on a few mountain peaks with an altitude above 900 m and it is threatened by Phytophthora dieback and the warming climate. The biggest population is found on Bluff Knoll, which is the highest and most accessible peak of the range. Hence, my field site was placed on the top of this mountain, which takes ~ 1.5 – 2 hours to climb up.

In the weeks before my departure, I was planning my field trip. As I have never done this before all by myself, I was naturally worried that I would forget important things. And I sure did learn some important lessons:

Lesson 1: Don’t worry! Everything will work out eventually.

Whether it was the accommodation, the food or the hired car – everything went smoothly. Neither my volunteers nor I went to bed hungry, got dehydrated or got bitten by a snake. The latter being the biggest fear of my relatives from Europe. Please be advised: I haven’t even seen a snake – not a glimpse of a tail.

Lesson 2: 20$ of food per day and person is a lot of food!

I don’t know who came up with the idea that you need to allow 20$ per day and person for food: I bought the food for 3 people and 2 weeks, resulting in a truckload full of food and I stayed well below the allocated budget. Of course, we camped and admittedly my volunteers were not the biggest eaters.

Lesson 3: For field work in Australia: it is not sufficient to buy a Telstra Sim card.

The Stirling Ranges are quite isolated from bigger settlement, the closest being Mount Barker about 60 km away from our campsite. We stayed in Donga rooms at the park camp ground Moinup Springs, which has no mobile reception whatsoever, but a few km north there is a parking area where you get Telstra (biggest Australian network company) reception, when sitting on the table – in case anyone needs this information. However, I wasn’t aware of the fact, that you need a special phone to actually receive this network. Therefore, we were limited in our communication to the times when we were working up on the peak. Up there, the Vodafone reception was the best followed by virgin. The Telstra 3G reception was negligible.

Lesson 4: You can plan everything ahead – the weather will change your plans.

Initially I planned to have 10 field days allowing 2 days of bad weather with no data collection. I organized two volunteers and estimated around 60 hours of observations. However, this year was one of the wettest Decembers in history and after the first week I only had 3 days of data collection. As the forecast for the 2^nd week wasn’t promising at all, I decided to return to Perth and wait for better weather. Back in Perth I organised a second trip, with another rented car and two fresh volunteers. Due to time restrictions, being Christmas and that all, we could only go back for 3 days, with one being to wet and cold again.

Lesson 5: However good you are prepared, it will never be the only field season

Yes, I was naïve to think one field season would be enough, but I got caught by the worst weather in years. Luckily I still haven’t gone through my entire budget and there is enough left for a second time. Once you check out my photos you will agree that it is not the worst place for field work. In this spirit, come December the tune will be similar “It’s Christmas…eh field work time”

Bluff Knoll – good weather

Bluff Knoll bad weather

scenic views

Lasioglossum bees were very fond of L. gnaphalioides

The proof – I was there

And I even carried a leaf blower with me to tidy up the mess

The south coast of NSW is a beautiful part of Australia and contains vast areas of native vegetation within National Parks and State Forests. The patchy landscape of dense forest and cleared agricultural land near the coast provides a wide range of habitats for fauna species and the bird life is highly diverse.

The highlight of my brief twitch was stumbling across a pair of Tawny Frogmouths (Podargus strigoides) roosting in a Broad-leaved Apple tree (Angophora subvelutina). At first, one of the pair was looking directly at me, but as soon as I lost eye contact it took up the position of pretending to be a branch and was very difficult to see.

Tawny Frogmouth (Podargus strigoides) pair: one front on, the other in cryptic pose

Tawny Frogmouth (Podargus strigoides) pair: both in cryptic pose

A second highlight was locating a nest of Rainbow Lorikeets in a very tall Sydney Blue Gum (Eucalyptus saligna). There were at least 8 adult birds congregating around a tree hollow – or what looked like a pair of hollows – around 15m from the ground. The birds would come and go throughout the day but often in this large group.

Rainbow Lorikeet (Trichoglossus haematodus) nest

I’ll finish here with a group of some other birds that I saw. I’m not sure if I have started something here but I am definitely feeling the twitch a little more after this trip. Hopefully I will see some more interesting things on the next trip.

A stream in Melbourne’s water catchment, carrying water that has flowed from mountain ash forest.

The effect of timber harvesting on water yield from mountain ash forest has been studied for decades. It is topical because mountain ash forests supply a large amount of water to Melbourne, a city of more than 4 million people.

Mountain ash forests are also the main source of wood for the timber and pulp industry in the state of Victoria. There seems to be an enduring debate about the extent to which timber harvesting influences water yields.

Let’s start with an aspect of the topic that is reasonably well resolved. Because of different rates of evapotranspiration, the amount of water that arrives in streams depends on the age of the trees. When the trees are old, their rate of growth is low, the trees are widely spaced, and water use by the forest is low; streamflow is at it highest. Immediately after a fire that kills the trees or after a timber harvesting event, water use is also low; again streamflow if high.

However, as the trees grow in dense stands, their water use increases and streamflow declines. However, some trees die as the forest ages, and the rate of water use of the survivors does not increase to compensate, and eventually water use of the forest declines again; streamflow begins to increase.

Water yield (as a proportion of the maximum achievable) versus the time since fire for mountain ash forest. This assumed curve is based on the “Kuczera curve”.

The change in streamflow with forest age can be represented by a curve – it is commonly referred to as the “Kuczera curve”, named after one of the researchers who first documented the relationship between forest age and water yield. Such a curve can be represented by the mathematical equation:

Here the water yield , y(x), is expressed as a proportion of the maximum possible yield (the yield achieved in an old growth forest or immediately after timber harvesting or a fire). The age of the forest stand is x, and the parameters b and c control the shape of the curve. I’ll use b = 0.022 and c = 0.07 for the remainder of this post*. This results in the yield curve shown above.

In this case, the water yield reaches its smallest level when the forest is about 20 years of age, at which point the yield is almost 50% less than the maximum achievable.

So, this shows that if we were able to keep the forest as old growth (or at zero years of age), then we could maximize the water yield. However, as noted in my previous post, it is overly optimistic to assume that mountain ash forest won’t burn.

The chance that a forest will survive for x years given fires occur randomly in time with a mean interval of m years is:

or equivalently the cumulative distribution function of the time of fire is:

From this we can get the expected age structure from the derivative of F(x), which is defined by the probability density function:

The above is all basic survival theory assuming a constant rate of fire with forest age (McCarthy et al. 2001, email me for a copy of the paper).

Now, we can get the expected water yield E(Y) by integrating the product of the expected age structure f(x) and the water yield curve y(x). The range of the integration is zero to infinity, which are all possible ages for the forest (OK, an infinite age is impossible – I am about to impose an upper limit on forest age via timber harvesting).

This might look a bit complicated, but it is simply the average water yield, but it is a weighted average based on the amount of forest that is expected to be of different ages.

For the water yield curve I have used (y(x)), solving this integral leads to:

Plotting this expected yield versus the mean fire interval, we have:

Expected water yield from mountain ash forest (as a proportion of water yield from an old growth forest) versus the mean fire interval.

So, we see that as mean fire intervals increase above about 20 years, the expected water yield increases. If we assume the average fire interval in mountain ash forests is 100 years (assuming tree-killing fires), then the expected water yield is about 20% less than that obtained for an old growth forest.

Now, the main question was about how timber harvesting might influence the water yield. The influence will occur via the effect on the forest age structure. If we assume that forests are harvested when they reach the rotation age R, then the age structure of the forest becomes truncated at R. The probability density of f(x) above the value of R needs to “redistributed” to values below R. Hence, the truncated probability density function is given by:

Using the same logic as above, the expected water yield given a mean fire interval of m and a rotation age of R is:

One can solve this integral. I’m not going to bore you with the solution – it is not particularly simple. And rather than focusing on this, I calculate the ratio:

This is the expected water yield from a forest under a rotation age of R years relative to the expected water yield of an unharvested forest. In both cases I am accounting for unplanned fires that occur with a mean interval of m years.

When p is less than 1, timber harvesting reduces the expected water yield from the forest. When p is greater than 1, timber harvesting increases water yields. Plotting the relative yield p versus the rotation age R gives:

Expected water yield from a harvested forest as a proportion of the expected water yield from an unharvested forest versus the rotation age. This assumes that the average fire interval is 100 years. The dashed line shows where the water yield from a harvested forest matches that of an unharvested forest.

This shows that as the rotation age increases above approximately 40 years, the water yield increases. Increasing the rotation age from 50 to 100 years increases expected water yield by approximately 10% in those areas exposed to timber harvesting. A further ~10% increase would be obtained by increasing the rotation age from 100 years to 200 years. At a rotation age of 200 years, the water yield is only about 5% below what would be expected in the absence of timber harvesting (but in the presence of fires).

Of course, these increases in water yield need to be weighed against the other possible benefits (e.g., more large trees for wildlife, possibly more sawlogs) and costs (e.g., fewer stands reaching rotation age due to fire) of increased rotations. Also, note that this analysis is limited to those parts of the forest exposed to timber harvesting. Effects will be moderated proportionally depending on how much of the forest is harvested.

Interestingly, water yields can be increased by reducing rotation ages below 40 years. In fact, when the rotation age is 8 years or less, timber harvesting actually increases water yield above that obtained in the absence of timber harvesting. This is because it keeps the water yield curve near it maximum at x=0. Such short rotation ages might not be feasible because sawlogs would not grow within this time. Water quality problems due to frequent harvesting might also be problematic, as might regeneration due to paucity of seed.

Given that nominal rotation ages in mountain ash forests are greater than 50 years, the right hand section of the graph is probably more relevant to the question about effects of timber harvesting on water yields. This simple model shows that it is reasonable to claim that clearfall timber harvesting reduces water yield in mountain ash forest.

*If you would like to investigate different parameters for yourself, I have created an Excel spreadsheet that does the various calculations displayed here.

In case you are interested in trying different values for the parameters b and c, note that the minimum yield occurs at x = ln[(b+c)/b]/c. At this point, the reduction in streamflow below the maximum is:

In our paper, we use a stochastic simulation of cane toad spread to show that excluding toads from just one hundred of these artificial waterbodies could significantly reduce the probability that toads could reach the Pilbara. Importantly, this corridor connects extensive patches of suitable habitat for toads, and thus excluding toads from artificial waterbodies could prevent them from occupying 268,000 square kilometres of their potential range in Western Australia; an area of land larger than the State of Victoria! However, before we can implement this strategy, we will need to conduct a much more holistic analysis that considers not only ecological costs, but economic and societal costs as well. The optimal strategy for doing so is a topic for another day…

You can read the article free of charge courtesy of the kind folks at Journal of Applied Ecology. Also check out my co-author Ben Phillips chatting about our work over at The Conversation and on ABC Radio National, and head over to the Faculty of 1000 to read a review of our work by Mark Lonsdale and Hazel Ruth Perry.

Reid Tingley, Benjamin Phillips, Mike Letnic, Gregory Brown, Richard Shine and Stuart Baird (2013). Identifying optimal barriers to halt the invasion of cane toads Rhinella marina in arid Australia, Journal of Applied Ecology, 50, 129-137. doi: 10.1111/1365-2664.12021.

First and foremost, it was the day I met ‘the man’ in Victorian herpetology. AJC had just stepped down from a stint at the Museum spanning 50 years. An incredible length of time, and an incredible contribution to Victorian herpetology. Over that time he had amassed and curated an extraordinarily detailed herpetological collection, discovered several new species, recorded others that were new to the state, and conducted an extensive study of the herptofauna of the Victorian Mallee that laid the ground work for much ecological research in this region. But this is just a snippet of AJC’s activities. Peruse the acknowledgements section of just about any herpetological paper, thesis or report with Victorian origins from the second half of last century and there he will be, receiving copious thanks for his time, support and expertise. For a young man besotted with herpetology, you can imagine that meeting AJC was somewhat intimidating.

The second thing that I remember very clearly from that day in 2001 is the sight of AJC catching the biggest and meanest Brownsnake I’ve ever laid eyes on. We were knee deep in grass on a stony knoll adjacent to the Craigieburn Grasslands – prime Brownsnake territory. On this particular day we were working with a crew of Botanists, one of whom suddenly yelped and headed off at break-neck speed. AJC knew exactly what it meant. In flash he had pounced, emerging from long grass with 6ft of writhing, hissing fury. It was a joy to behold.

Over the next six years, courtesy of Peter Robertson, I had the great pleasure of joining AJC on field trips far and wide. AJC was a font of herpetological knowledge, and gave freely of his wisdom. But he was also, quite simply, a fantastic person to be around. His humour was unique – a combination of ancient ‘Dad’ jokes and cheeky boyish naughtiness. Yet he was statesman too. I would marvel at his inner discipline and resolve. It didn’t matter if he had just spent his fifth night sleeping on a Banana lounge in a wheel rut in the middle of nowhere, AJC would be up at the crack, boiling the billy, with a glint in his eye and readiness to lead us off for another day chasing lizards in the sun. And this was after two hip replacements! Incredible stuff.

Fittingly, AJC’s final days were spent in the Mallee doing what he loved. I was there again courtesy of AJC’s great friend and colleague, Peter Robertson. I’ll be eternally grateful to Peter for taking me on that trip. We were searching for Heath Skinks (Liopholis multiscutata) on the tall dunes of the Big Desert. Years before, Peter and AJC had discovered a population of this species in the heart of the Desert, and our objective was to clarify its regional distribution. AJC was in fine form, tearing up and down dunes, leaving me in his wake. We found several new populations, and went home happy with our achievements.

Sadly, not long after arriving home and retiring for the night AJC’s heart gave out. Peter rang me the next morning with the terrible news, and I bawled and bawled.

But AJC’s legacy lives on. As well as his many scientific contributions, AJC leaves behind a network of Mallee parks whose gazetting sought to protect the biological diversity he played no small part in documenting. Next time you find yourself in the Big Desert marveling at the vista of Mallee heath, think of AJC. If there is a Heaven, his is there, sitting beside a Mallee root fire, with billy boiling and a sea of stars overhead.

AJC laying a pit line, Big Desert, 1979 (Photo: P. Robertson)

Dale Gibbons and I seek some wisdom from the great man. (Photo: P. Robertson)

Sleeping rough didn’t faze AJC. Nor Nick Clemann for that matter (background). I was the only one who took a tent.

AJC inspects the damage after a fire swept Heath Skink habitat in the Big Desert, 2002

Peter Robertson and AJC: thick-as-thieves. Prungle, 2002

AJC’s last morning in his beloved Big Desert.

These might happen, but no. The point of doing a PhD is learning to be a researcher. We’re not really students (except in the technical sense, where we enrol in an educational institution … and are called students) — we’re apprentice researchers*

OK, so yes, a thesis is often centred on solving a problem. But I found that focussing on my thesis as ‘the point of my PhD’ drove me a little mental. Because if it failed, then I failed. If I didn’t solve the problem in 3 (or slightly more) years then it was all over. Every molehill became a mountain. A mountain called ‘You-may-as-well-just-quit-now’.

But shifting to ‘apprentice researcher’ mode, things become a bit more manageable. If the goal is learning about research, then it’s pretty impossible to fail. Your project might not come out how you hoped. But you should. You should come out with all the skills and experience you need. And if the project doesn’t work you’ll have simply learned how to deal with that earlier in your career than others.

With that in mind, here are a few lessons I’m trying to focus on for the rest of my PhD…

Learn to recognise procrastination.

Very apt

Learn when to say no. Here’s a hint – when you find yourself bursting into tears because there’s no milk in the fridge, it’s probably time to say no to that meeting or reading group you offered to run next week. Be extra-curricular in moderation.

Learn to explore. Your project is very focussed and specialised but your way of thinking doesn’t have to be. Go to seminars that are just on the edge of your field. Read papers that are only loosely connected to your work. After you’ve finished your thesis you’ll need to be able to come up with creative new projects to work on, so it helps to broaden your scope.

Extreme ironing. Google it, it’s a real thing.

Learn the oboe. Or origami. Or extreme ironing. Whatever floats your boat. Just have a hobby. Have something you love just as much as doing your research. Like these guys.

Learn what works for you. A PhD should be part of your life, not an obstacle to having one. It took me a while to figure out that crazy-long hours are actually not compulsory. And also that they weren’t particularly productive. So now I try to work well for six hours instead of wasting ten. Figure out the how to get things done in the hours that suit you. Bad habits set now will probably haunt you for the rest of your career.

So, that’s my little ‘PhD skills to-do list’ (except maybe the bit about the oboe). What’s on yours?

Related links
The Thesis Whisperer blog and e-book has awesome tips on time management and productivity in general (best $3.99 I ever spent). Same goes for Get a Life, PhD blog. Conservation Bytes recently laid out the top 5 skills for grad students here, as does Next Scientist here.

*This post was inspired by recent coffee-break chats and conference discussions with fellow QAEco lab students apprentices about the nature of the PhD program. Special thanks to Inka Veltheim for the ‘apprentice researcher’ gem.

The Regent Honeyeater is critically endangered. Check out the Regent Honeyeater Project.

Friends! Welcome back to A Beginner’s Introduction to R: Written by a Beginner for a Slightly More Beginning Beginner. In Lesson 1 we learned how to load our data into R and in Lesson 2 we learned how to navigate our data. It’s time to learn how to plot our data.

Let’s investigate the effect of tree cover on the abundance of Regent Honeyeaters. Here’s an excel file of the data. This data is made up from the Box-Ironbark Woodlands of Victoria, Australia; a highly fragmented and often degraded woodland in an agricultural landscape. Open R, set the working directory, load in the data, have a look at the data and then we’ll get started.

(Hint: save the .xlsx as a .csv and then load it in.)

setwd("/Users/LizMartin/Documents/MSc/Blog/R_plotting")
data<-read.csv("regent_honeyeater_data.csv")
dim(data)
[1] 10  4
names(data)
[1] "site"       "species"    "abundance"  "tree_cover"
data
   site           species abundance tree_cover
1     1 regent honeyeater         0 0.02442748
2     2 regent honeyeater         0 0.03027990
3     3 regent honeyeater         0 0.03727735
4     4 regent honeyeater         1 0.07382952
5     5 regent honeyeater         2 0.10661578
6     6 regent honeyeater         4 0.15617048
7     7 regent honeyeater         2 0.21501272
8     8 regent honeyeater         5 0.42245547
9     9 regent honeyeater         5 0.45305344
10   10 regent honeyeater         7 0.58517812

Righteo, we have 10 sites with regent honeyeater abundance and proportional tree cover at each site.

Let’s get plotting! To make a plot in R we can use the function plot( ). Now, R isn’t a mind reader so we have to tell it what we want to plot and how we want to plot it. We do this using function ‘arguments’. What on this great green planet is a function argument, you ask? Let me break it down for you. You want a coffee, so you ask the barista for a coffee. You say “Barista, may I please have a coffee?” But you don’t just want any type of coffee, you want a latte. So, you say “Barista, may I please have a coffee? Type = latte”. But you don’t want just any milk, you want soy milk. So, you say “Barista, may I please have a coffee? Type = latte, milk = soy.” It turns out you’re exceedingly picky about your coffee (perhaps you’re from Melbourne?), so what you really want to say is: “Barista, may I please have a coffee? Type = latte, milk = soy, strength = double, size = large, mocha = TRUE, sugar = 1.5, temperature = hot.” All these qualifying statements are function arguments; they specify the what and how of a function.

Let’s explore the relationship between regent honeyeater abundance and tree cover. We put our predictor variable, tree cover, on the x-axis and our response variable, abundance, on the y-axis. We do this within the brackets of plot( ) with function arguments, like so: plot(x=data$tree_cover , y=data$abundance).(Remember from Lesson 2 that we can refer to one column of data by using the $ operator.)

plot(x=data$tree_cover , y=data$abundance)

Tree cover versus regent honeyeater abundance. Looks like there might be a positive association. Whodathunkit?

Hip-horrah, we’ve made our first plot in R! You can see that the axes are labelled with the words we wrote for the function arguments, namely “data$tree_cover” and “data$abundance”. We can change the axes labels by using arguments for axis labels: ‘xlab=…’ and ‘ylab=…’, as I’ve done below. Don’t forget the quotation marks around the labels.

plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance")

We can also change what points our data are plotted with using the argument ‘pch=…’. I don’t like empty circles, I like full circles (pch = 19). You might like your full circles to be blue (col = “blue”) or yellow (col = “yellow”). Perhaps you want squares (pch = 22). And you might like these squares to be blue (bg=”blue”). The argument ‘bg’ stands for the ‘background’ colour of the points. You can also specify the line colour of the squares, perhaps red (col = “red”). And the width of that line (lwd = 4). You might also like your points to be big (cex = 1.5) and bigger (cex = 2.5) and really big (cex = 4). Check these out:

plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=19)
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=19, col="blue")
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=19, col="yellow")
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=22, bg="blue", col="red")
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=22, bg="blue", col="red", lwd=4)
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=22, bg="blue", col="red", lwd=4, cex=1.5)
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=22, bg="blue", col="red", lwd=4, cex=2.5)
plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=22, bg="blue", col="red", lwd=4, cex=4)

R can plot colourful and exciting graphs too, but it’s not recommended.

That looks terrible. What were you thinking?

Sometimes we want to draw line graphs instead of scatterplots, for instance, when looking at time series data. If we imagine our abundance data were sampled over 10 years (instead of 10 sites) we can plot a line graph. We do this by using the argument ‘type=…’. For line graphs the argument is type=”l”. We can also change the colour of the line (col=…), line type (lty=”dotted” or lty=”dashed”) and line width (lwd=…). Bung these into R and see what happens:

plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l")
plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l", col="purple")
plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l", col="orange")
plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l", col="purple", lty="dashed")
plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l", col="purple", lty="dotted")
plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l", col="purple", lty="dashed", lwd=3)

An increase in regent honeyeater abundance? Please keep in mind this is not a real time series.

You’re almost an R-plotting pro now. Lastly, we’re going to plot two sets of data on one graph, maybe we have two species we want to compare. Download this excel file for the Australian Magpie and we’ll get cracking. Remember to assign it to an object other than ‘data’ because otherwise you’ll overwrite the honeyeater data. Perhaps ‘magpie_data’?

magpie_data<-read.csv("magpie_data.csv")
magpie_data
   site           species abundance tree_cover
1     1 Australian Magpie         4 0.02442748
2     2 Australian Magpie         6 0.03027990
3     3 Australian Magpie         3 0.03727735
4     4 Australian Magpie         5 0.07382952
5     5 Australian Magpie         4 0.10661578
6     6 Australian Magpie         7 0.15617048
7     7 Australian Magpie         5 0.21501272
8     8 Australian Magpie         2 0.42245547
9     9 Australian Magpie         4 0.45305344
10   10 Australian Magpie         6 0.58517812

To plot two sets of data on the same graph we first create a plot with one dataset using the plot( ) function and any arguments we wish. Then we use the function points( ) to add extra points onto that graph. The function points( ) has the same arguments as plot( ), the only difference is it doesn’t make a new graph, rather it plots onto the graph you have open at the time.

plot(x=data$tree_cover , y=data$abundance, xlab="Tree Cover", ylab="Abundance", pch=19, col="blue)
points(x=magpie_data$tree_cover , y=magpie_data$abundance, xlab="Tree Cover", ylab="Abundance", pch=24, bg="red", col="red")

Regent Honeyeater (blue dots) and Australian Magpie (red triangles) abundance versus tree cover.

We can do the same thing for line graphs using the function lines( ):

plot(x=data$site , y=data$abundance, xlab="Year", ylab="Abundance", type="l", col="blue", lwd=3)
lines(x=magpie_data$site , y=magpie_data$abundance, xlab="Year", ylab="Abundance", type="l", col="red", lwd=3)

Regent honeyeater (blue line) and Australian Magpie (blue line)abundance over time. (Honeyeaters are still critically endangers. This is simulated data. #context)

And that’s it for today, folks. I have some data entry I need to do, which is one thing Excel is still useful for. For a break down of all the arguments available for plot( ) and associated functions like points( ) and lines( ) use the help( ) function in R. E.g. help(plot).

• During a delicious vego lunch, I enjoyed a productive back and forth about Multi-dimensional Scaling (MDS) and it’s misapplication in analyses of changes in community composition. MDS axes have no units – can we quantitatively compare something that has no units? Are MDS analyses only useful for data exploration? Should we be creating a new, statistically and philosophically robust method? Perhaps multi-species heirarchical models of abundance can solve the problem?

• At the conference dinner, passions were flying about the relative benefits and disadvantages of mechanistic versus correlative species distribution models (SDMs). On one hand mechanistic SDMs take process into account, while correlative SDMs only identify patterns. On the other hand, correlative SDMs implicitly incorporate species interactions while their mechanistic counterparts map the fundamental niche and don’t account for interactions. Correlative SDMs are extrapolated into future environments (sometimes frivolously), but perhaps the projection of mechanistic SDMs into future environments is unrealistic because it doesn’t account for species interactions and how those interactions might change in new environments.

• Over tea and bikkies, there was a rowdy discussion about the lines drawn around environmental units. Is the concept of an ecological community a farce? How can we manage communities as environmental units if the definition is just a blurry line in the sand? Should we be managing at the species level because the concept is more tangible? But, is it really more tangible or are we kidding ourselves there too? Even the line between invasive and native is blurred by naturalization.

• Over a seemingly bottomless glass of wine, we got to talking about the interaction between social disadvantage and imposter syndrome in academia, a topic close to my heart. Being from a disadvantaged background can make one feel out of their depth at all stages of academia: undergraduate, post-graduate and even as a post-doc. We asked if there is more pressure to hide the bogon within at more ‘prestigious’ universities? We agreed that ‘faking it to you make it’ is an effective strategy because, in the end, it doesn’t matter how you got there, you got there and you deserve to be there.

• “So, what do you do?” Time and time again, I was challenged with the ’30 second thesis’. I’m happy to say I’ve got it downpat: I’m working on a new method to model the distribution of rare species, or species for which we don’t have much occurrence data, using community level data. I received some great feedback on assumptions that I’m making in my models and how they might play out statistically, which helped me to refine my ideas.

To me, this is why we get together: these great after-hours conversations that get us thinking, building networks, creating collaborations and challenging our views. A big thank you to everyone who made my first ESA so much fun, and in particular, all the good eggs listed below for all the stimulating conversations listed above.

Will Morris, Mike Bode, Pete Vesk, Dale Nimmo, Jos Moore, Yung En Chee and others I’m sure I’ve forgotten.

Lots of cute animals were found, I’ve included photos of some of my favourites below.

The Growling Grass Frog, Litoria raniformis

Me holding a large female and a small male Growling Grass Frog

Common Spadefoot Toad metamorph, Neobatrachus sudelli

Common Spadefoot Toad displaying it’s little digging ‘spades’

Velvet worms!

An Eastern Long-necked Turtle, Chelodina longicollis, attempting to lay eggs on the road

A baby Jacky Dragon, Amphibolurus muricatus

A young Stumpy-tail Lizard, Tiliqua rugosa

Old growth mountain ash forest (for scale, note the Toyota Landcruiser and people near the base of the tree)

There is a report in today’s Age about the decline of Leadbeater’s possum in the face of fires and timber harvesting. Professor David Lindenmayer of the Australian National University notes that the situation is dire, and that timber harvesting should be ceased in mountain ash forests. It is worth noting here that these forest are the main source of wood for native forestry in Victoria. The forest industry representative, Lisa Marty, says that fire is the problem, not timber harvesting. Who is correct?

Leadbeater’s possum relies on old forest for its survival. While it can persist in recently burnt forest, it only does so where large hollow bearing trees that existed before the fire remain present. However, if these trees are killed by the fire, most will collapse within decades. As a consequence, in a forest that is exposed to unplanned fires, only a fraction of that forest will be high quality habitat – it will be those areas that are currently old, and those that are young but were previously old.

It is possible to do a back-of-the-envelope calculation to determine that proportion. Let’s assume for simplicity that high quality habitat only occurs in forest where the previous fire occurred more than 200 years ago. The average fire interval in these forest is approximately 100 years. If fires occur at any site randomly in time, then the probability that a site will escape a fire and reach 200 years of age will be exp(-200/100), which is approximately 14%.

This means if an area of forest is conserved from timber harvesting, in the long term we would expect that only ~14% of it would be old enough to support high quality habitat. The actual proportion would fluctuate around that number over time, but it is a realistic (albeit rough) assessment of the amount of forest we would expect to occur in an old state at some time in the future.

So, if we set aside 10,000 ha for conservation, we would expect less than 1,500 ha to be suitable. The more that is set aside, the greater the amount we expect to be suitable in the future. Further, a larger conservation reserve also provides a larger buffer against fluctuations caused by irregular large fires.

Therefore, in a world exposed to unplanned fires, any harvesting reduces the expected area of forest that will become old because harvesting reduces the age of trees. In this light, timber harvesting certainly reduces the production of older forest, thereby contributing to declines of old-growth dependent species.

This is basically an issue about how we should plan to conserve biodiversity in the presence of unplanned fires or other random (but expected) events. It is a topic I studied during my PhD, leading to the following paper:

McCarthy, M.A., and Burgman, M.A. (1995) Coping with uncertainty in forest wildlife planning. Forest Ecology and Management 74: 23–36. [Email for PDF]

In fire-prone environments, forest managers should assume that fires will occur, but at a time they cannot predict. Thinking of a fire regime as a stochastic process provides a framework for predicting impacts.

Livestock grazing has attracted the attention of ecologists for many decades. We have learnt a lot but the story is not complete. The effects of livestock grazing on vegetation remain a controversial and important subject. This is highlighted by the recent controversy in Victoria on grazing in the high country – if you are interested in this issue feel free to read the posts from some fellow QAEcologists here, here and here.

My work does not deal with alpine grazing but is concerned with livestock grazing in native vegetation within the floodplains of the Victorian Riverina. The Victorian Riverina subregion has the highest proportion of cleared vegetation in Australia, with only 5.1% native vegetation remaining in 2001 (NLWRA 2001). Much of the remaining vegetation occurs in narrow strips along roadsides and creeks (Figure 1).

Figure 1. Aerial photo of Broken and Boosey Creeks in the Victorian Riverina

The remnant riparian vegetation is under significant threat from weed invasion and grazing. Many of the riparian remnants are licenced to be grazed by sheep or cattle. Some sites are also fenced from adjacent farms. However, a licence does not necessarily mean a remnant is grazed (or how heavily grazed) and a fence does not necessarily mean a remnant is more heavily influenced by the adjacent farm. I used three separate indicators of grazing for analysis: presence/absence of a fence, presence/absence of grazing licence, and presence/absence of recent heavy grazing.

Based on some of my previous work showing the change in vegetation cover with increasing distance from the creek edge (see my post on sampling for details), I separated remnant sites into two zones – Inner (first 25 m) and Outer (all outside 25 m) – for examining the effect of livestock grazing.

Using the zoned sites, I wanted to firstly determine the specific effect of grazing on each of the vegetation life forms. This is not directly possible with these data as I’m not modelling the specific change of cover at a given site that is grazed or not grazed. Instead I’ve done this by comparing vegetation attributes of sites that have been grazed to sites that have not been grazed. Boosted Regression Tree (BRT) analysis was used to do this, since they are an effective tool for modelling the influence of a particular variable while accounting for effects of other variables. I built models including management and environmental variables and used bootstrapping to perform multiple runs of the model (let me know if you want any more information about the modelling process).

These models estimated the cover of native vegetation life forms when grazed and ungrazed. All but the smallest native life forms had lower cover in grazed sites, as did plant litter (Figure 2). Bare ground was higher in grazed sites. Small native shrubs showed the weakest response to grazing, perhaps due to their prostrate growth form. Exotic life forms were much less sensitive to grazing, and in the case of small herbs increased in cover when grazed (Figure 2). These general findings are well supported in the literature, which highlights that over time, intense grazing will result in vegetation dominated by species with traits suited to grazing, such as short lived (annuals), low-growing perennials, small seeds, high regrowth potential, plasticity in response to grazing, and high fecundity (Landsberg et al. 1999; McIntyre and Lavorel 2001). The data in Figure 2 only include comparisons of inner zones.

Figure 2. The modelled comparison between cover in grazed (grey) and ungrazed (black) sites accounting for variation in other environmental variables. Life forms are: MS: medium shrubs, SS: small shrubs, MTG: medium tufted graminoids, MH: medium herbs, SH: small herbs, MNG: medium non-tufted graminoids, BG: bare ground and LT: litter. Appending letters refer to natives (.N) and exotics (.EX).

So this is great, we can detect correlative effects of grazing on vegetation cover in this system from my data. However, this is not particularly helpful if these effects are much weaker than other factors. For management to be effective it must have a relatively strong influence on the vegetation. This is partially taken into account in the BRT models above, however I did an additional test of this influence by comparing the management variables (grazing, licencing and fencing) against a range of environmental variables using hierarchical partitioning.

This method determined the contribution of each variable to the goodness of fit for the response variable (native vegetation cover). These individual contributions were summed into ‘management’ and ‘environmental’ variables to compare the combined influence. Grazing management was less influential than environmental variables but was higher and more variable in the inner zone (Figure 3). This suggests that the vegetation (and ground layer) cover is more susceptible to grazing management near the creek.

Figure 3. The combined contributions of management (dark) and environmental (light) variables on explained variance of native vegetation cover.

Hopefully I have explained this satisfactorily so that you can see that within this study site I have detected correlative responses of understorey life forms to grazing variables, and that these management indicators appear to be significant drivers of understorey cover.

Feel free to ask if anything is unclear. Also, for a great brief summary of grazing effects see Ian Lunt’s latest post in the Conversation here.

Chris

References:

Landsberg J., Lavorel S. & Stol J. (1999) Grazing response groups among understorey plants in arid rangelands. Journal of Vegetation Science 10, 683-96

McIntyre S. & Lavorel S. (2001) Livestock grazing in subtropical pastures: steps in the analysis of attribute response and plant functional types. Journal of Ecology 89, 209-26

NLWRA. (2001) National land & water resources audit: Australian native vegetation assessment. National Heritage Trust

I am sandwiched between two Bayesian talks. I’ve heard a rumour that Gerry Quinn and Ralph Mac Nally will present an entertaining double act (prior to me), while Brendan Wintle will be talking about Bayesian detection (posterior to me; sorry about that. I couldn’t resist…).

How many surveys should be done to account for variation in detectability?

My talk will discuss how to optimize the number of times one should survey a site to maximize the chance of observing a species. This might be the number of nights to survey a stream to find a frog species, or the number of plots to search to detect a species in a region, or even the number of searchers to send to a site.

The key is that because of variation in detectability (over time, over space, or among people), a single visit to a site (or a single site within a region) might have, by chance, a low probability of detection. So there is an incentive to search more than once – that will increase the chance of having a higher detection rate at least once. However, going more than one incurs greater travel costs, which eats into the time available to actually search.

The tension between more shorter searches (with extra travel time) and fewer longer searches (with less total travel time) creates a trade-off. We developed and analyzed a model to optimize this trade-off. And we used a field experiment to evaluate it.

If you want a sneak preview, a pdf file of my slides (minus the best jokes) is posted here.

I’ll also be auto-tweeting this talk using the hash-tag #micktalk in case you can’t make it to Melbourne and want to follow along with the slides. For overseas folk, my talk is on Thursday 1415 Melbourne summer time, which is Thursday0315 UTC (I’m not expecting too many folks following along in London!), Wednesday1915 in Los Angeles, and Wednesday 2215 in New York.

I was one of those newbies in the scene and the task posed some decisions: One was about the level of scientific depth. The audience was a mixed bunch of students and  researchers of various botanical fields. Ecology, although it dominated, is sort of the odd one out, because we use many different tools and investigate a multitude of questions that sometimes makes it hard even to understand your colleagues. Therefore, I started to explain the background and reasoning of my study right from the basics.

So, what did I talk about? Mostly the dodo. And here were my two slides and some brief description to it.

The dodo and the rat (Macleari’s Rat) were both endemic to islands and went extinct after the arrival of European settlers. However, the rat did not go alone, instead it was accompanied by a specialised tick, Ixoden nitens (pic only shows a tick from the same genus). This process of the simultaneous loss of dependent and host is called coextinction.

The IUCN lists 19,817 species (August 2012) in their red list of threatened species and with the magnitude of species interactions, it is easy to imagine the potential of species loss due to coextinction.

To illustrate the concept of translocation and the underlying problems that emerge when translocating 2 species together, I used the very unrealistic, but – at least from my perspective – quite vivid example of the simultaneous translocation of the dodo and a specialised bird louse. Actually, it’s not even unrealistic, it’s just not feasible anymore, but potentially…

Anyways, there’re quite a few questions that appear, e.g. How many individuals to translocate, without extirpating the original population. Is it possible to translocate both at the same time or does the dodo population needs to be established first, to support the louse population. Many, many questions, which I will have to answer in the next few months. And yes, I will use a realistic example – it’s just not as cool as the dodo.

And next time? I’ll probably stick to the science and leave the experimental stuff – the dodo caused some mixed feelings with the judges.

* Melbourne University Botany Graduate Society

** If you have given a three minute talk in front of a broadly interested audience, I’d love to read about it!

I’ll be speaking on beta-diversity in the deep-sea. My talk will focus on briefly describing the bathyal environment around Australia and New Zealand, the methodological approach I’ve used to create maps of beta-diversity based on presence-only records and I will attempt to test the hypothesis does deep-sea biogeographical patterns resemble those seen in shallow marine environments.

If you can’t make it to Wellington and are keen to have a look at a few of my maps and slides they are available for your viewing pleasure.

eDNA may revolutionise surveys for cryptic invasive species such as the cane toad, but is eDNA more cost-efficient than traditional survey methods?

QAECO is seeking an honours or postgraduate student with an interest in genetics and environmental monitoring. The successful candidate will work with researchers from Bio21 and the School of Botany at the University of Melbourne to develop and test novel methods for conducting biodiversity surveys using environmental DNA (eDNA). eDNA can be used to detect species from water samples, and thus holds great promise for reducing survey costs (see a recent blog on this topic over at qaeco.com). However, there are numerous uncertainties associated with this technology. How many water samples should we take? How quickly does eDNA degrade, and what conditions accelerate DNA denaturation? How does a species’ abundance influence detection?

The successful candidate will attempt to answer these types of questions in the laboratory using invasive and endangered aquatic vertebrates as model systems. Optimal protocols determined in the laboratory will then be applied to field surveys around greater Melbourne. The results of these field surveys will then be used to compare the cost-efficiency of traditional and eDNA sampling strategies.

If this sounds like something that interests you, drop me a line at reid.tingley@unimelb.edu.au.

I will be speaking on “Science for Conservation Decisions”. The lecture will introduce environmental decision theory, and describe an experimental test of a particular example of this theory (optimal searching for a species; it seemed to work quite well in that instance). I will finish by briefly noting how decision theory helps clarify thinking, and that is perhaps its greatest strength.

If you are interested in seeing the slides I am planning to use, I have posted them here.

It seems odd that the grazing leases would be extended before the study were designed. Surely any scientific study of these effects would dictate which areas should be grazed and which would not? And the confidence that one more study will answer the questions “once and for all” seems bold. As in Victoria, initiating the grazing before the study is designed gives an appearance of using science as a cloak for a policy decision.

We will wait and see what happens. In the meantime, check out this article by Ian Lunt about how livestock grazing can help conservation in some cases, and harm biodiversity in others. He also writes a little about the difference between science and sham.

What a good idea!  I was completely blown away by what I saw…

Common everlasting (Chrysocephalum apiculatum) and Blue grass-lilies (Caesia calliantha).

Blue pincushions (Brunonia australis) and Blue grass-lilies, with Common everlasting in the background.

A diverse patch, including Lemon beauty-heads (Calocephalus citreus), Scaly buttons (Leptorhynchos squamatus), Curved rice-flower (Pimelea curviflora), Kangaroo grass (Themeda triandra), and Blue devil (Eryngium ovinum).

I started this post thinking it was a good excuse just to put up some pretty grassland pictures.  But it’s turned into something a little more than that – for me, at least.

Lately, I’ve been thinking a lot about grasslands, why they’re important and the role they play in our city.  This afternoon’s visit involved a bit of a detour – not much, but it felt like an effort after a weekend away.  It would have been pretty easy to drive past the turn-off and head straight home.  I’ve done that before.  Just decided that it didn’t seem worth the effort on that particular day.  But today I was determined.

So, was it worth the effort?

Absolutely.  I got a real kick out of seeing this urban grassland in such good condition.  I felt a sense of satisfaction from detecting and identifying the many beautiful plant species that I see so rarely in my day-to-day life.  And spending an hour wandering around the grassland provided a sense of calm and time-out that is hard to find in the inner north.  So, while I don’t get there often enough, it is important to me to know that these places exist.

It’s no secret that Melbourne’s grasslands have taken a hammering.  There’s now less than 1% of the original grasslands of the Victorian volcanic plain remaining.  And only about 10% of that is in good condition.  More grasslands are certain to be lost under recently approved plans to further expand the city.  To offset this loss, 15,000 hectares have been set aside to the west of Melbourne for the Western Grassland Reserves.  Much of the grassland in the new reserves is not in great condition.  Significant investment in restoration and research is required.  In my books, this is not a great outcome – yet.  We shouldn’t be trading certain losses of remnant grasslands now for uncertain future gains.  But it has the potential to be great.  Large areas of diverse native grassland would be a significant achievement.

But we should remember that small grasslands in urban environments are worth protecting too.  And just in case you need some more convincing … here are a few more photos:

Yellow rush-lily (Tricoryne elatior) with Blue devil (Eryngium ovinum)

Slender speedwell (Veronica gracilis)

Common rice-flower (Pimelea humilis)

A sea of colour. Common everlasting, Blue pincushions and Chocolate lilies (Arthropodium strictum).

Blue pincushions and common everlasting. Late afternoon, Sunbury.

A freetail bat (Mormopterus sp.) which we trapped so we could build a key of the local dialect.

Bats are fascinating creatures – there are about 1,200 species, and they are the second most abundant mammal (behind the rodents). They occupy every continent aside from Antarctica, and are the only mammal capable of true flight.

In spite of their incredible diversity and abundance, there is still much we don’t know about bat ecology and management: they are quite cryptic which can make them tricky to study!  I had the opportunity to carry out some research on bats in agricultural areas as part of my PhD, the findings of this which have just been published in PLoS ONE (which you can find here and download for free). Here’s a summary of what we did, what we found, and some of the other details that get lost in the scientific publication process.

Why bats in agricultural areas? Well for a couple of very important reasons. Bats are relatively hardy little creatures, and often make up a large proportion of the mammalian fauna in agricultural areas when other species have become locally extinct. As insectivores, they provide a vital ecosystem service to farms by controlling insect pests, and consume 40-100% of their body weight in a single night.

The difference between land sparing and wildlife friendly farming.
Image: Phalan et al (2011), Food Policy 36, S62

Also, there is ongoing disagreement as to how we should be managing farming landscapes to balance food production and conservation. Is it better to segregate large fields of intensive production from conservation areas, or have lower-input system which integrate the two? The so-called “wildlife-friendly farming vs. land sparing” debate has primarily focussed on what is best for plant and bird species, but evidence for bats remains inconclusive. Given their role in controlling pests, I thought they warranted more attention.

Finally, I was concerned with the potential loss of travelling stock routes (which was the focus of my PhD research). Originally used for moving stock, stock routes now form large wooded corridors in the fertile flat portions of otherwise heavily cleared eastern Australian farming landscapes. Bats require old trees with hollows for roosting and breeding, and access to water. So although we strongly suspected that stock routes would form important habitat for bats, we wanted to quantify this.

The aim of this research was to answer a couple of questions:

1)      Are travelling stock routes providing important habitat for bats in the highly fragmented wheat-sheep belt?

2)    Is there evidence that bats will benefit from schemes designed to make the landscape more “friendly” to wildlife?

A bat detector on a large remnant tree in a heavily grazed exotic pasture. The wooded stock route can be seen in the background.

Now, bats use echolocation to locate their prey, and their calls are ultrasonic so we can’t hear them. Each bat species produces a slightly different type of call, so automated keys can be used to count the number of species flying past a “bat detector”. The only problem is that there are regional differences in these calls, sort of like the dialects of a language. Because we weren’t fully confident that we had an adequate grasp on the bat dialect of the slopes region I was working in, we set out to collect more calls. With the help of a couple of seasoned bat researchers, Caragh Threlfall and Anna McConville, we managed to trap nine species from our study area and collect local reference calls.

Armed with our regionally-specific key, we then set up bat detectors in travelling stock routes, as well as paddocks which represented all the common land uses in the area: native and exotic pastures, and lucerne, wheat and canola crops.

A bumper crop of insects in the UV light trap.

We also set out black-light traps to determine whether differences in bat activity between paddocks could be driven by the amount of insects active each night. Field work during the summer of 2010/2011 was made all the more fun by the La Niña conditions which brought drought-breaking rains, and associated shutting of roads, bogging of the field vehicle, and bumper collections of insects in the light traps.

It might look like just a weathered old tree in a paddock, but we recorded over 500 calls and 10 species around this Bimble box (Eucalyptus populnea) in a single night!

During this process I had many an interesting conversation with landholders, who were keen to know what I was investigating, and why I was stomping around their paddocks with buckets, lead batteries, plastic boxes and UV lights. After explaining the aims of the project, some were sceptical about whether I would record any bats at all (including the owner of the tree on the right) because they didn’t think that they occurred in the area. This just goes to show that in spite of their great numbers, bats and the services they provide still often go overlooked because of their nocturnal nature and inaudible calls.

Over 228 detector nights and 2,475 survey hours, we recorded 91,969 bat calls representing 13 taxa. We also collected 5.14 kg of dried insects. The data told us that all of the land uses supported the same amount of insects yet:

• There was higher bat activity in stock routes, and more bat species in wider stock routes;
• There were more trees characteristic of bat roosts in the stock routes;
• There were more bat species, and higher bat activity in the native pastures, the lowest-intensity form of farming; and
• There was also more species and activity in paddocks with retained habitat structures such as trees and logs.

This just goes to show that potentially economically important groups such as bats can benefit from low-intensity farming and agri-environment schemes, and that the retention of existing wooded areas such as travelling stock routes should be encouraged. I hope I’ve also convinced you that bats are worthy of your attention: want to learn more? Check out the Australasian Bat Society or Bat Conservation International.

Although I suggested in my last post that my next one would be about the hierarchical bit of hierarchical models, I am instead taking this opportunity to briefly present a little group called ‘MUBoGS’ and a little conference called ‘The MUBoGS Mini Conference’.

The postgraduate students in The School of Botany, here at Uni Melb, are getting active.  The Melbourne University Botany Graduate Society, affectionately known as MUBoGS, is holding the first Graduate School of Botany Mini Conference.

Our MUBoGS flyer. Note: mini plant (Wow!)

Most postgraduate members from the School (including Hons, Masters and PhDs) across all the lab groups will be presenting their thesis topics in 3 minutes and with the aid of up to 2 (non-animated) slides.

3-minute presentations, or ‘speed talks’ are becoming more frequent, both in conference settings but also within academic institutions. Annually, the University of Queensland runs a very successful 3MT competition that is open to students from many Universities from around the world.

The idea is to give postgrads a chance to practice effective communication of their research to an educated but unspecialised audience (see here for more info and some examples of cool, calm, collected and concise speakers, and here for good advice about preparing for a 3MT comp).

Putting together a 3-minute talk takes a whole lot longer than just 3 minutes.
The aim is to communicate what you’re researching, why you’re doing it and why it’s important.  Often these talks are judged firstly on your communication style, specifically whether you can get a non-specialised audience to understand the significance of your research, but also on how engaging you are.  Condensing the fundamentals of your research into something that can be delivered and understood efficiently is both surprisingly challenging and surprisingly rewarding.

Can you make people understand and care about your research in 3 minutes?

The postgraduate students here in The School of Botany will soon find out.  Our first Mini Conference is shaping up to be an excellent day.  As well as our students presenting, we are having panel discussions of PhD and Postdocs talking about transitions from PhD to ‘Post-PhD-Work’ and sharing some hints about ‘things I wished somebody had told me…’  We are also planning some social trivia and there will be many ‘Mini Plants’ to win throughout the day.

Stay tuned for more Mini Conference news…

And… here’s a photo of Eucalyptus incrassata (+ ants) flowering in the mallee…

Eucalyptus incrassata from Murray Sunset

What traits make some plants easier to see than others?

If you’ve seen previous posts, you would realise that I am interested in the topic of imperfect detectability in field surveys. I’m interested in what influences detectability, how to account for it in analyses, and what it means when designing surveillance and survey programs.

Well, we have some new research on the topic, in which Georgia Garrard leads a paper in Methods in Ecology and Evolution that investigates relationships between plant traits and rates of detection. Georgia has written a blog post about the paper. Check it out!

You can get the author-submitted version of the manuscript here. Or please auto-request a copy of the printed version of the manuscript by clicking here.

This post is about a new paper now available online in Methods in Ecology and Evolution.  Please email me at georgia.garrard@rmit.edu.au if you’d like a copy.

Imperfect detectability of plants and animals during ecological surveys is now widely recognised.  If unaccounted for, imperfect detectability can lead to biased estimates of abundance or occupancy, impaired ability to detect change or response to management action, poorly informed management decisions, escape of invasive species and increased risk of extinction of endangered species.

A range of methods exist for estimating detectability, including distance sampling, mark-recapture, N-mixture models, zero-inflated occupancy models and time-to-detection models.  Each of these models has its own set of assumptions, data requirements and applications.   Often, the data requirements of these models are heavy; data, including abundance counts, presence-absence observations and times-to-detection, are variously required from multiple sites and multiple observers.  All this means that, to date, estimates of detection probability are available for relatively few species.

In this paper, we ask whether we can learn about the influence of species traits on detectability, and use trait-based models to predict the detectability of species for which no species-specific model exists.

Using a time-to-detection model, we investigate the influence of a range of species traits on the detectability of grassland plant species.  Examples of the traits investigated were local abundance, height, likelihood of flowering at the time of survey, flower colour, leaf area, number of similar grassland species and whether the species grows in clumps.

We found that local abundance has a clear influence on detectability, with species that occur in higher numbers having lower detection times (higher detection rates) than those occurring in small numbers (Figure 1).  Species are also more likely to be detected if they are unique or in their peak flowering month at the time of survey, although these results are less definitive.

Figure 1. The relative size of the influence of traits on detection rate (mean and 95% credible intervals).

Our results also show that flower colour may have a large effect on detectability, with pink and red flowered species potentially more easily detected than those with inconspicuous or yellow flowers.  This makes sense in native grasslands, where there are many yellow flowers and few pink or red flowers.  The influence of flower colour is still very uncertain – it will be interesting to see whether more objective measures of flower colour (we used only coarse categories) can help to resolve this uncertainty.

Using trait-based detectability models, we were able to predict average times-to-detection reasonably well to new species (Figure 2).

Figure 2. Comparison of ln(ave detection time) estimates for eight species withheld from the model-fitting data set, as predicted by the trait-based (x-axis) and single-species (y-axis) models.  Filled diamonds are predictions for expert observers and open circels are predictions for intermediate observers.  Error bars are 95% credible intervals.

I’m probably biased, but I think trait-based detectability models are an exciting development in the field of detectability research.  With more than 1300 nationally-listed threatened plant species and another 400 animal species in Australia alone, it’s impossible to consider constructing a species-specific detectability model for every threatened species.  While they may not perfectly predict individual species’ detection probabilities, trait-based models should provide sensible bounded estimates of detectability on which to base survey design and effort requirements.

Mick McCarthy has made an author-submitted copy of the manuscript available here.

So here goes folks, the first of a series of blogs sharing these pics with you. Where better to start than with a research project on the endangered (and magnificent!) Inland Carpet Python in Victoria, which I was lucky enough to take part in during 2001…..

A tough old girl from the Murray Sunset National Park. What a cracker.

Her kind of country – Red Gum and Black Box woodland

And her kind of roost site – old hollow-bearing Black Box over lignum

Warby Ranges granite, where I did my Honours research. Fantastic country.

Some more…

My trusty Feroza put in the hard yards. God bless that car.

No. 4, who I tracked for 6 months, had a penchant for the Homestead garden. Here he is resting up after a big meal.

Spot the python…. (centre top)

Another haunt of No. 4′s: a big White Box. He was in the hollow jutting out centre right.

No. 4 liked to hunt rabbits in the summer. Here’s a burrow he spent sometime down. Part of my work was to investigate the timing of rabbit burrow ripping to minimise effects on python populations. The answer: don’t do it in summer!

Another threatening process. A logging coupe in Red Gum west of Mildura. Logging and firewood collection remove vital homesites for pythons, in the form of old hollow bearing trees and fallen hollow timber.

Even small hollows like this one are extremely important to individual pythons – they return to them year-after-year. And they take years to form. If anyone reading this blog heats their house with Red Gum firewood, please, please, switch to plantation timber.

I’ve getting a few more Aaaaaa’s than ARARRARRRR’s of late so I thought it’d be good to pass on some of my wisdom. One great part of any computer language is the ability to batch tasks that would normally involve a brain numbing amount of mouse clicking of command prompt typing.

In R there are a number of ways to get a loop going.

The most basic loop can be achieved using for:

for (i in 1:10)
{ mat=matrix(c(i,i))
print(mat)
}

The above loop will produce a two column matrix that contains the values one to ten for increasing with each row.

If you wanted you could turn the above loop into a function. That’s right! You can easily create your own functions in R.

Matrix.Gen <- function(x)

{

for (i in 1:10)

{ mat=matrix(c(i,i))

print(mat)

}

}

Matrix.Gen ()

To run your own function you just type in Matrix.Gen () into your R command line. When creating a loop in R it is
important to make sure you understand how the loops are subset in your function. This is especially important when
you have more than one loop.There are two schools of thought when sub setting code one where you just make sure
you close a bracket once you open it.

Eg:

Matrix.Gen <- function(x) {

for (i in 1:10) {

mat=matrix(c(i,i))

print(mat)  }

}

I personally don’t like this method as if you lose a bracket it can get really confusing! I prefer making sure that my brackets line up and you indent the code as you introduce a loop.

Eg:

Matrix.Gen <- function(x)

        {  for (i in 1:10)
            {   mat=matrix(c(i,i))
                print(mat)
             }
        }

If you set the open and close brackets in a line then they are a lot easier to track. I also like to place my brackets under the function or for loop so you can easily link the appropriate bracket to loop.

A good use for a loop in r is creating a function that runs through your model fitting or model selection. Using a loop to do this can speed up the process of manually entering a single model at a time. The code I will display below run an all model selection method for 249 individual species. This bit of code runs two loops, often referred to as a nested loop. A nested loop is a loop within a loop and this is where you need to start taking care with your bracketing, otherwise it can get really confusing.

Import data as csv files. There are three files that will be called into this loop. Environmental data, ##Species Data and the models I want to fit

Env_AusNZ360 = read.csv ("EnvPoints10m360lon.csv)
Oph_spp = read.csv ("Spp20_Oph_AusNZ.csv")
mods = read.csv("PPM_mod_select_lin_poly.csv")

This next bit of code creates a factor R can recognize based on my species data. I do this so I can get R to call in a single species at a time. These individual species will be called in the first loop (the j loop).

Oph_spp = Oph_spp[,c(1,2,3,4,5,6,7,9,11,12)]
WhichSpp = factor(Oph_spp$Spp, exclude = NA)
classes = levels(WhichSpp)
nClasses = length(classes)

fit.ppm <- function(x)
 { for (j in 1:nClasses)
     { print(classes[j])
       dat2add <- data.frame(who = Oph_spp[,1], Oph_spp[,2:3], depth = Oph_spp[,5], Oph_spp[,6:9])
         for (i in 1:78)
          { ft.ppm0.4 =  ppm(mods[i,2], dat2add[dat2add$who==classes[j],], Env_AusNZ360, 0.5)
            mods[i,2] <- mods[i,2]
            mods[i,3] <- classes[j]
            mods[i,4] <- ft.ppm0.4$ll
          }
          print(mods)
          write.table(mods, "C:\\Possion PP\\PPMresults_Spp_10m_05scale.txt",j, sep=".", quote=F)
      }
   print("Finished")
 }

So we’re going down a level ( and if you’ve seen inception you’ll get that link) To clarify: the  i loop is nested within the j loop so It’ll run 78 models for the number of species I have. I’ve then closed the i loop and asked it so save my model results to my mods csv file within the j loop. This will mean it’ll append the result of each species as it is solved by R.

So loops can be confusing, but they can greatly speed up your outputs if you get the basics right. Another way to increase the speed of doing calculations in R is to introduce parallelization.  Most modern computers have more than one core or processor which means instead of running a single loop like the above code you can set your computer to paralleling your code. The speed you can increase your loops is based on the number of cores you have. Most modern computers have 4-6, which means if you have a large data set or something that is slow, you can increase the output speed by 4-6 times! I know what you’re thinking that sounds complicated?… In fact that sounds way above my head?… You’d have to be a computer programmer or something like that to get parallelization working? We’ll luckily in R all the hard work has been done.

All you need is two software packages that you can download with ease. I use linux for parallelization, but there are windows versions available.

For linux the packages are:

install.packages("foreach")
install.packages(“doMC”)

For windows:

install.packages("foreach")
install.packages(“doSNOW”)

Then load packages:

library(“foreach”)
library(“doSNOW”)

This tells R how many cores to use, the more cores you use the faster the loop will run, but the harder your machine will have to work.
For the purposes of this example I’m going to set it at four.

registerDoMC(cores=x)
registerDoMC(cores=4)

For windows it’s slightly different:

cl <- makeCluster(4)
registerDoSNOW(cl)

To get R to start paralleling all you need to do is convert your for loops to foreach loops:

Matrix.Gen <- function(x)
   { for (i in 1:10)
       { mat=matrix(c(i,i))
         print(mat)
        }
   }

Will become:

Matrix.Gen <- function(x)
     { foreach(i=1:10) %dopar% 
         { mat=matrix(c(i,i))
           print(mat)
          }
      }

Now the foreach call will parallelize your loop based on the number of cores you assign. This should now run 4 times faster!
The same can be done for the nested loop example above.

fit.ppm <- function(x)
   { foreach(j=1:nClasses) %dopar%
       { print(classes[j])
         dat2add <- data.frame(who = Oph_spp[,1], Oph_spp[,2:3], depth = Oph_spp[,5], Oph_spp[,6:9])
            foreach(i=1:78) %dopar%
               { ft.ppm0.4 =  ppm(mods[i,2], dat2add[dat2add$who==classes[j],], Env_AusNZ360, 0.5)
                 mods[i,2] <- mods[i,2]
                 mods[i,3] <- classes[j]
                 mods[i,4] <- ft.ppm0.4$ll
               }
          print(mods)
          write.table(mods, "C:\\Possion PP\\PPMresults_Spp_10m_05scale.txt",j, sep=".", quote=F)
        }
      print("Finished")
    }

I hope this has been useful. Looping and paralleling your code should speed up your analyses. And while your loop is running you can continue to work on other projects or enjoy a cup of tea and a Monte Carlo biscuit. Like I’m going to do now. Until next time. Skip.

This post has been a long time coming. My excuse is my readership reduced to zero after my mum, when faced with insoluble Excel-madness, got half way through Lesson 1, gave up and paid me to do her stats for her. And like every honest ecology student, I took that money, bought a case of beer shiny, new sleeping bag and went hiking. With positive reinforcement like that, I don’t think I’ll tell her about this post. Mum’s the word, readers – I’ve got my eye on a new thermarest!

Excursion with mum-funded sleeping bag (not pictured). Mt Sugarloaf summit, Cathedral Ranges, looking over the Acheron Valley.

In Lesson 1, we became acquainted with R. We discovered that R uses the R console to communicate with us and we use R scripts to talk to R. We learned how to execute commands (tell R to do something), set our working directory (tell R where to look for files we want to load), how to load and look at our data.

The plural of platypus is platypus or platypuses, not platypi.

In this post we’re going to figure out how to navigate our data by executing commands instead of clicking our mouse, like we would in Excel. And if all goes well, we’ll overcome the anxiety of not being able to constantly see our data. Or instead, like me, you’ll just develop an incessant compulsion to use the function head( ) to look at your data.

Alrighty, let’s open R, set our working directory, load our data and have a look at it, exactly like we did in Lesson 1. I’m going to load a .csv file containing platypus biometric measurements. (You can download a .xlxs file of this data here. WordPress doesn’t allow .csv files to be uploaded, so you will need to open the file in Excel and then save as a .csv. What is a blog post without a little irony, ey?)

setwd("/Users/LizMartin/blog/R lesson 2")               ##set working directory
platypus<-read.csv("platypus_biometrics.csv")        ##load in data and save to object 'platypus'

The first thing we’ll want to know about our data is how much we have. To find out the number of rows and columns in our dataframe we use the function dim( ) (short for ‘dimensions’) on the object of our interest:

dim(platypus)              ##dimensions of dataframe
[1] 10  3

The output tells us that there are 10 rows and 3 columns in our dataframe.

The next thing we’ll want to know is what variables we have. To find out the names of our columns we use the function names( ), like so:

names(platypus)               ##names of columns
[1] "body_length"    "bill_length"    "tail_fat_index"

By now, we’ve got a little bit of separation anxiety from our data because we can’t see it. So, we’ll use the function head( ) to show us the first 6 rows of our dataframe, just to check in:

head(platypus)                  ##top 6 rows
  body_length bill_length tail_fat_index
1        45.0         9.5              1
2        41.4        10.8              4
3        48.1        11.1              3
4        48.2        10.2              3
5        46.5        11.7              2
6        54.5         8.7              5

Phew, it’s still there! We can also check the bottom 6 entries using the function tail( ). Give it a go, you’ll feel better:

tail(platypus)                  ##bottom 6 rows
   body_length bill_length tail_fat_index
5         46.5        11.7              2
6         54.5         8.7              5
7         49.1        11.9              4
8         39.5         8.3              3
9         51.0         8.2              2
10        34.8         8.7              3

The next thing you should know is R indexes dataframes first by row and then by column. This means that the 7th row in the 1st column is cell 7,1. And the 2nd row in the 3rd column is cell 2,3. We can tell R to show us the contents of individual cells by using these indexes, like such:

platypus[7,1]           ##contents of row 7 in column 1
[1] 49.1
platypus[2,3]           ##contents of row 2 in column 3
[1] 4

If we want to look at all the biometrics for just one individual, say the 7th one, then we leave out the column index. Like so:

platypus[7,]                       ##all data in row 7
  body_length bill_length tail_fat_index
7        49.1        11.9              4

Or we might be interested in all rows in just one column. In which case, we leave out the row index. For example:

platypus[,1]                       ##all rows in column 1
[1] 45.0 41.4 48.1 48.2 46.5 54.5 49.1 39.5 51.0 34.8

Maybe we want to look at a group of individuals, say the first five. Then we tell R to show us rows 1 to 5 using a colon, like this ’1:5′, and we put this in the row index place:

platypus[1:5,]           ##all data in rows 1 to 5
  body_length bill_length tail_fat_index
1        45.0         9.5              1
2        41.4        10.8              4
3        48.1        11.1              3
4        48.2        10.2              3
5        46.5        11.7              2

We can do the same thing for columns:

platypus[,1:2]           ##all data in columns 1 to 2
   body_length bill_length
1         45.0         9.5
2         41.4        10.8
3         48.1        11.1
4         48.2        10.2
5         46.5        11.7
6         54.5         8.7
7         49.1        11.9
8         39.5         8.3
9         51.0         8.2
10        34.8         8.7

If we’re interested in non-sequential rows, say individuals 3, 5 and 9, we use the concatenate function c( ), for instance, c(3,5,9). And, you guessed it, we can do the same thing for columns. Try this on for size:

platypus[c(3,5,9),]       ##rows 3, 5 and 9
  body_length bill_length tail_fat_index
3        48.1        11.1              3
5        46.5        11.7              2
9        51.0         8.2              2

We could keep going for any combination of rows and columns that tickle our fancy using these indexes. Just remember, row first and then column – write it on a post-it and post it on your computer monitor. If you’re into algebra, which let’s face it not many people are, you can think of it in terms of row i and column j. So any combination of row and column can be represented as [i,j].

But, really, I’ve led you down the garden path in the name of wholesome education, because the easiest way to look at one column of data is to use the $ operator and the name of the column. For instance, to look at the column ‘bill_length’ in our dataframe called ‘platypus’ we can simply type ‘platypus$bill_length’, as below:

platypus$bill_length        ##column 'bill_length' in dataframe 'platypus'
[1]  9.5 10.8 11.1 10.2 11.7  8.7 11.9  8.3  8.2  8.7

So much easier, isn’t it?

Now we know how to access and navigate our data by executing commands in R, we can start to analyse it. So, I’ll stop banging on and leave you with some example functions to try out. Adiós!

sum(platypus$bill_length)          ##sum
mean(platypus$body_length)         ##mean
sd(platypus[1:5,1])                ##standard deviation
var(platypus[c(1,3:9),2])          ##variance
length(platypus$tail_fat_index)    ##length of column (or count in Excel)
median(platypus[,1])               ##median

we’re all a little bit scared of being ‘found out’

Over the years imposter syndrome has come up time and time again in PhD pep talks and discussion groups. I’ve never, not once, heard someone say they’ve never felt it. Even senior, high achieving, silverback/legend-status academics admit to feeling it from time to time. Imposter syndrome, it seemed, was everywhere.

Couldn’t go past this imposter meme for fans of ‘The Office’

I worry that calling it a ‘syndrome’ gives it too much power and almost makes it more debilitating. If we think of it as a syndrome, as something abnormal, then we can’t get on with our work until we’ve found a cure. “Sorry, I won’t be able to submit that chapter on time. I’m off sick with imposter syndrome… “

But we’re not sick, we’re just feeling a little scared and insecure. We all feel this way at least some of the time. The trick is not to let it fester and become destructive.

So, next time you get that sinking, anxious feeling, take a deep breath and remind yourself that you’re here to learn. Talk it out with your friends or advisors – remember, there’s a good chance they’re feeling the same way at that very moment. And be patient with yourself. Yes, that academic is amazing and has a lot of publications, but they’ve also been working in the field since before you even started highschool. Give yourself (and your ego) a break.

Also, don’t forget what you’ve already achieved – big or small. Admittedly, it’s a little dorky, but I have a big, yellow ‘happy place’ folder. Every time I attend a workshop or conference, get a grant, have an abstract accepted, even have work show up in the local newsletter it goes into the folder. Flick through it whenever you feel like you haven’t achieved anything yet (or after your supervisor requests 100+ revisions on one of your chapter drafts).

Take an ego boost from this puppy. He knows the score.

An invasive exotic species, Sagittaria platyphylla, dominating a wetland in Cobrawonga, Victoria (2008)

About this time last year, I wrote an article for H2O Thinking, a water management magazine published by eWater (until recently the eWater CRC). While the turnaround time is nothing to envy, the piece found its place on the web earlier this week.

In the article, I focus on two questions that anyone* who has spent any time along a river will surely have asked:

Why are river banks, floodplains and floodplain wetlands so susceptible to alien species invasion?

And what can we do about it?

Well, I’m not going to give the game away, but lets just say that the words “flow” and “regulation” do make an appearance. Click here for more scintillating reading (?!).

A very nice wetland near Cobram, Victoria (2007)

*OK, anyone like me would surely have asked…

In our project we created a Bayesian State-and-Transition model that was informed by the literature and could be used to inform management decisions. This model included the most common grassland management options and assessed their success at improving plant species richness and decreasing weed cover. It also considered the financial and social costs associated with conducting the different management actions. From our research it was evident that in terms of financial and social costs as well as benefits in terms of species richness and weed control, burning was the best management option.

On Friday night we were awarded with first place in the Australian branch of the competition.

Our job last week was to collate data on the distribution and demography of several Australian frogs, and to use those data to develop models of population dynamics under climate change. The objective: to assess extinction risks, and identify management options to mitigate those risks.  See more on our group page.

Now, here’s the great thing about ACEAS. Autecologists like me are keen to tackle these sorts of projects, but often lack the modelling skills required. On the flip side, ecological modellers interested in these questions are often constrained by their knowledge of the species under study, particularly where to find the information needed to parameterise their models (which frequently hides in Honours or PhD theses, unpublished reports, someone’s hard drive or a dusty filing cabinet). ACEAS brings these two groups together, and harnesses their collective skills to tackle important questions in applied ecology.

Want a workshop of your own? Well, it just so happens that the next round of ACEAS funding is in October. You can enquire with ACEAS direct from here, and find instructions of how to apply here. I can highly recommend it!

I’ll write more about the frog project in blogs to come. For now, let me say a big thanks to Tracey Regan and David Keith for leading the charge and inviting me along, to all the group members for a great week (including our most recent member, Reid Tingley), and to ACEAS for their fantastic hospitality, and their broader efforts to bridge the gap between those with the data, and those with the models.

The all-nighter timeline. Image: PhD comics.

Here are some attempted-quotes from the Forum: (‘attempted’ because I can’t write very quickly and didn’t get them down verbatim… but the general gist should be right!)

“We are paid for the work we do during the day, and promoted based on the work we do at night”

 “We are doctors of philosophy – yet many grants are awarded on the basis of whether there will be some sort of monetary outcome at the other end. What happened??”

“I think we need to ask if science is exploitative”

So – is there any hope?

Well, I’m trying remain optimistic. There was a strong drive during the forum to identify (and look at implementing, or at least advocating) some solutions to the above problems. Longer contracts. Joint appointments for researcher couples. Better metrics. Better links between science and industry. Better communication of our science. Flexible work options. A change of culture so that those who move out of academia into industry or teaching aren’t viewed as ‘failures’.

Final dodgy phone photo: Chuwen Keynote address, Prof. Brian Schmidt

Ultimately, one should not have to sacrifice their health, personal relationships and sanity for the sake of their job (though many do, and this problem is certainly not limited to research). And to put it all into perspective, Brian Schmidt pointed out that there are very few unemployed PhDs out there, so we shouldn’t let it get us down. He himself was fourth in line for the position which he eventually won the Nobel Prize for; the three candidates ahead of him just happened to pull out. If he hadn’t succeeded in getting that job, he was considering moving on to teaching.

Cathy Foley, who is Chief of CSIRO Materials Science and Engineering, told us to remember that our skills are very transferrable, and realise we can sell that to employers outside of academia. Due to the nature of our work, we have generally mastered written and oral communication, time and project management, teaching and mentoring, analytical thinking, are good at problem solving, and are flexible and dedicated to our work.

But, more than anything else, the problem is that we as young researchers are inherently very competitive and proud (if trying to get a word in edgeways during the discussions was anything to go by!) So, whether we can be happy in these alternative non-academic positions is another thing, and this really will require a fundamental culture change where we are told by our mentors that teaching and industry positions are jobs to be proud of. Of course I’m being a huge hypocrite here, because I’m not putting my hand up to leave – I love doing research at a university, and if someone out there could make it a little easier for me to continue doing so I’d really appreciate it!

A final added note: If you’re an EMCR who would like to be kept up-to-date on funding and training opportunities and receive newsletters from Academy of Science, you can sign up to their mailing list here.

Dodgy phone photo #2: Panel discussion (Prof Andrew Hopkins, Dr Marnie Blewitt, Prof Tony Peacock)

The gist of the Forum was that frankly, the situation for young scientists in Australia is pretty dire. Here are some themes that came up time and time again (these do not necessarily reflect my views, just what I was hearing at the Forum. If you were there and think that I got it wrong, feel free to comment).

Job security: Young researchers (who aren’t teachers/lecturers) can expect to wait until they’re about 50 to get a permanent position, if they’re lucky. In the meantime, they jump between 1-5 year contracts, and spend the last 6 months of those contracts trying to get their hands on the next pot of money to keep them going. And you have to work bloody hard to get that next pot of money, because there is nowhere near enough to go around and funding is very very competitive.

During this process, you are expected to continually progress in your career, accumulating ever more students, money and staff underneath you. There are no options for those who do not want to eventually manage a big lab, because there is no money for “staff scientist” or “rolling postdoc” positions. You are also expected to move A LOT, from lab to lab and institution and institution, preferably between countries to gain skills and meet people. If you have family and other commitments to consider, tough luck. This makes it especially hard if there are two researchers in a relationship (not uncommon), and one usually has to sacrifice their career for the sake of the other.

Women in science: We were told the story Ruby Payne-Scott, a pioneer of radio astronomy who worked for CSIRO during the 1940s. Ruby kept the fact that she was married a secret, because married women were not allowed full-time positions in the public service. In 1951, she was forced to retire after she fell pregnant with her first child Peter Hall; a huge loss to her field (though Peter was a big gain to the field of statistics. Not that that justifies anything). So have things gotten any better for Australian women in science?

They certainly have at Walter and Eliza Hall Institute, where they have introduced some fantastic initiatives to address the gender equity issue. These were outlined by Marnie Blewitt: special fellowships for female laboratory heads, family rooms, childcare support, parent-friendly meeting times, and technical support for those on maternity leave (more detail available here). These initiatives should be loudly applauded, and I hope they are taken up by many more employers.

Ruby Payne-Scott, Alec Little, and “Chris” Christianson at Radiophysics’ Potts Hill Reservoir field station, probably in late 1948. From ATNF Historical Photographic Archive, Miller Goss

Sadly though, another piece of advice that I was given was that if I do have children, I should not bother going part time. If I want to keep my job or get the next job, I’m going to end up working the equivalent of full-time hours anyway, sneaking in paper writing while my baby is sleeping. “It’s sexist to expect women to go part-time when they have children, when they end up working the same hours as men”. While I think this person’s heart was in the right place, this, to me, is mortifying. Why do women in our research culture feel they have to work the equivalent of full-time hours after they’ve just given birth?!?

In reality, parenting responsibilities are only part of the problem, because there are still some serious fundamental cultural issues that need to be addressed. Do we just implicitly assume female scientists are less competent? (this new paper in PNAS shows that we probably do, at least in some fields).

Poor science education and communication: Unfortunately, there is something of stigma attached to the move from research to teaching, and particularly teaching below the tertiary level. If you’re teaching, it’s because you weren’t successful enough as a full-time researcher. Scientists don’t want to be teachers these days, and the level of qualification of science teachers drops with age, most acutely for physics. This is having an impact on what should be the next generation of scientists, as the number of high school students interested in a career in science continues to decline. Ian Chubb noted that in a survey, 33% of year 12 science students, and only 1% of non-science students thought that science may be useful for their future.

Dodgy phone photo #3: Young scientists asking questions at the Forum.

Clearly, we’re not doing a very good job of communicating the importance of our work! And this is having serious flow-on effects, with murmurs of delaying all contracts funded through the ARC for a year as a means of bringing the federal budget back to surplus (presumably because it was believed that if you put science off for a year, there wouldn’t be any great disadvantage or loss to society).

The lack of mobility between academic science and industry: Another issue is that young scientists apparently don’t want to work in industry or the public service, which may be perceived to be something of a cop-out (much like teaching). So there aren’t enough jobs in academia, but we’re hesitant to take them anywhere else! This may be partly due to the fact that PhD supervision is somewhat one-dimensional, with supervisors and universities grooming students to become academics and not providing diverse training opportunities for positions elsewhere. There is also the (justfied) fear that if you leave full-time research, you won’t be able to get back in.

There is also the fact that industry doesn’t really want us. Apparently we’re over qualified, can’t work in teams, don’t understand how to balance books, and can’t work outside of an academic time frame (i.e. we like to have a bit of a think about things before making recommendations and decisions). We also only go and talk to those in industry when we need partners for linkage grants, so we’re really missing out on some key sources of non-ARC funding and collaboration.

Where does all that leave us? Read on.

Home of the Australian Academy of Science, the Shine Dome (aka “Martian embassy“). Image: Wikipedia/Bidgee.

I have spent the last couple of days attending “Science Pathways: Getting science on the national agenda”, which was the “inaugural meeting of the Early-Mid Career Researcher Forum”. That’s a bit of a mouthful really, so what does it mean?

“Early-Mid Career Researchers” (EMCRs) are individuals who are within about 15 years of finishing their PhDs or other higher degrees. The Australian Academy of Science, which is made up of ~450 of Australia’s most esteemed scientists, thought it was time to cast their eye to this next generation who will take their place once they retire (and in academia, this translates to “when they die”) to see how they were feeling about their careers in research. The aim was to address two key questions: what are the key issues facing young scientists today, and how can we make their voices heard by the policy makers that fund their future?

Dodgy phone photo #1: Chief Scientist, Prof. Ian Chubb

Along with the 140-odd young scientists that attended from a wide range of research institutions across the country, there were also some heavyweights present to share their experiences and thoughts. Chief Scientist Ian Chubb. Nobel Laureate Brian Schmidt. Former CEO of the ARC, Margaret Sheil. CEO of the CRCs Association, Tony Peacock, and CEO of Science and Technology Australia, Anna-Maria Arrabia, just to name a few.

Now, I’ll start by saying that I was feeling quite good about life as a young researcher before heading up to the Shine Dome in Canberra. I have recently started a position at the University of Melbourne, where l get paid to do what I love and work with intelligent, friendly, and interested people. I optimistically (or naively) had the perception that if you work really hard, tick all of the boxes of what’s expected of you, you can continue to do research. Nuh-uh. One of the most telling pieces of advice I heard from a senior researcher (over drinks) went like this: “If you’re the second-best paediatrician in Australia you’re doing really well. If you’re the second best teacher, that’s great. If you’re the second-best researcher in your field: you’re f****d”

Why? Read on.

Clearly, the assumption that past refugia predict where future refugia will be depends on the nature of climate change. It is more likely if future climate change mimics past climate change, but may also be the case if past and future climates are different. Why? One reason is that local climates in refugia tend to be unique to their surroundings–buffering climate extremes for example. Or maybe they are mesic habitats in an arid landscape.

Populations of E. baxteri and closely related species on rocky, western slopes (‘West’ and ‘Endemics’) less diverse than eastern populations

In either case, we can better judge whether a past refugia will be a future one if we understand how local climates and landscape features are related to species persistence. We have a new paper out in the Journal of Biogeography that is an attempt at doing this. We looked within a regional refugium to find local conditions that match patterns of genetic diversity across the landscape.

We found populations of eucalypts on side slopes and valleys with deep, moist soils and protection from strong westerly winds were more genetically diverse than populations on exposed rocky slopes.

The Serra Range in the Grampians National Park in Victoria, Australia. The Grampians ranges are a cuesta formation with western dip slopes (left side of photo) and steep eastern slopes. Species may have persisted in the mesic east-facing slopes and disappeared from rocky west-facing slopes when climates were colder and drier

It seems fairly obvious that species would persist in nicer habitats, but this story was only revealed with genetics (not obvious from morphology or species composition). These findings help strengthen the case that mesic local climates in semi-arid Australia may serve as refugia.

The second part of this paper is about gene flow between species.. A complicated story for another day..

Caladenia alpina (orchid), Acacia sp. (wattle) and Diuris sp. (orchid)

A large part of my PhD will be building a hierarchical Bayesian model that incorporates plant functional traits as species level predictors of plant growth and reproduction.

In my next couple of posts I will endeavor to explain all these terms (but don’t worry there will also be a few posts about what is flowering around Melbourne). For my first post in this series about my research, I would like to start from the start and try to explain what a model is.

Needless to say, googling “model” gives you mixed results. I found it easier to ask my Supervisor for a good book on modelling.  The result of this was a book, ‘How to Model It, Problem solving for the Computer Age’ by Anthony Starfield, Karl Smith and Andrew Bleloch.

These authors suggest that models are formalisations of the relationships between things.

There are many types of models:

•  mental models
• graphical models
• statistical models
• mathematical models

A mathematical or statistical model, concerns the relationships between quantifiable things or values. Values that are actively changing are called variables, values that are less likely to change are parameters (these can mediate the effect of variables). Parameters that can’t be changed are called constants.

Hardenbergia violacea (is flowering outside at the moment)

How to model it begins with the authors posing a problem; “how long will it take you to read this book?” A length of time immediately pops into ones head. But, if you go to the trouble of really thinking about it (which the book makes you do), you might find that the time it will take may depend on a few things.

Maybe the time it takes to read the book will depend on how fast I read one page, and then on how many pages there are in the book. Maybe it will depend on the number of tasks I have to do on each page and the time it takes me to do each task (maybe how many things are in flower outside?).

If I take my mental model a bit further, I can formulate a simple equation:

T = pP + wWP

Where:

T = time taken to read the book,
p = time to read one page of the book,
P = number of pages in the book,
W = number of ‘tasks’ per page, and
w = time taken to complete each task

P,  W, and  w changes between different books and so these are variables (because they are actively changing).  p may not change as much from book to book, as everyone probably has an their own average reading time. Therefore, p is a parameter.

The authors of How to model it show that we all make (mental) models all the time.

A model is a purposeful representation of something. Models can be used for synthesising information, for estimating values of important variables, for prediction, decision-making and explanation.  One major benefit of using a formal model is that you’re forced to be explicit about your thought processes. It can be quite challenging and very revealing when you’re confronted with all your assumptions.

In my research I will be a using a type of statistical model called a hierarchical  Bayesian model.  My aim is to model plant growth and reproduction incorporating the influence of species-specific plant functional traits.

Of course, an important consideration should always be what the point of your model is.  Indeed the authors of How to model it suggest you should always ask yourself three questions:

• What (exactly) am I doing? (I should be able to describe it precisely)
• Why am I doing it? (How does it fit into the solution)
• How will it help? (What will I do with the outcome once I have it)

Acacia pycnantha inflorescence

Great questions to ask oneself.  What’s the point?   Why do I want to model plant growth and reproduction?  Why don’t I just spend all my time taking photos of orchids and Acacias?   The answers to all this and more will pop up in a future post or two.

In my next blog, however, I will attempt to give an overview of the ‘hierarchical’ bit of my modelling approach.

Stay tuned…

Ref:

Starfield AM, Smith KA & Bleloch AL (1994) How to model it, problem solving for the computer age.  McGraw Hill, New York.

Thanks to W.K Morris for proof reading.

This time of year (late winter into early spring) is a great time to have eyes and a nose.  One of the main reasons for this, is because it is the flowering time for many Acacias.  For those poor souls who have not been acquainted with, or have not paid much attention to this group of flowering plants, they often look like this (a bright, bright mass of yellow):

A bright mass of sweet smelling yellow

There are many species of Acacia flowering around Melbourne at the moment, and with only a little effort they are relatively straightforward to identify.  For a successful ID of Acacias you need to know something about their foliage and flowers.  A copy of ‘Native trees and shrubs of south-eastern Australia’ by Leon Costerman is also highly recommended.  This comes in a big comprehensive book covering common trees and shrubs of NSW, Vic and SA, but also a very handy smaller field guide ‘Trees of Victoria and adjoining areas’ (‘Mini Costermans’).

Underneath a bipinnate leaf. Note pinnae and pinnules.

Acacias have two broad types of foliage :

Bipinnate leaves  are ‘feathery’ and divided into pinnae. The pinnae further divided into pinnules.  All acacias have bipinnate leaves to start with.
In some species bipinnate leaves are replaced as the plant grows older by leaf like bodies called phyllodes.  Some species have broad, flat phyllodes whilst others can have spiny, needlelike or very slender phyllodes.

For ID in the field, the type of foliage (bipinnate or phyllode) is a good spotting characteristic.  After this, other foliage characteristics such as the number and placement of phyllode veins or the position and number of glands are significant characters for identification.

Different types of phyllodes

Look closely and you might see the individual flowers that make up this globular inflorescence. You can see the number of closed flower buds on the inflorescence on the right.

Individual flowers of Acacia species are tiny.  These tiny flowers are aggregated together to form an inflorescence and Acacias also have two broad inflorescence types: globular or cylindrical.

This division of inflorescence structure is also an important grouping characteristic for ID in the field.  Most Acacia species flower from late winter to early spring.  Time of flowering can be useful to distinguish between some species.

Single globular inflorescence, cylindrical spike and a raceme of globular flower heads.

An Acacia seed pod, a typical pea pod

Finally the fruiting pods are important features.  Size, shape and surface texture are important.

In mini Costermans (which you should put in your pocket before leaving the house, always) the first major groupings occur between :

- Wattles with bipinnate leaves and globular flower heads (go to page 132)
- Wattles with phyllodes and cylindrical inflorescences (go to page 135)
- Wattles with phyllodes and globular flower heads (go to page 138)

Once you have identified one of these broad groups, you can move to another section of the field guide and start looking in more detail at spotting characters such as placement of glands or veins, distribution and flowering times.  Clear drawings with detailed inserts of relevant features such as leaf glands are super helpful and make Acacia ID relatively straightforward.  So give it a go, it is surprisingly satisfying.

For example:

Acacia oxycedrus, photographed July in The Grampians

Acacia oxycedrus (photo left)

Foliage: Stiff green flat phyllodes 2-4 cm long, slightly curved, tapering to a very sharp point, 3-4 raised veins.
Flowers: (Jul-Oct) Yellow, crowded on axis in cylindrical spikes to 3 cm long.

Acacia longifolia (photo below)

Foliage: Flat  green phyllodes, spreading to erect on stiff branches, 2 (-4) main veins, gland near base.
Flowers: (Jul-Oct) Yellow, in spikes.

Acacia longifolia, photographed August at Seaford

Acacia verticillata (photo below)

Foliage: Green phyllodes mostly in whorls of about 6, often needle-like (or flattened) with sharp point, 8 – 25 mm x 1-2 (-5) cm long.
Flowers: (Jul – Oct) Bright yellow, in soft ovoid or cylindrical spikes 1 – 2 cm long.

Acacia verticillata, photographed in October near Bendigo

Refs

Costermans, L (1994) ‘Native trees and shrubs of south-eastern Australia’.  5th Ed. New Holland Publishers, Australia.

Costermans, L (2006) ‘Trees of Victoria and adjoining areas’.  6th Ed. Costermans Publishing, Australia.

(‘Mini Costermans’ = buy this).

All photos by Me!  For more Acacia love see http://www.flickr.com/photos/cracoo/sets/72157627391145521/with/7970371684/

A baby Acacia displaying both bipinnate leaves and phyllodes

This is just the sort of qualified optimism I intend to carry all the way through my PhD (Photo courtesy of my friend Hugh Rabinovici).

My PhD: Learning, planning and decision making for vegetation management

Over the next few years I plan to tackle (other than crippling poverty)… Um… lots of things? Well, so far only one thing in particular.

Chapter 1: Value of information for Box Ironbark woodland management

My first chapter continues on from work colleagues and I published last year and which I blogged about a while back.

In this chapter I am undertaking a value of information analysis. In essence, a value of information analysis is a decision theoretic tool for assessing the cost effectiveness of doing science to aid decision making. The issue of doing science (or not) before making a decision has come up quite a bit lately. How often have you heard a politician say “before we do such-and-such, we need to find out more about such-and-such… blah, blah, blah”. I suspect they rarely stop to ask whether the ‘science’ needs to be done or even it is worth doing in the first place.

Central to a value of information analysis are the quantities expected value of sample information (EVSI) and expected net gain of sampling (ENGS). EVSI is the average gain you expect to achieve given you have performed an experiment (or made some observations) and are subsequently able to make a better decision. ENGS is then the difference between EVSI and the cost of obtaining the new information.

More formally, a value of information analysis asks: given a set, , of possible, , actions you could take, which could result (probabilistically) in any outcome, , of a set of outcomes, , what is the value of performing an experiment, , from a set of possible experiments, , given the experiment could have any outcome from a set of experimental outcomes . In the mathematics of decision theory the aim is to maximise the utility function:

   .

In other words, if you are faced with a decision problem under uncertainty what experiment should you undertake (which includes not doing any experiment at all) to maximize the difference between the change in your expected gain from taking an action and the cost of doing the experiment itself.

I am applying a value of information analysis to the management of Box Ironbark Woodlands. An objective of Victoria’s Box Ironbark Woodland managers is to improve the biodiversity value of the forests they manage. They have a number of management actions at their disposal, each of which has a different cost. But the effect of management, and the dynamics of the system itself, are highly uncertain. My aim is to help them decide whether doing experimental work to resolve some uncertainty is a cost-effective strategy that can aid them in achieving their objectives.

Further reading:

• Raiffa, H. & Schlaifer, R. (1968) Applied statistical decision theory, MIT Press

• Czembor, C. A., & Vesk, P. A. (2009). Incorporating between-expert uncertainty into state-and-transition simulation models for forest restoration. Forest Ecology and Management, 259(2), 165–175

• Moore, J. L., & Runge, M. C. (2012). Combining structured decision making and value-of-information analyses to identify robust management strategies. Conservation Biology

I found this story deeply disturbing. But it wasn’t because the Leadbeater’s Possum – Victoria’s faunal emblem – is at imminent risk of extinction. And it wasn’t because some sections of our community are far more interested in maximising timber quotas than preserving this lovely creature and the ancient forests upon which it relies. I’ve known both of these things for some time.

What really disturbed me about David’s announcement this week was that it shook my faith in applied ecology to drive change. Applied ecologists are motivated by the pursuit of knowledge, for the betterment of environmental management. Data = knowledge = better environmental decisions. Simple as that. But David’s experience shows that this can be naive view.

The fact is, David Lindenmayer and his colleagues worked out how to save the Leadbeater’s Possum years ago. In a portfolio of work that has become legendary in ecological circles, David combined years of hard won field data with cutting edge modelling to identify the forest management regimes needed to conserve the LBP. But even more impressively, David paid close attention to the economic, social and political aspects of this issue. His work hasn’t been some idealistic crusade to protect the cute and cuddly LBP. Instead, David’s work has been dedicated to identifying forest management regimes that not only protect this endangered species, but also allow continued use of the forest by the community (be that for timber extraction, recreational activities, water harvesting etc).

Despite this impressive body of research, and the dissemination of the resulting knowledge through copious papers, books, talks and reports, David feels that the decisions being made about LBP management are now so poor that he can no longer be involved in them. Clearly, in this case, data and knowledge have not driven better environmental decision-making. On the contrary, David’s experience shows that even the best science will have little impact when decision-makers acquiesce to the parochial positions of a few.

So what is an applied ecologist to do? Do we carry on regardless, consoling ourselves that while our work may ultimately have no impact, we’ll publish some nice papers along the way? Or do we throw up our hands, renounce the pursuit of data, and join the shouting match instead? I very much hope I don’t have to do either, and that we ecologists can fulfill the role that society has given us: to gather data and broker knowledge for the advancement of environmental decision-making.

Some (very select!) further reading:

Lindenmayer D. B., Cunningham R. B., Tanton M. T., Nix H. A. & Smith A. P. (1991) The conservation of arboreal marsupials in the montane ash forests of the central highlands of Victoria, south-east Australia: III. The habitat requirements of Leadbeater’s Possum Gymnobelideus leadbeateri and models of the diversity and abundance of arboreal marsupials. Biological Conservation 56, 295-315.

Lindenmayer D. B. & Possingham H. P. (1994) The Risk of Extinction: Ranking Management Options for the Leadbeater’s Possum using Population Viability Analysis. Centre for Resource and Environmental Studies, Australian National University, Canberra.

Lindenmayer, D.B. (1996). Wildlife and Woodchips. Leadbeater’s Possum – A Test Case for Sustainable Forestry. UNSW Press, Sydney.

Lindenmayer D. B. & Possingham H. P. (1996) Ranking conservation and timber management options for Leadbeater’s Possum in southeastern Australia using population viability analysis. Conservation Biology 10, 235-51.

Lindenmayer, D.B. & Franklin, J.F. (2002). Conserving Forest Biodiversity: a Comprehensive, Multiscaled Approach. Island Press, Washington.

I chose the paper for our recent lab Reading Group and we discussed McGill et al.’s (2006) paper on community ecology and functional traits.  General principals in community ecology are notoriously hard to find.  McGill et al. (2006) believe that a focus on functional traits and environmental gradients can lead to a more quantitative, general and predictive science.  The authors propose four broad research themes based on traits, environmental gradients, the ‘interaction milieu’ and performance currencies that are intended to move community ecology beyond simple pairwise species interactions and towards a research agenda focused on a physiological approach with well defined units of measurement.

Overall our group agreed that traits are a powerful way to make generalisations in ecology.  We thought the discussion of big questions and clear future research directions in the paper gave a nice introduction to the field of functional ecology and to future applications of this method.

SLA (Specific Leaf Area) is a commonly used plant functional trait

Our discussion of this general approach led to some questions around the definitions and consistency of use for terms such as ‘functional trait’, ‘life history trait’, ‘functional type’, ‘vital rates’ and ‘attributes’.  This subsequently led to the acknowledgment of different scales in trait based research and questions surrounding levels of variation within traits, between traits, across species as well as between sites, regions and biomes.

We were not entirely clear on the detailed approaches being used to ‘rebuild’ community ecology based on traits, particularly in the face of multiple interactions between large numbers of species, but our attention was drawn to literature (not cited within this paper) that focuses on mechanistic and physiological trait based modelling (reviewed in Kearney and Porter 2006).  We thought that better theory about which traits are important, why, and under which circumstances will make trait-based approaches more applicable to conservation science.  In particular, we concluded that it would be nice to see more quantitative tests of the predictive ability of trait-based research, along the lines of Keith et al. (2007).

Hakea sp. with a different SLA

Within QAECO there are a few Qaecologists who work with functional traits of both plants and animals, with a general focus on applications of trait-based ecology to conservation problems.  See our QAECO blog here for more details.

Keep checking our main lab blog for more of our Reading Group papers (for our discussion on Restoration Ecology, see here).

Refs:

McGill, B.J., Enquist, B.J., Weiher, E., & Westoby, M (2006) Rebuilding community ecology from functional traits. TRENDS in Ecology and Evolution.

Kearney, M., & Porter, W.P (2006) Ecologists have already started rebuilding community ecology from functional traits.  TRENDS in Ecology and Evolution.

Keith, D.A., Holman, L., Rodoreda, S., Lemmon, J., & Bedward, M (2007) Plant functional types can predict decade scale changes in fire prone vegetation.  Journal of Ecology.

Arboreal animals, like possums, squirrels and monkeys, are finely adapted to life in the tree tops. Their feet are almost permanently poised to grab onto branches, strong claws sink into thick bark, and curly, ‘prehensile’ tails give extra grip in tricky situations. These animals are so good at moving from tree to tree that most species rarely come to the ground – this is where roads (and cars) can become a problem.

Squirrel Glider – falling, with style.

The Hume Freeway in north-east Victoria

But based on all this research came some action. In 2007 rope bridges and glider poles were installed to help squirrel gliders move safely across the road. Rope bridges (also called rope ladders or canopy bridges) are exactly what they sound like – a bridge made of rope that links habitat trees on either side of the freeway. Glider poles are tall wooden poles that can be placed in the roadsides and median as surrogate trees, allowing gliders to safely cross the road in a few short glides. These structures were installed at five sites (2 canopy bridges, 3 glider poles) where research showed the road was causing problems for squirrel gliders. Similar structures were also used in the recent upgrade of the Hume Highway in New South Wales in 2009.

A rope bridge over the Hume Freeway in Victoria (left) and glider poles on the Hume Freeway in New South Wales (right)

So, that’s what they are and that’s what they’re there for. But that’s not the question you all want to know, is it? What you want to know is, do they work? Me too. And this is where my research comes in – figuring out just how successful these structures have been. I’m still working on getting all the answers and will give more details on the monitoring side of my research in future posts (there are just too many cool details to squeeze into one short post!). I will answer one question now though – yes! animals are using these structures to cross the Hume Freeway and at this point, I think a picture is worth a thousand words.

Squirrel Glider on a rope bridge over the Hume Freeway in north-east Victoria.

References

van der Ree, R., Cesarini, S., Sunnucks, P., Moore, J.L., Taylor, A., 2010. Large gaps in canopy reduce road crossing by a gliding mammal. Ecology and Society 15.

McCall, S.C., McCarthy, M.A., van der Ree, R., Harper, M.J., Cesarini, S., Soanes, K., 2010. Evidence that a highway reduces apparent survival rates of squirrel gliders. Ecology and Society 15.

The main focus of my work has always been in systematic conservation planning and climate change. I am particularly interested in conservation optimization but also in bringing together different types of data to better understand what is really going on with this climate change business. I haven’t really specialized myself to any specific species or taxa, but somehow I often end up working with birds. I also spend way too much time in front of a computer than any self-respecting biologist should.