Thursday, June 30, 2016

The pessimistic and optimistic view of BEF experiments?

The question of the value of biodiversity-ecosystem function (BEF) experiments—their results, their relevancy—has become a heated one in the literature. An extended argument over the last few years has debated the assumption that local biodiversity is in fact in decline (e.g. Vellend et al. 2013; Dornelas et al. 2014; Gonazalez et al. 2016). If biodiversity isn't disappearing from local communities, the logical conclusion would be that experiments focussed on the local impacts of biodiversity loss are less relevant.

Two papers in the Journal of Vegetation Science (Wardle 2016 and Eisenhauer et al. 2016) continue this discussion regarding the value of BEF experiments for understanding biodiversity loss in natural ecosystems. From reading both papers, it seems as though broadly speaking, the authors agree on several key points: that results from biodiversity-ecosystem functioning experiments don’t always match observations about species loss and functioning in nature, and that nature is much more complex, context-dependent, and multidimensional than typical BEF experimental systems. (The question of whether local biodiversity is declining may be more contested between them). 

Biodiversity and ecosystem experiments typically involve randomly assembled plant communities containing either the full complement of species, or subsets containing different numbers of species. Communities containing lower numbers are meant to provide information about the loss of species diversity a system. Functions (often including, but not limited to, primary productivity or biomass) are eventually measured and analysed in relation to treatment diversity. Although some striking results have come out of these types of studies (e.g. Tilman and Downing 1996), they can vary a fair amount in their findings (Cardinale et al. 2012).

David Wardle’s argument is that BEF experiments differ a good deal from natural systems: in natural systems, BEF relationships can take different forms and explain relatively little variation, and so extrapolating from existing experiments seems uninformative. In nature, changes in diversity are driven by ecological processes (invasion, extinction) and experiments involving randomly assembled communities and randomly lost species do nothing to simulate these processes. Wardle seems to feel that the popularity of typical BEF experiments has come at the cost of more realistic experimental designs. This is something of a zero-sum argument, (although in some funding climates that may be true...). But it is true that big BEF experiments tend to be costly and take time and labour, meaning that there is an impetus to publish as much as possible from each one. Given BEF experiments have changed drastically in design once already, in response to criticisms about their inability to disentangle complementarity vs. portfolio effects, it seems they are not inflexible about design though.

Eisenhauer et al. agree in principle that current experiments frequently lack a realistic design, but suggest that there are plenty of other types of studies (looking at functional diversity or phylogenetic diversity, for example, or using random loss of species) being published as well. For them too, there is value in having multiple similar experiments: this allows metaanalysis and comparison aggregation, and will help to tease apart the important mechanisms eventually. Further, realism is difficult to obtain in the absence of a baseline for a “natural, untouched, complete system” from which to remove species.

The point that Eisenhauer et al. and Wardle appear to agree on most strongly is that real systems are complex, multi-dimensional and context-dependent. Making the leap from a BEF experiment with 20 plant species to the real world is inevitably difficult. Wardle sees this is a massive limitation, Eisenhauer et al. sees it as a strength. Inconsistencies between experiments and nature are information that highlight when context matters. By having controlled experiments in which you vary context (such as by manipulating both nutrient level and species richness), you can begin to identify mechanisms.

Perhaps this is the greatest problem with past BEF work, is that there is a tendency to oversimplify the interpretation of results – to conclude that ‘loss of diversity is bad’ but with less attention to ‘why’, 'where', or 'when’. The best way to do this depends on your view of how science should progress. 

Wardle, D. A. (2016), Do experiments exploring plant diversity–ecosystem functioning relationships inform how biodiversity loss impacts natural ecosystems?. Journal of Vegetation Science, 27: 646–653. doi: 10.1111/jvs.12399

Eisenhauer, N., Barnes, A. D., Cesarz, S., Craven, D., Ferlian, O., Gottschall, F., Hines, J., Sendek, A., Siebert, J., Thakur, M. P., Türke, M. (2016), Biodiversity–ecosystem function experiments reveal the mechanisms underlying the consequences of biodiversity change in real world ecosystems. Journal of Vegetation Science. doi: 10.1111/jvs.12435

Additional References:
Vellend, Mark, et al. "Global meta-analysis reveals no net change in local-scale plant biodiversity over time." Proceedings of the National Academy of Sciences 110.48 (2013): 19456-19459.

Dornelas, Maria, et al. "Assemblage time series reveal biodiversity change but not systematic loss." Science 344.6181 (2014): 296-299.

Gonzalez, Andrew, et al. "Estimating local biodiversity change: a critique of papers claiming no net loss of local diversity." Ecology (2016).

Tilman, David, and John A. Downing. "Biodiversity and stability in grasslands." Ecosystem Management. Springer New York, 1996. 3-7.

Cardinale, Bradley J., et al. "Biodiversity loss and its impact on humanity."Nature 486.7401 (2012): 59-67.

Tuesday, June 14, 2016

Rebuttal papers don’t work, or citation practices are flawed?

Brian McGill posted an interesting follow up to Marc’s question about whether journals should allow post-publication review in the form of responses to published papers. I don’t know that I have any more clarity as to the answer to that question after reading both (excellent) posts. Being idealistic, I think that when there are clear errors, they should be corrected, and that editors should be invested in identifying and correcting problems in papers in their journals. Based on the discussions I’ve had with co-authors about a response paper we’re working on, I’d also like to believe that rebuttals can produce useful conversations, and ultimately be illuminating for a field. But pragmatically, Brian McGill pointed out that it seems that rebuttals rarely make an impact (citing Banobi et al 2011). Many times this was due to the fact that citations of flawed papers continued, and “were either rather naive or the paper was being cited in a rather generic way”.

Citations are possibly the most human part of writing scientific articles. Citations form a network of connections between research and ideas, and are the written record of progress in science. But they're also one of the clearest points at which biases, laziness, personal relationships (both friendships and feuds), taxonomic biases, and subfield myopia are apparent. So why don't we focus on improving citation practices? 

Ignoring more extreme problems (coercive citations, citation fraud, how to cite supplementary materials, data and software), as the literature grows more rapidly and pressure to publish increases, we have to acknowledge that it is increasingly difficult to know the literature thoroughly enough to cite broadly. A couple of studies found that 60-70% of citations were scored as accurate (Todd et al. 2007; Teixeira et al. 2013) (Whether you can see that as too low or pretty high depends on your personality). Key problems were the tendency to cite 'lazily' (citing reviews or synthetic pieces rather than delve into the literature within) or 'naively' (citing high profile pieces in an offhand way without considering rebuttals and follow ups (a key point of the Banobi et al. piece)). At least one limited analysis (Drake et al. 2013) showed that citations tended to be much more accurate in higher IF journals (>5), perhaps (speculating) due to better peer review or copy editing. 

Todd et al (2007) suggest that journals institute random audits of citations to ensure authors take greater care. This may be a good idea that is difficult to institute in journals where peer reviewers are already in short supply. It may also be useful to have rebuttal papers considered as part of the total communication surrounding a paper - the full text would include them, they would be automatically downloaded in the PDF, there would be a tab (in addition to author information, supplementary material, references, etc) for responses. 

More generally - why don't we learn how to cite well as students? The vast majority of advice on citation practices with a quick google search regards the need to avoiding plagiarism and stylistic concerns. Some of it is philosophical, but I have never heard a deep discussion of questions like, 'What’s an appropriate number of citations – for an idea?'; 'For a manuscript?'; 'How deep do I cite? (Do I need to go to Darwin?)'. It would be great if there were a consensus advice publication, like the sort the BES is so good at on best practices in citation.

Which is to say, that I still hope that rebuttals can work and be valuable.

Friday, May 27, 2016

How to deal with poor science?

Publishing research articles is the bedrock of science. Knowledge advances through testing hypotheses, and the only way such advances are communicated to the broader community of scientists is by writing up the results in a report and sending it to a peer-reviewed journal. The assumption is that papers passing through this review filter report robust and solid science.

Of course this is not always the case. Many papers include questionable methodology and data, or are poorly analyzed. And a small minority actually fabricate or misrepresent data. As Retraction Watch often reminds us, we need to be vigilant against bad science creeping into the published literature.



Why should we care about bad science? Erroneous results or incorrect conclusions in scientific papers can lead other researchers astray and result in bad policy. Take for example the well-flogged Andrew Wakefield, a since discredited researcher who published a paper linking autism to vaccines. The paper is so flawed that it does not stand up to basic scrutiny and was rightly retracted (though how it could have passed through peer review is an astounding mystery). However, this incredibly bad science invigorated an anti-vaccine movement in Europe and North America that is responsible for the re-emergence of childhood diseases that should have been eradicated. This bad science is responsible for hundreds of deaths.

From Huffington Post 

Of course most bad science will not result in death. But bad articles waste time and money if researchers go down blind alleys or work to rebut papers. The important thing is that there are avenues available to researchers to question and criticize published work. Now days this usually means that papers are criticized through two channels. First is through blogs (and other social media). Researchers can communicate their concerns and opinion about a paper to the audience that reads their blog or through social media shares. A classic example was the blog post by Rosie Redfield criticizing a paper published in Science that claimed to have discovered bacteria that used arsenic as a food source.

However, there are a few problems with this avenue. First is that it is not clear that the correct audience is being targeted. For example, if you normally blog about your cat, and your blog followers are fellow cat lovers, then a seemingly random post about a bad paper will likely fall on deaf ears. Secondly, the authors of the original paper may not see your critique and do not have a fair opportunity to rebut your claims. Finally, your criticism is not peer-reviewed and so flaws or misunderstandings in your writing are less likely to be caught.

Unlike the relatively new blog medium, the second option is as old as scientific publication –writing a commentary that is published in the same journal (and often with an opportunity for the authors of the original article to respond). These commentaries are usually reviewed and target the correct audience, namely the scientific community that reads the journal. However, some journals do not have a commentary section and so this avenue is not available to researchers.

Caroline and I experienced this recently when we enquired about the possibility to write a commentary on an article was published and that contained flawed analyses. The Editor responded that they do not publish commentaries on their papers! I am an Editor-in-Chief and I routinely deal with letters sent to me that criticize papers we publish. This is important part of the scientific process. We investigate all claims of error or wrongdoing and if their concerns appear valid, and do not meet the threshold for a retraction, we invite them to write a commentary (and invite the original authors to write a response). This option is so critical to science that it cannot be overstated. Bad science needs to be criticized and the broader community of scientists should to feel like they have opportunities to check and critique publications.


I could perceive that there are many reasons why a journal might not bother with commentaries –to save page space for articles, they’re seen as petty squabbles, etc. but I would argue that scientific journals have important responsibilities to the research community and one of them must be to hold the papers they publish accountable and allow for sound and reasoned criticism of potentially flawed papers.

Looking over the author guidelines of the 40 main ecology and evolution journals (and apologies if I missed statements -author guidelines can be very verbose), only 24 had a clear statement about publishing commentaries on previously published papers. While they all had differing names for these commentary type articles, they all clearly spelled out that there was a set of guidelines to publish a critique of an article and how they handle it. I call these 'Group A' journals. The Group A journals hold peer critique after publication as an important part of their publishing philosophy and should be seen as having a higher ethical standard.



Next are the 'Group B' journals. These five journals had unclear statements about publishing commentaries of previously published papers, but they appeared to have article types that could be used for commentary and critique. It could very well be that these journals do welcome critiques of papers, but they need to clearly state this.


The final class, 'Group C' journals did not have any clear statements about welcoming commentaries or critiques. These 11 journals might accept critiques, but they did not say so. Further, there was no indication of an article type that would allow commentary on previously published material. If these journals do not allow commentary, I would argue that they should re-evaluate their publishing philosophy. A journal that did away with peer-review would be rightly ostracized and seen as not a fully scientific journal and I believe that post publication criticism is just as essential as peer review.


I highlight the differences in journals not to shame specific journals, but rather highlight that we need a set of universal standards to guide all journals. Most journals now adhere to a set of standards for data accessibility and competing interest statements, and I think that they should also feel pressured into accepting a standardized set of protocols to deal with post-publication criticism. 

Wednesday, May 25, 2016

Thoughts on successful postdoc-ing

Unlike grad school, postdoc positions start and end without much fanfare. If grad students are apprentices, postdocs are the journeymen/women of the trade. (Wikipedia defines journeymen as… “considered competent and authorized to work in that field as a fully qualified employee… [but] they are not yet able to work as a self-employed master craftsman.”) Though short compared to a PhD, postdoc jobs are an important stepping stone towards a 'real' job, be that another postdoc, or a position inside or outside of academia. There’s less advice out there about being successful as a postdoc, and often you are on your own to figure things out. I’m finishing a first postdoc this week, and moving on to a second one, and while I think the last 2 years worked out well, they took their own, unexpected path. Some of this is good advice that I was given, some comes from experience or observation, some I even manage to follow :-) *

Choose carefully. If you have some choice, be strategic in choosing a postdoc job. Decide what the position is going to accomplish for you: that may be expanding your skill set, such as by learning a new experimental system or additional analytical techniques; improving your current skills by working with an expert; being involved in high profile research; or being in a certain locale for various reasons. Beware projects too far from your current skill set – the risk is that the learning curve may be so steep that you will be barely competent at the end, and have little to show for your time. Of course, you might decide to use a postdoc to pursue interdisciplinary work, or move away from your dissertation work, in which case this is a risk worth taking.

Because postdocs are short, it may seem as though having a good fit with your supervisor is less important. Don’t assume that your new supervisor be broadly similar in approach to your previous supervisor (or an improvement). Mismatched expectations between supervisors and postdocs seem pretty common and it’s important to get an understanding of what your role is beforehand. The variation in expectations from supervisor to supervisor is huge - from those that require time sheets and expect strict hours, to those that give you total autonomy. Does your supervisor see postdocs as colleagues? 9-5 employees? Advanced students? Lab managers? Talk to friends, colleagues, and students. This may depend on the source of funding as well - will you be working on a specific existing project with specific timelines (common in the US where many postdocs are funded off of NSF grants), or are you funded by a fellowship and therefore more independent?

Get to know your neighbours. Once you’ve chosen and started your postdoc, the most important thing to do is to establish connections in your lab and department immediately. I cannot emphasize this enough. Don’t wait to settle in, or get on top of some papers, or hope people in the hallway will introduce themselves. Postdoc positions are short, and in many departments postdocs are isolated, not students but not really faculty. This can lead to feelings of disconnection, loneliness, and frustration. Seek out the other postdocs - join or organize postdoc social events, go to lab meetings and journal clubs, get the department to maintain an active postdoc email list. Not only will this give you a sense of belonging, but now you have people to talk to (and sometimes rant to), with whom to navigate administrative issues, and potential collaborators. Postdocs are an invaluable resource for job applications as well: they usually have the most up-to-date experience on the job market, and can provide great feedback on job applications and practice job talks. For example, the postdocs in my current department built an exhaustive list of potential questions asked during academic interviews, and shared interview horror stories over drinks.

Mental health and life balance. Postdocs don’t get the kinder, gentler approach sometimes given to grad students and people expect you to stand on your own. This can reignite imposter syndrome. There is no easy solution to this, but some combination of taking care of yourself, working on that mythical thick skin, and highlighting the positive events in your life can help.

Time management continues to become more important, at least for me. More than in grad school, you have to actively decide how much work you want to be doing. There is always something that you *could* be working on, so start scheduling when things will get done based on priority, energy, etc, is important. In addition, people start inviting you to things or asking for you input on projects. Learn to say no. Be strategic about your time management – it’s flattering to wanted, but time is limited and not all invitations are of equal value towards your specific goals.

Practice professional networking. On the other hand, don’t say no to everything: networking and the opportunities it creates are very helpful. Focus on the professional areas that are of interest to you, but consider joining and being active in ESA sections (including the Early Career section) or other relevant organizations; organize workshops or symposia at conferences; host invited speakers. If your department hosts an external seminar series, take advantage (nicely!) of the revolving cast of scientists. They are a great way to make connections with people whose work you admire, and even speakers you have less in common with are great practice for networking skills. From experience, if you have breakfast with a different visiting speaker every week, you will quickly improve your description of your research and your ability to keep a conversation going (also, you will become an expert on your city’s breakfast places). These are helpful skills to have for faculty interviews, for talking to the media and press, even for telling your family what you do.

Take initiative. You are your own advocate now. If you wish you could learn something, or be invited to a working group, or get teaching experience, look into making it happen yourself. This may include organizing working groups (many provide competitive funding, for example, iDiv/sDiv, CIEE (Canada), the new NCEAS, SESYNC), applying for small grants and other project funding on your own, recruiting undergraduates and mentoring them, organizing or co-teaching courses.

Similarly, don’t stop learning new things. Inertia gets higher the less time you have, and it can be hard find the time to pick up the next skill.

Publish.
Focus on publishing (if you are interested in academic jobs)– this may be obvious, but publishing is more important than ever as a postdoc. You need to show that you are independently able to produce work after leaving your PhD lab. This counters the ‘maybe they just had a good supervisor’ concern. It can be hard to find time to work on both current and past projects, but try to. From experience (and illustrated by the periodic emails from my PhD supervisor), the longer your dissertation chapters sit around, the less likely they are to ever be published…

Know what your dream job is, and apply for it if you see it. Be willing to move on if something better comes up. Postdocs usually have to think in the short-term, because most funding is in 1-2 year increments. So keep an eye on new sources of funding/positions. Make decisions based on your needs (be they career-related, family-related, whatever): it’s easy to feel guilty moving on from one unfinished position to another, but the reality is that postdocs are temporary and fleeting.

I was told to start applying for jobs as early as I felt reasonably qualified. The logic was that the best practice for job interviews is doing actual job interviews, and further, it is better to fail when it doesn’t matter, rather than when it is your dream job.

well, sort of...
*Obviously nothing is one-size-fits all, and this is mostly aimed at people who plan to apply for faculty jobs eventually. Other advice, especially for non-academic tracks, would be welcome in the comments!

Friday, May 6, 2016

What’s so great about Spain? Assessing UNESCO World Heritage inequality.

Some places are more valuable than others. We often regard places as being of high or unique value if they possess high biological diversity, ancient cultural artefacts and structures, or outstanding geological features. These valuable places deserve special recognition and protection. The sad reality is that when we are driven by immediate needs and desires, these special places are lost.

The natural world, and the wonderful diversity of plants and animals, is on the losing end of a long and undiminished conflict with human population growth, development, and resource extraction. We don’t notice it when there is ample natural space, but as nature becomes increasingly relegated to a few remaining places, we place a high value on them.

The same can be said for places with significant cultural value. Ancient temples, villages, and human achievement are too valuable to lose and we often only have a few remnants to connect us to the past.

In either case, natural or cultural, when they’re gone, we lose a part of us. That is because these special places tell us about ourselves; where we come from, how the world shaped us, and what unites all of humanity. Why did the world cry out in a united voice when the Taliban destroyed the Buddhas of Bamiyan in 2001, even though many of those concerned people were not Buddhist? The answer is simple –the expansion of Buddhism out of India along ancient trade routes tells us why many Asian nations share a common religion. They tell us about ourselves, the differences that interest us, and the similarities that bind us. The same can be said about the global outcry over the recent destruction of the ancient city of Palmyra by ISIS.

Before and after photos of the taller of the Buddhas of Bamiyan. Image posted by Carl Montgomery CC BY-SA 3.0.

Similarly, the natural world tells us about ourselves. The natural world has constantly shaped and influenced what it means to be human. Our desires, fears, and how we interact with the natural world are products of our evolution. If I flash a picture of a car to my 500-student ecology class, very few students, if any, screech in fear. But if I flash a photo of a hissing cobra or close-up of a spider, invariably a bunch of students squirm, gasp, or scream. Rationally, this is an odd response, since cars are the leading cause of death and injury in many western countries. Snakes and spiders kill very few people in Canada.

These special places deserve recognition and protection, and that is what the UNESCO World Heritage designation is meant to achieve. To get this designation for a site requires that countries nominate ones that represent unique and globally significant contributions to world heritage, and are adequately protected to ensure the long-term existence of these sites.  World Heritage sites are amazing places. They represent the gems of our global shared heritage. They need to be protected in perpetuity and should be accessible to all people. Though some I have visited seem like they are loved too much with high visitation rates degrading some elements of Heritage sites.

Examples of UNESCO World Heritage sites. A) The Great Wall of China. B) The Gaoligong Mountains, part of the Three Parallel Rivers of Yunnan. C) Angkor Wat in Cambodia. D) An example of a site that may be too loved -Lijiang in Yunnan. All photos by Shirley Lo-Cadotte and posted on our family travel blog -All The Pretty Places.

UNESCO World Heritage sites should also be representative. What I mean by this is that they should be designated regardless of national borders. Heritage sites are found on all continents across most countries –though a number of politically unstable countries (e.g., Liberia, Somalia, etc.) do not possess Heritage sites, likely because they lack the organization or resources to undertake the designation application process, and they lack the governance to ensure a site is adequately protected. But there are substantial differences in the number of World Heritage sites across nations[1]. Some countries, because of inherent priorities, national pride, resources or expertise, are better able to identify and persuade UNESCO that a particular place deserves designation.

The distribution of the number of UNESCO World Heritage sites across countries and the top ten.

Why do we see such disparity in the number of World Heritage sites -where many countries have few sites, and a few countries have many sites? This is a difficult question to answer, and to do so I took an empirical approach. I combined data on the number of sites per country with Gross Domestic Product (GDP)[2], country size[3], and country population size[4]. I then ran simple statistical analyses to figure out what predicts the number of Heritage sites, and identified those countries that are greatly over-represented by Heritage sites, and those that are very under-represented. A couple things to note, the best statistical models included variables that were all log-transformed, I excluded the World Heritage sites that spanned more than one country, and I did not include countries that did not have any Heritage sites. The data and R code have been posted to Figshare and are freely available.

All three of GDP, area, and population size predicted the number of World Heritage sites. It is important to note that these three country measures are not strongly correlated with one another (only moderately so). So, larger, richer and more populous countries had more World Heritage sites. This makes sense –big countries should contain more unique sites due to random chance and more populous countries tend to have longer historical presence of organized states, and so should possess more cultural relics (especially China). GDP is more difficult to assign a reason, but high GDP countries should have robust national parks or other bureaucratic structures that assess and protect important sites, making them easier to document and justify for UNESCO.  GDP is quite interesting, because it is the single best measure for predicting the number of Heritage sites, better than population size and area. Further, neither country density (population/area) nor productivity (GDP/population) are strong predictors of the number of Heritage sites.

The relationships between the number of World Heritage sites and GDP, area, and population. Note that the axes are all log-transformed.

While these relationships make sense, it is also clear that countries are not all close to the main regression line and some countries are well above the line –meaning they have more Heritage sites than predicted; as well as some below the line and thus having fewer sites. When I combine the different measures in different combinations and look for the best single statistical explanation for the number of World Heritage sites, I find that the combination including GDP and population size, and their interaction (meaning that population size is more important for high GDP countries) is the best. For aficionados, this model explains about 65% of the variation in the number of Heritage sites.

Now, we can identify those countries that are over or under represented by UNESCO World Heritage sites according to how far above or below countries are from the predicted line (technically, looking at statistical residuals).

The deviation of countries from the predicted relationship between the number of sites and GDP and population (and their interaction). The top 5 over-represented and under-represented countries are highlighted.


The top five over-represented countries are all European, which means that given their GDP and population size, these countries have more World Heritage sites than expected. At the other extreme, countries under-represented come from more diverse regions including Africa, the Middle East and Southeast Asia.

An interesting comparison to think about is Germany and Indonesia. Germany has more World Heritage sites than expected (residual = +0.61) and is a moderately sized, high GDP country. Let me say, I like Germany, I’ve been there a half a dozen times, and it has beautiful landscapes and great culture. However, does it deserve so much more World Heritage recognition than Indonesia, which has fewer sites than expected (residual = -0.63)? Indonesia has spectacular landscapes and immense biodiversity and great cultural diversity and history. To put it in perspective, Germany has 35 World Heritage sites and Indonesia has just 8.

To answer the question in the title of this post: what’s so great about Spain? Well, it not only has beautiful and diverse natural landscapes and cultural history, but it appears to have the infrastructure in place to identify and protect these sites. It's place at the top of UNESCOs relative (to GDP and population) ranking of the number of World Heritage sites means that Spain's natural and cultural wonders are in good hands. However, for the countries at the other end of the spectrum, having relatively few World Heritage sites probably is not a reflection of these countries being uninteresting, or that they have little to offer the world, rather it is something more alarming. These places lack the financial capacity or national will to fully recognize those places that are of value to the whole world. The problem is that the globally important heritage that does exist in these places is at risk of being lost. These under-represented countries serve as a call to the whole world to help countries not just identify and protect heritage sites but to aid these countries with infrastructure and human well-being that empowers them to prioritize their natural and cultural heritage.

Wednesday, May 4, 2016

The future of community phylogenetics

Community phylogenetics has received plenty of criticism over the last ten years (e.g. Mayfield and Levine, 2010; Gerhold et al. 2015). Much of the criticism is tied to concerns about pattern-based inference, the use of proxy variables, and untested assumptions. These issues are hardly unique to community phylogenetics, and I think that few ideas are solely ''good or solely 'bad'. They are useful in moulding our thinking as ecologists and inspiring new directions of thought. Many influential ideas in ecology have bobbled in confidence through time, but remain valuable nonetheless [e.g. interspecific competition, character displacement (Schoener 1982; Strong 1979)]. But still, it can be hard to see exactly how to use phylogenetic distances to inform community-level analyses in a rigorous way. Fortunately, there is research showing exactly this. The key, to me at least, to avoid treating a phylogeny as just another matrix to analyze, but to consider and test the mechanisms that might link the outcome of millions of years of evolution to community-level interactions.

A couple of potential approaches to move forward questions about community phylogenetics are discussed below. The first is to consider the mechanisms behind the pattern-inference analyses and ask whether assumptions hold.

1) Phylogenies and traits - testing assumptions about proxy value
As you know, if you have read the introductory paragraph of many community phylogenetic papers, Charles Darwin was the first to highlight that two closely related species might have different interactions than two distantly related species. People have tested this hypothesis in many ways in various systems, with mixed results. The most important directions forward is to make explicit the assumptions behind such ideas and experimentally test them. I.e. Do phylogenetic distances/divergence between species capture trait and ultimately ecological divergence between species?

From Kelly et al. 2015 Fig 1b.
Because evolutionary divergence should relate to feature divergence (sensu Faith), the most direct question to ask is how functionally important trait differences increase with increasing phylogenetic distances. For example, Kelly et al. (2014) found that “close relatives share more features than distant relatives but beyond a certain threshold increasingly more distant relatives are not more divergent in phenotype”, although in a limited test based only on patristic distances. This suggests that at short distances, phylogenetic distances may be a reasonable proxy for feature divergence, but that the relationship is not useful for making predictions about distant relatives.

Phylogenies and coexistence/competition. Ecological questions about communities may not be interested in traits alone. The key assumption behind many early analyses was that closely related species shared more similar *niches*, and so competed more strongly than distantly related species. Thus the question is one step removed from trait evolution, asking instead how phylogenetic divergence correlates into fitness differences or interaction strength. Not surprisingly, current papers suggest there is a fairly mixed, less predictable relationship between phylogenetic relatedness and competitive outcomes.

Recent findings have varied from “Stabilising niche differences were unrelated to phylogenetic distance, while species’ average fitness showed phylogenetic structure” (California grassland plants, Godoy et al. 2014); to, there is no signal in fitness or niche differences (algae species, Narwani et al. 2013); to, when species are sympatric, both stabilizing and fitness differences increase with phylogenetic distance (mediterranean annual plants; Germain et al. 2016). Given constraints, tradeoffs and convergence of strategies, it is really not surprising that the idea of simply inferring the importance of competition from patterns along a phylogenetic tree is not generally possible (Kraft et al. 2015; blogpost).

2) Phylogenies and the regional species pool
Really more interesting than testing for proxy value is to think about the mechanisms that tie evolution and community dynamics together. A key role for evolution in questions about community ecology is to ask what we can learn about the regional species pool—from which local communities are assembled. What information about the history of the lineages in a regional species pool informs the composition of local composition?

The character of the regional species pool is determined in part by the evolutionary history of the region, and this can in turn greatly constrain the evolutionary history of the community (Bartish et al. 2010). The abundance of past habitat types may alter the species pool, while certain communities may act as 'museums' harbouring particular clades. For example, Bartish et al. 2016 found that the lineages represented in different habitat types in a region differ in the evolutionary history they represent, with communities in dry habitats disproportionately including lineages from dry epochs and similar for wet habitats. Here, considering the phylogeny provides insight into the evolutionary component of an ecological idea like 'environmental filtering'.

Similarly, species pools are formed by both ecological processes (dispersal and constraints on dispersal) and evolutionary ones (extinctions, speciation in situ), and one suggestion is that appropriate null models for communities may need to consider both ecological and evolutionary processes (Pigot and Etienne, 2015).
Invasive species also should be considered in the context of evolution and ecology. Gallien et al. 2016 found that “currently invasive species belong to lineages that were particularly successful at colonizing new regions in the past.”

I think using phylogenies in this way is philosophically in line with ideas like Robert Ricklef's 'regional community' concept. The recognition is that a single time scale may be limiting in terms of understanding ecological communities.

References:
  1. Mayfield, Margaret M., and Jonathan M. Levine. "Opposing effects of competitive exclusion on the phylogenetic structure of communities." Ecology letters 13.9 (2010): 1085-1093.
  2. Gerhold, Pille, et al. "Phylogenetic patterns are not proxies of community assembly mechanisms (they are far better)." Functional Ecology 29.5 (2015): 600-614.
  3. Schoener, Thomas W. "The controversy over interspecific competition: despite spirited criticism, competition continues to occupy a major domain in ecological thought." American Scientist 70.6 (1982): 586-595. 
  4. Strong Jr, Donald R., Lee Ann Szyska, and Daniel S. Simberloff. "Test of community-wide character displacement against null hypotheses." Evolution(1979): 897-913. 
  5. Kelly, Steven, Richard Grenyer, and Robert W. Scotland. "Phylogenetic trees do not reliably predict feature diversity." Diversity and distributions 20.5 (2014): 600-612.
  6. Godoy, Oscar, Nathan JB Kraft, and Jonathan M. Levine. "Phylogenetic relatedness and the determinants of competitive outcomes." Ecology Letters17.7 (2014): 836-844.
  7. Narwani, Anita, et al. "Experimental evidence that evolutionary relatedness does not affect the ecological mechanisms of coexistence in freshwater green algae." Ecology Letters 16.11 (2013): 1373-1381.
  8. Rachel M. Germain, Jason T. Weir, Benjamin Gilbert. Species coexistence: macroevolutionary relationships and the contingency of historical interactions. Proc. R. Soc. B 2016 283 20160047
  9. Nathan J. B. Kraft, Oscar Godoy, and Jonathan M. Levine. Plant functional traits and the multidimensional nature of species coexistence. 2015. PNAS.
  10. Bartish, Igor V., et al. "Species pools along contemporary environmental gradients represent different levels of diversification." Journal of Biogeography 37.12 (2010): 2317-2331.
  11. IV Bartish, WA Ozinga, MI Bartish, GW Wamelink, SM Hennekens. 2016. Different habitats within a region contain evolutionary heritage from different epochs depending on the abiotic environment. Global Ecology and Biogeography
  12. Pigot, Alex L., and Rampal S. Etienne. "A new dynamic null model for phylogenetic community structure." Ecology letters 18.2 (2015): 153-163.
  13. Gallien, L., Saladin, B., Boucher, F. C., Richardson, D. M. and Zimmermann, N. E. (2016), Does the legacy of historical biogeography shape current invasiveness in pines?. New Phytol, 209: 1096–1105.

Friday, April 22, 2016

More ways to understand traits in ecology

It seems that increasingly, ecology is moving away from relying primarily on summary statistics and approximations, to considering measures that recognize the often meaningful variation in ecological data. Using only the mean of a variable, for example, may be informative in some ways, but insufficient in others. Indices of diversity increasingly reflect that ecologically relevant information is not restricted to a single moment (as seen in the framework for measuring trait diversity (Villeger et al. (2008)) and the analogous framework for phylogenetic diversity (e.g. detailed in Pavoine and Bonsall (2011); also Tucker et al. (2016)).

Particularly, the functional ecology literature has developed increasingly complex and integrative methods for measuring and comparing trait diversity. The literature has gone from descriptions of general types or traits (e.g. Whittaker 1956), to measuring measuring individual traits and relating them to particular ecologically relevant variables (e.g. Gaudet and Keddy (1988)); to calculating community-weighted values for individual traits (e.g. D Schluter, (1986)); to incorporating multiple variables into single measures (e.g. FD package); to a framework reflecting mathematical moments in data (Villeger et al. (2008); and to the use of multivariate hypervolumes to describe the multi-dimensional shape and volume of trait space to be measured (e.g. Blonder et al. 2014).

A new paper in TREE does a nice job of summarizing and integrating these developments with yet another addition: a ‘trait probability density’ approach. In  “Traits Without Borders: Integrating Functional Diversity Across Scales", Carlos P. Carmona, Francesco de Bello, Norman W.H. Mason, and Jan Lepš nicely illustrate a way to capture the complexity inherent to a concept such as the ‘functional niche’. [The "region of the functional space containing all the trait combinations displayed by the individuals of a species"].

The truth about traits is that there is meaningful variation at every scale at which we measure them (including variation between individuals, variation between populations, variation between species, and variation between communities). Often decisions are made to ignore or collapse unwanted levels of variation (such as using a mean value across several individuals to calculate a single species-level value). The authors suggest that we can instead incorporate this variation appropriately. A probability density function can be defined for the multi-trait space, with probabilities representing the relative abundances of each combination of trait values. Thus, for a species, the curve (Figure IA) would show the multivariate trait space seen across all measured individuals, with uncommon combinations of traits seen in few individuals shown at the tails of the distribution. Outliers and extreme values are incorporated but not overemphasized as they can be in convex hull approaches.

The probabilistic approach reflects that a niche *is* probabilistic for a species - after all, it is unlikely that the niche is simply a fixed set of traits that is identical for all individuals or populations. However, not all combinations of trait values (niche dimensions) are equally likely for members of a species, and these curves reflect that. And when probabilities are incorporated into trait measurements, greatly different conclusions may be made about how similar or dissimilar assemblages may be (e.g. Fig IC).

Reproduced from Fig I., Carmona et al. 2016 TREE.

One concern--one that is pretty much universal to all analyses in functional ecology--is about how the biases and limitations of available data will affect this type of measure. Some species are better described, some traits are not available for most species, some studies lack interspecific measures, some lack local measures (relying instead on general databases of trait values). In addition, some intraspecific variation arises from other sources of noise like stochasticity and measurement error. This is all part of a bigger question about sufficient data: not only do we need to know how many traits are needed to define a species, but we need to decide how much and what kind of data is necessary to understand a trait…

Fig. 2 from Carmona et al. TREE 2016. It is possible to incorporate existing measures of functional diversity (richness, evenness, divergence) into the probabilistic definition.

Wednesday, April 13, 2016

Who should communicate the policy implications of ecological research?

Ecology is a science that tries to understand the world. How is the diversity of organisms distributed around the world? How do extreme climate events influence populations of animals and plants? How does the diversity of organisms in a landscape influence its function and the delivery of services to humanity? These are all questions routinely asked by ecologists and, importantly, are topics that most academic ecologists would believe are necessary for providing evidence for policy and management of habitats and natural resources. Yet policy makers, managers and practitioners seldom access ecology research. There is a research-policy divide that needs to be overcome.

Spanning the chasm between academic research and policy (from http://www.adventureherald.com/8-scary-suspension-bridges-you-do-want-to-cross/)
 
Many ecologists are reluctant to promote the policy implications of their research because they do not feel comfortable or connected enough to talk to non-academics. But if not them, then who is responsible to communicate the policy repercussions of their research?

The romanticized view of an untouched, pristine ecosystem no longer exists. We now live in a world where every major ecosystem has been impacted by human activities. From pollution and deforestation, to the introduction of non-native species, our activity has influenced every type of habitat. But this is where management and applied ecology have relevance. The study of human physiology has direct relevance for health science –that is, the value of this basic biological science is measured in its ability to help sick people, and not necessarily in its ability to better understand how healthy people function. So to does ecology need to be relevant for our ‘sick people’, that is, human-impacted landscapes. We have spent much of our collective effort studying intact, semi-natural systems, and this is necessary to understand the basic operations of nature. But now we are required to apply this understanding to improve ecological integrity and human wellbeing. We are surround by sick ecosystems and ecology is desperately needed to influence policy and management.

I just attended the joint symposium “Making a Difference in Conservation: Improvingthe Links Between Ecological Research, Policy and Practice”, put on by the British Ecological Society and the Cambridge Conservation Initiative. This meeting was attended by a nice mix of academic researchers and practitioners, and covered a broad range of ideas, issues and solutions to overcoming barriers to implementing evidence-based policy. Overcoming these barriers requires communication, and scientists need to be at the table. In arguing the case that scientists need to communicate the policy implications of their research below, I take ideas and information passed on in a number of excellent talks, including from: John Altringham, Malcolm Ausden, John Beddington, Ian Boyd, Fiona Fox, Georgina Mace, Andrew Miller, E. J. Milner-Gulland and Des Thompson, and my own workshop on communicating research to maximise policy impact.

A guy who probably doesn't know what he is talking about, talking about policy. Perhaps a bit outside my comfort zone. (photo by Martin Nunez)

The Hurdles

The hurdles to the uptake of research and evidence into policy decisions are complex and multifaceted. On the scientists’ side, the hurdles are mainly a lack of training, experience and comfort promoting the policy implications of their work. In graduate school, very few scientists-in-training take journalism and media courses, and so are not well versed in the ways to communicate in a broadly approachable way. Instead, we are taught to communicate in technically precise ways that can only be understood by similarly trained experts.

On the practitioner side, there are a number of pragmatic and systemic limitations to the uptake of evidence into policy and management decisions:

1.       Structural: There is a lack of resources and time to read and synthesize scientific research. A lack of access because of expensive subscription fees, is a pervasive problem for individuals and small organizations.
2.       Systemic: Big organizations and agencies are complex and communication of best practices or idea sharing might be lacking. Frequent staff turnover means that research understanding and institutional memory is lost.
3.       Relevance: Practitioners need research relevant to their problem and trolling the impossibly large literature is not an efficient way to find the necessary information.
4.       Timescale: Practitioners and policy makers work at a variety of speeds, dictated by priorities, contracts, etc., and looking for resources may not work within these timeframes.

These limitations and the lack of relevant research uptake result in policies and management strategies that are not adequately informed by research, which can waste money and may not produce in the desired results. We heard about the requirement to build bat crossings across new highways (to avoid car collisions), costing millions of dollars, but research has not supported their efficacy. 

Random bat picture to break up the flow (from http://www.bugsbirdsandbeasts.co.uk/go-batty)

Should scientists engage policy makers? 

I do think that scientists have a responsibility to communicate, and perhaps advocate, for evidence to be used in policy decision-making. There is a line between being seen as objective versus as an advocate, and scientists need to do what they are comfortable with, but remember:

  1. You are an expert on your research; you are uniquely position to comment on it.
  2. Related to the previous point, you may not want other, untrained, people to represent and communicate your work.
  3. You have an obligation to the public. You are likely paid by tax dollars and your research is funded by public grants. A part of the responsibility then is to not only do research but to ensure that it is communicated and if the people who ultimately pay you would benefit from learning about your findings, you owe it to them to communicate it.
  4. There are positive feedbacks for your career. Being seen as a scientist who engages and does relevant work will mean that you achieve a higher profile.


Citizens and policy-makers get the most out of their new information (which forms the basis for their opinions) from media news. If the only voices being heard are advocates and interest groups, then evidence will be lacking or misrepresented. Scientists’ voices are needed in the media, and here you can educate many concerned people. The former British Education minister, Estelle Morris, when speaking about the Fukushima reactor meltdown, said that she learned more about radiation from scientific experts in the media than she had during her education.

Of course it is important to remember that science is only a part of the solution, human needs, economics and social values are also important. But without scientists’ involvement, evidence will not be an important part of solutions to crises. 

How to communicate

Scientists are often driven by immediate career concerns and they need to publish high profile, impactful papers in peer-reviewed scientific journals. And this won’t change. But as Georgina Mace said in her presentation, overselling the implications of research in papers diminishes their value and confuses practitioners and policy makers. Policy implications contained within publications is one avenue to influence policy makers, but rather than tacking on broad policy recommendations, consider consulting them before writing the paper, or even better, include them in the planning stage of the study. One speaker commented that instead of asking for a letter of support for a grant proposal from a non-academic partner at the 11th hour, discuss the ideas with them at the outset.

How should scientists communicate their research?
  1. Discuss finings with local interest groups (e.g., park managers).
  2. Give a public lecture to community organizations (e.g., naturalist club).
  3. Talk to local politicians.
  4. Use social media –create a persona that acts as an information broker.
  5. Write opinion articles for magazines or newspaper editorials.
  6. Be accessible to journalists (e.g., get yourself listed in your university expert database).


The UK as a model

The UK provides one of the best examples of meaningful interactions between scientists and policy makers. Perhaps a better way to state it, is that there is a gradient of engaged individuals from pure scientist to local practitioner. There are robust organizations that span government agencies, NGOs, and universities that bring scientists and practitioners into contact with one another. They have Chief Scientific Officers and advisory groups at multiple levels of government. These groups develop the risk registry that estimates the likelihood and the potential repercussions of environmental and biological disasters or emergencies (e.g., influenza pandemic, severe drought, etc.). There is a well respected and effective Science Media Centre that organizes briefing sessions that bring scientists together with journalists on timely and important topics. These briefings result in influential news stories that sometimes challenge government policy or public sentiment (e.g., badger culls, links between vaccines and autism, etc.). This is a system to be emulated.

So, should scientists communicate their findings and engage policy makers, managers and the public. Absolutely. It may seem like you are entering uncharted territory, but believe me, your voice is desperately needed.

If you want advice, encouragement or more information, feel free to contact me.