#ibstucson – International Biogeography Society 2017 meeting

I returned this weekend from the IBS 2017 meeting in Tucson. It was a great meeting. The organizers moved it on fairly short notice from Brazil to Tucson due to concerns about Zika. This resulted in a lot of extra work for the organizers, but it didn’t show. It was a well-run meeting. And it was my favorite type of a meeting a few hundred people organized around a fairly specific topic.

I’m not going to repeat individual talks – check out the twitter feed for many great talks (#ibstucson). As is usual with me, such meetings inspire big-picture musings. This one probably more than most, since the last time I was able to attend IBS was the inaugural meeting in Mesquite Nevada in 2003. I noticed a lot of differences in the 14 year gap.

Four quick observations of how things have changed since 2003:

  • Collaborative data – biogeography is a wonderfully collaborative field when it comes to sharing data. I do not know of a field in ecology that is better at assembling large databases using data assembled by often dozens of scientists than biogeography. The sociology of data sharing (or not) fascinates me. It is obvious that some of this collaborative instinct exists because the nature of biogeography makes it hard to answer the questions without big datasets. But this is too glib. Fourteen years ago most of the data was either a few government sponsored datasets (e.g. Breeding Bird Survey which I believe my poster was on) or else data collected from a lot of natural history reports assembled by one person (I recall Dov Sax, now the president of IBS, then was a student with a poster next to me who had laboriously built a dataset of invasive species on islands) or else phylogenies based on careful sample collection and sequencing by one or a few individuals. By contrast, in 2017 there were huge public, multi-collaborator databases such as GloNAF, ReefLife, the compilation of animal and plant matrix data (Comadre & compadre) and dozens more. And that is not including all the new environmental data layers Iearned about (just e.g. hires down-scaled climate data, Chelsea, and SoilGrids1K). These are all freely shared data built by collections of researchers. Something good is happening in biogeography more than just the scale of the questions. Steve Jackson gave a hint in his wonderful introduction of Margaret Davis (winner of the Wallace prize for career achievement) when he described how open and collaborative she was and how this led to collaborations across a whole generation of palynologists. What ever is in the drinking water in biogeography should be shared.
  • Computation – I’m not sure I saw a t-statistic or ANOVA in the entire conference! (a slight exaggeration but just barely). The level of statistics and analysis was incredibly complex. Many many statistics on phylogenies were used. Complex null models were used. And many computations on GIS raster layers were made. It may be hard to be a biogeographer in this day and age without strong computational skills. But as a longtime proponent of ecoinformatics, it was refreshing to see. There are oodles of up-and-coming highly capable informaticians.
  • Community ecology – have ecologists banned the word community ecology from their vocabularies? I know it was passe in the 1990s and early 2000s. But with two books out on community ecology and a couple more coming, I thought we had got past that. But I saw a lot of talks that were asking basically community ecology questions but avoiding the words and intellectual legacy. I think this is unfortunate. Community ecology is one of the five or so core disciplines of ecology and disowning its rich past is not helpful.
  • Climatic gradients – the one novel twist on community ecology that I saw, and it was a very good thing, was lots of studies with communities arrayed along an environmental gradient. For a long time community ecology was obsessed with single site studies of the endogenous dynamics (i.e. species interactions). That remains an important topic. But I personally think we won’t get far until we start doing what several people called to me “comparative community ecology”. This has ancient roots from the study of biomes and Whittaker’s elevational gradients and even Paine’s manipulated tide pools along a latitudinal gradient. But it seems to me ecology has finally embraced the gradient and that is a good thing.

And one contrast between biogeography and macroecology:

  • Consensus question – I have blogged a couple of times about how macroecology is a little unsure of what defines the discipline with some resulting angst (here and here). Biogeography seems to have none of that. At the first IBS meeting in 2003 there seemed to me to be a clear divide between the evolutionists who were mostly tracing evolutionary histories using phylogenies and the ecologists. But that divide seemed gone in 2017 – community phylogenetics seems to have successfully bridged the gap. But even in the old (2003) days, nobody argued about what was or wasn’t biogeography. It has always had a clear organizing question: why do these species grow here but not there? or as I like to say, “what grows where, why?”. I think this organization around a central question is really beneficial. Even at the IBS meeting, one of the half day symposia was really about definition of macroecology (specifically what experimental macroecology would look like – Miguel Matias did a great job organizing it). As I noted in my talk, macroecology is really defined primarily by scale and by method (statistical) rather than a question. That doesn’t make it a poorly defined field. But it makes it a fuzzier field.

What did other people at IBS see or think? I would be curious to hear other’s perceptions.

20 thoughts on “#ibstucson – International Biogeography Society 2017 meeting

  1. Interesting, cheers for this. Now wishing I could have gone. I think I’ve decided that in future I need to start going to some biogeography/macroecology meetings, since that’s where a lot of the community ecology (that dare not speak it’s name?) is these days.

    Plus, if people are thinking about experimental macroecology/biogeography, I’d be very interested to hear about it and talk to them about collaborating. Indeed, maybe I should do a little post sometime on previous work on “macroecology in microcosms”. As the scientific equivalent of putting a sign on the sidewalk advertising that my shop is open for business (“business” meaning “biogeographers/macroecologists who’d like to run experiments in model systems but who need a collaborator to do so”).

    So, what did that old symposium decide that experimental macroecology consisted of? What’s the leading edge?

    • You’ll be happy to hear that in my talk on experimental macroecology I identified three main kinds of experimental macroecology:
      Large Grain/Large Extent but feasible due to small organisms
      Large Grain/Large Extent but feasible due to lack of replication
      Small Grain/Large Extent (think quadrats spread over 1000 km).

      I argued the first two were both underutilized/underappreciated in macroecology.

  2. I found contributions in the symposium on experimental macroecology a bit disconnected from the large picture (biogeography/macroecology), and I was left unconvinced that experimental macroecology can deliver anything substantial at large spatial grains.

    • I take this comment as a challenge! šŸ™‚

      In seriousness: can you elaborate? What did you find unconvincing or disconnected about the symposium contributions? That such experimental studies would ask different (and less interesting/important?) questions than those currently being asked with non-experimental methods? That it wasn’t clear how experimental data would aid interpretation of non-experimental data? That it wasn’t clear how experiments in model systems (e.g., laboratory microcosms) tell us anything about what’s going on in nature? Very interested in your thoughts.

      • Its not about what is going in nature. Its about what is going on at large grain sizes. The processes involved in a small grain (e.g. quadrat, thus stochastic arrival & preemption, close but direct competition with neighbors, a pathogen field influenced by neighbors) don’t necessarily have a lot to do with what is going on at a large grain (metapopulation dynamics, rare colonizations of new species, a much vaguer more diffuse form of competition and pathogen – i.e. playing the hwole field or what I once called the competitive millieu). When people look the examples where not just the processes but the actual patterns change with scale are legion (see Wiens 1989, Rosenzweig 1995, Shmida & Wilson 1985, McGill 2010, Powell et al 2013 and etc).

        Thus like Petr, while happy to see large extent, small grained experiments feel it is not at all clear it will tell us much about large extent, large-grained questions.

        All that said, like Mike said below, you may well be working at large grain sizes in your microbial systems (with better odds of generalizing to a large-grained plant system than would say a small grained quadrat of plants)

      • @Brian:
        Re: grain size within a single protist microcosm, hmm. I could see arguing it more than one way. Which may just show that I have a shaky grasp of what you macroecologists mean by “grain size”!

        On the one hand, yeah, a single 50 ml jar can easily contain several hundred to hundreds of thousands of individuals of any given protist species (and many more bacteria, of course). So on pheidole’s operationalization of “grain size” below, even a single microcosm is pretty coarse-grained, much more so than, say, a 1 m^2 quadrat that might only contain a few individuals of any given grassland plant species. And a single jar definitely contains spatial heterogeneity to which protists respond behaviorally. For instance, bacterial densities are highest on the bottom, around particulate detritus, where the bacteria grow in loose flocculent biofilms. So that’s where the bacterivorous protists tend to hang out.

        But on the other hand, many species of protists are sufficiently mobile to potentially cross the entire jar within a single generation if they so chose. And population dynamics within a single jar often (*not* always!) can be fit reasonably well with some simple model assuming a well-mixed system, like a Lotka-Volterra model. Which says that, if a single jar does contain a lot of spatial heterogeneity at small spatial grains relevant to an individual protist, well, one can still average across or summarize that small-grain heterogeneity reasonably well with a model that treats the entire jar as a spatially-homogeneous, well-mixed system.

        On the third hand, one thing protist microcosms let you do is manipulate grain size. One could imagine a microcosm system comprising a bunch of patches (each patch is a jar or a well on a 96-well plate or whatever), and then aggregating those patches to do a more coarse-grained analysis. Indeed, probably somebody’s done this and I’m forgetting just now.

      • Yep, partly I provoke, partly I mean it. Brian nailed it in his comment below: The experimental studies may be global in extent, but their grain is still extremely small. What does an experiment executed at a 1 m x 1 m botanical plot tells about what’s going on over a 100 km x 100 km plot? The upscaling problem we face is enormous: we need to scale up 10,000,000,000 times! On top of that, too many macroecological quantities have weird (non-proportional) scaling properties in principle. So it is a huge challenge.

  3. Brian’s talk crystallized for me that Community Ecology and Macroecology lies on either end of the quantitative gradient (N*G) where N is community abundance (number of individuals in the system) and G is generations (number of generations, or, perhaps, the inverse of generation time). Both fields ask mostly the same questions–what regulates diversity, abundance, form, and function? But at low values of N*G, extirpation driven by drift, filtering, the ability to increase when rare, and rescue effects predominate. At high values of N*G you have diversification by evolution coming to the fore since you can capture more mutations with more individuals, and simultaneously extinction rates are lower.

    So you can study Macroecology on a lab bench if you work with bacteria and on a landscape if you work with oribatid mites. But it takes a continent to get at Macroecological processes for birds and mammals.

    Likewise, much of what we describe as microbial community ecology is largely Macroecology. We need to get way, way, way smaller to study microbial communities (One of my favorite results from Rainy and Travisano’s famous study on adaptive radiation in Pseudomonas is how the diversification becomes less predictable as the microcosms get smaller). It is both daunting and exciting to consider a 50 ml mesocosm of sand and silt as a meta-community, with each grain occupied by strongly interacting assemblages of bacteria, occasionally exchangin propagules via a water bridge, and stalked by roaming packs of nematodes and amoebae.

    Of course for an aging ecologist seeking to avoid an increasingly sedentary lifestyle, this could very well be a trap.

    • I’m skeptical of Rainey and Travisano’s conjecture for two reasons. First, they don’t actually show the data to support this hypothesis in the Nature paper and instead just mention it in passing. Second, I was unable to replicate the finding (using their stated culture volumes) and instead observed predictable evolution of of the morphotypes regardless of culture volume (though strong nutrient limitation seemed to do the trick). I think this is because even in very small volumes, you still have hundreds of millions of cells turning over very rapidly and an O2 gradient sharp enough to drive selection.

      I completely agree with you re: microbial community ecology, though. This is an issue for field studies of microbial communities, which do not often consider the smallest sampling unit (say, a g of soil) to be heterogeneous in the extreme.

    • We had our first discussion group meeting related to Mark Vellend’s book today. At that, we spent a fair amount of time trying to decide how to define community ecology vs. macroecology (we also concluded it was a spectrum), and then trying to figure out how microbial ecology links with community ecology. Sounds like we should have been at IBS!

  4. Good wrap-up, Brian, and I agree with your observations. As a first time IBS attendee, I found the overall cohesion of the conference to be really helpful. Also, I liked the fact that during symposium sessions there were no other parallel sessions, so everybody got to hear the same talks. It made it easier to discuss them with lots of people.

    From an outsider’s perspective, it seems like paleoecology is a big part of biogeography/macroecology. Being a spatial ecologist, I’m always a bit worried about the generality of studies that infer a lot from a very long term record of a single point in space, so it was encouraging to see the amount of effort put into combining paleo data from different sources into larger databases. On a related note, I really liked Jessica Blois’s first talk, in which she showed (using paleo data from multiple locations) that predicting future distributions of species is a rather unreliable endeavor. I think that studies that predict future distributions have become too ‘easy’ or generic (i.e., fit occurrence data to environmental variables using a SDM, plug in future climate predictions, generate future distribution), while ignoring the multiple problems with such analyses (e.g., non-analogue conditions, effects of biotic interactions). I wish that people were more careful in how they approached this question!

    • I think all macroecologists and probably all community ecologists should at least be aware of the paleoecology literature. I think most biogeographers already are. It brings a lot of value and additional information. And I agree Jessica’s talk was a pretty powerful example (there is a reason she won the MacArthur award for mid-career achievement!).

  5. Brian defines biogeography as “why do these species grow here but not there?”. This sounds pretty similar to a traditional definition of ecology: “distribution and abundance of species”. (By traditional I mean seminal texts and books 1950s-1980s.)

    I certainly don’t want to start a territorial dispute about which discipline owns which question, but I am interested what people think about the broader topic how much it’s been a plus or a minus to identify separate named disciplines? Has it been a plus or a minus to identify “macroecology” separate from ecology in general? Or ecology separate from biogeography from geobotany? And so forth.

    • That question got asked at the symposium on experimental ecology to the panel at the end. I think the answer was very mixed. Some felt that rigid boundaries did a disservice to science. Others felt that have a tradition and a center point to build from was productive for science.

      Personally, I think conversations about field and boundaries are themselves thought producing and constructive. All of ecology is a bit of an overwhelming place to try to span, be an expert in, generate ideas on. And almost by definition some of the next frontiers will occur on the boundaries of today’s disciplines. But getting hung up on definitions or trying to police it or getting territorial is certainly not constructive.

      And Mark an interesting point about all of ecology having been defined as “distribution and abundance of organisms”.I am of course aware of those questions. And have even used them in job talks to define my interests. But I guess starting ecology grad school in the late 90s it never struck me as an accurate description of all of ecology what with behavioral ecology, evolutionary ecology, ecosystem ecology, many aspects of physiological ecology, many people who stopped calling them community ecologists and started calling themselves as studying species interactions (e.g. mutualisms) etc. Could they all be traced back to distribution and abundance? Sure, everything can. Was that sincerely the primary goal of a majority of people calling themselves ecologists in 1997? It didn’t feel like it to me. It feels more like a subdiscipline of ecology that might well touch all of community ecology, macroecology, biogeography, paleoecology (which all obviously have a non-trivial area of overlap in Venn diagrams). Curious to hear your thoughts?

      • @Brian:

        You and I clearly hung around different community ecologists back in the late 90s and early oughts! I wouldn’t claim I and my friends are representative, necessarily. But FWIW, I and the people I hung out with were all happy to call ourselves community ecologists.

    • To answer your question, Mark, I don’t think it’s been very useful to identify different named disciplines, it leads to territoriality and exclusion, but I think it’s almost inevitable as broad fields mature and coalesce around different points of focus.

Leave a Comment

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s