Positive Multifunctionality ≠ All Functions Are Positive

Positive Multifunctionality ≠ All Functions Are Positive

I was dismayed this morning to read Bradford et al.’s recently accepted paper Discontinuity in the responses of ecosystem processes and multifunctionality to altered soil
community composition
in PNAS for several reasons.

The paper itself is really cool. They manipulated community complexity and nutrient conditions in the Ecotron, and then looked at five soil ecosystem functions. They then looked at whether complexity influenced multi functionality (as well as N), and found, indeed, it did! They went on and, as recommended in our paper on multifunctionality analyze single functions to understand what is driving that multifunctionality relationship, and then…

Then they fall off the boat completely.

Disappointment #1
They find that, while some functions were affected positively, some were not, and one more was affected negatively. They conclude, therefore, that multifunctionality metrics are not useful.

…multifunctionality indices may obscure insights into the mechanistic relationships required to understand and manage the influence of community change on ecosystem service provision.

The mismatch between our community and fertilization effects on multifunctionality and the individual processes, however, cautions against using the framework as a predictive tool for achieving desired levels of functioning for multiple, specified ecosystem services.

What is frustrating about this is that the authors completely miss what multifunctionality actually tells us.

I’m going to say this once very simply, and then in much more detail –

high multifunctionality ≠ every single function performing well

To quote from my own work, multifunctionality is “simultaneous performance of multiple functions.” No more, no less. A positive relationship between a driver and multifunctionality does not imply a positive relationship between that driver and every function being monitored. But rather that said driver will be able to increase the performance of more functions than are decreased.

Some More Detail
Indeed, in the example in Byrnes et al. 2014, we look at the data from the German BIODEPTH experiment. Some of the functions have a positive relationship with richness. Some do not. One has a trending negative relationship. But, put together, multifunctionality is a powerful concept that shows us that, if we are concerned with the simultaneous provision of multiple functions, then, yes, biodiversity enhances multifunctionality.

In our paper, we advise that researchers look at single functions – precisely because they are likely not all related to a driver in the same way. We state

The suite of metrics generated by the multiple threshold approach provide powerful information for analysing multifunctionality, especially when combined with analyses of the relationship between diversity and single functions.

We say this because, indeed, one has to ask – is the driver-MF relationship as strong as it could be? Why or why not? How can we pull the system apart into its component pieces to understand what is going on at the aggregate level?

The approaches are not in opposition, but rather utilizing both provides a much more rich picture of how a driver influences an ecosystem – both through an aggregate lens and a more fine-scale lens. The similarities and differences between them are informative, not discordant.

Disappointment #2
UPDATE: See comments from Mark and Steve below. This #2 would appear incorrect and a tale of crossed paths not scene. While I cannot find anything in my various inboxes regarding communication, it’s possible either a bad email address was used, or it went missing in my transition between nceas and umb. If this is the case, I’m in the wrong on this. An interesting quandry of how do we resolve these things outside of the literature, and worth pondering in this our modern age of email. I leave my comments below for the sake of completeness, and as there are still some ideas worth thinking about. But, wish that email hadn’t disappeared somewhere into the ether! Now the more my disappointment in technology!

Perhaps the bigger bummer is that, despite this being a big critique of the idea of multifunctionality that our group spent a *huge* amount of time trying to figure out how to quantify in a meaningful and useful way, as far as I know, none of us were contacted about this rebuttle. The experiment and analysis of the experiment is excellent, and it gets into some really cool stuff about soil biocomplexity and ecosystem multifunctionality. But the whole attacking multifunctionality as a useful concept thing?

That entire controversy could have been resolved with a brief email or two, tops. For this group to go so far off base is really kind of shocking, and dismaying.

Dismaying because the advice that would seem to stem from this paper is to go back to just looking at single functions individually and jettison the concept of multifunctionality (no other alternative is provided). That places us squarely back in 2003, with fragmented different types of analyses being used in an ad hoc manner without a unifying framework. Precisely what we were trying to avoid with our methods paper.

And all it would have taken to prevent is a little bit of communication.

Bradford, M. A., S. A. Wood, R. D. Bardgett, H. I. J. Black, M. Bonkowski, T. Eggers, S. J. Grayston, E. Kandeler, P. Manning, H. Setälä, and T. H. Jones. 2014. Discontinuity in the responses of ecosystem processes and multifunctionality to altered soil community composition. PNAS. link

Byrnes, J. E. K., L. Gamfeldt, F. Isbell, J. S. Lefcheck, J. N. Griffin, A. Hector, B. J. Cardinale, D. U. Hooper, L. E. Dee, and J. Emmett Duffy. 2014. Investigating the relationship between biodiversity and ecosystem multifunctionality: challenges and solutions. Methods in Ecology and Evolution. 5: 111-124. doi: 10.1111/2041-210X.12143

Spring is sproinging earlier than we thought

ResearchBlogging.orgClimate change is speeding up the timing of plants growing & flowering. It’s an old trope we’re all used to hearing by now. Spring warms up earlier, and plants shift their timing – their phenology. This shift in phenology is going to have big consequences for things like agricultural production, the ability for wildlife to find food, the ski season, etc. Important, right? Something we might want to predict. So, we’ve been running experiments for years, looking at climate change scenarios, and finding out how much earlier plants will do their thing, and we’ve gotten some pretty good results.

Or have we.

New work published online in Nature by Lizzie Wolkovich and colleagues (go-go NCEAS working group headed by a former postdoc!) shows that…ah, hell, I think I can summarize it with their figure 2.

Results from a meta-analysis showing sensitivity to warming in experiments and what we've observed. Panel a shows all plant species, b is just those that are in both the observed and experimental data sets.

Basically, after collecting an enormous database of changes in leafing and flowering times in both experiments and what has been observed out in the world, we’ve found that experiments UNDERPREDICT how sensitive plants are to climate change.

That’s right. Something has moved these plants to leaf and flower ever EARLIER than predicted. Climate change’s effects on plans are even STRONGER than we’d predict from simple warming experiments.

To me, this suggests a whole lot of interesting directions – why are things moving even earlier? What indirect effects is warming having on the Natural world that might speed up plant timing? What crazy interactions are there that we’re missing when we try and isolate out just one signal of climate change?

It’s a little disheartening – we canot predict climate impacts on Life on Earth from changes in one or two simple variables – and a little scary – we have only begun to understand the impacts of climate change. But I think it’s fascinating in that it suggests we really need to understand the how the complex system of nature is going to be affected by climate change to alter even the most basic properties of the world around us.

Wolkovich, E., Cook, B., Allen, J., Crimmins, T., Betancourt, J., Travers, S., Pau, S., Regetz, J., Davies, T., Kraft, N., Ault, T., Bolmgren, K., Mazer, S., McCabe, G., McGill, B., Parmesan, C., Salamin, N., Schwartz, M., & Cleland, E. (2012). Warming experiments underpredict plant phenological responses to climate change Nature DOI: 10.1038/nature11014

Diversity Loss v. Environmental Change: The Story of the Paper

ResearchBlogging.orgJust over a year and a half ago, I was lucky enough to be in a room with some of the world’s top authorities on the consequences of species loss at the National Center for Ecological Analysis and Synthesis. We were worried. When Brad Cardinale asked, “So, where does diversity loss rank? As important as climate change? Rampant nutrient runoff from agriculture? I mean, come on, how important is it, really?”, none of us knew the answer.

But now we do.

As our group has just published in Nature, species loss matters quite a bit – as much as many of the major drivers of environmental change.

Will losing one or more of these algae matter as much as the rapid rise in global temperature?

Let me back up. As species continue to go extinct across the globe, one has to ask – will there be any consequences? We’ve spent the last 15 years answering this question, and it is, yes. Losing species, at the bare minimum, reduces the ability of fields of plants or algae to most efficiently turn sunlight and nutrients into new production. Losing some of the myriad of species responsible for munching their way through dead detrital material will slow that whole process down. More questions remain, but diversity loss seems to be altering a wide variety of ecosystem properties.

But how important is that loss, really? In the grand scheme of climate change, pollution, and other forms of environmental change, is losing species important? Or is it just the icing on the cake of environmental degradation?

What would results from experiments like this one with flumes full of many different combinations of algae tel us about how diversity and the environment both shape primary production?

Realizing that there was no good answer to this, we sat down, rolled up our digital sleeves, and got to work. Led by the ever steady Dave Hooper and always insightful Carol Adair, we began to dig into the literature – to see what information was out there about the impacts of these different forms of environmental change. We were armed with the hugely revised and updated version of Brad’s monster database (in no small part due to the Epic efforts of the unflappable Kristin Matulich at UC Irvine) documenting nearly every experiment that has measured the consequences of diversity loss ever conducted. Most of the data was from experiments describing changes to plant or algal productivity & decomposition. So we had a starting point.

How will diversity loss compare to the plethora of other forms of environmental change in altering these basic processes?

The answer came from what we called the Meta-Meta. (Come on, say it 10 times fast). We assembled a huge database of meta-analyses – statistical summaries of previous experimental results much like our own diversity set – that examined the effect of, say, nutrient addition or warming on plant productivity and decomposition. Carol then did some statistical wizardry, calculating bootstrapped effect sizes of each from these previous meta-analyses. Hence, we had Meta-Meta-Analytic results that we could compare with our own meta-analytic results. There was a lot of meta.

But we had a problem. The measurements in our diversity experiment database looked at the consequences of having 1 species versus many species in a plot for productivity. That number of ‘many’ species varied from experiment to experiment. And, besides, what’s the right diversity loss scenario to compare to climate change. 10% species loss? 50%? 95%? Only 1 last species left, clinging tenaciously to the earth?

We actually argued about this for a long time as only scientists can when they want to get something dead-on right. But the solution ended up being delightfully elegant.

Why not look at the whole range of species loss scenarios. Bruce Hungate, Carol, Dave and I figured out a way to look at slices of the data that simulate some relevant range of species loss and applied a nonlinear model shown to fit species loss data quite well to generate a loss-productivity curve. The curve showed that, if you lose a few species, meh! Some productivity is lost, but no big whoop. But as you start losing more and more, the problem of species loss starts to compound. Like the interest rate from hell, more species lost means exponentially more plant production lost.

We still struggled with visualization (hey, infographics people, any ideas), but the figure that we ended up producing I think gets the point across nicely.

Figure 1 from the paper

Changes in primary production as a function of per cent local species loss or the application of an environmental change treatment. The line with the confidence interval around it shows the effect of diversity loss. The think no-CI curve is its mirror image so that we can easily compare effect magnitude to types of environmental change that differ in the sign of their effect. Anything in red is a negative effect on productivity. E.g., drought and diversity loss both have negative effects. Anything in blue is a positive effect. E.g., if you add nitrogen and phosphorous, you boost productivity by quite a bit. Y-axis is in units of log response ratios. Any designers in the house with an even slicker way of presenting this, email me!

So, now you can compare the effect of some environmental change to some percentage of species loss. For example, ~50% species loss has an equivalent impact to dousing a plot in acid rain.

Of all of the types of environmental change out there, a lot of them are roughly equivalent to between 30-70% species loss for production and decomposition. And, some additional review work we did showed that predictions of local extinctions (i.e., within a plot) range pretty widely, but, the upper end of that range is ~41-60% – so, right in our range of observations.

“Oh-ho!”, I hear some of you saying (and some in the group said – I’m looking at you, Lars). “Aren’t the effects of environmental change in your Meta-Meta actually including any loss of species in experiments? I want a cleaner result!”

Well, we thought of that.

Kristin, Dave, and I did some data-sorting kung-fu on the diversity experiment data set, and turned up several studies that not only manipulated species number, but also crossed that with a manipulation of some form of environmental change – things like drought, nutrient enrichment, and more. This was a much smaller set of data, and a smaller set of environmental change manipulations. But, the results are compelling, and show that we’ve got the story right.

Figure S4 from the supplement. All is as it was before in the above figure, but, this data is from experiments that cross changes in number of species with one or more other forms of environmental change.

Those careful factorial crossed experiments show that once the diversity signal is taken out, the size of many of those environmental effects gets smaller relative to diversity. For example, the average effect of a CO2 increase is of the same magnitude as 20% species loss (although opposite in sign). The effect of drought, one of the biggest from the Meta-Meta, drops waaaay back as well – although for both of these, the confidence intervals overlap 0 due to low sample sizes (and the 2 or the three drought datapoints come from mosses, so, more factorial experiments are needed!). Still, it’s pretty darned suggestive that there are some interesting indirect effects hidden in the meta-meta.

At the very least, the experimental data pretty clearly backs up the conclusion that loss of species appears to have effects on primary productivity that is similar to other sources of environmental change. And more intriguingly, it suggests that if environmental change ALSO causes loss of species, ecosystem functions like productivity are going to get hit with a 1-2 punch.

The team noted a few caveats and open questions to this. When multiple kinds of environmental change were present – say, multiple kinds of fertilizers – the effects are way stronger than diversity loss. But we don’t have a great handle on those kinds of synergistic effects overall. We also don’t know how they’ll affect diversity and then indirectly alter function. Also, because we were working with meta-analyses, we don’t have, say, the ability to compare 30% species loss to a 30% increase in temperature, or something comparable. It would be awesome to be able to compare two continuous curves, but we’re not there yet. Our results (deep inside the supplement) also show that composition – which species go extinct – can matter a great deal. But that was ancillary to our overall question of diversity loss. And there’s all sorts of ancillary questions about scaling and interaction between diversity loss and environmental change – but that’s work that Andy Gonzalez and Mary O’Connor are following up on in a massive multilevel modeling extravaganza.

Overall, we felt like this told a pretty tight story. Sure, as we came up with question after question, we accumulated a lot of figures and explanatory materials that made their way into the supplementary material, but the big story was pretty clear and compelling. In working group meeting 3, we really shaped this sucker, and then bounced revisions back and forth, wrenching the text back and forth. I’ve never been part of such a large collaboration (save for our earlier first paper), so it was an inspiring process to see big ideas hashed out, thrown aside, revised, and made into clearer cleanly crafted pieces all on the screen of my word processor.

I’m pretty psyched that the paper is now out, and I hope you all enjoy it as well. There’s more to come from the working group, so stay tuned!

The working group at our first meeting.

(Oh, and, final note – to anyone who wants to look at this data, or for R-nerds who want to see how we created these awesome figures, it’s all available here.)

Cardinale, Bradley J.; Matulich, Kristin L.; Hooper, David U.; Byrnes, Jarrett E.; Duffy, Emmett; Gamfeldt, Lars; Balvanera, Patricia; O’Connor, Mary I.; Gonzalez, Andrew. 2011. The functional role of producer diversity in ecosystems. American Journal of Botany. 98:572-592. 10.3732/ajb.1000364

Hooper, David U., Adair, E. Carol, Cardinale, Bradley J., Byrnes, Jarrett E.K., Hungate, Bruce A., Matulich, Kristin L., Gonzalez, Andrew, Duffy, J.E., Gamfeldt, Lars, O’Connor, Mary I. 2012. Biodiversity loss ranks as a major driver of ecosystem change. Nature. 10.1038/nature11118

Dancing Yeti Crabs! Now with a Soundtrack!

ResearchBlogging.orgSeveral years ago, when the Yeti crab, Kiwa hirsuta was first described, the world looked at a crustacean for the first time and went, “AWWWWW!!!”

I mean, how could you know love crabs from the Kiwa genus? They have fuzzy arms! And are adorable! People immediately began paying tribute with plush toys of all manner and even decorative food arrangements.

So what could be better than a plain ole’ Yeti crab?


That’s right – marine ecologist, deep sea biologist, and all around good egg Andrew Thurber has discovered a new species if Yeti crab, Kiwa puravida that appears to be farming methane consuming bacteria living on the hairs on its arms. How does it farm them? It waves its arms around in the air (like it just don’t care!) – er, water – around of methane seeps. Then periodically scrapes the bacteria off of its arms as food. The waving action serves to amp up supply the bacteria with more methane and other compounds that otherwise would be limited due to boundary layer conditions around the crabs hairy arms. And it looks like they’re having a great time doing it. In fact….I couldn’t resist grabbing the creative commons video on the PLoS paper and, um, adding a soundtrack.

(I really really couldn’t resist.)

(OK, maybe I saw him give a talk on this a few months ago, and have been waiting this entire time for it to be published so I could put a soundtrack on this video. Maybe. Maybe definitely.)

OK, ok, aside from this awesome behavior, what I love about this paper is Andrew takes what could have been a neat behavioral observation with a hypothesis that makes a nice just-so story, and then he tackles it with some really hot science. He uses detailed fatty acid and isotope analysis that shows, definitively, that the Yeti crabs are getting their nutrition from the bacteria on their arms. The symbiosis is real and biologically important. It’s a compelling solid story that gives us a new insight onto the unique life that lives in the deep sea.

Moreover, as Thurber writes, if anything, it highlights how little we know about life in the deep sea. If we have only just discovered that Yeti crabs must dance in the deep to make a living, what other fascinating discoveries are out there?

Update: See Doctor Zen’s version Yeti crabs can dance if they want to (Safety Dance). Anyone else want to take a swing at it?

Thurber, A., Jones, W., & Schnabel, K. (2011). Dancing for Food in the Deep Sea: Bacterial Farming by a New Species of Yeti Crab PLoS ONE, 6 (11) DOI: 10.1371/journal.pone.0026243

Can We Reduce the Carbon Cost of Scientific Mega-Meetings?

ResearchBlogging.orgI admit it. I love big scientific meetings. There’s something about the intense intellectual hubbub of thousands of my fields greatest minds gathered in one place for a few days of showing off the latest, greatest, flashiest work that just fills me with joy. Also a need to sleep for a week afterwards due to my brain going at a Matrix-like pace to keep up with all of the new and interesting information while spouting off ideas, critiques, beginning collaborations, and constantly questing to understand the growing shape of the research fields that interest me. It’s quite simply an intellectual smörgåsbord. But like all such dining experiences, there is a cost. A cost I’ve been wrestling with in this new piece in Enthobiology Letters with my collaborator Alexandra Ponette-González.

It’s a carbon cost. A cost for climate change.

Simply, there a lot of people at these Mega-Meetings. A LOT. And they are rarely local. Most of us fly in, from across the country, from another country, or even another continent. Those flights put out CO2 emissions – a lot of it. Heck, even driving the full distance to some of these meetings would have a high emissions profile given the distances. And it makes you stop and wonder – we ecologists who are so environmentally conscious, what is the carbon cost of our engagement in big Mega-Meetings? Could we be doing better? How?

A map of the location of the last several ESA meetings and the 2010 AGU meeting (triangles with Carbon Cost next to them) as compared ti the distribution of attendees (circles proportional to number of attendees from that area over all meetings). Costs are in per capita metric tons.

A few years ago, this issue came up at the DISCCRS conference – an annual interdisciplinary gathering of early career climate researchers that is truly amazing. During the coffee afterwards I got to talking with a fellow attendee, and we began brainstorming. How could the big scientific societies of the world – the ESAs, AGUs, or, heck, maybe even the AAASs – still conduct their vital business of intellection discourse while reducing their carbon footprint from meeting travel?

Travel is the key – if attendees, even the same number of attendees, didn’t have to travel to far and use air travel, it’s possible that we could dramatically lower carbon costs. Merely limiting the number of meetings or restricting the number of possible attendees seemed draconian and not possible. Carbon offsets have proven to be unreliable. Telecommuniting to meetings limits the real value of live social interaction (so far). It seemed like there wasn’t a good solution. But then we began to think about a second kind of meeting that some, but not all, of us attend.

I’m talking about meetings that are smaller, cozier, with researchers rarely from more than a few states away. Grad students have piled into cars, trucks, vans, llama-powered motor-scooters, and more to make the pilgrimage for the meeting’s weekend of showing their stuff and finding new colleagues, collaborators, and mentors. These are the meetings where you form deep relationships that you come back to year after year – relationships that slowly bear great intellectual fruit. Meetings like The Western Society of Naturalists, for example.

True, Mega-Meetings are quite different from these smaller more local meetings – like the big flash of molecular gastronomy to the simple elegant nourishment of slow food – elBulli to Chez Panisse. Therein, however, lies their intrinsic value – a value that attendees of only Mega-meetings may actually be missing.

So we began to ponder – what if societies alternated between Mega-meetings and a large number of smaller more regional meetings? Could this be a possible solution? Intellectually, sure, I’m sure some would still argue against it, but that would be moot if the carbon savings were trivial. So we sat down over the next few months and did the computational equivalent of some back-of-the-envelope calculations of carbon as currently emitted versus carbon emitted based on several different scenarios of meeting distributions.

And then we sat back, pretty surprised.

Assuming that pretty much everyone drives, but that no one carpools (or uses llamas), carbon savings under our most pessimistic set of assumptions were around 50%. That’s right, halving the carbon emissions.

Granted, this is back-of-the-envelope, but, the idea is pretty compelling. And yes, there are other costs – administrative, logistic, etc. But thinking from a carbon perspective alone, this result is pretty stunning. Not only are there large carbon benefits, but local meetings confer other benefits – contribution to regional economies, better ties to regional organizations and NGOs, and quite likely a higher degree of participation from graduate students (and lower attendance barriers to undergraduates and the community).

We also considered other alternatives – lowering carbon costs by taking the distribution of members into account, reducing international participation, etc. But the drawbacks in these seemed to be ones that most people would not, at least currently, accept, when we floated ideas to others.

So, this local-regional alternation seems to be something worth thinking about. Would you be willing to participate in an alternative society structure – one where meetings alternated between being large and international and then small and regional? What would be lost for you? What would be gained? Would it be too much of an additional burden on organizers? Would that burden be justified by carbon savings?

Also of note, we had a hard time getting this published. We had a lot of wonderful comments from editors and reviewers who were very positive about this work, but then would say, “Oh, but, you know, we just don’t have a venue for this.” (sometimes followed weeks later by editorials stating “THIS IS A PROBLEM! WHERE ARE THE SOLUTIONS?” which we thought curious) We tried multiple generalist and specialist journals, journals for societies and by regular publishers. I’d like to thank Ethnobiology Letters for going out on a limb and publishing this, as conversations like this need to be had in the peer reviewed literature.

Ponette-González, Alexandra G, & Jarrett E Byrnes (2011). Sustainable Science? Reducing the Carbon Impact of Scientific Mega-Meetings Ethnobiology Letters, 2, 65-71

The Story Behind the Paper: Climate Change and Kelp Forest Food Webs

ResearchBlogging.orgYay! First paper of my postdoc is out in the August 2011 issue of Global Change Biology!


So, what have I been doing for the past few years of my life?

In brief summary: Kelp. Food webs. Climate change. A potent combination.

And if you want the punchline without digging into the rest of this article, it is this: the diverse, complex food webs of southern California kelp forest will likely be greatly simplified if climate change leads to big storms every year.

I could give you all of the gory details of the paper, how I arrived at that conclusion, the science-y nuggets, etc. Instead, I thought I’d give you the slightly longer, more human, but more meander-y version of how this paper was created – the story of this particular story. It’s not something we always talk about in science. Science isn’t always a nice linear process. What we set out to do it not always what ends up happening. In the end, though, we want a nice linear story that leaves the juicy bits of exploration out. And the process behind this paper was, indeed, intellectually juicy.

So, how does a paper on kelp forests, food webs, and climate change, come to be?

Like all projects in my life, this was not one I was expecting to do. It’s kind of a theme for me. I came to work at the Santa Barbara Coastal Long-Term Ecological Research site to work on potential feedbacks between the diversity of life on the seafloor and grazing pressure by urchins. Essentially, I wanted to test a theoretical model that I’d put together with colleagues in a 2007 paper. I was convinced that feedbacks between diversity and function were going to be the next big thing. And I still think they’re pretty neat. The experiment was pretty fun, led me to revisit some old concepts in new ways, and ultimately produced some great data which is going to be submitted soon.

Five year running average of extreme (i.e., storm-driven) winter non-tidal residual wave heights from the San Francisco Tide Guage. Starting ~ 1945, we see an increase. Figure adapted from Bromirski et al. 2002 J. Climate.

But early on in all of this, the project PI called me to his office and laid out the following.

The SBC LTER has a buttload (metric, not Imperial) of data. They’ve been sampling thirty-five 80 m2 transects at reefs along the Santa Barbara Chanel coastline every summer for nearly a decade, counting the heck out of everything. At the same time, we’ve noticed two interesting things in the climate literature: 1) climate change projections say that the strongest storm of the winter should get bigger in the future, and 2) if you look at the data, the largest winter waves in California have gotten bigger over the last 50 years.

We know that big waves rip out kelp to life on the seafloor.

We don’t know what will happen if kelp is ripped out every year, or lost altogether.

The LTER had started a project to simulate this annual storm disturbance – big 200m2 plots where we went through and trimmed the giant kelp every winter with hedge clippers. But, coming up on year 2, no one had an idea of how to put the whole story together.

Hello, opportunity, how’s it going?

My facebook network. My mom is actually fairly well connected - but mostly to my High School cluster.

I’d long been fascinated with a network approach to studying communities. Basically, you can visualize life on the seafloor by thinking about Facebook. No, really! There are a ton of tools out there you can use to visualize your network of friends, with the links between them showing friendships, and you at the top. There’s a ton of information there. Who are the hubs? Who connects disparate groups of friends? How does the size of, say, your group of college friends influence the number of other random groups of friends you’ve accumulated through life?

And who has your mom friended, anyway?

OK, now, instead of friendship, think of your friends as eating each other. And you’re at the top!

OK, ok, now replace the people in your network with different species. And you’re a shark. Or giant squid. (or theoretical giant mutant Pycnopodia). Voilá! Food web! And all of that structure and complexity, it has real meaning describing the stability and function of a community of organisms. (well, ok, the function part is what I’m tackling in my current postdoc at nceas)

So, I went back to my PI and said, OK, hey, let’s take a look at how changing the annual frequency of storms can shift the network structure of kelp forest food webs. It would be an indirect effect, so we can use my favorite tool, Structural Equation Modeling, for the analysis. And we can even bring in two awesome other bits of data – transect-level wave height projections from the Coastal Data Information Program and awesome new measurements of kelp beds right after storm season taken by satellite (and developed by Kyle Cavanaugh – an awesome grad student at UCSB who uses Landsat images to count kelp).

He said, sure! But, I might want to see about that food web. You see, no one has actually put together a full kelp forest food web. Or even one for the 250 species that we sample. So, can this be done? Really?

And thus began a 6 month quest. Of living and dying by google scholar. Of talking to experts. Of driving up and down the coast to marine labs, riffling through their libraries of unpublished masters theses, or appendices to undergraduate student reports. Just to find out, who indeed, eats who?

It’s that kind of basic natural history that is necessary to inform sexy fun theory-based analyses. And without it, the sexy-fun-stats-nerdery is really meaningless.

I emerged with a nice solid web, a good sense of uncertainty, and a decent idea of how I’d put these models together. The next part was shockingly simple. Use plyr to smoosh our data with a master kelp-forest food web to get individual transect-level food webs (e.g., what the structure of a place with two seaweeds, an urchin, and a lobster, versus a full-blown hyper-diverse kelp forest)? And then use Structural Equation Modeling to fit models that looked like so:

Path diagram of an SEM showing how waves indirectly influence the species richness and linkage density of a kelp forest food web.

This is all very interesting, and one can contrast the strength of different indirect pathways by which waves influence kelp. It was not immediately intuitive, however, as to what this means for the future of kelp forests under a climate change scenario with annual large storms. Fortunately, as I was fitting Structural Equation Models, which are really just a system of linear equations, I could turn my models into dynamic simulations.

Yes yes. Lots of coding.

I then used these simulations to make predictions about how increasing the annual frequency of large storms will affect the network structure of kelp forest food webs. I could reproduce the table of results for you, and discuss each individual result, but I think the following image more or less sums it up

Basically, frequent large storms will simplify food webs in the end. What’s interesting, though, is that just one storm after years of calm – our current scenario – may actually increase complexity. Everything gets a little mixed up as sunlight streams in and lets suppressed algae establish a beachhead, even while top predators may decrease in diversity. And, shockingly, results from our large-scale field experiment – at least the first two years – appeared to match this pattern beautifully.

And thus, blammo! Publication!

Well, OK, no. Honestly, after the initial results it took a few meeting with my co-authors to get everyone on the same page. In no small part this was due to the atypical methods I was using. Also, while the final paper has about nine or ten different measures of community complexity in it, my initial analyses looked at about thirty. I had some winnowing to do in order to establish a good story. Good clear story is king. What is a scientific paper, after all, than good storytelling backed up by data and then confused with jargon.

After we all got on the same page (and I had tried my story out via a few talks and posters), I wrote it up for Science. Because, well, why not. Even so, it took multiple rounds of revision, before submission. And sweet sweet rejection. Thus followed attempts to submit to three different other journals (What? I wanted to try and be in one of the glossy magazines! I’m thinking’ about my career, here, and other postdocs will back me up on this!) And a major re-write for the format of each. Then, finally, I realized that GCB really was a logical and perfect fit for the piece, and the reviews I got were most helpful in clarifying the last few pieces.

In the end, I’m a pretty proud Papa on this one. I think it’s a nice solid piece of science. It’s got a massive chunk of natural history in it, filling what I see as a key gap. It uses some fancy-pants statistics – and allowed me to go on a deep statistical odyssey in learning the ins and outs of some arcane parts of SEM, such that I’m now an SEM package developer in R. And it coupled the analysis of a smokingly hot large-scale observational dataset (go-go LTER power!) with an intense and awesome mongo-effort field experiment (Clint, Shannon, and Christine, you guys are underwater animals!) It’s basically everything I want in a paper. And yet, it all coalesces around a single story:

The diverse, complex food webs of Southern California kelp forest will likely be greatly simplified if climate change leads to big storms every year.

BYRNES, J., REED, D., CARDINALE, B., CAVANAUGH, K., HOLBROOK, S., & SCHMITT, R. (2011). Climate-driven increases in storm frequency simplify kelp forest food webs Global Change Biology, 17 (8), 2513-2524 DOI: 10.1111/j.1365-2486.2011.02409.x

Should We Eat Fish? The Online Discussion

One of the main reasons science blogs excite me is the possibility of communication between scientists. It allows for a medium that scientists can use to hash out ideas and do so publicly. This has the added advantage that those who are not big names in a field or somesuch can listen in, as it were, or even step in and participate.

The discussion of science between scientists here, online, also has the added advantage of allowing the public to see inside of scientific debates and discussions in realtime. How do we think? What conclusions can we reach when we talk online rather than through the slow-moving medium of the peer reviewed literature?

I think we have been luck to have just witnessed a great example of how scientists talking through a problem in an online milieu can lead to an interesting conversation – one worth seeing from both the inside and the outside. It concerns the ever contentious issue of overfishing and the eating of (delicious) fish.

I can’t really do better than Emmett at summing the whole thing up, but, I’ll give a rough timeline of what’s been talked about online so far. Enjoy the conversation.

It all started… well, it actually got kicked off by three posts in The Nature Conservancy’s Cool Green Science.

These posts and the corresponding comments are well worth a read. Then came an NYT editorial that really got things roiling – and this is where things got interesting.

Scientists! Of! Fishiness! Online!

It’s very cool stuff that is a must-read, I think. If you read nothing else, really, go see Forum on fish, food, and people right now. Enjoy!

Diversity: It Matters! (for plants & algae)

ResearchBlogging.org Geological time has witnessed 5 extinction crises. Now we’re in the middle of the 6th – this time driven by man. What’s unprecedented, however, is the rate we are driving species extinct. For many taxa, it’s faster than we’ve seen in geological time (see here for discussion).

So? Will vast reductions in the diversity of species on earth matter? I mean, heck, maybe we only need two or three of each taxa, and we’re all good. Or maybe not…

A summary of the types of studies linking plant species richness to other functions found in the literature.

This question has been the driver behind the field of diversity-function research – a young (20-ish) discipline in ecology. But the pace of research has been furious. While a series of meta-analyses peered into the field in 2006, the number of publications has since tripled, allowing us to obtain a clarity not capable just five years ago, particularly when it comes to plants and algae. These results are interrogated by a new meta-analysis by Brad Cardinale and colleagues (including me!) (and some other ocean bloggers!) where we wask some hard questions and come up with some intriguing new answers.

So what do we know?

First, yes. Despite some near vitriolic arguments in the past, plant and algal diversity does increase ecosystem productivity and nutrient use. On average (What? I’m a scientist!).

On other functions, things are more murky. While on average it alters rates of decomposition (think – where do all of those fallen leaves go?), the results are fairly ecosystem dependent. Moreover, we STILL after 20 YEARS do not have enough experiments to tell whether diversity influences rates of herbivory.

Kind of shocking, no? And, having looked at the fullness of the dataset, don’t even get me started on whether diversity of animals influences ecological phenomena. We just need more data.

So that’s the net answer, but what are the ooey-gooey nuances? This is where things get interesting.

Why does diversity matter? Is it that different species really do different things? Or is it that one species does it all, and, by including many species in a plot you increase your likelihood of including that strongly performing species? This is another question that has caused arguments, recriminations, and no small amount of teeth gnashing.

And yet, the answer, unsurprisingly, is both. As Lars Gamfeldt, a co-author and all-around good egg puts it, “Plant communities are like a soccer team. To win championships, you need a star striker that can score goals, but you also need a cast of supporting players that can pass, defend, and goal tend. Together, the star players and supporting cast make a highly efficient team.”

OK, so, it’s both, but…What about the old saw of the rivet hypothesis? The rivet hypothesis states that ecosystems are like the wings of a plane and species are rivets. You can loose a number of rivets and witness no change in function. But, once you start getting to those crucial few…well, things fall apart. We test this idea by fitting three different curves to the data – one consistent with the rivet idea (saturating curve), a linear curve (every species matters), and an exponential curve (species are synergistic, to use business-speak).

Indeed, looking at the data, diversity functions like rivets. There is a point of saturation – where more diversity gives diminishing returns.

A comparison of three models of how biodiversity influences ecosystem function with the percentage of studies that are best described by each model. In summary, diversity: it's like the rivets on an airplane wing.

This leads to a logical next question: how many species are needed to maintain an ecosystem? This question is big, huge, daunting, and any answer we could ever give may have some frightening implications in its use. It’s one of those quantities that scientists are, honestly, loath to name – and, indeed, as a group we’ve had a lot of arguments about how and why to make this calculation.

But, it’s the logical next question, so, we decided to take a stab at it, and Brad used our curve fits to generate some estimates that ask, what is the fraction of species, relative to what we’ve used in our experiments, necessary to maintain 50-90% of production, nutrient use, and decomposition?

A probability distribution based on modeled results asking, what is the fraction of species from an experiment needed to maintain 90% of plant primary production. The arrow points to the mean at 8.27X more species than were included in the maximum diversity treatment.

This is a total back-of-the-envelope rough ballpark estimate. Yet, what we see is kind of surprising. Namely, if we want to achieve 90% of function in our experiments, we have not yet even sampled the amount of diversity necessary. This means that, while our experiments are doing a good job at looking at the diversity of species needed to hit 50% of an ecosystem’s total possible function, we just don’t have the data necessary to peer beyond that – not yet – and that the number of species is quite possibly higher than we have used in the majority of experiments to date.

A word of caution, though, from the text: “We would caution against taking these numbers too literally at this point, since it is hard to imagine that a researcher could place 10× more species into a small experimental plot and expect those species to coexist”. While diversity may need to be higher to achieve this maximum productivity, we simply don’t have the data necessary to get an accurate result here. More studies like BIODEPTH Switzerland, with it’s huge diversity of species, may be necessary to sample this area of the productivity curve.

Still, though, we find that diversity effects get stronger as spatial and temporal scales are increased. Yet most experiments are conducted for no longer than a year at the scale of a bucket (yes, I, too, am guilty of this!) This appears to be a general phoenomenon.

So, what’s the takehome? After 20 years, we know that plant diversity matters. Even at small scales. And not just because of including a single species at high diversity.


If someone tells you otherwise, they likely have an agenda.

But how many species do we really need? How much does diversity matter relative to other drivers of global change (e.g., climate change, nutrient pollution, etc.)? Does diversity loss interact with these other drivers? How do these effects scale? Is space or time more important? How do the functions measured here connect to human services? And while we have a full picture of plants, we do not have a similar clear picture for animal diversity, or, indeed, even something as simple as plant and algal effect on herbivory.

Agh! So much to be done! So much to be synthesized (this paper was part of an NCEAS working group – out first meeting – and having just finished our second meeting, man, we’re cooking!).

But we’re getting there. And the answers seem to be falling in line with theory.

Cardinale, B., Matulich, K., Hooper, D., Byrnes, J., Duffy, E., Gamfeldt, L., Balvanera, P., O’Connor, M., & Gonzalez, A. (2011). The functional role of producer diversity in ecosystems American Journal of Botany, 98 (3), 572-592 DOI: 10.3732/ajb.1000364

The Fingerprint of Fishing

ResearchBlogging.orgHow is fishing changing the ocean? This simple question has motivated a slew of fantastic research. One of the most pervasive ideas has been that of Fishing Down Marine Food Webs. Popularized by Daniel Pauly and colleagues in their 1998 paper, the idea simply states that when humans began fishing, we hit the top predators first. Gradually, as we depleted those stocks, human fishing moved down to the next trophic level. And the next. And the next.

Fishing down marine food webs. A conceptual diagram (left) and data from Pauly et al. 1998 (right). At right, the figure shows change in mean trophic level - the position in the food web - of fisheries from 1950 to 1994 in open ocean (top) and inland (bottom) areas.

This idea has a great deal of intuitive appeal and has guided quite a bit of research. Heck, its one of the primary reasons I focused my dissertation on the consequences of losses of predator diversity in the oceans. But, the idea has been challenged. Essington and colleagues formalized an alternative with Fishing Through Food Webs. The idea here was not that we were serially depleting different parts of the web, but, rather, that as demand for more seafood grew, we added more and more lower trophic levels. In many ways, this idea was slightly more terrifying. If true, we are essentially asking different levels of ocean food webs to simultaneously maximize production for our needs.

So, fishing down? Fishing through? Or, are neither of these right – is fishing purely based on what’s available? Or are we just fishing the heck out of everything?

A new paper by Branch and colleagues in Nature elegantly demonstrates that, in most cases, we’re just fishing everything. They begin by looking at multiple global fisheries and laying out concrete predictions for the fingerprint of fishing as proposed by each of the above scenarios. They then go on to say, ok, what do we see? How has the average trophic level of fisheries changed over time?

Predictions for the fingerprint of fishing under different scenarios. Predictions cover average trophic level of catch (top row), the whole ecosystem (middle row), and the porportion of fisheries collapsed (bottom row).

They find that, overall, we’re fishing everything. Sure, there are exceptions. In the Northern Atlantic in the US and Canada, we’ve fished down the web. And in a few places, it’s all about what species are easiest to catch. But on the whole, exploitation targets the whole food web.

Catch per year of species in different trophic levels. Fishing pressure on all trophic levels has increased since 1950.

These results conflict a bit with the earlier work of Pauly and colleagues. However, the explanation is fairly simple – the way we calculate the position of a fish in a food web has changed. As an aside as a food web ecologist, I’d say, let the buyer beware in terms of what your trophic level calculation is telling you. They different techniques all have various assumptions. If you are going to draw a strong conclusion, make sure it is conditioned upon your assumptions or conduct a sensitivity analysis.

Not only that, but fishing data is not always a reliable indicator of the status of whole marine food webs. There is greater complexity out there, and we need to be careful when using fishing records as proxies for the health of our oceans.

The results of Branch and colleague’s work have critical implications for assessing the health of our oceans. First off, the situation is potentially even more precipitous than what would be possible if we were simply fishing through food webs. Patterns of exploitation are more complex, and in overfished areas, the consequences may be far more unpredictable. Second, just assessing the average position of fished species in a food webs is not sufficient to determine ecosystem health. Rather, we need to look at ecosystem diversity, stability, and the vulnerability of the resources provided by the ocean. We need to not only think about individual species, but how ocean ecosystems work as a whole.

It’s a daunting challenge, and a grim picture. But the analysis here provides some helpful guiding lights through the murk of assessing the health of our oceans. And the future looks hopeful with some major initiatives tackling the problem.

Branch, T., Watson, R., Fulton, E., Jennings, S., McGilliard, C., Pablico, G., Ricard, D., & Tracey, S. (2010). The trophic fingerprint of marine fisheries Nature, 468 (7322), 431-435 DOI: 10.1038/nature09528

Essington, T. (2006). Fishing through marine food webs Proceedings of the National Academy of Sciences, 103 (9), 3171-3175 DOI: 10.1073/pnas.0510964103

Pauly, D. (1998). Fishing Down Marine Food Webs Science, 279 (5352), 860-863 DOI: 10.1126/science.279.5352.860

Finding Truth in a Messy World

ResearchBlogging.org*-note, this was derived from a combination of emails between myself and my former phd advisor. See if you can pick out who is arguing what and where. It’s fun – well, for some of you, anyway.

How do we know the world?

This is a seemingly simple and vast question – one with no answer. And yet, it is at the core of every single scientific endeavor. We make choices. We frame it in pretty language – that we are designing specific tests of mechanistic hypotheses in order to better parameterize our quantitative or qualitative models of natural processes.

Well that language does no one any favors. (Or, as one colleague of mine is fond of saying “Horseshit!”)

We are grappling with the simple issue of how we can gain real true knowledge of the world around us. Now that’s a meaty quandary.

As a PhD student, you begin to wrestle with this problem intuitively, but using the pat answers given to you by the mentors around you. As you grow up, as it were, one day you have to take ownership of it. And it can spawn some interesting discussions. Two papers, seemingly with little to do with each other recently provided their authors takes on this age-old issue, albeit under two different guises. And while they are talking about different pieces of Ecology, I’ve witnessed this argument playing across vast swaths of the natural and social sciences. So, substitute in your discipline of choice in the discussion of these two papers.

The first, Macroecology: Does it ignore or can it encourage further Ecological syntheses based on spatially local experimental manipulations? by Bob Paine is a harsh rebuke of the field of Macroecology. Paine argues that we can only know the world by kicking the can, as it were. Small scale manipulative experiment are the essential building blocks of knowledge. Macroecology, he argues, seeks to substitute large-scale observational analyses of unclear patterns and trends with vague ideas behind them for detailed mechanistic insights.

But what about those detailed models? Or even their application to large-scale experiments. Ellison and Dennis argue in their recent piece, Paths to statistical fluency for ecologists, that it’s time for ecologists to grow up statistically. We’re still grounded in the agricultural statistics of the 1940’s and 50’s, they say. Statistics has evolved so much. It’s time for us to both get more statistically savvy and require a strong mathematic and statistical education for every incoming ecological student. This is true not only for big observational analyses, but even so that we can design and properly analyze better experiments.

These papers and the different camps they represent have a lot to say to one another, really. At first, they may seem to be in disagreement, but there’s an intriguing core underneath both.

An overly zealous devotee of the Ellison and Dennis logic would likely read Paine and argue that his is an argument that is stuck. That the low-hanging fruit of strong interactions that are easily detectable by simple experiments have largely been, well, picked. However, we have made a mistake – one that has haunted us from forest to fisheries management. The world is a complex place. One can look at one particular mechanism functioning with a simplified world in a small little corner of the world and derive some truth. But in the real world, processes modify each other or change in strength – often in sharp and unpredictable ways – across vast swaths of space and time in ways experiments alone can never detect. Similarly, there vitally important patterns and processes in nature that cannot be tested in a square meter plot. Statistical models informed by nature may be the only way to tease apart natural variation and divine real meaning. But, as ecologists, we lack the skills and training to do this correctly – or, worse, to be able to tell when our fellow colleagues are misapplying modern statistics. This must change.

Conversely, a Paine partisan might read the Ellison and Dennis paper and argue that many of our most important insights have come from basic studies where the result is easily observed and convincingly related without fancy statistics. To play devil’s advocate (and make sure to see Paine’s note regarding his own soul in his acknowledgements – cracked me up), one could write a rebuttal to Ellison and Dennis that argued the converse: complex statistical analysis is a crutch. We need to focus our work on elucidating patterns and processes that are plainly observable. These are likely to be the strongest and most important drivers in natural systems. If they’re not that strong, how likely is it that a fancy statistical result showing that they are distinguishable from expectations or a null distribution only serve to allow publication of essentially meaningless results?

This is not an argument against statistical fluency, mind you, but for many there is an inverse relationship between the complexity of the statistics used and the believability of the results. If ecology is interested in influencing management decisions, policy, etc., complex statistical analyses need to be reduced to simple depictions or the results won’t be usable. Will training our students in math/stats classes designed for engineers really help that goal?

I mean, who DOESN’T want incoming students to be more quantitatively literate than we are? But is the program by Ellison and Dennis the way to go? Or will their proposed program extend graduate school by another year and likely deter many good students from a program (“I want to DO ecology, not sit in a classroom and take math classes pitched at engineering students!”).

Clearly, there is an in between – a sweet spot.

The fallacy of Ellison and Dennis’s argument for training rests in what happened to me in my first quarter of Alan Hastings’s excellent Mathematical Methods in Population Biology class. I suddenly understood calculus. In blinding clarity, calculus and linear algebra jumped off the page, and said “Hello! Aren’t we awesome!” It was a similar clarity that gripped me while reading Ben Bolker’s book – a clarity I do not get from reading most of the primary/non-ecological sources when I seek statistical wisdom.

To force students to take engineering/maths classes to understand these things is but a stopgap. The real goal should be to have Ecologists teaching applied math/stats/comp-sci classes that meld these concepts with ecology and the practical analysis of data. For example, I am super-excited about this new book. Imagine combining that with Bolker’s text, or even Gelman and Hill. What wonderful team-teaching opportunities there are! And not just for grad education – what unique classes we could design that build the statistical and quantitative literacy of undergraduates. These are concepts that are necessary far outside the realm of just ecology, but by teaching them in an ecological context, we can give them an entrance point that they would not otherwise find.

Or, at least, that’s my utopian/give-me-a-job view.

This same chain of logic is true of the Paine paper. Microecology is great. But even he admits at the end of the paper that the REALLY strong kind of inference is derived from small-scale experiments replicated across large scales and/or coupling small-scale experiments with observational data. The former is what I’ve read Bruce Menge and others call the ‘Comparative Experimental’ approach or Jon Witman define as ‘Experimental Macroecology’. The latter is similar, but, actually requires the techniques Ellison and Dennis would have us use.

This is a slim point of agreement, but, this combined approach really is very compelling. Kick the system. But never trust your results unless they are true in the world – using the best inference possible. If they are not, ask why not? Kick the system again – but do it everywhere. And if you’re going to do such an experiment right, here in this crazy complex world, you will require more sophisticated machinery than we have been taught to use.

It all leads to some funny places – like my desk. If you were to look at my hellaciously disorganized desk right now, on one side you’d find papers detailing Conditional Likelihood (a non-parametric version of maximum likelihood) and some scribbles of pseudo-code on its relevance and how to implement it for complex SEM models. On the other side is a copy of Morris, Abbott and Haderlie sitting next to a copy of Leighton’s thesis on consumption rates and nutritional demands of purple urchins and abalone.

And damn if I don’t feel like that will all get me closer to Truth with a capital T in my own work. At the same time, this is not the way for everyone. It takes all kinds – from the most math/stats-y of us to those who live and die by their weed-whacker. And when all of these kinds of scientists agree, well, that’s when you know you’ve got something worth paying attention to.

Paine, R. (2010). Macroecology: Does It Ignore or Can It Encourage Further Ecological Syntheses Based on Spatially Local Experimental Manipulations? The American Naturalist, 176 (4), 385-393 DOI: 10.1086/656273

Ellison, A., & Dennis, B. (2010). Paths to statistical fluency for ecologists Frontiers in Ecology and the Environment, 8 (7), 362-370 DOI: 10.1890/080209