Showing posts with label complexity. Show all posts
Showing posts with label complexity. Show all posts

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.

Tuesday, March 29, 2016

What are important directions for ecology?

I was recently asked “what is the most important problem in ecology?”. I was dissatisfied with whatever I ended up answering, so it has been on my mind. I think there is an analogy with medicine here – it’s a little like asking a medical scientist “what is the most important disease to cure?” Similarly, there are multiple possible answers, and the one you give will depend on your area of interest/what type of doctor you are. (I also assume this is a question about basic research, and the answer is not as simple as saying, stop extinctions or prevent habitat loss).

Levels of biological organisation.
The medical analogy breaks down a little because ecology is *far* more complicated than medical science. Medicine has a foundation in anatomy and physiology, which in turn rely on basic sciences like cell biology and genetics. This creates a reasonably constrained framework within which further learning/investigation can be organized. Medicine typically stops at the level of the individual, but ecology inherently involves many additional levels of organization (from individuals, to populations, to species and communities, to ecosystems, and beyond). Within any one of these higher levels of organization (population, community, ecosystem), there can be such an immense amount of variation in outcomes and dynamics that ecologists can lose sight of connections with lower and higher levels. For example, community ecology encompasses so much complexity on its own, that also considering the impacts of population level processes and on ecosystem level processes is a tall order. But, we should also appreciate, given these barriers to understanding, just how far ecology has actually advanced in the last 100 years. The combination of reductionist experiments and descriptive work at all scales has been immensely successful (e.g. see this blog post for a partial list). Many general tools have been developed that we can then use to answer specific ecological questions (the integration with statistics with ecology has been highly successful; the use of specific mathematical models). Still, the ability to reconcile multiple levels of organization and scales still limits ecology.

This is a problem that cell biology has also experienced, and is now approaching via systems biology: "The reductionist approach has successfully identified most of the components and many of the interactions but, unfortunately, offers no convincing concepts or methods to understand how system properties emerge...the pluralism of causes and effects in biological networks is better addressed by observing, through quantitative measures, multiple components simultaneously and by rigorous data integration with mathematical models"(1): to me, this quote rings so true for ecology as well. Systems biology uses mechanistic, mathematical and computational models to attempt to represent multi-scale complexity.

Of course, the optimism about systems biology might be premature in that it hasn’t produced many useful models yet, such that it may be “more of an agenda than a body of results.”. Some of the best “systems ecology” (e.g. meta-ecosystem models) are very system specific and data-heavy (e.g. 2). Can they inform us about generality in ecology?

All of which is to say, I think the most important problems in ecology relate to this need to make the connections between studies and systems and levels of organization. But, doing so may be difficult.

More specific problems

1. The scaling of ecological processes. Many ecologists include a line about being ‘interested in questions of scale’ on their website blurbs. Despite this, our understanding of the aggregate outcome of multiple processes that are occurring at different spatial or temporal scales remains limited, and poorly predictive. There have been a few useful starts (particularly in Peter Chesson’s scale transition papers (3, 4)), but recent theoretical interest seems to be low. We have data at the community scale, and data at the macro-scale. How do we connect these (and can we)? Models describing how processes occurring at smaller scales produce larger scale dynamics can be complex: they may include non-linearities, autocorrelation between regions, the combination of discrete and continuous events, and multiple attractors.

2. Mechanisms maintaining multi-species coexistence in the real world. Hutchinson’s paradox of the plankton remains unsolved*. Community ecologists have invested a lot of time and energy into understanding species interactions as seen in natural communities. To explore the mechanisms behind coexistence, usually (but not always) ecologists have focused on two-species interactions (or maybe 3): understanding coexistence in larger groups tends to be mostly restricted to theory. But fitting the individual pieces into the larger puzzle is exponentially more difficult: in observed large groups of interacting species, what is the relative contribution of the many coexistence mechanisms identified? Which mechanisms are most important, and how do they change through space and time?
*Perhaps not surprisingly, given it is a paradox...

3. Moving farther away from species. In so many ways, focusing on ‘species’ as the unit of measurement is limiting, because ‘species’ is a discrete term and ecology is interested in quantitative measures. Important advances have been made by redefining ecology as the outcome of species traits and species interactions (5). But I think our ability to connect these ideas more closely to species’ multidimensional niches can still improve. In particular, understanding that traits and interactions can change in context-dependent ways (plasticity, ontogeny, environment) will be important (6, 7).

4. Reproducibility of ecological research. This is more of a philosophical question - how do we achieve reproducibility in a science where context-dependence, alternative stable states, chaos and stochasticity all affect results? How do we differentiate between reproducibility (same results under identical conditions) and generality (same results under similar conditions) in results?

References:
1) Sauer, Uwe; Heinemann, Matthias; Zamboni, Nicola. Genetics: Getting Closer to the Whole Picture. Science 316 (5824): 550–551. doi:10.1126/science.1142502. PMID 17463274.

2) Dominique Gravel, Frédéric Guichard, Michel Loreau and Nicolas Mouquet. Source and sink dynamics in meta-ecosystems. Ecology 91(7): 2172-2184.

3) Chesson, Peter. Scale transition theory with special reference to species coexistence in a variable environment. Journal of biological dynamics 3.2-3 (2009): 149-163.

4) Melbourne, Brett A., and Peter Chesson. The scale transition: scaling up population dynamics with field data. Ecology 87.6 (2006): 1478-1488.

5) McGill, Brian J., et al. Rebuilding community ecology from functional traits. Trends in ecology & evolution 21.4 (2006): 178-185.

6) Poisot, T., Canard, E., Mouillot, D., Mouquet, N., Gravel, D. & Jordan, F. (2012) The dissimilarity of species interaction networks. Ecology letters, 15, 1353–61.

7) Siefert, A., Violle, C., Chalmandrier, L., Albert, C.H., Taudiere, A., Fajardo, A., Aarssen, L.W., Baraloto, C., Carlucci, M.B., Cianciaruso, M.V. and L Dantas, V. A global meta‐analysis of the relative extent of intraspecific trait variation in plant communities. Ecology letters 18.12 (2015): 1406-1419.

Wednesday, April 17, 2013

Progress on the problem of pattern, process and scale

Jérôme Chave. 2013. The problem of pattern and scale in ecology: what have we learned in 20 years? Ecology Letters. DOI: 10.1111/ele.12048.

Why do patterns get so much attention from ecologists? MacArthur (1972) suggested it was because patterns imply repetition, and repetition implies predictability. And prediction is the Holy Grail of ecology. Of course, patterns are meaningless without consideration of spatial or temporal scale. As Levin put it in his MacArthur lecture (1992) "the description of pattern is the description of variation, and the quantification of variation requires the determination of scales". Observing, modelling, and predicting ecological patterns at differing spatial scales has dominated much of ecological thought since Levin’s paper – today, entire subfields heavily focus on patterns through space or time (species-area relationships, macroecology, biogeography, etc).

When ecological research focuses on pattern, but lacks attention to process and scale, it has received much (deserved) criticism. Even when patterns are considered at the appropriate scale and with regard to process, the ability to understand how these processes and patterns translate from one scale to the next (i.e. how do we explain the differing relationship between invasion success and community diversity at local compared to regional scales?) is still limited. And yet clearly connecting processes across scales is a central goal. In the upcoming issue of Ecology Letters, a review article by Jérôme Chave looks at how ecology has progressed in dealing with patterns and scale in the last 20 years.

Chave does a great job of placing current ecological thought into historical context. Sometimes we forget that one of the benefits of ecology’s youth is that ecology has developed concurrently with necessary technological advancements and demand for ecological knowledge. As a result, the need for ecological knowledge and the ability to provide it are tightly linked in time. As a result, Chave suggests that ecology is making noticeable progress, particularly in four focal areas: 1) coupling ecology and evolution, 2) global change, 3) modularity in interaction networks, and 4) spatial patterns of diversity.

The first two topics reflect ongoing issues in ecology. The incorporation of evolutionary dynamics into ecology is an increasingly popular topic (for example), and it is not uncommon for ecological and evolutionary dynamics to have similar temporal scales. Explaining temporal patterns then may require coupling models of ecology and evolution: for example a study of Darwin’s finches found that for one period evolutionary dynamics were occurring on a more rapid temporal scales than ecological dynamics. Global change has dominated ecological research and the problem of scaling processes up from local to global or from global to local effects (of temperature on productivity, etc) is another clear area of growth. This may be the most successful attempts to scale, since models of global carbon cycles have progressed from empirical data and models to predictive models. An apparent example of what can be achieved when demand and appropriate technology are both present.

The remaining two foci relate to networks, and spatial patterns of diversity. The first, modularity in interaction networks, allows groups of interactions to be incorporated into larger scale networks; for individual variation could be incorporated into interactions between species. More generally, Chave suggests that the “abstracted multidimensional space of an interaction network” might be one way to simplify temporal and spatial scales. He suggests that this is where ecology could learn from other studies of complex biological systems such as cellular networks and networks of human governance and management. Finally, spatial patterns of diversity – a striking and oft-considered issue in ecology – are suggested as an area in ecology that has seen advances. Biological diversity is patchy through space, and the amount of patchiness is dependent on the scale of observation. Planktonic blooms might be patchy on a global scale while tropical trees might be patchy over meters. Scaling from local patterns to global has been difficult – for example, models of local dispersal don’t necessarily predict regional dispersal patterns. Chave suggests that one problem in the past was the ignorance of processes at larger scales (i.e. systematics, biogeography) and a predominant focus is on local processes. He provides a few examples that have bridged this issue, for example neutral theory includes both regional and local processes, while ecophylogenetics incorporates evolutionary history.

The review focuses attention on several relevant or insightful approaches to the problem of pattern and scale, and suggests possible connections between ecology and other areas of work (for example, interaction networks and metabolic networks). Although it provides interesting examples, it offers little synthesis or ideas for reconciling issues of pattern and scale, and while the four foci are valid and appropriate, they feel like a rather patchy way of covering a larger and more general issue. This may simply be too complicated and large a topic to cover in a single short review. Chave seems a little generous is giving props to approaches which at their best do incorporate multiple scales (e.g. neutral theory and ecophylogenetics), but which arguably have relied heavily on pattern analyses without a strong focus on process, something that seems to go against the spirit of the review. In addition, some of the explicitly general attempts to reconcile scale and pattern in community ecology are missing. For example, a series of papers from Brett Melbourne and Peter Chesson used 'scale transition theory' to model dynamics across multiple scales. This framework has been applied at least to a few fisheries-related papers. In addition, research on predator-prey dynamics has long considered the question of how functional responses scale up (one review). That said, it's clear that ecology has made progress in some areas and that there are options for moving forward.

Ultimately, Chave seems to suggest that the question of how well ecology can deal with patterns and scale depends on whether complexity is reducible or intrinsic to understanding natural systems. He goes so far as to state “This suggests that in approaching novel frontiers of the study of complex ecological systems we need to pause about the challenge ahead of us...Once we enter the realm of complex systems, neither physics nor biology are well equipped to progress.” This is obviously a pessimistic take on the future for ecology. Is it true?