ESA (Ecological Society of America) is celebrating its 100th anniversary in 2015. This will culminate in the 100th annual conference in Baltimore in August 2015. As part of the buildup, ESA has asked various people to discuss today (Dec 3) via social media “big ideas or discoveries that have had the greatest impact on ecological science over the last century”. So I’m sharing my thoughts today. Meg & Jeremy will add their own over the next few weeks. Check out Twitter hashtag #ESA100 as well. (And the Brits reading this don’t have to remind us that 2013 was the 100th anniversary of the British Ecological Society and they got there first as I was also at and enjoyed their 100th conference).
A couple months back I took a stab using Wordles at how ecology has changed in 25 years. For this longer time frame of 100 years, I’m not going to pass the buck to technology and going straight to my own (lengthy!) opinions. I am going to divide this into three sections – core ideas that spanned most or all of 1915-2014, ideas that emerged over the latter half of that 100 year period that are dominate now, and ideas that I predict will dominate the next 100 years. I will also divide each section into tools/methods and ecological concepts.
- Differential equations as models of population abundance – without a doubt this has been one of the most dominant ideas for the last 100 years. It started as a way of modelling dynamical chemical reactions and then moved into ecology in the 1920s in the work of Lotka and Volterra (although Verhulst presaged this work with the logistic equation of human population growth in the 1830s rediscovered by Pearl in 1920). By the 1930s full treatises applied to competition, predation and mutualism appeared such as Lotka’s excellent 1925 book Elements of Mathematical Biology (worth reading still today!), Gause’s 1934 book Struggle for Existence and the review paper by Gause and Witt 1935 (Am Nat). If you look at any modern theoretical ecology book (or any undergraduate ecology text book) you will see these core ideas explained in great detail and then elaborated on with age structure, stochasticity, time lags etc. In the 1970s there was a movement led by Robert MacArthur and EO Wilson to define this use of differential equations focused on populations as the sine qua non of good ecology with profound effects (the highly influential population biology graduate program at UC Davis as one example). I would argue population-level differential equations has been THE dominant tool for the last 100 years. I am personally a little ambivalent about this. While I think quantification and math are important in science, I don’t think populations are the only important scale to study (and its not obvious what one variable would be at the center of differential equations at other scales), we have only been able to capture parameters for rates of change of populations in highly phenomenological (almost circular) fashion, and the differential equation approach has led to an overemphasis on equilibria (something easy to solve for in differential equations but not so obviously a prominent feature of nature)
- Succession – succession of plants in the Indiana sand dunes was the 1898 thesis topic of Henry Cowles, founder of the ESA. Frederic Clements also worked on this in the early decades of the 20th Succession has been at the center of one of ecology’s great, ongoing defining debates: individualistic responses vs. species interactions and community structure (Gleason vs Clements). Succession played a central role in Odum and Whittaker’s undergrad textbooks in the 1970s and you can still find a full chapter devoted to succession in every popular textbook today. At the same time succession has become passé in the last 30 years (e.g. the 2009 Princeton Guide to Ecology has almost 100 entries but not one on succession). Deserved death or a pendulum swing that will come back? (Can I say both of the above?)
- Competitive Exclusion, Limiting Similarity, Niche overlap – and etc – The fact that one of four outcomes of the Lotka-Volterra differential equation model of competition leads to competitive exclusion followed in short order by Gause’s 1934 experimental confirmations in a microcosm has led to a central role for competitive exclusion and related ideas like limiting similarity, niche overlap, body size ratios, closely related species not co-occurring in communities in phylogenetic community ecology, and etc. If I were to pick one concept that dominated ecology from the 1930s to the present day, this would be it. Indeed, I would probably go further and say it crossed the line to become an obsession. Competition incorrectly received primacy over predation, disease and mutualism. And the blindingly obvious fact that species coexist outside our window even if they don’t in homogenous bottle systems has not prevented an over focus on how two species coexist instead of the more important question of what controls whether it is 2, 5, 20, or 200 species coexisting.
- Food webs – the food web idea loosely goes back to Stephen Forbes, arguably the first ecologist in America with his 1887 essay on “The lake as a microcosm”. Food webs sensu strictu as a graph of who eats who have run as a key idea through the work of Charles Elton, Robert Paine’s starfish removals and keystone species, Joel Cohen, Stephen Carpenter’s trophic cascades, work on alternative stable states and right into the present day with efforts to model the population dynamics in a food web context using differential equations. Network theory is hot these days and a clear extension of food webs. Food webs sensu latu have also served as a metaphor for the idea both in research ecology and the environmental movement that everything is connected to everything. We love these stories – remove one little insect and watch the whole ecosystem collapse.
- Ecophysiology culminating in mechanisms driving biomes – naturalists all the way back to von Humboldt and Darwin, some of the early German founders of ecology (Warming, Schimper) and running through Robert H Whittaker and his 1975 book have noted that there is very systematic variation in vegetation structure and type across the globe with climate (tall trees in wet tropics, savannas in dry tropics, thorn scrub in deserts, grasslands in dryish summer wet places, Mediterranean in dryish winter wet places, etc). This topic remains active into the present day with attempts to include realistic models of vegetation in global circulation models and carbon models. I am hard pressed to pinpoint a single turning point (although Gates 1965 book Biophysical Ecology is a good stab) , but we have gradually worked out the core physiological principals driving this (water balance, heat balance, photosynthesis controls, etc) and some of the biggest names in ecology (Hal Mooney, Stuart Chapin, Christian Korner, Monteith, Graham Farquhar, Ian Woodward) have worked in this area. The animal people have not been quite as successful in prediction of distribution and abundance, probably because there is not as much variation in growth forms as in plants, but great progress has still been made, especially in lizards and/or thermal ecology, by the likes of Ray Huey, Warren Porter, Bruce McNab etc. Whether you call this field physiological ecology, functional ecology, biophysics or something else it is one of the few areas of ecology to become predictive from first principles.
- Importance of Body Size – if you could only know two simple facts about an organism, probably the two things you would want to know is which taxonomic class (bird, mammal, angiosperm, fern) and body size. Body size makes good predictions about who will eat who, degree of thermal stress, etc but also, in a relationship that is unusual precise in ecology, metabolic rate and a whole host of things that are connected including calorie requirements, growth rates, life span, age of maturity, intrinsic rate of population increase, dispersal distance, etc. The central role of body size has been understood at least since 1932 (Kleiber in a German publication and a 1947 paper “Body size and metabolic rate”in English). Two 1980s books (Peters 1983 The ecological implications of body size and Calder’s 1984 Size Function and Life History) showed just how central body size is. This work received significant recent attention through some of the most highly cited papers in ecology by Jim Brown, Brian Enquist and Geoff West among others. Although the potential of this discovery to inform about poorly understand species of conservation is in my opinion still underappreciated there have been some very clever applications including a paper by Pereira and Gordon 2004 in Ecological Applications, a fun one on dinosaurs by Farlow 1976 in Ecology and John Lawton’s 1996 calculation of the population dynamics of the Loch Ness monster in Oikos. Mechanisms are still hotly debated but the sheer statistical predictive power of body size is rare in ecology.
Tools 1950s to 2050s
This section and the next contain ideas on tools and concepts that got their start in the latter half of the 20th century and are arguably still in their infancy today but with much work going on.
- Stable Isotopes – I’ve never personally used this technique, but the ability to quantify the ratio of different isotopes of an element (say oxygen 16 and 18) in small samples has revolutionized what we can measure. We can measure how old things are (when they died) (carbon), where they came from (strontium), how hot or cold it was when tissue was laid down (hydrogen among others), where/when water used by a plant came from (again hydrogen among others), how high up the trophic chain a species eats (nitrogen), and on and on. I’m sure we’re nowhere near the end of novel measurements that can be done with stable isotopes.
- Phylogenetics – starting with Willi Hennig’s 1950 book on cladistics and Felsenstein’s early 1970s and 1980s papers and software on methodology followed by many others, the ability to unravel the precise evolutionary history of species has changed not just evolution but ecology. Like any such tool, it has created some bandwagons, but there is no denying it has changed our ability to ask meaningful ecological questions in a macroevolutionary context (how fast do different traits evolve, is higher species richness in the tropics due to speciation or extinction, how do coevolving clades speciate, and etc).
Concepts1950s to 2050s
- Space– there has been exponential growth in the study of the role of space in structuring populations and communities. Arguably this started in the 1950s with Andrewartha’s 1954 ecology textbook that had what we would now call a metapopulation on the cover and Skellum’s 1951 diffusion equation models. This was followed by Levins 1969 paper on metapopulations, Hanski’s development and popularization of metapopulations through the 1980s and 1990s, MacArthur and Wilson’s 1969 island biogeography, Simon Levin’s 1970s work showing space can be a coexistence mechanism, Monica Turner and many other’s launch of landscape ecology as a subsdiscipline, the increasing interest in the role of regional pools in structuring local communities (accelerated by Hubbel’s 2001 neutral theory), and the growing interest in dispersal ecology and the role of dispersal limitation. The recent recognition of the importance of scales is closely tied to finally putting our understanding into a spatial context. I don’t think we’re done with space yet.
- Evolutionary Ecology/Optimality – Hutchinson wrote a textbook in 1965 on The ecological theater and the evolutionary play, arguing for the need for stronger links between the two fields (or more precisely observing they exist whether we ignore them or not). Judging by the proliferation of journals in the field of evolutionary ecology I think he was heard! The backlash against Wynne-Edwards 1962 book (Animal dispersal in relation to social behavior) containing group selection arguments certainly focused our collective minds as well. A great deal of individual behavior and life history as well as sociality are now evaluated through the lens of evolution. So are species interactions (i.e. coevolution). And although it is only a short cut, optimality with constraints is a very useful short cut to understand the evolutionary outcome of behavior ranging from foraging to habitat selection to various forms of game theory.
- Mutualisms and Facillitation – competition and predation were dominant ideas for the last 100 years (and remain dominant ideas). But it seems mutualism didn’t get much love until recently. While the existence of mutualism was well understood 100 years ago (and the aforementioned Gause and Witt paper gave a model of mutualism population dynamics in 1935), understanding mutualism as a fundamental structuring force of communities came much more recently. The growth of tropical ecology certainly fed an interest in mutualism as has the increasing study of pollination as an ecosystem service and the idea of facilitation (a gradient from competition to mutualism depending on the harshness of environmental conditions). The role of the microbiome mutualism is likely to be part of Meg’s answer to great conceptual advances.
- Macroecology – I may be biased on this one … but I think the snapping out of amnesia induced by the population only approach to return to our roots and look at some of the oldest questions in ecology like the controls of species richness, the controls of abundance, species ranges, distribution of body sizes etc has been a very good thing. Not in a replacement sense (of e.g. population biology), but in an addition sense of we have to tackle these questions now and not work up to them in 100 years when population ecology is all figured out. And I think it has happened just in time with the looming challenge of global change. It is interesting that the arc of the careers of many of the most famous community ecologists (Rosenzweig, Brown, Mittelbach, Ricklefs, Lawton, May and, ahead of his time, MacArthur) all included a turn toward macroecology. And macroecology seems to be at a magic scale such that it has produced many of the most law-like, universal principles in ecology (abundant species are rare, big-bodied animals are rare, species area relationships, decay of similarity with distance, etc). I could go on for many pages on this topic alone, but I’ll stop here for now!
Tools 21st century
- Remote sensing – Using digital images taken from elevation so as to cover large areas (usually from airplanes or satellites, but increasingly towers too) has been slowly creeping into ecology for decades. At the moment, remote sensing is more informative about the environment (e.g. topography) and the ecosystem aspects (e.g. NDVI as a proxy for productivity or greenup). And these will be continue to be growth areas. I am part of a project supplementing ground weather stations with satellite measurements of weather to fill in the gaps on the ground, and I suspect it won’t be too long until we can dispense with ground measurements entirely for coarse scale measurements of climate at remote sites. And Greg Asner’s work among others on using hyperspectral (100s of channels or frequencies instead of the usual 3 or 4 – think very fine gradations of color) allowing detecting of nitrogen levels in leafs and etc is impressive. But I also think we’re within a decade or two of remote sensing being able to identify individuals to species in some settings like canopy trees or ungulate herds. And that will open up whole new spatial scales to abundance questions.
- Biodiversity informatics – Linnaeus became famous for being the first to formally catalog biodiversity. We have had museums and collectors working at this goal ever since. The rapid movement of this data into online databases is opening up whole new vistas. These include generating species ranges for entire classes of organisms (e.g. all vertebrates by NatureServe and others and soon 100,000+ plants of the New World by BIEN). And changes in species ranges, phylogeny and traits like morphometrics over the last 100 years or so are being evaluated by using dates on collection records. And even at the most basic level, having online, real-time updatable standardized taxonomies are a great boon to those studying poorly known systems. We’re also finally starting to get a handle on some surprising basic trends in biodiversity metrics that make us realize how little we really know about biodiversity trends in response to the Anthropocene.
Concepts 21st century
- Species Richness – Will the 21st century finally answer one of the greatest questions in ecology first raised in the 19th century – why are there more species in the tropics? And more generally will we get traction on the question of what factors control species richness at different scales and along different gradients? I am optimistic – I just hope it happens in my lifetime.
- Species Ranges – the species range is one of the most fundamental properties of a species, and there is a pressing need for prediction of how ranges will respond to climate change, habitat destruction, etc. Yet we have mostly just danced around the edges of this problem and really only have anecdotes for specific species and specific range edges plus a giant cottage industry of predicting range shifts using correlative niche models that aren’t that well validated. We’ve got to do better on this one!
- A predictive theory of the response to global change – I’ve harped on the need for ecology to become more predictive, and I personally think the biggest intellectual challenge and the biggest test of whether ecology has any value for society is being able to predict how the biosphere will respond to human-caused global change (i.e. the Anthropocene).
What I left out
I have left a number of obvious choices out. Some of these are statements of my ignorance. For example, I don’t know enough about ecosystem science to name the big concepts (probably global nutrient cycles and controls of productivity globally belong in the 1950s to 2050s key concepts category, but I don’t have anything intelligent to say on them). And the same for DNA barcoding as a tool for the 21st century.
Other omissions are more intentional. For example, a lot of energy has gone into niches, the diversity stability debate, disturbance ecology, population cycles, and etc but I’m not sure they have yet earned their keep as key concepts that have fundamentally changed our view of ecology (which is not to say they couldn’t still do it). I’d be curious to hear other nominations for this category in the comments (or disagreement with my intentional omissions). I’ve left out some obvious tools too. For example, I didn’t mention a single statistical method as a key tool. This probably shouldn’t surprise readers who know I’m a pretty firm believer that ecology should be in the driver’s seat and statistics is just a tool. The same with big data. Any number of topics I mentioned above intersect with big data (all the way back to von Humbold and Kleiber!). But I just don’t see discussion of big data in isolation as a useful way forward – big data is just a tool letting us finally crack controls of species richness, biodiversity trends, etc. I could have suggested the computer as a key tool, for it certainly has been – it has enabled larger data, much better statistics, null models, complex simulations, not to mention big data and etc. But it’s a little trite and too general to list here.
What do you think? What is missing from my list? What should I have left off? Answer in the comments or Tweet with #ESA100.