The MacroEcological Theory on the Arrangement of Life (or METAL theory) states that biodiversity is strongly, and in a deterministic way, influenced by climate and the environment. This influence mainly occurs through the interactions between the species’ ecological niche and both climatic and environmental changes. Using the ecological niche allows the consideration of underlying processes (genetic and physiological) that are difficult to identify and parametrise for a large number of species. Interaction between the niche and climatic and environmental changes propagate from the species to the community and ecosystem levels and are detectable from the smallest ecosystems to the whole ecosphere. This theory offers a way to make testable ecological and biogeographical predictions to understand how life is organised and how it responds to global environmental changes, including climate change. At the organismal level, the theory predicts species phenology and biogeography, phenologic and biogeographic shifts as well as local and large-scale responses in species’ abundance to climate change and at the community level, it explains the large-scale organisation of life (biodiversity patterns) and community shifts, including abrupt shifts. All these phenomena, which have been separately investigated in the context of climate change, can therefore be connected through the METAL theory.
Bay of Somme (English Channel) Photos: Dr Grégory Beaugrand
Etretat (English Channel)
Bay of Somme (English Channel)
1. The macroscopic elementary brick of METAL is the concept of the ecological niche (sensu Hutchinson)
Thomas Malthus showed in an essay in 1798 that all species have the possibility to increase in number indefinitely. This was summarised in the Gause’s book ‘the struggle for existence’ by the following mathematical equation (Gause 1934):
Where Nt is the number of individuals of a given species at time t and b the exponent of the geometric progression. From Equation, it is clear that a species could increase indefinitely if there were no external mechanisms to control its progression (Figure).
Population growth predicted by the exponential model for different values of b. From Beaugrand (2015, Earthscan)
All organisms may populate the whole planet at a different speed depending on b in the above Equation. The rate of reproduction was already known to be a function of organism size (Gause 1934). Therefore, plants or animals with high dispersal capabilities could have a pandemic distribution if they were not limited by the constraints of the environment affecting their physiology. This is summarised by the Becking/Beijerinck’s law, which states that “everything is everywhere, but the environment selects” (De Wit and Bouvier 2006; Lomolino et al. 2006). This law is especially true for plankton and probably explains why these organisms adjust very quickly their spatial distribution to changing climatic and environmental conditions. Species with high dispersal capabilities are rapidly constrained by the effect of the environment on their physiology. It is likely that when the stress imposed by the environment becomes high, the effect of interspecific relationships such as competition and predation (or resistance to parasite or disease) become more patent, reinforcing the direct effect of changing climatic and environmental conditions on the individuals. Therefore, the environment exerts a control on each individual species. This has well been recognised for decades. For example, the law of minimum of Justus von Liebig (1840) stipulates that production fluxes of an organism is an exclusive function of a limiting factor (Baar 1994). This law has received large empirical support (Kooijman 2001). The law of tolerance (Shelford 1913) extended the law of Justus von Liebig by stating that a species is limited by its range of tolerance for environmental factors (Lynch and Gabriel 1987). However, many refinements can be added. The tolerance range varies during the life cycle of the organisms, younger organisms being generally more sensitive than older stages (Reygondeau and Beaugrand 2010). The tolerance range varies among ecological factors. A species can be ‘euryecious’ to many factors while being ‘stenoecious’ to another. The range of species tolerance to a given factor may also be modified by another (interactive effects).
Both the law of Minimum and the law of tolerance led to the concept of the ecological niche. The first idea of the niche was envisioned by Joseph Grinnell (Grinnell 1917) as the habitat requirement of a species. In his paper, Grinnell related the small spatial distribution of the California thrasher (Toxostoma redivivum) to the adaptation of this bird through physiological and psychological processes to a narrow range of environmental conditions (e.g. temperature and atmospheric humidity). Charles Elton envisioned differently the niche concept (Elton 1927). The zoologist defined the niche in term of functional attributes. The niche is the place a species occupies in the food web and the influence it exerts on its environment (Chase and Leibold 2003). The concept of the niche of Hutchinson (Hutchinson 1957) is probably the definition of the niche that has been the most widely applied by ecologists. The Hutchinsonian niche represents the combination of environmental tolerances and resources required by an organism. Hutchinson (1957) conceptualised this notion with the so-called n-dimensional hypervolume, in which n ideally corresponds to all environmental (biotic and abiotic) factors. This way to define the niche was considered to be revolutionary because it was operational, enabling a straightforward quantification of the species’ niche. The figure below shows a simplified representation of the niche of a hypothetical species characterised by only one dimension (one environmental factor).
Hypothetical one-dimensional niche showing how environmental conditions may affect a species. Note that the niche may not always be Gaussian. Modified, from Helaouët & Beaugrand (Helaouët and Beaugrand 2009).
Only optimal conditions generate high abundances and allow for successful reproduction (Helaouët and Beaugrand 2009). When the environment becomes less favourable, this affects consecutively offsprings production, growth and feeding. Extreme conditions become critical and may eventually affect survival (Schmidt-Nielsen 1990). Towards the niche extremities, the energy taken from the environment become more assigned to maintenance and to ensure homeostasis. The energy is therefore dissipated as heat. Towards the niche centre, energy is allocated to productivity and stored in organic matter as biochemical (endosomatic) energy, becoming accessible to other organisms (Connell and Orias 1964). The Hutchinsonian niche can also include biological factors, although the operational concept rarely incorporates biological interactions. The Hutchinsonian niche is close to the niche of Grinnell and some authors consider that the Hutchinsonian niche is simply a better (and mathematical) conceptualisation of the niche of Grinnell (Jackson et al. 2009).
Many authors have worked on the niche concept (Macarthur 1972; Tilman 1982). Chase and Leibold (Chase and Leibold 2003) proposed a revised definition of the niche as “the joint description of the environmental conditions that allow a species to satisfy its minimum requirements so that the birth rate of a local population is equal or greater than its death rate along with the set of per capita effects of that species on these environmental conditions”. They then pursued in a second definition by stating that the niche is “the joint description of the zero net growth isocline (ZNGI) of an organism along with the impact vectors on that ZNGI in the multivariate space defined by the set of environmental factors that are present”. This definition is close to the Huchinsonian niche but perhaps more difficult to implement.
2. An important assumption in METAL: the Gause's principle
The Gause’s principle of competitive exclusion stipulates that in a community at equilibrium two species with a similar ecology (i.e. similar niche) cannot live in the same place (Gause 1934; Hardin 1960). Or in a few words, “complete competitors cannot coexist” or “ecological differentiation is the necessary condition for coexistence” (Grubb 1977). This principle is said to be a “corollary of the process of evolution by natural selection” (Grubb 1977). The contention of the Russian microbiologist (Gause 1934) has been at the origin of many works and reviews to explain how so many species coexist in the same place (Hardin 1960). This principle, often used when species assembly rules are examined, is applied in the models developed as part of the METAL theory.
3. Rationale of the METAL theory
The responses of marine ecosystems and their biodiversity to climate and the environment involve many organizational levels and cross many scientific disciplines from the lowest biological organisational level –the gene– to the biosphere (Figure).
Effects of climate-induced environmental changes on ecosystems and their biodiversity at different organizational levels. From Beaugrand (2015, Earthscan).
Species are set up genetically to experience a specific environmental or a climatic regime. When random or seasonal changes in environmental conditions occur, it has to be adjusted by a corrective response. The law of requisite variety formulated by Ashby (Ashby 1958) stipulates that a system that can respond in a great variety of ways can better compensate for a variety of perturbations. Therefore, a species can only occur where it has the physiological potential to resist to climatological or environmental variability.
Huntley and colleagues (Huntley et al. 2006) hypothesised from first principles that species may respond to climate change in two different ways. First, the response can be through natural selection of genotypes with selection of individuals best adapted to the new climatic or environmental conditions. Second, species might exhibit a temporal (phenological) or spatial (biogeographic) response, allowing species to match their bioclimatic envelope. How can genetic difference alter the ecological niche? Long-term climate changes from the tectonic frequency to the historical band can engender genetic changes. Genetic changes at the species level more often observed at the tectonic scale cause undoubtedly changes in the limits of the ecological niche. Toward higher frequencies from the orbital frequency band, genetic changes manifest themselves at the population level. However, these changes are unlikely to affect the species ecological niche. If climate change is too rapid, it might be above the species capability to adapt (Barnosky and Kraatz 2007). How changes in genotypes and phenotypes will alter the limits of the species bioclimatic envelope remains an open question. Does this complexity mean however that we need to consider all processes to anticipate the effect of climate change on marine species and ecosystems? Genetics set up the limits of the physiological and the biological responses of an individual to climate change. Processes such as mutation rate, selection, gene flow, inbreeding depression are classically examined because they influence either phenotypic plasticity or adaptation and affect species survival on long-term time scales (Hewitt 2004) (Figure).
Interpretation of the effects of environmental changes along a hypothetical one-dimensional gradient at three biological levels: (A) genetic, (B) physiological and (C) populational levels. Note that the shape of the niche may not always be Gaussian. From Beaugrand (2015, Earthscan).
To understand the effects of climate change at the individual level, physiological responses are particularly important (Pörtner et al. 2001) (Figure), not least processes contributing to thermal tolerance (e.g. oxygen limitation, ventilation and circulation processes) are particularly scrutinized (Pörtner 2001). At this level, behavior is also significant and some species may reduce the effects of climate change on their physiology using a specific behavior (e.g. the search for microrefugia or microniches (Chapperon and Seuront 2011)). At the population level, processes such as reproduction, growth, dispersal, mortality rate, all become relevant (Speirs et al. 2005) (Figure) and individual-based models (IBMs), also known as process-based models, have been used to investigate the effect of climate change at both temporal and spatial scales. When the whole species range is investigated, views diverge on how to tackle the problem (Beaugrand et al. 2013; Pearson and Dawson 2003). A common school of thought considers that nearly all the above mechanisms are important and applies process-based models (the reductionist approach). Mechanistic models have therefore been developed providing a reasonable fit to actual data and improving our understanding of the spatial distribution of some marine species (Speirs et al. 2005). A reductionist approach is not possible for many marine species however, because the biological responses of marine organisms to climate-induced changes in their environment are rarely known (Beaugrand et al. 2013). Furthermore, the integration of all biological processes into a model (sometimes based on the sum or the product of hundreds of equations) may lead to information overload and increase the uncertainties on the estimates, counterproductively. In contrast to this reductionist method to study ecological phenomena, another school of thought takes the macroecological approach that proposes to develop macroscopes to reveal emergent patterns and processes (Brown 1995; Gaston and Blackburn 2000). The macroscope is a theoretical concept, coined by de Rosnay (De Rosnay 1979), which looks for properties of a system at a macroscopic level. In the same way a microscope is needed to observe the infinite small or a telescope is needed to observe the infinite big, the macroscope is a theoretical concept to investigate complex systems (De Rosnay 1979). The concept of emergence in ecology stresses that an ecological entity or unit cannot be predicted from the study of its elements. This is also called non-reducible property (Odum 1971). Emergent patterns and processes are therefore not simple extrapolations from microscopic studies to larger scales. In a complex adaptive system (CAS), even when the components and the assembling rules are known, it is rarely possible to predict the details of the resulting system because of the emergence of new properties at a greater level (Levin 1998). At the macroscopic level, prediction of the structure and behaviour of CASs may be possible by focusing on emergent patterns, some of which being statistical (Brown 1995; Gaston and Blackburn 2000). For example, the concept of the ecological niche sensu Hutchinson (Hutchinson 1978) can be considered as an elementary emergent macroscopic mechanism because it integrates the sum of all physiological and demographic processes occurring at both the individual and population levels, which is not possible to realistically implement in process-based models for most species (Beaugrand et al. 2013).