Geography and its subdiscipline biogeography are multidisciplinary sciences that will always be the subject of continuous discussions about their specific role in science. There is, however, a consensus view that a major aim in biogeography is to elucidate the factors that influence the diversity of life and its spatial distribution on the globe. Biogeography is a rapidly growing science, and the number of studies on species diversity is growing particularly quickly.

If one evaluates the research on species diversity by the Popperian criterion that progress is made by falsification of hypotheses, then research on diversity is in a very bad state. There seem to be too many hypotheses explaining variation in species richness, and these are not actually mutually exclusive (Palmer, 1994). It is as if the diversity issues begat a diversity of potential hypotheses. This is very different from the situation for other unsolved scientific phenomena that have normally been accompanied by a limited set of potential hypotheses.

Thus, most biogeographers and ecologists see the need for theory reduction if there is to be any scientific progress on the subject of diversity. A few attempts have been made to reduce the number of hypotheses (e.g. Palmer, 1994; Scheiner & Willig, 2005). Palmer (1994) listed over 120 different hypotheses on species richness and then elegantly demonstrated that many of them were more or less synonymous. He grouped the hypotheses according to seven different ways that violate the competitive exclusion principle and its ultimate result, namely monocultures. The approach was adapted from population genetics and its successful use of the Hardy–Weinberg equilibrium: there is no evolution unless the equilibrium is violated.

Richness gradients exist over a continuum of scales from a few metres to the global scale. Richness will always be determined by processes external and internal to the species assemblages. It is mainly the puzzles over fine-scale internal mechanisms that have resulted in the myriads of related hypotheses, based on the many loose concepts in community ecology. These fine-scale internal mechanisms may cause only minor variations in richness along the broad-scale species richness gradients. Thus there is a shorter list of mechanisms that have been hypothesized to account for broad-scale richness patterns. Currie (1991) summarized eight main factors that influence species richness on broad spatial scales, whereas Hawkins et al. (2003) refer to 30 hypotheses. The strong covariation between climate and the broad-scale (e.g. latitudinal and elevational) diversity gradients has been well documented, and thus the majority of the broad-scale hypotheses are related to climate variables or to factors related to climate such as productivity (Hawkins et al., 2003). Mechanisms that are not related to climate and productivity are either non-environmental factors, such as area and geometric constraints, or are related to dispersal, speciation, and evolutionary history.

The most ambitious approach to theory reduction is to find the single factor that makes most other hypotheses redundant, such as the energy richness hypothesis (Currie, 1991) and the metabolic theory of ecology (Brown et al., 2004). These authors use ambient energy as the main predictor variable and assume that this principal factor should explain variation in species diversity for all types of organisms, or at least for a wide range of organisms. These theories convincingly explain why there are so many species in the equatorial zone and so few at the cold poles. But why are there relatively few species in energy-rich deserts? Too much energy and a lack of water: hence, water must be an essential factor for diversity in general. Hawkins et al. (2003) found that water variables usually represent the strongest predictors in the Tropics, Subtropics and warm temperate zones, whereas energy variables (animals) or water–energy variables (plants) dominate at high latitudes. They concluded that analyses that do not include water–energy variables are missing a key component for explaining broad-scale patterns of diversity (Hawkins et al., 2003).

This key component is the starting point for a recent publication by O'Brien (2006) that elaborates the idea of water–energy dynamics (WED: O'Brien, 1998). She defines WED as changes in the state, form or location of matter and/or in the energy of a system caused by changes in the state, form, location or internal energy of water doing work over space and time. The idea is labelled ‘biological relativity to water–energy dynamics’, and is an elaboration of WED that should relate to the geography and evolution of life in general. The WED mechanism underpins the Interim General Models (IGM I and II), which are statistical models that test the idea that optimal energy and liquid water are the fundamental factors linking climate and tree richness. The IGMs are able to predict the potential for woody species richness with a fairly good accuracy at continental scales (O'Brien, 1998; Field et al., 2005), based on the idea that woody species richness is a linear function of water and a parabolic function of energy. Water is represented by mean annual precipitation, and minimum monthly potential evapotranspiration is used as the energy variable (the predictions are aimed at a grain size of 25,000 km2).

Biological relativity to WED refers to the relative nature of biotic dynamics to the operation and outcome of WED as defined above (O'Brien, 2006). O'Brien's earlier work, in developing the IGMs, aimed to explain the variation in richness of one type of organism, but the new ‘biological relativity to WED’ should be used to explain both current diversity in general and the evolution of life's diversity. O'Brien (2006) starts her theoretical reasoning with water and the work done by water. The rationale is:

There is no life without liquid water.

There is no life without energy either, but energy itself does not give life.

If energy provides water in a liquid state then the abiotic condition for life is present.

This is formulated as the first biogeographical principle: ‘water–energy-dynamics in particular, determine the initial abioitic environmental conditions available for life to exist and do work, everywhere and always’ (O'Brien, 2006).

It is modestly presented as an idea, although it bears all the characteristics of a developing theory. This idea has the potential to reduce the number of hypotheses explaining variation in species richness, but its success will depend on how strong the synthesis is and its predictive power for a wide range of organisms, not only trees. The most important results in science are sometimes the generalizations and theories that derive from the basic factual material, i.e. first principles. O'Brien uses basic facts from standard textbooks on physics, hydrology, climatology, geomorphology, biology, and ecology. She underlines the abiotic part, since most potential readers are experts on the biotic component. O'Brien (2006) lists five comprehensive implications of this idea: in summary, (1) the biosphere is a subsphere of the hydrosphere, (2) life depends on liquid water to do the necessary work to sustain the abiotic world, (3) WED is a fundamental mechanism of evolution and a constant mechanism in natural selection, (4) from WED one can predict global patterns of biotic dynamics (e.g. species richness) that will dissolve into chaos locally, and (5) this theory provides a framework for operationalizing hierarchy theory and trans-scalar explanations for the geography and evolution of life's diversity.

The first two points are intuitively easy to accept. Water is regarded as a resource gradient and energy as a regulator of water, that is, a regulator gradient. A key phrase in the paper is ‘work done by water’, but its actual meaning may be work done by WED, since water and energy are in fact both resources and simultaneous regulators of each other. The third point has significant implications for evolutionary biology, which already has a mature theoretical framework, compared with ecology and biogeography. Nevertheless, evolution started in water, and the constant challenge for life on dry land is the water balance in general, and successful sexual reproduction without living in water. O'Brien (2006) infers increased evolutionary rates in locations with plenty of water and high energy (Tropics). There is little empirical evidence for this at present, but the rapidly advancing genetic sciences may soon be able to test this prediction (Currie et al., 2004). The fourth point is empirical, and demonstrated through tests of predictions deduced from the IGM based on the WED idea (O'Brien, 1998; Field et al., 2005). An interesting potential use is to compare and analyse predicted and actual values of richness. For instance, it will be possible to elucidate if all the potential trees have recolonized the area covered by the last glaciations, or if northward migration is still going on. If the model overpredicts the number of trees, it may indicate that migration is an ongoing process, although it will be confounded by the effects of global warming.

The second part of the fourth point states that these predictions are constrained to a certain grain size, and that patterns in richness ‘will dissolve into chaos locally’. It is true that the predictions based on IGM may not be very good when the scale of resolution is increased (Vetaas & Ferrer-Castan, unpublished data), but this seems to be in contrast to the fifth point, which promises a trans-scalar explanation and the operationalization of hierarchy theory.

It follows from the theory that WED does not cease to function at a certain scale. There are, for instance, species richness gradients only a few metres long caused by tree canopies that intercept both solar energy and water vapour, generating a gradient in local evapotranspiration and available water (Vetaas, 1992). There are huge differences in the actual temperature from the xeric ridges in windy alpine landscapes to the moist snow-beds on the leeward slopes a few metres away. This causes a clear richness gradient. Thus models other than IGMs have to be developed within this new framework if trans-scalar explanations are to be achieved.

It may be easy to understand how WED operates at a short spatial extent in extreme environments as described above, but it is obviously more difficult to understand this in a dense moist evergreen forest (O'Brien, 2006; Fig. 2). In such a landscape, the potential internal mechanisms are many, and the theoretical framework will be filled with loose community concepts. However, to link this microbiosphere with the broad-scale system models that use geophysical parameters is a major challenge if trans-scalar explanations are to be achieved.

The reasoning behind biological relativity to WED has provided some conceptual structures that are based on a few general propositions. From this conceptual framework one may derive different types of models that generate testable predictions. This work appears to mark a significant step in consolidating ideas on water and energy as predictors of biological diversity.


  1. Top of page
  2. References
  • Brown, J.H., Gillooly, J.F., Allen, A.P., Savage, V.M. & West, G.B. (2004) Toward a metabolic theory of ecology. Ecology, 85, 17711789.
  • Currie, D.J. (1991) Energy and large-scale patterns of animal- and plant-species richness. The American Naturalist, 137, 2749.
  • Currie, D.J., Mittelbach, G.G., Cornell, H.V., Field, R., Guegan, J.-F., Hawkins, B.A., Kaufman, D.M., Kerr, J.T., Oberdorff, T., O'Brien, E.M. & Turner, J.R.G. (2004) Predictions and tests of climate-based hypotheses of broad-scale variation in taxonomic richness. Ecology Letters, 7, 11211134.
  • Field, R., O'Brien, E.M. & Whittaker, R.J. (2005) Global models for predicting woody plant richness from climate: development and evaluation. Ecology, 86, 22632277.
  • Hawkins, B.A., Field, R., Cornell, H.V., Currie, D.J., Guegan, J.-F., Kaufman, D.M., Kerr, J.T., Mittelbach, G.G., Oberdorff, T., O'Brien, E.M., Porter, E.E. & Turner, J.R.G. (2003) Energy, water, and broad-scale geographic patterns of species richness. Ecology, 84, 31053117.
  • O'Brien, E.M. (1998) Water-energy dynamics, climate and prediction of woody plant species richness: an Interim General Model. Journal of Biogeography, 25, 379398.
  • O'Brien, E.M. (2006) Biological relativity to water-energy dynamics. Journal of Biogeography, 33, 18681888.
  • Palmer, M.W. (1994) Variation in species richness: towards a unification of hypotheses. Folia Geobotanica, 29, 511530.
  • Scheiner, S.M. & Willig, M.R. (2005) Developing unified theories in ecology as exemplified with diversity gradients. The American Naturalist, 166, 458469.
  • Vetaas, O.R. (1992) Gradients in field-layer vegetation on an arid misty mountain plateau in the Sudan. Journal of Vegetation Science, 3, 527534.