Credits – Valerio Giacomini (1958) La Flora – Conosci L’italia, Touring Club Italiano. Milano
It’s a fact. Biodiversity is not uniformly distributed over the Earth’s surface. Some regions are lush with a rich, heterogeneous flora, others are homogeneously covered by only a few plant species. A recurrent pattern is the decrease of the number of animal and plant species from the equator to the poles, as well as from to low to high elevation. What happens when rather than considering the number of species, one focuses on the variability in species composition (=beta-diversity) and compares this variability across geographical regions?
We have just published a new study in Ecography to understand how beta-diversity varies along elevation gradients. There is evidence that, similarly to species richness, beta-diversity also decreases with increasing elevation and latitude. But what are the mechanisms behind this pattern?
Whereas classical biogeographical studies mainly focused on latitudinal and elevational patterns in the number of species, other facets of biological diversity received less attention. Indeed, diversity is an incredibly vague concept. Not only there are different dimensions that one could measure, but these are also intimately related to the spatial scale of the analysis.
Now, the problem is that comparing beta-diversity among sites or regions is difficult, as beta-diversity is mathematically defined as the ratio between the number of species in the species pool of a region (=gamma-diversity) and the average number of species that can be found in a given sampling plot (=alpha diversity). How big is a ‘region’ or a ‘plot’ depends on the focus of a study, but let’s leave this aside here. Since both alpha and gamma diversity vary along geographical gradients, how can we test whether the variation in beta-diversity does not simply depend on the fact that the species pool of a region (i.e. gamma diversity) decreases from the equator to the poles, or from lowlands to highlands?
Well, a few years ago Nathan Kraft and coauthors came out with idea of using null-models to control for the confounding effect of gamma diversity. A null-model, is simply a model that randomizes ecological data to generate specific patterns. With the words of Gotelli and Graves (1996), “Certain elements of the data are held constant and others are allowed to vary stochastically. […] The randomization is designed to produce a pattern that would be expected in the absence of a particular ecological mechanism”. Basically what Kraft et al. did, was to shuffle their real data of occurrence (which species occurs in which site, and with what abundance?) in a way that maintained the species pool of each given region being compared constant, but disrupted all the other mechanisms that lead to patterns of co-occurrence between species. Giraffes and polar bears don’t usually occur together (except in zoos), and for good reasons. In a randomly generated community they could. In short, a (properly designed) null model disrupts the community assembly mechanisms, which is just a complicated way of calling the set of rules determining who lives with whom and why in ecological communities.
Now, in this famous but controversial paper, Kraft et al. concluded that beta-diversity decreases with increasing elevation and latitude just as a consequence of the decreasing size of the species pool. They suggested that there is no need to invoke differences in the mechanisms of local community assembly to explain these patterns. However, some aspects of Kraft et al.’s study were successively criticized and later studies reported contrasting results, reporting that local community assembly mechanisms are actually likely to change along geographical gradients (Mori et al. 2013; Tello et al. 2015).
Clearly more research is necessary, and here we are. Although studies of β-diversity patterns have been frequently limited by the lack of data, nowadays vegetation-plot databases are becoming more available and accessible to ecologists. As these databases contain huge collections of plant data, they allow us to perform studies at unprecedented spatial scales and are ideal for exploring large-scale biogeographical patterns.
To test the hypothesis that the variation in the species pool size is the sole factor responsible for variation in β-diversity patterns along a 1,200 m elevation gradient, we used plant data on as many as 8,795 forest sites in Czech Republic, contained in the Czech National Phytosociological Database. We are talking of data patiently collected by hundreds of researchers during the last decades, and recently collated and standardized in a common repository. Our praise to such an effort!
Using such a large database we could also replicate the study across different subregions, to exclude the possibility of getting idiosyncratic results due to the specific characteristics of a given area. Unfortunately, the data contained in this database were collected without an a priori sampling scheme. That’s why we compared the results obtained using different strategies and resampling schemes to check whether they were ‘robust’, i.e. to extent to which our results were being influence by the geographical bias in the data. We refer the readers interested in these technical details to the publication.
Interestingly, even after controlling for the confounding variation in species pool size using null-models, beta-diversity declined with elevation: observed communities were more diverse than expected at low elevation, and less diverse than expected at high elevation. This means that the magnitude of different local community assembly mechanisms changes along the elevation gradient, and that the gradient of β-diversity was not caused exclusively by the decline in species pool size.
There is a number of possible explanation for this result. At low elevation, for instance, communities may be more strongly shaped by environmental filtering and dispersal limitation than at high elevation, because of higher habitat heterogeneity or wider geographical area. The higher than expected β-diversity at low elevation may also be the results of other mechanisms being stronger in productive, species rich communities at lower elevation, including competition or priority effect (Chase 2010), or depend on a more widespread and heterogeneous effect of human disturbance and forest management.
Whatever the reason, our study provides empirical evidence that the pattern in beta-diversity is real. Communities at higher elevation are less heterogeneous than those at lower elevation, and this does not depend on the fact that beta-diversity is mathematically related to the species pool size. Another jigsaw of the big puzzle of biodiversity distribution is in place. As usual, much more work is needed to understand the hows and whys!
Sabatini, F.M., Jiménez-Alfaro, B., Burrascano, S., Lora, A. & Chytrý, M. 2017. Beta-diversity of Central European forests decreases along an elevational gradient due to the variation in local community assembly processes. Ecography: doi:10.1111/ecog.02809
Originally published on the blog FORESTS and CO
Chase, J.M. 2005. Towards a Really Unified Theory for Metacommunities. Functional Ecology 19: 182-186.
Gotelli, N. & Graves, G. 1996. Null models in ecology Smithsonian Institution Press. Washington, DC, USA.
Kraft, N.J.B., Comita, L.S., Chase, J.M., Sanders, N.J., Swenson, N.G., Crist, T.O., Stegen, J.C., Vellend, M., Boyle, B., Anderson, M.J., Cornell, H.V., Davies, K.F., Freestone, A.L., Inouye, B.D., Harrison, S.P. & Myers, J.A. 2011. Disentangling the Drivers of β Diversity Along Latitudinal and Elevational Gradients. Science 333: 1755-1758.
Mori, A.S., Shiono, T., Koide, D., Kitagawa, R., Ota, A.T. & Mizumachi, E. 2013. Community assembly processes shape an altitudinal gradient of forest biodiversity. Global Ecology and Biogeography 22: 878-888.
Tello, J.S., Myers, J.A., Macía, M.J., Fuentes, A.F., Cayola, L., Arellano, G., Loza, M.I., Torrez, V., Cornejo, M., Miranda, T.B. & Jørgensen, P.M. 2015. Elevational Gradients in β-Diversity Reflect Variation in the Strength of Local Community Assembly Mechanisms across Spatial Scales. Plos One 10: e0121458.