Understanding species richness gradients informs projected responses to climate change

Species richness is a common response variable in biogeographic studies that aim to generate fundamental scientific understanding or to apply science to management. Researchers have long explored patterns of species richness and potential natural and anthropogenic drivers of species richness within and among regions. Numerous studies have investigated whether observed species richness is related to environmental gradients such as temperature, precipitation, elevation and primary productivity and to gradients in land uses such as urbanization or intensity of agricultural production. If observed associations are strong, then it may be possible to project future responses of species distributions to alternative changes in land use and climate. Data on patterns of species richness and underlying mechanisms have frequently affected conservation planning, education and marketing by non-governmental organizations and other groups. Butterflies are among the best-known groups of invertebrates in terms of their ecology, and in some regions—most notably Great Britain—their historical distributions and abundance trends. They are popular among biologists and the public in part because they are diurnal, often colourful and relatively easy for trained observers to detect or trap. More recently, the identification of mechanisms by which abiotic factors directly and indirectly affect butterflies has allowed generation of specific, testable predictions related to climate change (Hellmann, 2002; Parmesan, 2003). Projections of the response of butterflies to climate change encompass the effects of altered means and variances in temperature, precipitation and the frequency and magnitude of extreme weather events. Projections also encompass the responses of butterflies to changes in human infrastructure and appropriation of natural resources such as water and primary productivity. In a recent issue of this journal, Hawkins (2010) explored whether tropical niche conservatism explains gradients in species richness of butterflies in seven geographic regions. According to the tropical niche conservatism hypothesis, ancestral inability to tolerate temperatures below freezing prevents Neotropical clades from becoming permanent residents of temperate regions. More derived clades, by contrast, are able to diversify and colonize higher latitudes. Hawkins & DeVries (2009) previously found that patterns of species richness of butterflies in the USA and Canada were consistent with the hypothesis that ancestral Neotropical species have not adapted well to cold temperatures. Here, Hawkins found that tropical niche conservatism is generally consistent with patterns of species richness globally, although drivers of species richness vary somewhat among Mexico, Europe and north-west Africa, trans-Baikal Siberia, Chile, South Africa and Australia. Gradients of species richness of butterflies are widely recognized to be driven directly (physiologically) and indirectly (mediated through the composition or structure of vegetation) by climate. Hawkins (2010) investigated relationships between two response variables, species richness and mean root distance (a measure of phylogenetic diversification) of butterflies, and five environmental covariates: average annual temperature, annual actual evapotranspiration (a measure of aridity), range in temperature, annual global vegetation index (a measure of vegetation greenness that is generally associated with primary productivity) and geographic region (the six listed above plus the USA and Canada). Hawkins found that associations between species richness and environmental covariates varied among regions. Accordingly, species richness cannot be projected for all biogeographic regions on the basis of a single functional relationship. He commented that some regional differences might reflect historical factors that covary with climate. Differences may also result from widely divergent sampling history and methods within and among regions. When data for each region were examined separately, annual actual evapotranspiration had the strongest association with species richness of butterflies in Mexico, Europe and north-west Africa, South Africa and Australia. As aridity increased in those regions, species richness decreased. Species richness of butterflies in the USA, Canada and trans-Baikal Siberia was most closely associated with temperature: as temperature increased, so did species richness. Species richness of butterflies in Chile had a close, positive association with the global vegetation index. In general, the proportion of butterfly species from more derived subfamilies increased from the equator to the poles, and was strongly and negatively correlated with temperature. Thus, temperature appears to have influenced phylogenetic diversification strongly, whereas annual actual evapotranspiration, primary productivity and unknown historical factors had the greatest influence on species richness. Inferences about relationships between species richness and environmental gradients often depend on the resolution and extent of observation and on variation in disturbance regimes (Pickett & White, 1985). The resolution of a global analysis by necessity is coarse (27.5-km to 55-km pixels in this case), and unexplained variation among regions clearly affected relationships between species richness, phylogenetic history and environmental gradients. There appear to have been time lags in compilation of species distributions for different regions. Publication dates of the data sources cited by Hawkins (2010) ranged from 1986 (USA and Canada) to 2007 (Mexico). Furthermore, it is unknown whether drivers of regional patterns in species richness of butterflies may have changed over time. Data on at least one of the environmental covariates (global vegetation index) were compiled 20 years ago. Regional caveats notwithstanding, there was a fairly strong signal that species richness of butterflies was associated with annual actual evapotranspiration. Recent literature suggests that climate change is affecting butterflies in terms of their phenology, the location and size of their geographic ranges and abundance. Most evidence focuses on the responses of population dynamics to temperature, for which records are generally more complete than for precipitation (e.g. Parmesan et al., 1999; Dell et al., 2005). A small number of studies have emphasized relationships between the population dynamics of butterflies and patterns of precipitation (e.g. Roy et al., 2001; McLaughlin et al., 2002). The latter observations are not necessarily linked to deterministic climate change. However, changes in the amount, timing and variability of precipitation are consistent with projections of climate change. Hawkins’ analyses of winter faunas (residents only) versus summer faunas (residents and seasonal migrants) provide some insight into dynamic patterns of species occurrence and associated responses to climate. Not only current latitudinal and elevational distribution but also phylogenetic history may affect the extent to which species can adapt to deterministic shifts in temperature and aridity. For the few species of butterflies that qualify as agricultural pests, or in other taxonomic groups from which non-native invasive species have been derived, it may be worthwhile to explore whether phylogeny is associated with probability of colonization of novel systems. Many apparent associations between butterflies and weather or climate are mediated through larval host plants, nectar sources or structural aspects of vegetation that affect microclimate. A positive association between species richness of butterflies and ‘greenness’ of vegetation suggests that the two taxonomic groups are responding in similar ways to abiotic and biotic gradients. In Hawkins’ work, the global vegetation index covaried strongly with annual actual evapotranspiration. Increasing evidence suggests that relationships between particular species of butterflies and their host plants may be more plastic than traditionally understood. At a local level, the larvae of many species of butterflies are restricted to one or a few closely related species of host plants, and adults of some species are linked closely with certain species of plants from which they derive nectar. Nevertheless, the breadth of host-plant use and preferences sometimes differs dramatically in space, time and even among individuals in the same population (Singer, 1983). Straightforward presentation of associations between species richness, evolutionary relationships and environmental gradients can inform the development of testable predictions of responses to land use and climate change. Regional analyses complement work at finer scales and suggest ways to advance the science of biogeography and its application to practical management. Editor: John Lambshead

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.