Generalized dissimilarity models as a tool to predict Odonata assemblages species composition

Central Iran is considered one of the most water-stressed regions of the Central Plateau of Iran. In recentyears a combination of climate change, population growth, and economic development is placing enormousstress on freshwater ecosystems of this area. Thus the quality and quantity of freshwater habitats are decliningand freshwater biodiversity is threatened. However, information about the diversity patterns of freshwaterspecies and drivers of these patterns is scarce. Therefore assessing the effects of anthropogenic environmentalalterations would be challenging. One way to overcome this challenge is to use statistical and modellingtechniques. By using these techniques we can estimate species assemblages across a less explored region byextrapolating the results of a few surveys conducted in that region. Results of such analysis will have the bestchance of being effectively used in conservation decision-making if it is done for flagship taxa like Odonata.Generalized dissimilarity modelling (GDM) is a statistical approach that uses the pairwise dissimilarity ofsurveyed localities to fit a model predicting the biological distance between any given pair of localities as amultivariate non-linear function of differences in the environmental characteristics of those localities. In thisstudy, we used data of ۴۱ water bodies surveyed regarding Odonata species in Central Iran (Qom and Esfahanprovinces). We perform GDM using the R package gdm to extrapolate species assemblage composition inareas where the species composition data was not available. GDM provided a map that clusters the predictedcommunities into a discrete set of communities with a common profile. It shows how Odonata communitiesare structured along the altitudinal gradient of Central Iran in terms of their species composition. We can usethis map for priority conservation planning and to suggest the best distribution of potential locations forconducting future field surveys.

کلیدواژه ها:

Arid and semi-arid ، Dragonflies ، GDM ، Extrapolation ، Ecological Modelling