Skip to main content

Frontiers in invasive species distribution modelling: incorporating human-associations and intraspecific niche structure to improve risk predictions.

Periodic Reporting for period 1 - InvasiveSDM (Frontiers in invasive species distribution modelling: incorporating human-associations and intraspecific niche structure to improve risk predictions.)

Reporting period: 2017-07-15 to 2019-07-14

Biological invasions represent a major component of global change through their impacts on biodiversity, ecosystems and societies. Awareness of biological invasion impacts and the critical importance of evidence-based decision making have led to a persistent effort to understand the factors driving invasion success so as to be able to predict invasion outcomes. To this end, a range of modelling tools has been developed. Among them, species distribution models (SDMs) - correlative techniques using statistical associations between the environment and species occurrences to assess habitat suitability- play a critical role in invasion risk assessments. In this project, two unresolved challenges faced by SDMs when applied to the biological invasion process have be examined: how (1) species’ association with human-modified habitats in native ranges and (2) intraspecific niche variation shape the distribution of invasive species at biogeographical scales and how these effects influence the reliability of predictions of invasion risk.
Explanation of the research work carried

As a first step, we conducted the first global assessment of the relative importance of human-habitat associations in shaping the native distributions of species introduced worldwide, in relation to other key important drivers, that is climate and land-use. For this, we applied deviance partitioning analysis and species distribution models (SDM) to 776 introduced alien bird species from five continents. While an independent effect of climate, and a joint effect of climate and nonurban land uses, appear as major factors governing alien species distribution in their native ranges, significant independent contributions of anthropogenic variables were found for most species. The effect of anthropogenic variables was mostly positive, or concave with highest responses at intermediate values. Notably, human-habitat associations in the native distributions of alien birds were significantly higher than expected, relative to a pool of available species from the same bird families (N = 3,565). Thus, introduced alien birds are a non-random sample with respect to their association with human-altered habitats.

As a second step, we conducted a global assessment of niche conservatism in climatic and human disturbance spaces for bird species showing different responses to humans in their native ranges, and evaluate the effect of considering human variables on the performance of invasion predictions. For this, we compare the native and alien distributions of 158 bird species with established alien populations and enough available data. Invasion predictions built on climatic species distribution models (SDM) that either ignore, or incorporate, human-related variables. Alien bird species tended to invade areas with similar climatic and human disturbance conditions to their native range, but exceptions occur and relate to species responses to human-modified habitats in native ranges. Incorporating information about human-related variables in SDM results in a more accurate prediction of invasion risk, irrespective of species responses to human influence in native ranges.

Finally, we tested whether SDMs accounting for within-taxon niche variation in climate and human-habitat associations in the native range provide better fits and projected distributions for alien bird species than species-level SDMs. We focused on 15 alien bird species, selected because of the availability of robust phylogeographic information. We explored niche variation in climatic and human disturbance spaces occupied by different lineages in the native range and fit SDMs for species and their intra-taxon phylogeographic lineages. We compared the extent of projected distributions and their accuracy with respect to observed occurrence data in alien ranges. Niche analyses revealed significant intraspecific niche structuration within all species considered. Phylogenetic groups within species occupied partially overlapping but distinct portions of the climate and human spaces available in the native range. Notably, models accounting for intra-taxon variation projected larger habitat suitability at a global scale than models conducted at the species level. However, when compared to species occurrences in alien ranges, accounting for within-taxon niche structure did not significantly improve model predictions.


For each of the three work packages of the proposal a main article has been written - one already published in Journal of Applied Ecology. Two additional, more transversal papers have been also published and 3 papers more are currently under review. Main results of the proposal have been also disseminated by attending and participating at internal seminars (1 as invited speaker), as well as one national (British Ornithological Union 2018 Annual Conference, University of Nottingham, UK) and two international conferences (10th International Conference on Biological Invasions. European Group on Biological Invasions
The results of this research bring new understanding of the invasion process and provide a more solid framework to predict, and ultimately to prevent, the impact of invaders. This is in agreement with current EU initiatives, like Horizon 2020, which seek to address the problem of invasive alien species in a comprehensive manner. From a more general point of view, results obtained also allow furthering on the determinants of species distribution in native ranges and how accounting for human-habitat associations and intraspecific niche diversity can help to better understand species distributions.