COS 65-10 - The inclusion of positive interactions in modeling the niche space of desert annuals

Tuesday, August 8, 2017: 4:40 PM
B113, Oregon Convention Center
Alessandro Filazzola, Department of Biology, York University, Toronto, ON, Canada, Diego A. Sotomayor, Department of Geography, York University, Toronto, Canada and Christopher Lortie, Department of Biology, York University, Toronto, Canada

The niche for plant species is typically defined by their environmental requirements, but this ignores interactions among species. In deserts, positive interactions structure plant communities and are important drivers of biodiversity. Thus, models for the available niche of desert plant species should include these potential positive interactions and associations. We tested the hypothesis that shrubs increase the geographical distribution of desert annuals because through microclimate modification they can provide conditions that match necessary niche requirements. We used the database Global Biodiversity Information Facility to construct MaxEnt species distribution models for annuals with and without reported benefactor species in the deserts of California. We contrasted probabilities of occurrence in climate models with and without reported benefactor shrub species for annuals reported as beneficiaries (i.e. beneficiary) relative to those that have not been reported as facilitated (i.e. unreported). To determine associations with shrubs and annual plant species, we also compared density estimates for each the beneficiary and unreported plant groups.


We found the inclusion of shrubs into models significantly improved model predictability and suitable area for all beneficiary species. However, the inclusion of shrubs did not affect models for the unreported plant species. This suggests that shrubs provide climatic conditions that match the niche characteristics of the beneficiary species, but not the unreported species. We also found that shrubs strongly correlated with beneficiary annual species and slightly correlated with the unreported annual species. Therefore, shrubs and beneficiaries species are geographically associated and could be dependent on positive interactions for their occurrence. We highlight the importance of including facilitation in modeling climate scenarios that currently neglect biotic interactions. Including positive interactions in species distribution modeling is still a relatively novel concept that can enhance the predictability of climate models that estimate species loss.