Skip to main content

Construct and Collapse: Self-emergence and dynamics under global-change scenarios of mutualistic systems

Periodic Reporting for period 1 - CoCo (Construct and Collapse: Self-emergence and dynamics under global-change scenarios of mutualistic systems)

Reporting period: 2017-07-01 to 2019-06-30

Ongoing global change is driving many species to extinction. Natural ecosystems are composed of many species embedded in complex webs of interaction. Changes in the distributions, phenologies, and abundance of one species may have cascading effects on the rest of the community. This is particular worrying regarding nature's contributions to people, such as pollination. The disruption of pollination interactions is likely to have a large impact in ecosystem biodiversity and food production. But only with a deep understanding of the mechanisms driving such species interactions we will be able to forecast their response capacity to global change. By studying interaction networks we have seen that not all species in the ecosystem have equal importance - some have a more central role in the network, maintaining the general structure and functionality of the community. Therefore, the correct identification of these central species is a useful first step in order to delineate strategic and efficient conservation practices. CoCo's overarching goal was to use a well-studied pollination system (hummingbird-plant interactions) to identify the traits associated with central hummingbird species (e.g. those with highest conservation priority). With that information, we aimed at anticipating species importance in other communities based on species traits.
We have worked with a large database of hummingbird-plant interaction networks, that describe how frequently each hummingbird species visits a specific plant species in a given community. This dataset covers about half of the described hummingbird species, and over 1000 plant species, throughout America. In addition to interaction data, we also have collected data on species morphological traits, abundance, plant phenology or hummingbird migration patterns. Combining all that information with advanced network and statistical analysis, we have been able to identify several patterns in hummingbird-plant interactions. During this fellowship we have strengthen the HummLab, a group of different researchers working on complementary aspects of this dataset in order to obtain a deep understanding of the mechanisms that drive hummingbird-plant interactions across space. With CoCo, we have developed a new analytical framework that enabled us to overcome previous biases in the analysis of mutualistic network structure, and, with it, identified the core species in each community -- that is, species that are more important for the overall community structure and dynamics. Interestingly, we have found differences in the characteristics associated with core hummingbirds across biogeographical areas, illustrating a strong context-dependency and complexity in these interactions, and the necessity for extensive local data.

As a result of CoCo, I have published a manuscript, and have two more in the final stages of preparation. Within the HummLab we have four manuscripts published during this time, one under review, and another two in preparation. I have also presented diverse aspects of this work in seven international scientific conferences, and have focused my outreach activities on young children. I took several nursery and primary classes for tours at the Zoological museum, preparing, in addition to the exhibits, small displays with pollinators and insects. I have also visited two international schools/kindergartens in Copenhagen, where I implemented several games and activities about biodiversity, pollination systems, and hummingbird and bee ecology, and have advised in further nature-related activities.
CoCo has revealed a strong context-dependency of the interaction pattern of hummingbirds, which needs to be accounted for when forecasting species responses to climate change. Our results showcasing such spatial patterns are a first step towards a more comprehensive analysis of pollination systems. Moreover, this analytical framework and findings can be extrapolated to other mutualistic interactions and ecological systems. Our work covering fundamental issues of the analysis of ecological networks and providing clear protocols and tools for the analysis of these systems is already having an impact in the analysis of ecological networks, with new studies incorporating our practices. Furthermore, we have received an invitation to submit a final paper for a special issue, where we will provide with the largest-to-date database on the interactions of a mutualistic system, enabling further analysis from our peers, and a strong support to the open data movement. My mentor and supervision activities have also resulted in successful training of graduate students and preprints of their work. The outreach activities with kids have also proven a high success - follow ups of my school visits showed that the kids learnt about the importance of pollination for food production, and of biodiversity. Being international schools, I put a particular emphasis on how diversity at large (also among people) is important. In the follow up visit kids were excited to see me and shared new thoughts they had after my first visit and the activities they engaged in. I have also been asked to do a new set of visits in the coming months, proving there is interest in my outreach work.