Improved models help predict biodiversity
At present scientists’ capacity to predict ecological communities from a single species is limited. The SESAM-ZOOL (A novel modelling framework for predicting animal communities under global change) project addressed this problem. It improved existing models and developed new ones to provide a better understanding of assembly processes. An innovative framework for modelling species assemblages was developed, implemented and tested. This combined many pre-existing approaches for predicting biodiversity. It produced improved spatially explicit projections that could overcome the limitations of a single method. The SESAM (Spatially-explicit species assemblage modelling) framework was first tested on a highly robust community dataset for plants and animals collected over many years in the Swiss Alps. The model was used to obtain the potential composition of both plant and insect assemblages. Two biodiversity components, species traits and species’ probability of presence, were tested separately. The tests combined individual species distribution models, macroecological models and different approaches to set biotic rules. A second test was conducted on two insect groups (butterflies and grasshoppers) and explored the effect of different techniques for creating presence/absence predictions for individual species. The integration of new ecological assembly rules in the framework was also tested and their performance compared with the ′Probability Ranking′ rule. Use of the framework led to insights regarding the assembly mechanisms of the communities studied and suggested the main routes to follow for future research on community-level modelling. It also highlighted the importance of producing good individual species models in order to reconstruct communities. If the models are not of a high standard they will offer little information about the ecological potential of a species and the framework will be biased. Tests on multiple data sets showed that the SESAM framework can be used to reconstruct community composition. They also confirmed the use of the ′Probability Ranking′ rule in improving community predictions, when the previous bias was detected. The new analytical tools developed by SESAM-ZOOL will be particularly useful to those working in the conservation field, enabling them to explore the possible consequences of biodiversity changes due to human activities. They will also enable the industrial sector to explore ways of mitigating their impacts on the environment.
Keywords
Biodiversity, computer model, communities, SESAM-ZOOL, species assemblage, probability ranking rule, conservation
