Rare species can be mapped more accurately with new statistical approach

 

Information on Earth's biodiversity is increasingly collected using DNA-, image- and audio-based sampling. At the same time, new statistical analysis methods are being developed to make more out of the collected data, providing detailed information on Earth's biodiversity and its status. Now, an international research group has developed a new statistical approach for jointly modeling the distributions of up to millions of species. The new approach improves predictions especially on rare species that have remained difficult to model with earlier approaches.

Earth is home to several millions of species, and only a fraction of nature's diversity is currently known. One reason for this is that many rare species are difficult to detect, which means that we may not even know about their existence, or at least that there is a lack of information available to evaluate their endangerment status. This can lead to inaccurate and incomplete assessments of their abundance and the reasons for their endangerment.

"It is important to gather information on the distribution and habitat requirements of rare species to understand the current state and changes in biodiversity. This requires models that can, for example, accurately forecast how climate factors or land use changes will affect species of which we have only a few observations," says Academy Professor Otso Ovaskainen from the University of Jyväskylä.

Statisticalmodeling advances will improve predictions on biodiversity change

Researchers at the University of Jyväskylä, in collaboration with international research groups, have now developed a new statistical method that enables the analysis of data sets comprising millions of species.

"We have developed the CORAL method, which enables comprehensive modeling and prediction of biodiversity. We illustrated that CORAL leads to much improved prediction and inference in the context of DNA metabarcoding data from Madagascar, comprising 255,188 arthropod species detected in 2874 samples," clarifies Professor David Dunson from Duke University.

By borrowing information from a backbone model of common species, CORAL makes it possible to model even the rarest species in a statistically effective manner by combining an informative prior model with the limited data available for each rare species.

"In our sample analysis of Madagascar, we were able to understand the climatic and evolutionary factors affecting the occurrence of species and seasonal variation. This allows us to produce more accurate information on biodiversity, particularly for rare species, for both decision-makers and researchers," says Professor Brian Fisher from the California Academy of Sciences who led the data sampling in Madagascar.

📌 Visit Us 👇
🌐 Website: https://statisticsaward.com/
🏆 Nomination: https://statisticsaward.com/award-nomination/?ecategory=Awards&rcategory=Awardee
📝 Registration: https://statisticsaward.com/award-registration/