Plenty of bugs: Using open ecological data to predict interactions between rodents and their parasites in Eurasia

Background/Question/Methods: Managing and understanding ecological systems requires a
variety of data. Not all of these can be easily measured, either because they
span a large spatial scale, require considerable field work, or are simply too
expensive. One of the solutions to this challenge is to leverage existing data
and models to make predictions about the current and future states of ecological
systems. I present a case study that both generates new ecological knowledge,
and outlines the promises and issues of integrating open ecological data to make
new predictions. This case-study uses rodents and their parasitic fleas in
Eurasia, to show how (i) the development of tools to interact with open
ecological data allows to make predictions at an unprecedented spatial scale,
and (ii) these predictions generate new knowledge, by identifying areas that
should be sampled in priority because they have high uncertainty. Specifically,
this study integrates (i) informations about species interactions, (ii)
taxonomic ranking databases, (iii) point-occurence data, and (iv) reference
bioclimatic predictors.

Results/Conclusions: This analysis generated a map of the expected community
structure of rodents and their fleas in Eurasia, at a fine spatial scale.
Specifically, I identify hotspots of host-parasite interactions (in which hosts
are expected to carry a heavy parasitic load, which has public health
implications), as well as hotspots of taxonomic and phylogenetic diversity,
which can indicate where this system has a long-standing coevolutionary history.
More importantly, it underlines where increased collaborations between data
producers and data consumers is needed. Studies like this are going to become
more frequent in ecology at a large spatial scale, and it is important that the
community collaborates on good practices, and appreciates the promises and
pitfalls as early as possible.