Detecting potential impacts on ecosystem services related to ecological regime shifts – a matter of wording
(English)Manuscript (preprint) (Other academic)
Assessing which ecosystem services are likely to be affected by ecological regime shifts is one of the greatest challenges of current ecological research. Regime shifts are large, abrupt re-configuration of ecosystem’s structure and function; they are hard to predict, often difficult to reverse, and present potential changes on the benefits people receive from natural systems impacting human well-being. While a bulk of research comprises localized case studies where data is available and experimentation is possible, a global assessment of regime shifts consequences is missing. Here topic mining is presented as a complementary strategy to assess changes in ecosystem services at broader time and spatial scale than direct assessments. We explore an indirect approach by using latent Dirichlet allocation to automatically identify topics that align with ecosystem services, and compared it with the impacts reported by contributors to the Regime Shifts Database. We found that identifying ecosystem services is possible, successful detection is correlated to how well studied regime shifts are. While the majority of supporting and regulating services were successfully detected, pollination and cultural services were elusive. The technique provides a new indirect monitoring method for places where current data is scarce or data collection is challenging.
ecosystem services, topic modelling, regime shifts
Environmental Sciences Ecology Computer Science
Research subject Sustainability Science
IdentifiersURN: urn:nbn:se:su:diva-116893OAI: oai:DiVA.org:su-116893DiVA: diva2:809430
FunderSwedish Research Council Formas, 2009-6966-139149-41