Climate change effects on marine protected areas: Projected decline of benthic species in the North Sea. (January 2021)

• Record Type: Journal Article
• Authors: Weinert, Michael; Mathis, Moritz; Kröncke, Ingrid; Pohlmann, Thomas; Reiss, Henning
• Is Part Of: Marine environmental research. Volume 163(2021)
• Publication Date: 2021-01
• Subjects: Benthos -- Ecosystem management -- Climate change -- Species distribution models -- biomod2 -- Projections -- Ocean warming -- Benthic indicators -- Macrofauna; Marine pollution -- Environmental aspects -- Periodicals; Marine ecology -- Periodicals; Mer -- Pollution -- Aspect de l'environnement -- Périodiques; Écologie marine -- Périodiques; Electronic journals; 577.705
• DOI: 10.1016/j.marenvres.2020.105230
• Languages: English
• ISSNs: 0141-1136

Abstract

Abstract: Climate change is a global threat for marine ecosystems, their biodiversity and consequently ecosystem services. In the marine realm, marine protected areas (MPAs) were designated to counteract regional pressures, but they might be ineffective to protect vulnerable species and habitats, if their distribution is affected by global climate change. We used six Species Distribution Models (GLM, MARS, FDA, RF, GBM, MAXENT) to project changes in the distribution of eight benthic indicator and key species under climate change in the North Sea MPAs for 2050 and 2099. The projected distribution area of most species will be stable or even increase within the MPAs between 2001 and 2050. Thereafter, the distribution area decreased, especially within MPAs in the central North Sea by 2099, and some key species even disappeared from the MPAs. Consequently, the monitoring and protection of benthic species might not be possible within static MPA borders under climate change. Highlights: 1. Distributional changes will be most pronounced for the second half of the century 2. MPAs in the North Sea might not fulfil their objectives in the far future 3. Machine learning and regression-based methods showed high performance variation 4. Consensus models were an appropriate approach to reduce this variation