A novel multispecies approach for the detection of ecosystem regime shifts – a case study in the North Sea

Abstract. The physical environment both above and below the ocean surface has changed dramatically during the last century. Changes in the marine environment induced by increased release of greenhouse gases and direct exploitation of resources include increased ocean temperature, decreased salinity and pH, and removal of apex-predators. The risk of ecological regime shifts occurring has similarly increased. A variety of methodologies to predict regime shifts have already been used in the North Sea, which has become an important case study for the analysis of regime shifts in a semi-enclosed water body. The North Sea is regarded as a case study in part due to the operation of the continuous plankton recorder, which has provided detailed abundance records of phyto- and zooplankton for over 60 years. Here, we propose a new methodology to calculate regime shift likelihood for every month between 1958 and 2020. This methodology is unique as the model described produces a single time series of regime shift likelihood, using sequential abundance data of more than 300 plankton species. We show the model's ability to identify when regime shifts occurred in the past by comparing it to previous, less automated methodologies. We have validated the model for use in the North Sea by estimating how often false positives and false negatives are generated. Results from the model indicate evidence for three periods of high regime shift likelihood in various parts of the North Sea: between 1962 and 1972, between 1989 until 1999, and between 2002 until 2015. We show that these periods are consistent with previous estimates of North Sea regime shifts, and discuss possible applications of the model's output of a single time series.