Workshop on Time-Series Data relevant to Eutrophication Ecological Quality Objectives (WKEUT)

The Workshop on Time Series Data relevant to Eutrophication Ecological Quality Objectives [WKEUT] co-chaired by Ted Smayda, USA, and Gunni Ærtebjerg, Denmark, met on 11–14 September 2006 at Sankt Helene, Tisvildeleje, Denmark. Peter Henriksen, Denmark, served as Rapporteur. OSPAR originally co-sponsored the WKEUT Workshop, but then withdrew. WKEUT discussed 17 long-term phytoplankton data sets available from European and relevant North American coastal sites. The criteria used to select the time series for ecological comparison were (1) the data set was to be minimally 10 years in duration, and the sampling frequency of the physical, nutrient and phytoplankton parameters adequate for workshop objectives; (2) the time series habitats and associated phytoplankton processes were to be representative of the different European coastal water environments found; (3) western Atlantic sites of equivalent ecological value were to be included to allow evaluation of possible trans-Atlantic basin similarities in the trends observed at the selected European sites, particularly with regard to climate-driven commonalities. In total, 14 European and 3 North American sites were selected for ecological comparison. The 3 North American sites selected were the Bay of Fundy (Canada), Narragansett Bay (Rhode Island) and the Tampa Bay – Charlotte Bay complex (Florida). European sites included Irish coastal waters, Stonehaven (Scotland), Floedevigen and Gullmar Fjord (Skagerrak), the Kattegat – Oresund – Belt Sea complex, Sylt, Helgoland, the German and Dutch Wadden Sea ecosystem, Belgian coastal waters, Iberian coast and Thau Lagoon in French Mediterranean waters. This selection provided time series data that allowed a comparative analysis of phytoplankton dynamics in response to nutrification and weather-driven changes (proxied by the North Atlantic Oscillation Index) along a latitudinal habitat gradient in European coastal waters that extended from Ireland to the French Mediterranean. The time series sites group into four general habitats: (1) Large open coastal systems – the Skagerrak, Kattegat; Belgian, Dutch and German coastal waters in the southern North Sea (2) Fjord-like or well-mixed shelf waters – Bay of Fundy, Irish coastal waters, Stonehaven, Spanish rias; (3) Shallow systems – the Dutch and German Wadden Sea, Thau Lagoon; (4) Aquacultural sites – Irish coastal waters, Spanish rias, Thau Lagoon, Bay of Fundy.Narragansett Bay is a coastal estuary which does not have a close counterpart in the European sites selected, but has habitat features and phytoplankton responses that overlap with groups 1,2 and 3 listed above. Within the habitat groupings, the Wadden Sea ecosystem is under heavy riverine influence, while the Skagerrak and Kattegat systems are open to North Sea and Baltic watermass intrusions, i.e. to farfield effects of Wadden Sea, Southern Bight of the North Sea and Baltic nutrient loading. The workshop first focused on presentations of the long-term patterns and trends in physical features, nutrients and phytoplankton behavior at the time series locations selected for ecological comparison. This was followed by three invited talks: 1) on the need to evaluate the role of irradiance as a factor regulating the response of phytoplankton to elevated nutrient levels; 2). the influence of time series duration on the detection of the effects of long-term changes in nutrients and climate change on phytoplankton behavior; and 3). the role of modelling in time series analysis, with emphasis on the fact that statistical analyses may reveal parallel trends, but do not explain the underlying mechanisms which are more tractable by mechanistic modelling approaches. The Abstracts and the descriptions of the time series data sets considered are given as Annexes 6 and 7.The time series descriptions and invited lectures led to various conclusions that influenced the working group responses to the Workshop Terms of Reference (Annex 1) and Workshop Questions: (1) trend analyses should be supported by statistical analyses; (2) the techniques used in time series analyses are not standarized, but should vary with the intended use of the analyses, e.g. correlation, prediction, etc.; (3) time series analyses are vulnerable to interpretive error if the sampling frequency, or length of the time series, does not reflect the system components and dynamics; (4) an interaction between statistical analyses and modellers is required, and time series analysis can help to calibrate models.The papers presented at the Workshop will be published in a special issue of the Journal of Sea Research kindly being made available by Dr.Katja Philippart, Chief Editor.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.