Monitoring and prediction of phytoplankton dynamics in the North Sea

Phytoplankton forms the base of the marine food web, but when concentrations get too high, algal blooms can have adverse effects on ecosystems and aquaculture. Phytoplankton concentrations vary strongly in space and time. However, the nature and drivers of this variability are not yet well understood. For ecological assessments and for early warnings of harmful algal blooms, monitoring strategies are required with sufficiently high temporal and spatial resolution to capture the natural variability of phytoplankton. Novel monitoring methods, such as satellite remote sensing and automated moorings, can acquire information on phytoplankton abundance at a high resolution in space and time. In this thesis, high-resolution data sets from the North Sea are used to investigate phytoplankton variability at different time scales and to identify the mechanisms driving phytoplankton variability. We applied a range of statistical and modelling approaches to analyse data from various traditional and novel monitoring techniques. In this way we demonstrate the potential power of these novel techniques and gained a much improved understanding of phytoplankton variability. Based on this understanding appropriate monitoring strategies can be designed to assess the response of phytoplankton to changes in nutrient inputs and climate change. Furthermore, the improved understanding of drivers of phytoplankton variability supports the set-up and validation of ecological models to predict harmful algal blooms and the productivity of marine ecosystems.