Improving the Algal Bloom Prediction on the North Sea by Dual Ensemble Kalman Filter in GEM Model

Increasingly, the phytoplankton blooms (algal blooms) become a public health concern and a worldwide ecological problem. The Generic ecological model (GEM) plays a very important role to assess ecological quality of the Dutch coastal waters and the southern North Sea. Algal blooms can also be predicted by the GEM model mainly through chlorophyll-a concentration. However, GEM contains large uncertainties caused by errors in both the model itself and the input. Data assimilation techniques are being used to reduce these uncertainties and improve the model forecast for chlorophyll-a concentration. In the meantime, data assimilation also provides an efficient approach to estimate the parameters. Several data assimilation schemes based on the dual Ensemble Kalman Filter will be compared with each other in this work. Based on the synthetic measurement, the superior scheme will be carried out to assess the data assimilation result for a longer period. The forecasting result from proposed data assimilation scheme will be compared with deterministic GEM forecasting output from different point of view. Based on the comparison result analysis, the proposed data assimilation scheme is proved to be successfully implemented and satisfactory results are achieved.Applied mathematicsElectrical Engineering, Mathematics and Computer Scienc