Prediction of changes in summer phytoplankton concentration based on satellite data assimilation methods
DOI:
https://doi.org/10.17308/sait/1995-5499/2024/3/33-49Keywords:
prediction, summer phytoplankton, shallow water, Earth remote sensing, satellite data, numerical experiment, research and forecasting complexAbstract
The article is devoted to the application of mathematical modeling methods as the main tool for researching the water ecosystem functioning, namely, prediction of changes in phytoplankton concentration in shallow water in summer using satellite data. A systematic approach based on the synthesis of mathematical modeling with the remote sensing data assimilation methods made it possible to comprehensively analyze the changes in biogeochemical processes in time and space, taking into account the combined effects of physical and chemical, biological and anthropogenic factors on the researching water ecosystem. The developed mathematical model is correlated with satellite data. It makes possible to predict the behavior of summer phytoplankton in shallow water, changes in the density of isolated plankton populations in accelerated time, describe the redox water processes, sulfate reduction, transformation of biogenic substances (phytoplankton mineral nutrition), research the lock formation as a result of anthropogenic eutrophication, predict changes in the oxygen and biogenic water regimes. A research and forecasting complex, as well as an algorithm for its interaction with GIS were developed and implemented to predict the summer phytoplankton dynamics. The designed tools make it possible not only to assess the natural disaster degree (eutrophication, «blooming», water pollution of various types, etc.), but also to make short-term and medium-term predictions of its development in accelerated time to prevent negative consequences of economic and social character. An operational algorithm for restoring the water quality parameters of the Azov Sea was developed. It based on the Levenberg — Marquardt multidimensional optimization method. The spatial distribution of phytoplankton populations was used as input data. It is the result of applying the method of local binary patterns to satellite images obtained by the authors of this method.
References
Downloads
Published
Issue
Section
License
Условия передачи авторских прав in English













