Seasonal forecast of ocean mixed layer depth and cold intermediate layer volume in the Baltic Sea

Climate warming has resulted in a steady increase of air temperature over the world ocean. Ocean mixed layer directly interacts with the atmosphere. The Baltic Sea ecosystem is strongly impacted by changes in the mixed layer thickness because of primary production is mainly taking place there. Main aim of the research is to prepare seasonal forecasts of the mixed layer depth and cold intermediate layer volume in the Baltic Sea using transformer neural network models. The data used for the study consists of model reanalysis and seasonal forecast data from the Copernicus Marine Service database, ERA5 climate reanalysis database and ECMWF seasonal forecasting system’s SEAS5 data.

Climate warming has resulted in a steady increase of air temperature over the world ocean. Ocean mixed layer depth is one of the most important quantities of the upper ocean because it defines the quasi-homogeneous surface region that directly interacts with the atmosphere. In the brackish Baltic Sea, permanent vertical stratification of the water column is determined by the location of the halocline at 60-100 m depth. Water temperature plays a significant role in the formation of secondary stratification related to the upper mixed layer in the Baltic Sea. The latter has a profound seasonal cycle due to the annual course of solar radiation. Seasonal thermocline exists at a depth range of 10–30 m from spring to autumn, is strongest in summer and is eroded down to the permanent halocline in autumn due to vertical convection driven by the negative air-sea heat flux. The 20–50 m thick cold intermediate layer is present at a depth of 15–65 m from March to October. The depth of upper mixed layer plays a significant role in the salt flux to the surface layer of the Baltic Sea.

The importance of subseasonal and seasonal weather predictions is continuously increasing worldwide. Machine learning methods are rather extensive, and the topic is evolving very rapidly, which provides a ground for the development of machine learning models suitable for seasonal prediction of the mixed layer depth and cold intermediate layer volume in the seas. Combination of convolutional neural networks, transformer neural networks and long short-term memory models have been shown to be able to predict the characteristics of the Earth system components. Therefore, these methods will be used for the seasonal forecast of mixed layer depth and cold intermediate layer volume of the Baltic Sea.

Responsibilities and (foreseen) tasks

• Processing of the Copernicus Marine Service model reanalysis data, ERA5 climate reanalysis database and ECMWF’s fifth generation seasonal forecasting system’s SEAS5 data. Calculation of spatially mean monthly mixed layer depth and cold intermediate layer volume for different subregions of the Baltic Sea. Analysis of the ERA5 climate reanalysis data for net heat flux, winds and sea surface temperature. Preparation and analysis of the sea ice volume data and river runoff data for the Baltic Sea.
• Implementation, testing and validation of the machine learning clustering algorithms for the categorization of the dynamical regions of the Baltic Sea based on the mixed layer depth and cold intermediate layer volume. Analysis of the climatology of ocean freshwater content of the Baltic Sea subregions and the statistical relationships with global and local atmospheric and oceanographic drivers. Analysis of the SEAS5 seasonal forecast data as drivers of the upper mixed layer and cold intermediate layer volume of the Baltic Sea.
• Selection, testing, validation and application of deep machine learning models for the seasonal prediction of the mixed layer depth in the Baltic Sea subregions. Evaluation of the predictions, improvement and finalisation of the models. Preoperation implementation of the machine learning model for the mixed layer depth and cold intermediate layer volume predictions in the Baltic Sea.

Applicants should fulfil the following requirements:

• a master’s degree in physics, oceanography, data analysis or mathematics
• a clear interest in the topic of the position
• excellent command of English
• strong and demonstrable analytical skills
• capacity to work both as an independent researcher and as part of an international team

(The following experience is beneficial: )

• Basic knowledge of numerical modelling
• Basic knowledge of machine learning
• Programming in Matlab or Phyton
• Some experience of working with big data
• Some experience in HPC

We offer:

• 4-year PhD position in one of the leading research departments in oceanography and related numerical modelling in Estonia
• The chance to do high-level research in cooperation with European research institutes in the framework of the Copernicus Marine Service
• Opportunities for conference visits, research stays and networking with leading universities and research  centers in the fields of oceanography

About the research group

Research Group on Modelling and Remote Sensing of Marine Dynamics, Tallinn University of Technology, School of Science, Department of Marine Systems.

The research group is conducting oceanographic process research based on scientific analysis to find cause-and-effect relationships. Innovative (operational) methods for monitoring the marine environment and analyzing changes are being developed, incl. weather forecasting and climate models applied to supercomputers, to elucidate the mechanisms of atmospheric and ocean interactions; and machine learning based algorithms for satellite image processing and model data analysis. The research group has a long experience in developing applications / methods of operational oceanography, the outputs of which are information provided to the public and public authorities on water level variability, ice conditions and other parameters of marine physics. The research group is making a significant contribution to the pan-European Copernicus program. In scientific process research and applied research, the strength of the research team is the use of big data (mass processing) for climate studies and statistical analysis of the properties of the marine environment, as well as for finding dynamic relationships.

(Additional information)

For further information, please contact Prof. Urmas Raudsepp,urmas.raudsepp@taltech.ee