About this job

The Research Fellow will join the remote sensing team within the NWP department at ZAMG. The candidate will develop a severe weather nowcasting prototype application based on a fused data approach. With MTG FCI and LI (and precursor data until MTG data are available) together with other data sources, notably NWCSAF, radar and NWP, a machine learning based algorithm will be developed to predict severe weather. Data mining techniques will be applied to determine the parameters which contribute most to the nowcasting algorithm which will be based on a deep learning technique. The ultimate goal is to identify convective cells at an early stage, follow their development and estimate the severity level they may reach.

The candidate will work closely with experts on NWCSAF, radar processing, machine learning and remote sensing at ZAMG.

Responsibilities

• Preparation of datasets from satellite data (MTG and precursor), NWCSAF, radar and NWP, which will form the basis for the development of the nowcasting algorithm;
• Implementation of data mining techniques (to find the most relevant parameters) and machine learning algorithms (to predict severe weather);
• Statistical evaluation of the model, optimisation and tuning of the algorithms;
• Publication and presentation of results (conferences, meetings, journals)

Profile

• The Fellow should have a university degree in Physics, Mathematics, Meteorology, Remote Sensing, or equivalent
• Candidates must be able to work effectively in English, both verbally and written.
• Experience in satellite data analysis and/or machine learning is desirable.
• Good knowledge of scientific programming (C/C++, Python) and experience with statistical packages (e.g XGBoost, Keras, scikit-learn etc.) are significant assets.
• Experience in working with large data sets applying statistical methods is desirable.
• The ability to work as part of a team, a high level of commitment and interest in continuing education/training is required.
• The Fellow should have a university degree in Physics, Mathematics, Meteorology, Remote Sensing, or equivalent, and relevant research experience, including PhD or at the level thereof.
• Candidates must be able to work effectively in English, both verbally and   written. Good working knowledge of the German language (spoken and written) is an advantage.