Phd Position to Take Advance of Nasa’s Pace Mission for Sea Ice Studies

The Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (AWI) is a member of the Helmholtz Association (HGF) and funded by federal and state government. AWI focuses on polar and marine research in a variety of disciplines such as biology, oceanography, geology, geochemistry and geophysics thus allowing multidisciplinary approaches to scientific goals.
- PhD position to take advance of NASA’s PACE mission for sea ice studies (d/f/m)

**Background**
As primary producers, phytoplankton are integral to marine ecosystems and biogeochemical processes. Keeping track of their distribution and population levels is vital for assessing the health of the Arctic/Antarctic marine environment. NASA’s Plankton, Aerosol, Clouds, ocean Ecosystem (PACE) mission signifies a pivotal advancement in hyperspectral remote sensing for monitoring phytoplankton dynamics in polar regions.

This PhD project aims to explore the PACE's capabilities in retrieving near-surface phytoplankton properties in polar regions, specifically chlorophyll-a concentrations and community composition, while enhancing melt pond detection and identifying black carbon on snow and sea ice. Initially, the proposal seeks to enhance the spatial and spectral resolution of hyperspectral data from PACE by utilizing a super-resolution data fusion technique. This approach incorporates corresponding available multispectral satellite imagery with significantly higher spatial resolutions, thereby improving subsequent monitoring analyses using the spatially-spectrally exceptional PACE data. Then, an Integrated Deep Learning Framework, integrating Mixture Density Networks, hybrid 3-D-2-D Convolutional Neural Networks (CNNs), and hybrid Long Short-Term Memory (LSTM)-CNN networks within a meta-learner framework, will be developed to estimate Total Chlorophyll-a levels and Phytoplankton community composition in the Arctic/Antarctic Ocean, utilizing the previously enhanced hyperspectral PACE imagery and in situ measurements. The model will be trained by pairing the satellite and in situ data collected through the AWI expeditions measurements and other available data. Additionally, the approach will assess the model's effectiveness in detecting melt ponds and black carbon on sea ice, with necessary adjustments as required. For this purpose, training samples will be derived from PACE reflectance data classified for sea ice, melt ponds, and open water, supported by high-resolution World-View data. Model validation and classification evaluation will utilize imagery from Sentinel-2 and World-View satellites.

By employing super-resolved hyperspectral PACE data and an integrated deep learning framework, this study seeks to improve the accuracy of phytoplankton biomass estimates, differentiate melt ponds from surrounding ice, and detect black carbon deposits more effectively, serving as a vital tool for their management.

**Your Tasks**
- Evaluate how the proposed ensemble machine learning model compares to traditional radiative transfer models and Empirical Algorithms in accurately retrieving chlorophyll-a concentrations from PACE hyperspectral data in polar regions
- Identify the specific spectral signatures of different phytoplankton community compositions detectable by PACE hyperspectral data, and how can these signatures be leveraged to improve the classification accuracy of phytoplankton diversity in polar regions
- Evaluate to what extent the integration of ensemble machine learning models enhances the retrieval accuracy of phytoplankton biomass compared to traditional algorithms, and assess the implications for monitoring phytoplankton dynamics in response to climate variability
- Evaluate improvements in spatial-spectral resolution achieved through the data fusion super-resolution method, and how these enhancements affect the accuracy of phytoplankton community composition classification compared to heritage algorithms
- Present results at national and international science conferences and team meetings
- Publish results in peer-reviewed scientific journals

**Your Profile**
- PhD in Geophysics, Electrical Engineering, Physics, Oceanography, Climate Sciences or related fields
- Strong background in machine learning and remote sensing
- Experience working with large geophysical data sets
- Excellent programming skills
- Very good English communication skills (C1)

**Further Information**

This position is limited to 3 years. The salary will be paid in accordance with the Collective Agreement for the Public Service of the Federation (Tarifvertrag des öffentlichen Dienstes, TVöD Bund), up to salary level **13 (66 %)**. The place of employment will be **Bremerhaven**.

**The AWI is characterized by**
- our scientific success - excellent research
- collaboration and cooperation - intra-institute, national and international, interdisciplinary
- opportunities to develop - on the job and towards other positions
- an international environment - everyday contact w