Advanced Aerosol Data Assimilation of the GEO-LEO Constellation

About the role

The Advanced Aerosol Data Assimilation of the GEO-LEO Constellation opportunity is part of the NASA Postdoctoral Program (NPP). This program offers one- to three-year fellowships to scientists engaged in NASA's research projects.

Description

This project focuses on integrating multi-wavelength AOD observations from both geostationary (GEO) and low earth orbit (LEO) satellites to improve the representation of aerosols in the NASA GEOS Earth System Model. The activities involve implementing machine learning transfer learning methods, integrating UV AOD observations, and developing machine learning foundation models.

Responsibilities

• Implement machine learning transfer learning methods to homogenize AOD observations from the geostationary constellation of weather and atmospheric monitoring satellites (e.g. GEOS, HIMAWARI) utilizing well-calibrated AOD observations from low earth orbit (LEO) sensors.
• Integrate and test the impact of UV AOD observations from hyperspectral sensors in LEO and GEO orbits on the aerosol analysis and forecast utilizing the JCSDA-Joint Effort for Data assimilation Integration (JEDI) framework for atmospheric data assimilation.
• Develop machine learning foundation models based on simulated observations from high-fidelity atmospheric simulations for developing downstream tasks such as retrieving aerosol information in the presence of clouds, and utilizing multi-pixel spatial and temporal approaches.

Requirements

The applicant must have a Doctoral Degree. They should also be U.S. Citizens, U.S. Lawful Permanent Residents (LPR), Foreign Nationals eligible for an Exchange Visitor J-1 visa status, or applicants for LPR, asylees, or refugees in the U.S. at the time of application with valid EAD cards and pending I-485 or I-589 forms.

Qualifications

The successful candidate will have expertise in machine learning, aerosol data assimilation, and experience with satellite and ground-based observation data. Familiarity with the NASA GEOS Earth System Model and the JEDI framework is preferred.

Skills

Strong skills in programming languages such as Python, proficiency in data analysis tools like MATLAB or R, and experience with satellite and ground-based observation data are required.

Benefits

The NPP offers unique research opportunities, competitive fellowships, and access to state-of-the-art facilities and equipment. Fellows also receive professional development support and networking opportunities.

Pay

The fellowship provides a stipend and benefits package tailored to the individual's needs and career stage.

Schedule

The fellowship is typically one to three years, depending on the project and the individual's progress.