Overview
CISESS at the University of Maryland is seeking a Remote Sensing Scientist to develop advanced retrieval algorithms for precipitation and cloud properties using satellite observations across multiple spectral bands. This mid-level position offers the opportunity to work on cutting-edge machine learning applications in atmospheric science while collaborating with leading agencies including NOAA and NASA. The role is initially for one year with potential for renewal.
Job Description
CISESS at the University of Maryland seeks a scientist to develop retrievals of precipitation and cloud properties from satellite observations across passive microwave, infrared, and visible spectrum. The group works on characterizing uncertainty and validating against ground-based and spaceborne references. Active research directions include machine learning approaches to precipitation rate and type retrieval, quantification of retrieval uncertainty, and estimation of quantities that satellites do not observe directly. The successful candidate will work within a team collaborating closely with NOAA, NASA, and partner institutions. Position initially for one year with possibility of renewal.
Primary Responsibilities:
- Develop, train, and evaluate models for geophysical retrieval from satellite observations
- Construct and maintain collocated datasets spanning multiple satellite sensors and ground-based references
- Validate retrievals against ground and spaceborne sensors and characterize retrieval uncertainty
Required: Ph.D. in Atmospheric Science, Hydrology, Environmental Science, or closely related field; or M.S. in Computer Science with demonstrated ML application to scientific problems. Strong Python programming. Practical experience training ML models (PyTorch or TensorFlow). U.S. Citizenship or Green Card holder.
Preferred: Expertise in satellite observation of the troposphere across passive microwave, infrared, and visible spectrum.