Postdoctoral Researcher

Postdoctoral Researcher Opportunity The Terra D2I (Data To Insights) lab at the CUNY New York City College of Technology, led by Dr. Viviana Acquaviva, is seeking a highly motivated postdoctoral researcher to join a growing research group for the project "From sparse data to full spatio-temporal fields: surface ocean carbon and beyond", sponsored by the Simons Foundation. Project Overview This project aims to reconstruct the global surface ocean pCO2 field, starting from observations that are extremely sparse in space and time. Because of data sparsity, the reconstruction of the full field relies on additional information that can be measured from satellites, such as the temperature and salinity of the ocean. These become the features of a machine learning model that is trained to predict pCO2 using the available observations as a learning set. The predictions for the ML model are then used for "infilling" or reconstructing the full pCO2 field, which serves to estimate the global ocean carbon sink. This is a naturally difficult problem for ML methods, because there is an unsolvable distribution shift between the training domain (where observations are available) and the application domain (all other points in space and time). The project's objective is to improve this reconstruction, making it more accurate and robust. The tools that we use include classical statistics, Bayesian parameter inference, and machine learning. We collaborate with a broad community of researchers, from statisticians to physical oceanographers to climate modelers to cosmologists.

Tests of Generalization - The predicted global pCO2 field derived from the infilling is a crucial input for the Global Carbon Budget, but we can't test its accuracy directly. We use Earth System Models (ESMs) and Global Ocean Biogeochemistry Models (GOBMs) as testbeds to better understand the reconstruction process and to build resilience into our representation and ML modeling above. Testing ESMs and GOBMs: Through our work on optimal representation, we also plan to develop custom metrics to assess how well the relationship between feature variables and pCO2 is captured in the models, compared to the observations. Other related duties as assigned. The lab also anticipates hiring a post-baccalaureate researcher in Fall/Winter 2025 and a Ph. D. student with starting date in Spring or Fall 2026 to work on related projects. The postdoc will participate in co-mentoring at least one junior researcher and will have many opportunities for further professional development, decided together with the PI and according to their professional goals and interests. Additional responsibilities include occasional travel (once or twice a year) to conferences and workshops to present research.