Postdoctoral Researcher in Physics-Informed Machine Learning for Ice Sheet Modeling

Hanover, PA - USA

Monthly Salary: Not Disclosed

Posted on: 12 hours ago

Vacancies: 1 Vacancy

Job Summary

The Ice Future Lab in the Department of Earth Sciences at Dartmouth College invites applications for a postdoctoral researcher with expertise in ice sheet numerical modeling and machine learning. This is a 2-year full-time position with possibility to extend to three years preferably starting April 2026 and is funded by the NSF Collaborations in Artificial Intelligence and Geosciences program in collaboration with the Astera Institute. The position is based at Dartmouth College with primary advising and support provided by the Ice Future Lab.

The project is focusing on developing next-generation physics-informed AI tools to better understand and predict the contribution of the Greenland and Antarctic ice sheets to sea-level rise. Using the open-source PINNICLE framework we fuse multi-sensor remote sensing data into a model-ready living data product that provides physically consistent mesh-free fields of ice geometry flow and basal conditions for both ice sheets. We also use physics-informed neural networks and Bayesian uncertainty quantification to learn new data-constrained parameterizations of basal sliding and iceberg calving at major Greenland outlet glaciers and assess how these processes control future ice loss under different climate scenarios. The work will directly interface with leading ice-sheet models and international intercomparison efforts producing open datasets software and projections that reduce key uncertainties in sea-level rise estimates.

Major Duties/Responsibilities:

Develop and apply physicsinformed neural networks to fuse multi-sensor remote sensing data into a data product for the Greenland and Antarctic.
Design implement and evaluate machine-learned parameterizations of basal sliding and iceberg calving.
Maintain and extend open-source software PINNICLE including reproducible workflows documentation and example notebooks for the broader glaciology and AI communities.
Collaborate closely with an interdisciplinary team contribute to community training activities such as the Glaciology and Machine Learning Summer School.
Presenting research findings at national and international conferences and publishing results in peer-reviewed journals.

Postdoctoral researchers are advised and hosted in the Department of Earth Sciences. They are also supported by the Guarini School for Graduate and Advanced Studies including their community initiatives.

Dartmouth is committed to academic excellence and encourages the open exchange of ideas within a culture of mutual respect. Dartmouth welcomes people with different backgrounds life experiences and perspectives and believes that diversity in all its forms enhances academic excellence. Applicants should address in their cover letter how their research teaching service and/or life experiences prepare them to serve Dartmouths commitment to academic excellence in an environment that is welcoming to all.

A Ph.D. in Earth Sciences Glaciology Climate Science Applied Mathematics Computer Science or a closely related field or ABD with degree received by the start date.
Strong experience in numerical modeling of ice dynamics and/or machine learning methods particularly with neural networks.
Programming experience in Python or other relevant languages.
Proficiency with ice sheet modeling software and experience working with large geospatial datasets are preferred.
Familiarity with remote sensing datasets relevant to glacier and ocean studies as well as skills in data assimilation and model calibration.
Strong track record of relevant publications in peer reviewed journals.
Strong oral and written communication skills.

Required Experience:

Major Duties/Responsibilities:

Develop and apply physicsinformed neural networks to fuse multi-sensor remote sensing data into a data product for the Greenland and Antarctic.
Design implement and evaluate machine-learned parameterizations of basal sliding and iceberg calving.
Maintain and extend open-source software PINNICLE including reproducible workflows documentation and example notebooks for the broader glaciology and AI communities.
Collaborate closely with an interdisciplinary team contribute to community training activities such as the Glaciology and Machine Learning Summer School.
Presenting research findings at national and international conferences and publishing results in peer-reviewed journals.

A Ph.D. in Earth Sciences Glaciology Climate Science Applied Mathematics Computer Science or a closely related field or ABD with degree received by the start date.
Strong experience in numerical modeling of ice dynamics and/or machine learning methods particularly with neural networks.
Programming experience in Python or other relevant languages.
Proficiency with ice sheet modeling software and experience working with large geospatial datasets are preferred.
Familiarity with remote sensing datasets relevant to glacier and ocean studies as well as skills in data assimilation and model calibration.
Strong track record of relevant publications in peer reviewed journals.
Strong oral and written communication skills.

Required Experience: