Postdoctoral Scholar-Global AI/ML Modeling and Scenario Generation

Penn State; Pennsylvania; Presencial; USD 55.000 - 75.000; Jornada completa

Descripción de la vacante

The Pennsylvania State University invites applications for a full-time Postdoctoral Scholar to advance global AI/ML Earth system emulators and regional downscaling for large ensembles of extreme weather scenarios. You will work with a diverse team across Earth and energy systems to inform water and energy risk assessments.

Required qualifications include a Ph.D. in a related field, strong Python and ML skills, and experience with large datasets and HPC.

Formación

• Ph.D. in a closely related field conferred prior to start date.
• Demonstrated experience with scientific Python (NumPy, SciPy) and ML methods.
• Experience with large climate/weather datasets and HPC environments.
• Strong written and verbal communication skills; ability to collaborate across institutions.

Responsabilidades

• Deploy and analyze outputs from global AI/ML weather and climate emulators.
• Apply diagnostics to evaluate emulator fidelity and compare with physics-based benchmarks.
• Develop downscaling workflows and distill event archetypes for scenario libraries.
• Collaborate with multi-institution partners and contribute to publications.

Conocimientos

Scientific Python programming

ML techniques (clustering, etc.)

Large climate/weather datasets

Strong communication

Multi-institution collaboration

Educación

Ph.D. in atmospheric science / Earth science / CS / physics / applied math

Herramientas

NumPy

Xarray

Pandas

Scikit-learn

UNIX

HPC

Descripción del empleo

Postdoctoral Scholar-Global AI/ML Modeling and Scenario Generation### Postdoctoral Scholar-Global AI/ML Modeling and Scenario GenerationApply: Penn State University Park: Full time: Posted Yesterday: REQ\\_0000084025# **APPLICATION INSTRUCTIONS:*** ## CURRENT PENN STATE EMPLOYEE (faculty, staff, technical service, or student), please login to Workday to complete the internal application process. Please do not apply here, apply internally through Workday.* ## CURRENT PENN STATE STUDENT (not employed previously at the university) and seeking employment with Penn State, please login to Workday to complete the student application process. Please do not apply here, apply internally through Workday.* ## If you are NOT a current employee or student, please click “Apply” and complete the application process for external applicants. **Approval of remote and hybrid work is not guaranteed regardless of work location. For additional information on remote work at Penn State, see Notice to Out of State Applicants.**This is a term position; length of the term will be discussed during the interview process. Continuation past the term length discussed will be based on university need, performance, and/or availability of funding. **POSITION SPECIFICS**The Department of Earth and Mineral Sciences at The Pennsylvania State University invites applications for a full-time **Postdoctoral Scholar** to support the DOE-funded HyperFORMANCE project. The successful candidate will work under the supervision of Prof. Colin Zarzycki and contribute to a multi-institutional team. The postdoc will play a central role in developing and applying global AI/ML Earth system emulators and regional downscaling techniques to generate large ensembles of physically credible extreme weather scenarios for water and energy system risk assessment. **RESEARCH FOCUS** The postdoc's work will include: - Evaluate where global AI/ML emulators (ACE-E3SM, FourCastNet, NeuralGCM) credibly reproduce extreme weather characteristics and where physics-based models remain necessary - Combine physics-based simulations, existing model archives, and certified AI/ML emulation to build libraries of decision-relevant extreme events - Develop and apply machine learning clustering techniques to distill large event libraries into representative scenario sets and event archetypes - Generate large ensembles of spatially coherent, temporally compounded hazard scenarios at testbed-relevant scales - Work with hydrologic, energy, and urban impacts teams to ensure scenarios are suitable for downstream applications **KEY RESPONSIBILITIES** AI/ML Model Application and Evaluation - Deploy and analyze outputs from global AI/ML weather and climate emulators, including multi-decadal free-running simulations - Apply process-oriented diagnostics (e.g., TempestExtremes) to evaluate emulator fidelity for extreme precipitation, atmospheric rivers, tropical cyclones, and extratropical cyclones - Compare AI emulator outputs against physics-based benchmarks (E3SM, SCREAM) and observations - Apply regional AI/ML downscaling tools (CorrDiff, StormCast-WUS) to generate km-scale meteorological fields suitable for impacts modeling - Contribute to development and refinement of downscaling workflows across multiple testbed regions - Identify representative event archetypes spanning the distribution of extreme outcomes - Contribute to peer-reviewed publications documenting emulator evaluation, downscaling fidelity, and scenario methodology - Present results at scientific conferences (AGU, AMS) - Collaborate with multi-institutional partners on scenario development and evaluation - Participate in project workshops and cross-team coordination meetings **REQUIRED QUALIFICATIONS** - Ph.D. in atmospheric science, Earth science, computer science, physics, applied mathematics, or closely related field, conferred prior to the start date - Demonstrated experience with scientific Python programming (NumPy, Xarray, Pandas, Scikit-learn) - Experience with machine learning techniques (clustering, classification, dimensionality reduction, or emulation) - Familiarity with large climate or weather datasets and analysis of extreme events - Experience in UNIX and high-performance computing environments - Strong written and verbal communication skills - Ability to work collaboratively in a multi-institutional team environment **PREFERRED QUALIFICATIONS** - Experience with physics-based and/or data-driven climate/weather models (E3SM, CESM, SCREAM, ACE2, FourCastNet, NeuralGCM, StormCast, or similar) - Familiarity with feature-based extreme event detection and tracking - Knowledge of tropical cyclones, atmospheric rivers, mid-latitude storms and/or other synoptic-scale weather extremes - Interest in uncertainty quantification and ensemble modeling **APPLICATION MATERIALS** Applicants should submit the following materials: - Current curriculum vitae - Detailed statement of research interests and relevant experience (1-2 pages), addressing how your background aligns with the postdoc responsibilities - Contact information for three professional references **BACKGROUND CHECKS/CLEARANCES**Employment with the University will require successful completion of background check(s) in accordance with University policies.