Deep Learning for Earth System Modeling Evaluation - Postdoctoral Researcher

Lawrence Livermore National Laboratory (LLNL) is seeking a Postdoctoral Researcher in Deep Learning for Earth System Modeling. The role involves conducting research on AI-based Earth System models and collaborating with a multidisciplinary team to evaluate their performance against traditional models and observational data.ResponsibilitiesConduct research on the ability of Deep Learning Earth System Models (DL-ESMs) to accelerate Earth System scienceApply a set of standard metrics based on DL-ESM outputs, and design, develop and carry out innovative advanced experiments (e.g., storyline analyses, or implementing nudging methods) to evaluate the trustworthiness of DL-ESMs against conventional ESMs and observational datasetsEngage and actively contribute to the international initiative AI-MIP, an effort to define a standard set of experiments for evaluating and benchmarking state-of-the-art DL-ESMsPursue independent research and work closely with colleagues in a multidisciplinary team environment to advance research goalsPrepare comprehensive documentations of findings to guide future usersPublish research results in peer-reviewed scientific or technical journals and present results at external conferences and seminarsTravel as required to coordinate research with collaborators or participate in relevant hackathonsPerform other duties as assignedSkillsPhD in Atmospheric Science, Data Science, or related fieldExperience conducting research in atmospheric science or closely related fieldsAbility to manipulate and analyze large, and complex ESM output datasets, such as those collected in the Coupled Model Intercomparison ProjectProficient programming skills using Python and demonstrated experience with deep learning frameworks (e.g., PyTorch, TensorFlow)Experience using high-performance computing environmentsProficient verbal and written communication skills as evidenced by peer reviewed publications and presentationsAbility to work independently as well as effectively in a collaborative, multidisciplinary team environmentAbility to travel as requiredExperience developing and applying advanced statistical algorithms or machine learning models for one or more of the following applications: weather forecasting, subseasonal-to-seasonal (S2S) prediction, storyline analysis, nudging, green function, or dynamical adjustmentFamiliarity with the analysis of weather extremes, variability across time scales, or the impact of extreme events on infrastructure, natural, or human systemsExperience with one AI-based weather prediction model, for example, NeuralGCM, ACE2, GenCast, WeatherNext 2, is a plusBenefitsFlexible Benefits Package401(k)Relocation AssistanceEducation Reimbursement ProgramFlexible schedules (*depending on project needs)Company OverviewLawrence Livermore National Laboratory, a national security laboratory, provides transformational solutions to national security challenges. It was founded in 1952, and is headquartered in Livermore, California, USA, with a workforce of 5001-10000 employees. Its website is http://www.llnl.gov.



Apply To This Job