Public Abstract

DE-SCL0000258: EARL: Earth-Atmosphere Agentic Research and Learning -- Physics-Constrained AI Closure Development for Cloud Microphysics and Turbulence

Award Status: Active

Institution: Pacific Northwest National Laboratory (PNNL), Richland, WA
UEI: CWKJEXDG79A7
PM: Nasiri, Shaima

Most Recent Award Date: 08/01/2026
Number of Support Periods: 1
PI: Pressel, Kyle

Current Performance Period: 08/01/2026 - 04/30/2027
Current Project Period: 08/01/2026 - 04/30/2027

Public Abstract

EARL: Earth-Atmosphere Agentic Research and Learning — Physics Constrained AI Closure Development for Cloud Microphysics and Turbulence Pacific Northwest National Laboratory: Kyle Pressel (PI), Laura Fierce Michigan Technological University: Will Cantrell, Raymond Shaw University of Notre Dame: David Richter Vision. This project advances cloud microphysics and precipitation modeling on two fronts. First, we develop a physics-constrained machine-learning (ML) closure for unresolved subgrid-scale (SGS) turbulence effects on cloud-droplet size distribution (DSD) broadening, formulated within two- and three- moment Predicted Particle Properties (P3) microphysics. The closure is trained using direct numerical simulation with super-droplet microphysics benchmarks for Pi Chamber-like conditions and matched P3 box-model simulations, then implemented in the Energy Research and Forecasting (ERF) model. Second, co-developed with the closure, EARL (Earth-Atmosphere Agentic Research and Learning) provides a human-supervised workflow for ERF closure development and agentic experimentation to explain why a given closure succeeds or fails. Together, they deliver a machine-learned representation of the turbulence–microphysics coupling that drives cloud-droplet growth toward rain and a workflow that is extensible across model components. Alignment and Challenge. DOE’s water-for-energy mission depends on accurate forecasts of runoff, reservoir inflow, and cooling-water availability, all of which trace back to precipitation. Two barriers stand in the way. First, existing bulk microphysics schemes omit SGS turbulence-driven DSD broadening that drives droplets to sizes at which collision-coalescence takes over; the result is delayed drizzle onset and errors in rainfall timing and amount. Second, closure development and model experimentation are serial and manual, limiting the range and rigor of candidate evaluation and the pace of growth in process- level understanding. Objectives and AI Advantage. The project pursues two linked objectives, both demonstrating AI advantage. Under AI-for-physics, we develop an ML closure for SGS supersaturation effects on cloud condensational growth in two- and three-moment P3, formulated as a learned residual correction to analytic grid-mean tendencies. Candidate families include physics-constrained neural networks, symbolic regression, and hybrid parametric-plus-learned residual closures. AI advantage is shown by improved DSD-broadening skill and stable ERF integrations. Under AI-for-workflow, we build EARL as a portable, human-supervised agentic workflow that assembles data, generates and screens closures, and stages build–test–run–score cycles, and then expands it role to orchestrate model experimentation and hypothesis generation. AI advantage is quantified by efficiency gains in closure development and experimentation. Combined AI advantage: EARL’s agentic experimentation converts faster closure development into greater process-level understanding, delivering an advantage that neither front achieves alone. Expected Impact. The primary deliverable is a validated SGS condensational-growth closure deployed in the project’s P3 implementation in ERF, reducing DSD-broadening errors that propagate into water- for-energy predictions. Because P3 is also the microphysics scheme in the Simple Cloud-Resolving Energy Exascale Earth System (E3SM) Atmosphere Model (SCREAM) and other E3SM configurations, the closure offers a direct transfer path into DOE’s global model ecosystem. EARL extends the project’s reach further as a portable, human-supervised agentic workflow reusable across ERF model development and experimentation, positioned for Genesis Mission coordination, American Science Cloud readiness, and broader reuse.

EARL: Earth-Atmosphere Agentic Research and Learning — Physics Constrained AI Closure Development for Cloud Microphysics and Turbulence

Pacific Northwest National Laboratory: Kyle Pressel (PI), Laura Fierce

Michigan Technological University: Will Cantrell, Raymond Shaw

University of Notre Dame: David Richter

Vision. This project advances cloud microphysics and precipitation modeling on two fronts. First, we develop a physics-constrained machine-learning (ML) closure for unresolved subgrid-scale (SGS) turbulence effects on cloud-droplet size distribution (DSD) broadening, formulated within two- and three- moment Predicted Particle Properties (P3) microphysics. The closure is trained using direct numerical simulation with super-droplet microphysics benchmarks for Pi Chamber-like conditions and matched P3 box-model simulations, then implemented in the Energy Research and Forecasting (ERF) model. Second, co-developed with the closure, EARL (Earth-Atmosphere Agentic Research and Learning) provides a human-supervised workflow for ERF closure development and agentic experimentation to explain why a given closure succeeds or fails. Together, they deliver a machine-learned representation of the turbulence–microphysics coupling that drives cloud-droplet growth toward rain and a workflow that is extensible across model components.

Alignment and Challenge. DOE’s water-for-energy mission depends on accurate forecasts of runoff, reservoir inflow, and cooling-water availability, all of which trace back to precipitation. Two barriers stand in the way. First, existing bulk microphysics schemes omit SGS turbulence-driven DSD broadening that drives droplets to sizes at which collision-coalescence takes over; the result is delayed drizzle onset and errors in rainfall timing and amount. Second, closure development and model experimentation are serial and manual, limiting the range and rigor of candidate evaluation and the pace of growth in process- level understanding.

Objectives and AI Advantage. The project pursues two linked objectives, both demonstrating AI advantage. Under AI-for-physics, we develop an ML closure for SGS supersaturation effects on cloud condensational growth in two- and three-moment P3, formulated as a learned residual correction to analytic grid-mean tendencies. Candidate families include physics-constrained neural networks, symbolic regression, and hybrid parametric-plus-learned residual closures. AI advantage is shown by improved DSD-broadening skill and stable ERF integrations. Under AI-for-workflow, we build EARL as a portable, human-supervised agentic workflow that assembles data, generates and screens closures, and stages build–test–run–score cycles, and then expands it role to orchestrate model experimentation and hypothesis generation. AI advantage is quantified by efficiency gains in closure development and experimentation. Combined AI advantage: EARL’s agentic experimentation converts faster closure development into greater process-level understanding, delivering an advantage that neither front achieves alone.

Expected Impact. The primary deliverable is a validated SGS condensational-growth closure deployed in the project’s P3 implementation in ERF, reducing DSD-broadening errors that propagate into water- for-energy predictions. Because P3 is also the microphysics scheme in the Simple Cloud-Resolving Energy Exascale Earth System (E3SM) Atmosphere Model (SCREAM) and other E3SM configurations, the closure offers a direct transfer path into DOE’s global model ecosystem. EARL extends the project’s reach further as a portable, human-supervised agentic workflow reusable across ERF model development and experimentation, positioned for Genesis Mission coordination, American Science Cloud readiness, and broader reuse.