Projects
Featured Research
Quantifying Alaska Methane Fluxes via Geostatistical Inverse Modeling
Quantifying methane fluxes in Alaska using aircraft data coupled with a geostatistical inverse model and the STILT atmospheric transport model.
Problem
- Arctic warming at ~2× the global average, amplifying methane feedbacks
- Process-based models underestimate Alaska methane budgets
- Conflicting results on inter-annual variability
Method
- Geostatistical inverse modeling (GIM)
- Bayesian inference & BIC model selection
- STILT transport footprint calculations
- HPC-scalable computation pipelines
Impact
- Reduced regional uncertainty in emission budgets
- Validated against aircraft observations
- Fully reproducible analysis workflow
Python
MATLAB
Inverse Modeling
HPC
Uncertainty Quantification
Publication Under Review
Top-Down Constraints on U.S. Methyl Bromide Emissions After the Montreal Protocol
Estimated U.S. methyl bromide emissions using atmospheric observations and inverse modeling to evaluate Montreal Protocol effectiveness.
Problem
- Strong ozone-depleting compound with uncertain emissions
- Conflicting emission estimates from inventories
- Unclear post-2015 phase-out compliance
Method
- Geostatistical inverse modeling
- NOAA GHG Reference Network data
- STILT transport modeling
- Bayesian inference with REML
Impact
- Post-2015: 55–60% emission reduction
- California: >50% of remaining emissions
- Policy insights on ozone recovery
Atmospheric Chemistry
Inverse Modeling
Bayesian Inference
Policy Evaluation
Manuscript in Preparation
Global Wetland Methane Emissions from Satellite Observations
Spatial and temporal analysis of global wetland methane emissions using satellite observations during 2019–2020.
Python
GEOS-Chem
Satellite Data
Time Series
Evaluation of Global Process-Based Methane Flux Models
Large-scale assessment of wetland methane emissions in Arctic-boreal North America using atmospheric tower observations to evaluate 16 process-based models.
Python
MATLAB
R
STILT
HPC
Publications
Peer-Reviewed
Atmospheric Chemistry and Physics, 26, 1229–1247, 2026. DOI
Journal of Geophysical Research: Atmospheres, 2025. DOI
In Preparation
Manuscripts in Preparation
Quantifying Alaska methane fluxes via geostatistical inverse modeling using aircraft observations
In preparation for submission
No detectable inflection point in U.S. oil and gas methane emissions following the Inflation Reduction Act
In preparation for submission to Nature Climate Change
A decline but no cessation of methyl bromide emissions from the United States over 2007–2018
In preparation for submission
Presentations
Conferences
NASA ABoVE Science Team Meeting
NASA ABoVE Science Team Meeting
NASA ABoVE Science Team Meeting
American Meteorological Society 104th Annual Meeting
American Geophysical Union Fall Meeting
American Geophysical Union Fall Meeting
Experience
Research Experience
May 2022 — Aug 2026
Graduate Research Assistant
Johns Hopkins University · Advisor: Dr. Scot Miller
- Developed a geostatistical inverse model in Python and MATLAB to estimate regional wetland methane emissions, reducing uncertainty in emission budgets by 55% relative to existing process-based models
- Evaluated 16 global process-based wetland-methane models against tower observations across Arctic-boreal North America
- Redesigned large-scale matrix-computation workflows using vectorized operations and sparse data structures, reducing model processing time by 60% and shortening end-to-end analysis timelines by 20% on NASA-funded projects
- Communicated research to audiences of 100+ at AGU, AMS, and NASA ABoVE
May — Sep 2022
Research Data Analyst
University of Florida · Gainesville, FL (Remote)
- Analyzed epidemiological and survey data on medical-marijuana use among older HIV patients with chronic pain; applied statistical methods to evaluate treatment benefits and risks, contributing findings that informed ongoing clinical recommendations
- Designed standardized data-collection protocols and instructional materials, improving data quality and consistency and reducing entry errors across participant cohorts
Aug — Dec 2021
Undergraduate Research Assistant
The George Washington University · Washington, DC
- Collected and analyzed property-level financial data from SEC 10-K filings to study the relationship between climate-risk exposure and cost of capital for U.S. Real Estate Investment Trusts
- Built structured datasets with 22+ variables to support quantitative analysis linking climate risk to financing costs
Jan — Aug 2021
Research Assistant
Peking Union Medical College · Beijing, China
- Led a team of three data analysts conducting case-control and cohort studies on the relationship between serum prolactin levels and pneumothorax risk in Lymphangioleiomyomatosis (LAM) patients
- Analyzed 30+ clinical indicators and identified a statistically significant association that informed diagnostic screening protocols
Funding
Grants & Fellowships
NSF NCAR Graduate Visitor Program (GVP) Fellowship
National Science Foundation / National Center for Atmospheric Research · 2026
NASA Arctic-Boreal Vulnerability Experiment (ABoVE) Research Funding
National Aeronautics and Space Administration · 2023 – 2027
Invited
Invited Talks & Seminars
Introduction to DSIM: Event-Driven Simulation for Power Electronics
Advanced Universal Ventures (AUV) Webinar Series · Virtual · 2026
The Current and Future Application of AI Agents in Engineering
Tsinghua University, Dept. of Electrical Engineering · Beijing, China · 2026
Improved constraints on Alaska methane fluxes using aircraft observations
Johns Hopkins University, Dept. of Environmental Health and Engineering Seminar · 2025
Methane fluxes from Arctic and boreal North America: Comparisons between process-based estimates and atmospheric observations
Johns Hopkins University, Dept. of Environmental Health and Engineering Seminar · 2024
Interested in collaborating?
Open to research collaborations and conversations about climate science and data systems.