I am a fourth year graduate student in Dr. Susan Solomon's group at Massachusetts Institute of Technology. My research interest is in chemistry-climate interactions particularly focusing at the interface between physical oceanography and atmospheric chemistry. I'm also working with Dr. Paul O'Gorman using machine learning to improve subgrid parameterizations in climate models.

Prior, I did my undergraduate study at University of Wisconsin-Madison with double major in Atmospheric & Oceanic Sciences and Applied Mathematics, and a certificate in Computer Science. During my undergraduate study, I was advised by Dr. Tracey Holloway combining satellite, ground measurement and model to study atmospheric chemistry. I also did a summer research with Dr. Caroline Ummenhofer at the Woods Hole Oceanographic Institution (WHOI) to study how volcanic eruptions influence hydroclimate from climate model and tree-ring records.

I am a fan of traveling, nature exploring, photography, and cooking. My favorite place on Earth is Antarctica, since I like cold weather, and that is one reason I choose to live in both Madison and Boston.

CFC-11 ocean inventory and lifetime

Background: Man-made chlorofluorocarbons (CFCs) are the primary cause of the Antarctic ozone hole. CFC-11 is soluble in seawater, and has long been used as a passive tracer to study ocean circulation. However, the effect of the ocean on atmospheric CFC-11 lifetime has been overlooked in the past when anthropogenic emissions were large. Now as anthropogenic emissions have decreased about 50 times in the past decades due to the Montreal Protocol, ocean uptake is becoming more important.

We used a hierarchy of models (from simple box models to MITgcm) to study the ocean uptake of CFC-11 from 1930 to 2300. We found that after 2070s, the ocean will become a net source of CFC-11 emission. The ocean uptake/outgas also introduces a time-dependency to the atmospheric CFC-11 lifetime, which can affect its emission estimation.

Background: Formaldehyde (HCHO) is considered as one of the most hazardous carcinogens in the ambient air. It is also a precursor of ground level ozone that triggers respiratory diseases. There has a limited number of ground-based monitor stations that measure HCHO on a regular basis. There are also several satellite instruments that observe HCHO and has been used to support the assessment of the ozone production regime combined with satellite-derived nitrogen dioxide.

I am focusing on analyzing spatial-temporal patterns of ambient HCHO both from AQS (Air Quality System from EPA) ground-based monitors and OMI (Ozone Monitoring Instrument from NASA) satellite observations in the U.S. over the past 10 years. I also combine temperature and emission profiles to explain possible drivers of the HCHO patterns shown in both monitor and satellite data.

Summer time formaldehyde (HCHO) diurnal cycle from CMAQ model simulation

Composite SST and AMOC anomalies 2-5 years after volcanic eruptions

Background: Volcanic activity exerts a global cooling effect on climate and thus can counteract anthropogenic global warming on short time scales. However, its effect on and interaction with modes of climate variability are not well known. In particular, climate responses in the higher latitudes and remote areas are also less studied due to lack of high-quality instrumental data.

I work on combining the Community Earth System Model Last Millennium Ensemble (CESM LME) and a newly developed tree-ring network across Labrador in Canada to understand North Atlantic climate responses to volcanic eruptions in the last millennium. Particularly, I am looking at how different modes of climate variability change in response to volcanic eruptions spatially and temporally, such as changes in Atlantic Multidecadal Variability (AMV), North Atlantic Oscillation (NAO), Atlantic Meridional Overturning Circulation (AMOC), and lower tropospheric moisture transport, which as a combination change Labrador region precipitation and can be reflected in the tree-ring records.

Background: The understanding toward chemistry-climate interaction is still largely uncertain, particularly due to the complicated direct and indirect effects from aerosols. The 2014-15 Holuhraun eruption in Iceland could provide more insights on how atmospheric chemistry influences climate, given the relative pure environment around Iceland. During this eruption, it emitted about 120 kilo-tonnes of sulfur dioxide (SO2) per day into the atmosphere, which could form sulfate aerosols (SO4) and further influence aerosol-cloud interaction and climate.

The nudged GFDL model simulation AM3 (full chemistry) and AM4 (simple chemistry) have similar SO2 emission rate but totally different responses in SO2 and SO4 formations, leading to different climate responses. I am evaluating AM3 and AM4 with satellite observations from OMI, MODIS and CALIPSO to find what drives different responses in sulfur formation.

2014 Sept. SO2 anomalies

Cartoon demonstration of how AWS works for performing WHIPS

Background: Cloud computing is an emerging technology that allows individual to have the access to shared high performance computing machines remotely. This shortens the time of processing data and makes computationally expensive works in a relatively low cost.

I work on cloud computing in satellite data in atmospheric chemistry, particularly use Amazon Web Services (AWS) to perform WHIPS (Wisconsin Horizontal Interpolation Program for Satellites) to oversample level 2 satellite product to custom-gridded level 3 satellite product. More information of WHIPS could be found at WHIPS website. We have created a public AWS server with pre-installed WHIPS, please see the tutorial guide for running WHIPS on AWS on your own.