Jeff Wen is a PhD student in the Department of Earth System Science. His research interests are broadly focused on applying machine learning to understand the social impacts of climate change and make decisions under climate uncertainty. He was previously an Assembly Fellow at the Berkman Klein Center and MIT Media Lab studying the governance and ethics of AI and formerly a data scientist at Tesla. Jeff holds a Bachelors in Economics from Wharton and a Masters in Environmental Studies from the University of Pennsylvania.
Education & Certifications
Masters, University of Pennsylvania, Environmental Studies (Sustainability)
Bachelor of Science, Wharton School at The University of Pennsylvania, Economics (Operation and Information Management)
Data Science Team Lead, Pixability
Data Scientist, Tesla
Business Analytics Associate, ZS Associates
Lower test scores from wildfire smoke exposure
View details for DOI 10.1038/s41893-022-00956-y
View details for Web of Science ID 000863190700002
Wildfire smoke exposure worsens students' learning outcomes
View details for DOI 10.1038/s41893-022-00958-w
View details for Web of Science ID 000863190700005
Daily Local-Level Estimates of Ambient Wildfire Smoke PM2.5 for the Contiguous US.
Environmental science & technology
Smoke from wildfires is a growing health risk across the US. Understanding the spatial and temporal patterns of such exposure and its population health impacts requires separating smoke-driven pollutants from non-smoke pollutants and a long time series to quantify patterns and measure health impacts. We develop a parsimonious and accurate machine learning model of daily wildfire-driven PM2.5 concentrations using a combination of ground, satellite, and reanalysis data sources that are easy to update. We apply our model across the contiguous US from 2006 to 2020, generating daily estimates of smoke PM2.5 over a 10 km-by-10 km grid and use these data to characterize levels and trends in smoke PM2.5. Smoke contributions to daily PM2.5 concentrations have increased by up to 5 mug/m3 in the Western US over the last decade, reversing decades of policy-driven improvements in overall air quality, with concentrations growing fastest for higher income populations and predominantly Hispanic populations. The number of people in locations with at least 1 day of smoke PM2.5 above 100 mug/m3 per year has increased 27-fold over the last decade, including nearly 25 million people in 2020 alone. Our data set can bolster efforts to comprehensively understand the drivers and societal impacts of trends and extremes in wildfire smoke.
View details for DOI 10.1021/acs.est.2c02934
View details for PubMedID 36134580
Exposures and behavioural responses to wildfire smoke.
Nature human behaviour
Pollution from wildfires constitutes a growing source of poor air quality globally. To protect health, governments largely rely on citizens to limit their own wildfire smoke exposures, but the effectiveness of this strategy is hard to observe. Using data from private pollution sensors, cell phones, social media posts and internet search activity, we find that during large wildfire smoke events, individuals in wealthy locations increasingly search for information about air quality and health protection, stay at home more and are unhappier. Residents of lower-income neighbourhoods exhibit similar patterns in searches for air quality information but not for health protection, spend less time at home and have more muted sentiment responses. During smoke events, indoor particulate matter (PM2.5) concentrations often remain 3-4* above health-based guidelines and vary by 20* between neighbouring households. Our results suggest that policy reliance on self-protection to mitigate smoke health risks will have modest and unequal benefits.
View details for DOI 10.1038/s41562-022-01396-6
View details for PubMedID 35798884
- Wildfire Smoke Plume Segmentation Using Geostationary Satellite Imagery International Conference on Machine Learning (ICML) - Tackling Climate Change with Machine Learning Workshop 2021