This study was led by Dr. Qing Yao, with great collaboration with Victoria Lynch, Molei Liu, @wuxiao1993.bsky.social and @robbieparks.bsky.social.

Check out the press release from Columbia Mailman. www.publichealth.columbia.edu/news/hurrica...

As hurricanes are becoming stronger and expanding their reach due to climate change, people need to be prepared for disasters they've never experienced before.

The analysis informed a feature story in the Washington Post earlier this year 👉 www.washingtonpost.com/climate-envi...

This is a great opportunity to join an interdisciplinary team in NYC working on infectious diseases, modeling, and health. The position will remain open until filled. The first round review will begin on November 15th. The starting time is flexible. Please reach out if you have any questions!

This study is led by Dr. Qing Yao, in collaboration with Victoria Lynch, Molei Liu, ‪@wuxiao1993.bsky.social‬, and @robbieparks.bsky.social‬. Great pleasure discussing our findings with Sarah Kaplan and Kevin Crowe at the Washington Post. The preprint has not been peer-reviewed. ‪End/

Within the Helene-affected regions, coastal counties showed stronger evacuation than inland counties, which suffered the most devastating damages and losses. There is an urgent need to increase awareness and develop evacuation plans in places that are unprepared for impending climate threats. 4/

Milton primarily impacted coastal areas and prompted sharp increases in out-region travel prior to landfall and sustained elevated mobility in the post-disaster period. In contrast, Helene affected mostly inland areas, where mobility changes were modest and largely within natural variation. 3/

Using over 300M cellphone foot-traffic records, we analyzed human mobility during Hurricanes Helene and Milton in 2024. We observed marked differences in adaptive mobility responses across locations with varying levels of historical hurricane exposure. 2/

Preprint: www.medrxiv.org/content/10.1...

Model fitting suggests a sublinear increase in force of infection with crowdedness and dwell time. The model can also generate improved short-term predictions in retrospective forecasts, suggesting that aggregated mobility data are sufficient to support high-resolution epidemic forecasting.

We developed a model incorporating place-specific mobility, indoor crowdedness & dwell time, and seasonality of virus transmissibility. This model explained the heterogeneous transmission of SARS-CoV-2 across NYC neighborhoods in 2020.

Using aggregated mobile foot-traffic data, we measured the daily connectivity across 42 NYC neighborhoods in different activities. We also quantified contact patterns using indoor crowdedness and dwell time for each place category.

Daily activities like dining & shopping create opportunities for human contact, facilitating respiratory virus spread. Different contact patterns may lead to differential outcomes across neighborhoods, as seen in NYC. But can we use foot-traffic data to predict epidemics in each neighborhood?