Research

Our research sits at the nexus of climate change, air pollution, and human health. We are interested in how a changing climate and its extreme events affect human health, and in how to design effective policies to reduce the resulting burden.

The group uses atmospheric chemistry and climate modeling, econometrics and causal inference, machine learning, remote sensing, and epidemiological methods.

Overview of the group's three research themes

Wildfire smoke Exposure, health impacts, and solutions Energy & climate policy Emissions, air quality, and who benefits Climate & air quality How a warming climate reshapes exposure

1. Impacts of wildfire smoke on air quality and human health — and solutions

Wildfire smoke is an emerging health risk and air pollution source around the world. In the US it has reversed decades of air quality progress, and is one of the few major pollution sources still growing. Our research quantifies smoke impacts on air quality at the population scale: how much smoke people are exposed to, what that exposure does to health, and which interventions reduce the burden.

Smoke health burden under future climate change. We project how smoke exposure and the resulting mortality burden grow under future climate change (Qiu et al., Nature, 2025), and translate that into what climate mitigation is worth in lives saved (Qiu et al., PNAS, 2026).
Smoke impacts on ozone exposure. We quantify the impacts of wildfire smoke on ozone, the leading pollutant for air quality standard non-attainment in the US. Smoke ozone follows a pattern distinct from that of smoke PM_2.5, and its mortality burden can reach a comparable magnitude (Li et al., Science Advances, 2026). We have also created a national-scale wildfire ozone dataset at daily 10 km resolution — preprint, data.
Modeling smoke exposure at population scale. Studying the health impacts of wildfire smoke requires accurate exposure data. We contributed to the development of a national-scale smoke PM_2.5 dataset that is now widely used (Childs et al., 2022; 2024). Working at the intersection of machine learning and chemical transport modeling, we are pushing the frontier of smoke exposure estimation — integrating multiple methods and being more transparent about the uncertainty in current estimates (Qiu et al., EST, 2024; Zhou et al., 2026). We also quantify smoke PM_2.5 separately by fire type in California — wildfire, prescribed, and agricultural burning (Kelp, Qiu, et al., 2026).
Health impacts of smoke exposure. Our work has shown that wildfire smoke raises mortality even years after exposure. A main focus is the health impact of long-range smoke transport on places that historically see little smoke: with colleagues at Stony Brook Medicine, we find that smoke from the 2023 Canadian wildfires significantly increased emergency department visits for cardiovascular disease and hypertension (Danesh-Yazdi et al., 2026). We have also quantified the differential respiratory impacts of smoke from wildfire, prescribed, and agricultural fires.

Maps of annual mean smoke PM2.5 over the contiguous US, historical and under three future emissions scenarios — Smoke PM_2.5 rises sharply across the western US under every future emissions scenario *(Qiu et al., Nature, 2025)*.

Components of the US social cost of carbon, with wildfire smoke mortality added — Adding smoke mortality raises the US social cost of carbon by roughly 70% *(Qiu et al., PNAS, 2026)*.

Annual average smoke-attributable ozone across the contiguous US, 2006 to 2023 — Annual average smoke-attributable ozone across the contiguous US, 2006–2023, from our daily 10 km dataset *(Li et al., Science Advances, 2026)*.

2. Effects of environmental, energy, and climate policies on emissions, air quality, and human health

We develop methods that combine causal inference with atmospheric modeling to produce policy insights that neither approach delivers alone, and apply them across the energy transition.

Data-driven approaches to policy evaluation. Combining firm-level data with chemistry modeling produced results markedly different from what process-based projections had predicted for a Chinese energy policy, because baseline energy intensity was already falling and firms responded unevenly (Qiu et al., EST, 2020). Weather is the other confounder: our machine-learning meteorological correction cuts by more than 60% the bias that falsely credits emission reductions for air quality gains (Qiu et al., ACP, 2022).
Distributional benefits of climate policy. Clean energy delivers large aggregate health gains without necessarily reaching those most exposed — wind power does not meaningfully narrow pollution exposure gaps between population groups (Qiu et al., Science Advances, 2022), and US decarbonization pathways raise the same equity questions at national scale (Picciano, Qiu et al., Nature Communications, 2023).
Health and climate benefits across the energy transition. We quantify what specific transitions deliver: electric vehicle usage across Chinese cities (Ma, Qiu et al., Nature Cities, 2026), imported solar photovoltaics in the US (Qiu et al., One Earth, 2025), solar adoption and CO₂ reductions (Biswas et al., Science Advances, 2025), and the cost trajectory of nuclear power in China (Liu et al., Nature, 2025).

Generation, SO2 and NOx displaced per MWh of wind power across US grid regions, with maps of where the avoided emissions occur — Which fuels — and how much SO₂ and NO_x — a MWh of wind displaces depends strongly on the grid region, and the avoided emissions often occur far from the turbines *(Qiu et al., Science Advances, 2022)*.

Regional health and climate benefits of imported solar PV, and county-level health benefits of avoided PM2.5 — Imported solar PV delivers $180/kW in combined health and climate benefits, unevenly distributed across US regions *(Qiu et al., One Earth, 2025)*.

3. Influences of climate change on atmospheric composition and resulting health effects

Climate change threatens air quality through complex pathways, including its influence on anthropogenic and natural emissions, on transport, and on meteorology. Our work aims to quantify the role of air pollution in the climate–health relationship, pointing to adaptation pathways for reducing the health impacts of future climate change.

Climate extremes reshape emissions and air quality. Drought forces the western US electricity system away from hydropower, raising emissions from fossil fuel plants and worsening air quality (Qiu et al., PNAS, 2023).
Compound and interacting exposures. Heat and pollution increasingly arrive together, and we characterize these joint extreme temperature and PM_2.5 events across countries (Chen et al., GeoHealth, 2026).
The human health impacts of extreme events. Health impacts vary widely across countries and contexts (Chu et al., Annual Review of Public Health, 2025), and the consequences reach beyond health itself — tropical cyclones measurably reduce schooling across 13 low- and middle-income countries (Jing et al., PNAS, 2025).

Effect of low runoff on fossil generation across California, the Northwest, and the Southwest — Low-runoff years push fossil generation up sharply in California and the Northwest, with effects spilling across regions *(Qiu et al., PNAS, 2023)*.