Search NASA⌕ Search
Home/Archive/NASA NTRS

NASA NTRS · 20205008643

Ensemble-Based Deep Learning for Estimating PM2.5 over California with Multisource Big Data Including Wildfire Smoke

Lianfa Li·Mariam Girguis·Frederick Lurmann·Nathan Pavlovic·Crystal McClure·Meredith Franklin·Jun Wu·Luke D. Oman·Carrie Breton·Frank Gilliland·Rima Habre
View original source ↗Export metadata ↓

Abstract

Estimating PM2.5 concentrations and their prediction uncertainties at a high spatiotemporal resolution is important for air pollution health effect studies. This is particularly challenging for California, which has high variability in natural (e.g. wildfires, dust) and anthropogenic emissions, meteorology, topography (e.g. desert surfaces, mountains, snow cover) and land use.

Explore related subjects

air qualityPM2.5wildfiremachine learningair pollution exposurewildfiresremote sensingcaliforniahigh spatiotemporal resolution

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Lianfa Li, Mariam Girguis, Frederick Lurmann, Nathan Pavlovic, Crystal McClure, Meredith Franklin, Jun Wu, Luke D. Oman, Carrie Breton, Frank Gilliland, Rima Habre. 2020-09-24. Ensemble-Based Deep Learning for Estimating PM2.5 over California with Multisource Big Data Including Wildfire Smoke. https://ntrs.nasa.gov/citations/20205008643

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2025

Modeling the Air Quality Impacts of Future Energy Scenarios

Many greenhouse gas (GHG) emission reduction measures achieve simultaneous reductions in air pollutants. Human-Earth system models can estimate such emission changes in the energy system but using them in chemistry-transport models (CTMs) to study their air quality impacts involves resource-intensive emissions processing. This is greatly simplified by an emissions scaling approach linking state-level emissions estimated by a human-Earth system model to a CTM. A scenario continuing pre-2022 energy policy in the U.S. to 2050 shows widespread air quality improvements over the 2015 baseline from SO 2 and NO x emission reductions of 50–80% from electricity generation and light-duty vehicles. Scenarios of GHG mitigation and vehicle electrification at the state and national level add further benefits. However, PM 2.5 increases from increased use of wood heating and bioenergy suggest that additional PM 2.5 management may be needed when using biofuels. In conclusion, this approach helps assess multiple future energy scenarios efficiently without sacrificing chemical detail in the air quality simulations.

air quality↗
NASA NTRS2024

Overview of the International Space Station’s Water and Cabin Air Quality: A Five-Year Status

Since the beginning of the International Space Station (ISS), water and air quality have been monitored to ensure crew health and verify the performance of the regenerative Environmental Control and Life Support (ECLS) systems. Over the last 25 years, the ISS has evolved greatly with significant changes to operations, crew complement sizes, visiting vehicles, payloads, and upgrades within the regenerative hardware, seen through Technology Demonstration integrations. In particular, better assessment and prevention of volatile organic releases from payloads and crew hygiene products, and implementation of advanced sorbents both on the air and water strings have been successful in reducing contaminant loads. Data on air and water quality for the last five years on ISS will be presented (nominal and contingency air grab samples, in-flight monitoring for air and water quality, and water samples from all segments of the ISS water system), including some notable events. The available data demonstrate the performance of existing ECLS systems and overall status of how the approach to air and water quality have evolved through the new ISS architecture baseline operations.

air quality↗
NASA NTRS2023

Public Health Data Applications Using the CDC Tracking Network: Augmenting Environmental Hazard Information with Lower-latency NASA Data

Exposure to environmental hazards is an important determinant of health, and the frequency and severity of exposures is expected to be impacted by climate change. Through a partnership with the U.S. National Aeronautics and Space Administration, the U.S. Centers for Disease Control and Prevention’s National Environmental Public Health Tracking Network is integrating timely observations and model data of priority environmental hazards into its publicly accessible Data Explorer (https://ephtracking.cdc.gov/DataExplorer/). Newly integrated datasets over the contiguous U.S. (CONUS) include: daily 5-day forecasts of air quality based on the Goddard Earth Observing System Composition Forecast (GEOS-CF), daily historical (1980-present) concentrations of speciated PM2.5 based on the Modern Era Retrospective analysis for Research and Applications, version 2 (MERRA-2), and Moderate Resolution Imaging Spectroradiometer (MODIS) daily near real-time maps of flooding (MCDWD). Data integrated into the CDC Tracking Network are broadly intended to improve community health through action by informing both research and early warning activities, including (1) describing temporal and spatial trends in disease and potential environmental exposures, (2) identifying populations most affected, (3) generating hypotheses about associations between health and environmental exposures, and (4) developing, guiding, and assessing environmental public health policies and interventions aimed at reducing or eliminating health outcomes associated with environmental factors.

air quality↗