Clearing the Air: Aerosol Estimates in the Coastal United States

Category: Assessing Air Quality & Water Resources
Project Team: Coastal US Health & Air Quality
Team Location: National Oceanic and Atmospheric Administration National Centers for Environmental Information – Asheville, North Carolina

The Aqua and Terra MODIS mean July Aerosol Optical Depth for 2000-2015 demonstrates elevated aerosol levels (in pink) emitted from Kilauea Volcano, Hawaii. Image Credit: Coastal US Health & Air Quality Team

The Aqua and Terra MODIS mean July Aerosol Optical Depth for 2000-2015 demonstrates elevated aerosol levels (in pink) emitted from Kilauea Volcano, Hawaii.
Image Credit: Coastal US Health & Air Quality Team

Christie Stevens
Toni Strauch
Alec Courtright

Annette Hollingshead (Global Science & Technology, Inc.)
DeWayne Cecil (Global Science & Technology, Inc.)
Jesse Bell (Centers for Disease Control and Prevention)

Past/Other Contributors:
Alec Courtright (Center Lead)


Outdoor air pollution can have severe impacts on human health, endangering the lives of vulnerable children and adults. Particulate matter (PM) air pollution is particularly hazardous, contributing to an estimated 800,000 premature deaths every year, as recorded by the World Health Organization (WHO). U.S. agencies such as the Environmental Protection Agency (EPA) and the Centers for Disease Control and Prevention (CDC) have an increasing need to provide accurate air quality assessments to properly mitigate adverse health effects. The EPA monitors air quality levels for the United States with AirNow, an online tool that uses in situ data to calculate an Air Quality Index (AQI) for real-time updates and forecasts in air quality. This product allows for reliable air quality estimates but is limited to the specific location of monitoring stations and terrestrial regions within the contiguous United States. Though current projections estimate that about 48 percent of the U.S. population will live in coastal regions by the year 2020, there is no long-term record of air quality measurements for the coastal United States. This project will use the NASA Aerosol Optical Depth (AOD) and the National Oceanic and Atmospheric Administration’s (NOAA) Aerosol Optical Thickness (AOT) datasets to create a long-term analysis of Hawaii and coastal Florida air quality from January 1981 to December 2015. This analysis will be used to identify anomalous months, seasons, and other periods as well as identify trends in aerosol concentrations and patterns. Satellite and ground station data verification will provide statistical and quantitative support for incorporating satellite data into both the EPA’s and CDC’s current air quality reporting efforts and further serve managers’ needs by building the foundation for a comprehensive air quality tool.

Previous story | Main Page | Next story



Kelsey Herndon (Chaco Canyon Cross-Cutting) 08-09-2016, 21:29

Really great project! Just a few questions: How would you tease out the relationship between asthma and aerosols if you were to have a continuation project? I would also like to second Alison’s question on how the ground stations compared with the satellite data? How many ground stations are there and in what regions are they located? Were you able to include data from cloudy days? Were you able to correlate any specific events with anomolous data?



Christie Stevens 12-09-2016, 17:58

Thank you, Kelsey! Exploring the relationship between asthma and aerosol levels falls within the scope of applying CDC public health records, which would fall primarily as a highly collaborative effort in which our CDC contacts could apply their expertise in that arena. The correlation between ground station and satellite data was pretty low overall. AOD and PM2.5 did, however, follow similar seasonal patterns.

The number of ground stations varied in both Hawaii and Florida, with Hawaii having fewer stations for comparison and Florida housing roughly 50 stations. For this project, we had to sift through various PM monitors to find suitable PM2.5 ground stations. We took data from cloudy days into consideration, but it had little effect. We loosely analyzed specific anomalous events such as volcanic smog (VOG) in Hawaii and sugarcane burning in Florida. We would have liked to dive further into looking at specific years of such events, but we hope that will be looked further into in depth in continuing work.

Julia Marrs 07-09-2016, 17:14

Congratulations on a beautiful and informative video! I really enjoyed looking at the detail in your monthly time series, and it’s very cool that you were able to relate the signals you saw to historical events on the ground. I see your comment above about continuing this project by integrating certain kinds of public health data. Do you know if your partners at the EPA are planning to use similar visualizations to monitor any specific species or ecosystems at risk from air quality concerns in the future?

Alec Courtright 09-09-2016, 18:01

Thank you Julia! The EPA group that we worked with are focused primarily on human health concerns related to air quality. I doubt that they will use our results or similar visualizations for ecosystems. However, it sounds like you may be onto an interesting topic for a future DEVELOP project!

Dash Cruz 07-09-2016, 11:38

Great Video! How do you see your results helping your project partners? Are they looking to change laws? Do they want to increase the general knowledge of the public?

Sean McCartney 07-09-2016, 09:19

Terrific work Coastal US Health & Air Quality team! Interesting you decided to look at both 30- and 15-year monthly and seasonal climatologies. What were some of the trends you discovered, and how did they differ between both 15 and 30-year study periods? Also what were some of the biggest takeaways from the project?

Alec Courtright 09-09-2016, 19:00

Thank you Sean! That means a lot!

Our 30 year and 15 year climatologies were created with NOAA’s AVHRR AOT and MODIS AOD respectively. AVHRR AOT was only available over the oceans which is why we focused more heavily on the 15-year MODIS AOD climatologies. Correlations values were also a little higher with MODIS AOD.

The seasonal trends were relatively similar between the 30 year and 15 year climatologies. The 30 year climatologies showed less spatial variation, especially in the gulf of Mexico, but were great for seeing global trends. For example, with the 15 year climatologies you can see higher aerosol concentrations in the eastern and northern portions of the Gulf during the summer months. The 30 year climatologies “wash-out” these variations and the majority of the gulf yields similar aerosol values in the summer.

For Hawaii, both the 30-year and 15-year climatologies showed very similar trends with higher aerosols stemming off of the Kilauea Plume at the southern portion of the big island.

One of the biggest takeaways for me is the potential value in combining diverse satellite aerosol measurements. Multi-angle measurements, like those form MISR, combined with AOD would produce a better representation of aerosols at the surface, particle size,…etc.

Alison Thieme 07-09-2016, 08:57

Interesting video! How closely does the ground station PM 2.5 data match the AOD data from NOAA and NASA Earth observations?

Alec Courtright 09-09-2016, 18:12

Thank you Alison. The correlation between AOD and ground station PM2.5 varied by location and state but, were fairly low overall. Generally, correlation values were higher in Florida than in Hawaii. However, in both states the ground station PM2.5 and AOD followed similar expected seasonal trends.
Further studies should try and include other aerosol measurement methods such as MISR to get a closer relationship between point measurements from ground stations and satellite aerosol measurements. MISR multi-angle data which means that you can set limits on particle sizes (i.e. you could specify smaller particles such as PM2.5).

Sara Lubkin 02-09-2016, 15:17

Great job on your video – congratulations on making the finals! I always think of coasts as having clean air, but that could be because in California the wind comes from the ocean. Since the prevailing winds travel west to east, did you see a difference between the east and west sides of the state?

Alec Courtright 09-09-2016, 18:29

Thank you Sara! We really appreciate your kind words!

That is a great question. You can see some impact form west to east winds along the Florida panhandle. During summer months the panhandle experiences higher aerosol values than the rest of the western coast. During the spring months you can see higher aerosol levels originating further west in the gulf.
On the East coast there is a visible increase in aerosols that is most likely due to Saharan dust during the summer months (the increased aerosols coming from the Sahara are visible on our monthly averages pre-subsetting to Florida).

Sara Lubkin 09-09-2016, 18:46

I think it’s pretty amazing that the sources of aerosols can be tracked around the globe.

Daryl Ann Winstead (Mekong River Basin Agriculture) 18-08-2016, 15:44

Great project! Is a continuation of the project planned? If so, what would be the focus? Thanks in advance for your response!

Alec Courtright 18-08-2016, 16:43

Thank you Daryl! There is the possibility of a continuation project which will work closely with the CDC (possibly a group of CDC interns) to relate public health data (i.e. asthma attacks, coronary artery disease…etc.) to surface and remotely sensed aerosol data. We would also try and incorporate either MISR or Aeronet data.

Daryl Ann Winstead (Mekong River Basin Agriculture) 18-08-2016, 19:30

Hi Alec! This is great! I’m looking forward to hearing about the progress if the project is continued.

Emily Gotschalk 16-08-2016, 08:57

Interesting project! Were there any major differences in the statistical relationships and/or understanding the spectral signatures of the volcanic smoke vs. the residue from sugarcane burning?

Alec Courtright 16-08-2016, 13:42


Unfortunately we weren’t able to start identifying different spectral signatures for Vog and sugarcane burning particles. We mostly looked for higher concentrations in areas effected by vog/sugarcane burning. However, areas typically impacted by vog showed a poor correlation. Future terms should look into this further or use MISR data (since it has 9 different measurement angles that can be used to detect different particle sizes) to take a further look into vog and particulates from sugarcane burning.

Emma Baghel 12-08-2016, 13:01

Very cool images and great presentation! Has this project spiked your interest in public health, air quality, and possibly working for the CDC or EPA in the future? If not, how and what will you pull from your experience working with this data and end-users for your future careers?

Christie Stevens 16-08-2016, 12:32

Thank you, Emma! Yes, this project absolutely sparked our team’s interest in public health & air quality, as well as future possibilities within the EPA and CDC. This was a new topic for the entire team to tackle and the challenges we faced with understanding satellite aerosol data and the differences in ground station and satellite data measurement techniques opened our eyes to the difficulties faced by our end-users and understanding their goals with this project. Communication between the team and our end-users was key in this process, something we will undoubtedly take away for our future careers.


Leave a Reply