Información en español

• AQI at Monitored Locations

• The AQI Grid

• AQI Between Regulatory Monitors

• Key Changes from the Previous Website

• Frequently Asked Questions

How is my AQI determined?

Air Quality Index (AQI) values are determined using a technique developed by South Coast AQMD scientists and published in a peer-reviewed scientific journal that is considerably more accurate than other methods. This method blends measurements from high-quality regulatory monitors, quality-controlled and calibrated low-cost sensors, and an air quality model.

AQI at Monitored Locations

The current Air Quality Index (AQI) on the South coast AQMD website and app is calculated from measured and modeled concentrations of five pollutants:  ­particle pollution (PM_2.5 and PM₁₀), ozone, nitrogen dioxide, and carbon monoxide. The “Current AQI” uses measurements of these pollutants at regulatory monitors that are operated by the South Coast AQMD, along with regulatory monitors operated by other agencies within or near the South Coast AQMD jurisdiction. Regulatory monitoring locations within the South Coast AQMD jurisdiction are identified with markers on the AQI map. These regulatory monitors provide the most accurate concentration measurements, but due to finite resources, regulatory monitoring is not practical in every neighborhood. 

The AQI Grid

To report air quality to a neighborhood level, the AQI map is shown using 1,218 grid cells. Each grid cell is 5 km (3.1 mi) x 5 km and is identified by a unique name. Measurements from regulatory monitors are used to calculate AQI values in the grid cells where the monitors are located. South Coast AQMD staff have developed a state-of-the-art technique to calculate AQI values in every one of these grid cells, using a combination of data from regulatory monitors, low-cost fine particulate matter sensors, and an air quality model.

[Return to Top]

AQI Between Regulatory Monitors

Fine Particulate Matter (PM_2.5) 

We use model predictions from the National Air Quality Forecasting Capability operated by the National Ocean and Atmospheric Administration (NOAA) (https://vlab.noaa.gov/web/osti-modeling/air-quality) to help determine PM_2.5 concentrations between regulatory monitors. We also use PM_2.5 concentrations measured by low-cost sensors developed by PurpleAir (https://www2.purpleair.com/) and, starting in April 2024, the Clarity Node-S open data air quality sensors (https://www.clarity.io/). The PurpleAir low-cost sensors are installed at hundreds of locations by residents and agencies throughout the jurisdiction. Since PurpleAir sensors are not as accurate as the regulatory monitors, we do extensive quality control to ensure that we are not using data from bad sensors. We also calibrate the data to improve accuracy. Because PurpleAir sensors may be located near hyper-local sources such as a barbeque grill or chimney, we average all the sensor measurements in a neighborhood using a search procedure, only using the data if three or more sensors are in the neighborhood. The neighborhood search first searches within each 5 km x 5 km grid cell, expands to 15 km x 15 km and then 25 km x 25 km centered on each grid cell, and ends when three or more sensors are found in the search area.  

Most Clarity Node-S sensors in the South Coast AQMD jurisdiction are deployed at Los Angeles Unified Schools, away from hyper-local sources, and are spaced uniformly over the neighborhood (https://www.clarity.io/customer-stories/lausd-and-cca-build-the-nations-largest-school-based-outdoor-air-quality-sensor-network); these measurements are representative of the air quality in the neighborhood surrounding the sensors. Therefore we use the Clarity Node-S data even if there is only one sensor in a grid cell. Clarity’s air quality experts calibrate the measurements and ensure their data quality. 

The low-cost PurpleAir sensor data, Clarity Node-S data, and the model predictions are blended with the regulatory monitoring data using a method that takes into account the relative accuracy of each of these sources to determine PM_2.5 levels in the grid cells without regulatory PM_2.5 monitors. 

Ozone

We blend NOAA model predictions with regulatory monitor measurements to determine ozone concentrations in grid cells without regulatory ozone monitors. Similar to the approach used with PM_2.5, we use a blending method that considers the relative accuracy of these two data sources. Low-cost sensor measurements are not yet widely available for ozone, but the size and coverage of our ozone monitoring network along with excellent model performance is sufficient to accurately determine ozone levels between monitors.

Particulate Matter (PM₁₀)

PM₁₀ are particles with diameters of 10 microns or less. In order to determine PM₁₀ concentrations between monitors, we split PM₁₀ into the mass of particles of 2.5 microns or less (Fine Particulate Matter, PM_2.5) and the mass of coarse particles (particles bigger than 2.5 microns but of 10 microns or less). 

We then calculate coarse PM levels in the grid cells without measurements by using a mathematical technique that has been shown to perform well for these types of measurements. Once we have coarse PM levels at all grid cells, we add them to the PM_2.5 levels to determine PM₁₀ concentrations between monitors.

Nitrogen Dioxide and Carbon Monoxide

AQI values are almost always driven by concentrations of ozone, PM_2.5, and PM₁₀; however, we also incorporate nitrogen dioxide and carbon monoxide concentrations in the map. Nitrogen Dioxide and Carbon Monoxide levels at grid cells without monitors are determined using a mathematical technique that has been shown to work well with these types of measurements.

Key Changes from the Previous Website

In the past, the South Coast AQMD AQI map used a different method to estimate concentrations between regulatory monitors. The current method improves upon the older method in several ways. It shows the locations where concentrations are measured and estimated, presents air quality on a smaller scale (5 km), and is more accurate. The picture below compares the current and previous websites. The current website uses rectangular grid cells to display AQI while the previous method used larger zones with similar air quality conditions.

For additional information on the methodology used to generate the AQI map, please refer to our peer-reviewed journal publication at the following link:  https://iopscience.iop.org/article/10.1088/1748-9326/abb62b 

[Return to Top] 

Frequently Asked Questions

1. How do you convert concentrations to AQI values?

   AQI is used to report air quality, rather than the concentrations, because AQI is a better way to communicate the degree of heath concern and recommended precautions during periods of poor air quality. For each pollutant, concentrations are converted to AQI values using a function that is based on health studies and the federal standard . The website and app show the largest AQI from all the pollutants in a grid cell. To learn more about how AQI is calculated visit https://www.airnow.gov/aqi/aqi-basics/ . To learn more about what health concerns are associated with an AQI level and what to do when AQI is high, visit https://www.aqmd.gov/aqi.

    

2. Why are AQI values from other websites or apps different than the South Coast AQMD map?

   South Coast AQMD AQI values are calculated with high-quality measurement data and a scientifically demonstrated methodology that has been peer-reviewed by other researchers in the field. This methodology uses the best-available science to determine air quality concentrations between monitors in a formulation that is tailored for the South Coast AQMD jurisdiction. Concentrations are then converted into AQI values using EPA methodologies. Significant differences between South Coast AQMD AQI values and some private-sector apps may be due to the following factors:

   • In some apps, only one pollutant may be considered at a monitored location, which can produce incorrect AQI values when an unmonitored pollutant is driving the AQI.  For example, if your location only has a low-cost PM_2.5 sensor, your AQI value may not reflect ozone concentrations, which often drive the AQI in our region.  

   • Some apps incorporate lower quality measurement data without quality control and/or calibration.

   • Some apps rely on models that are designed to estimate AQI values across the world, but are poorly-suited to represent air quality on a neighborhood level in a widespread region with complex topography and many sources such as the South Coast Air Basin or the Coachella Valley.

   • Some apps employ methodologies to convert measured concentrations to AQI values that are not consistent with EPA guidance.

      

3. Why does the AQI forecast differ from the current AQI measurements?

   We issue two forms of the AQI forecast each day. A single AQI value is reported for each area, summarizing the conditions expected over the entire day, and therefore, it cannot be compared directly to the current hourly AQI value. Hourly AQI forecasts for each area are also available to simulate a real-time AQI measurement. These forecasts are useful for residents to help plan outdoor activities. Some extreme events that arise after the forecast is issued may not have been anticipated when developing an air quality forecast. Air quality is difficult to predict, especially during extreme events such as wildfires and dust storms. Therefore, hourly forecasts may not be available during these conditions. During wildfires or dust storms, visit www.aqmd.gov/advisory to view current air quality advisories.
   
   

4. Why doesn’t the grid cell containing my location have the name of my city or neighborhood?

   As grid cells can contain portions of multiple cities or neighborhoods, their names may not always be identified by a single city or neighborhood name. Therefore, the grid cell containing your location may correspond to the name of an adjacent city, neighborhood, or geographical landmark. However, the air quality information provided is still the most accurate for your location.  In the mobile app, users are free to assign a nickname to the grid cell containing your location.  

    

5. What averaging time do you use when reporting current PM2.5 air quality on the app and website?

We use a method developed by the U.S. EPA called NowCast, (More Information on the NowCast Algorithm ) which uses shorter averaging windows when concentrations are changing quickly and longer windows when concentrations are more stable. This is consistent with how EPA and other state and local air agencies calculate real-time AQI values. The conversion from a concentration to an AQI value is based on the federal air quality standards (which are based on health studies), so that we can make health recommendations based on particular AQI ranges.  Since the short-term PM_2.5 air quality standard and the health studies used to develop the standard are based on 24-hour average exposures, we need to simulate a 24-hour average concentration before converting to AQI. The NowCast methodology is designed to simulate 24-hour concentrations from shorter term averages with an emphasis on the most recent data so that our AQI map can be more responsive to local conditions. Hourly averages would be more responsive to changing smoke conditions during a wildfire, but there is not enough established health science to provide recommendations to the public for a given concentration range.

Current hourly measurements are available on the South Coast AQMD website (https://xappd.aqmd.gov/aqdetail/AirQuality). However, there is no PM₁₀ and PM_2.5 hourly standard and there is limited science relating 1-hour exposures to health effects, so it is difficult to provide context and recommendations for a given hourly concentration.

[Return to Top]