Chapter 4: How Satellite Data Complement Ground-Based

Monitor Data3:15 – 3:45

Sources of Data for Air Quality Analysis

• There are three major sources of data for air quality analysis:– Ground-based monitor data– Satellite data– Model forecast data

• For a complete picture of air quality, need to use all three sources, when possible!

Ground-Based Monitor Data

• Strengths:– Accurate, precise measurements of specific

pollutants (PM2.5, PM10, O3,etc.).– Measurements at surface, where people live and

breathe the air.

• Limitations:– Measurements only in areas where monitors are

located; typically, there are large gaps in any country’s monitor network.

– Monitors are expensive to purchase and operate.

Example: Ground-Based Monitor Data

Satellite Data• Strengths:

– Global, regional, and local coverage – satellites provide information for areas where no ground-based monitors exist.

– Routine measurements are made daily.– Data are “free” – no cost to users with access to the internet.

• Limitations:– Measurements are for entire vertical column of atmosphere,

not specific to surface.– Clouds block measurement of many air quality satellite

parameters.– Temporal and spatial resolution of satellites can limit data

availability.

Example: Satellite Data

Model Forecast Data

• Strengths:– Forecasts provide information on future air quality

conditions; advanced notice allows governments to take action to protect public health.

• Limitations:– Model forecasts are NOT measurements! They are

predictions of future air quality conditions.– Air quality models are complex combinations of numerical

weather models and atmospheric chemistry, physics, and emissions models – the sum is only as good as the individual parts.

– Model output can be difficult to interpret for novices or non-experts.

Example: Model Forecast Data

How Can Satellite Data Complement Ground-Based Monitor Data?

• Satellites make global measurements, so we can use them to supplement ground-based measurements from a monitor network.

• Since satellites measure the total pollutant concentration in a vertical column of atmosphere, it is necessary to correlate satellite data with ground-based monitor data.

• Once we know the relationship between satellite and ground-based data, we can apply satellite data to areas where there are no ground-based monitors and obtain and estimate of surface pollutant concentrations.

Example: Correlation between MODIS AOD and PM2.5 for One Location in U.S.

MODIS AOD (Collection 5)

Dai

ly P

M2.

5 (

g/m

3 )

0.00 0.25 0.50 0.75 1.00 1.25

010

2030

4050

60 y = 26.9 x + 8.2r 0.77

Terra

Baltimore, Maryland2005

Example: Correlation between MODIS AOD and PM2.5 for May 2007 in U.S.

• Correlation varies by: Region and season Vertical aerosol

distribution and properties

Meteorological conditions such as relative humidity and boundary layer height

• AOD retrievals are also less accurate over bright surfaces such as desert or snow.

Image generated by Giovanni , NASA GES DISC

Visualization of Satellite and Monitor Data

• Satellite imagery and ground-based monitor data can also be visualized together to provide a more complete picture of air quality conditions.

• Try using Google Earth or the mapping products on SERVIR.netto visualize air quality data!

Eastern United StatesJune 13, 2008

Smoke

Haze

• Quantitative for PM2.5 and NO2

• Qualitative for SO2

• Bottom line: use satellite images for information about the distribution and transport of pollutants, but not about the exact concentrations of pollutants.

• Don’t forget: if satellites indicate the presence of pollutants, it doesn’t mean concentrations are high at the surface. The pollution could be aloft in the atmosphere, above the surface.

Interpretation of Satellite Images

Air Quality Analysis: Final Thoughts• For best results, look at all three types of data

when you are conducting an air quality analysis:– Ground-based monitor data– Satellite data– Model forecast data

• The three data sources each convey different information!

• Don’t forget the strengths and weaknesses of each type of data.