Air Quality applications of Remote Sensing Data

1
Air Quality applications of Remote
Sensing Data
CMAS NASA Satellite Training October 14-16th,
2009
2
September8, 2002
September10, 2002
September11, 2002
Strengths of Satellite Remote Sensing
Monitoring Pollution Transport
3
Sept 9
Sept 10
Satellite View
4 day sequence showing transport of
regional pollution event. Posts show PM2.5
ground-based measuring site. Color contours are
MODIS aerosol optical depth
No monitors MODIS fills in
Sept 11
Sept 12
4
Air Quality Index

• Based on 24 hour PM2.5 mass concentrations air
  quality indices are derived color coded
• However, these ground monitors lack spatial
  coverage (Guidelines may vary by country)

5
Combining Satellite and Ground-Based Information
June 13, 2008
6
The satellite ground monitor problem?
Multi-spectral satellite data
Surface measurement
Satellite measured radiances are converted to
columnar Aerosol Optical Thickness
7
Satellite measurements are a total column, not
always correlated with surface Air Quality !!!
Northeast US - Smoke above the PBL - Satellite
shows pollution but PM2.5 shows good air quality
8
How well does the MODIS AOD Represent Air
Quality/Health Conditions ?

• In general MODIS does a good job of retrieving
  AOD over oceans and dark vegetated surfaces
  provided the AOD is not too low.
• The MODIS AOD is less accurate over
  non-homogeneous surface and bright surfaces
  (desert, snow).
• If aerosols are below a cloud layer, passive
  imagers such as MODIS cannot see them. Therefore
  there are no data in that part of the image. This
  can lead to biases when calculating aerosol
  burden at various time scales

9
How well does the MODIS AOD Represent Air
Quality/Health conditions ?

• AOD, like most satellite observations is a Total
  Atmospheric Column measurement and does not
  provide information about the vertical
  distribution of the pollution
• Satellite observed pollution does not mean the
  pollution is at the surface or that hazardous
  health conditions are present
• AOD imagery cannot differentiate between
  pollution types (dust, smoke, haze).

10
MODIS AOD/PM2.5 Comparisons
Pollution that is well mixed in the boundary
layer Provides a good AOD-PM2.5 relationship
11
Smoke over the southeastern U.S May 2007
Clouds
Smoke
Fires
12
True Color Time Series Images MODIS-Terra Southeas
tern United States
13
MODIS-Terra Aerosol Optical Thickness (AOT)
Colored circles represent air quality categories
derived from surface PM2.5 mass concentration
using EPA AirNow network
14
Terra-Aqua Daily Composite
15
AOT Time Series Images MODIS-Terra Southeastern
United States
16
Filling in the gaps

• If we know the PM2.5-AOD relationship then, can
  we fill in gaps in Air Quality Information where
  there are no ground monitors?

17
Air Quality Index
Use AOT PM2.5 relationship and get AQI
PM2.5 (µgm-3) (27.5 x MODIS AOT) 15.63
18
Air Quality Index
PM2.5 (µgm-3) (27.5 x MODIS AOT) 15.63
19
Summary Advantages and Limitations of Satellite
Imagery

• Advantages
• NASA satellite data are FREE and at no cost to
  users with access to the internet.
• Satellites provide local, regional, AND Global
  coverage
• Satellites help fill information gaps where
  surface monitor data on air quality conditions
  may not be available
• Daily images are available for most satellite
  instruments
• Limitations
• Column measurements are for the entire vertical
  column of the atmosphere and may not reflect
  surface/health impacts
• Many air quality satellite parameters CANNOT be
  measured when clouds are present or for certain
  land types (desert)
• Satellites cannot differentiate between pollution
  types

20
Web Tools for Comparing Satellite and Surface
Pollution
21
Giovanni Gridded PM2.5 Surface Monitor Data
AIRNow PM2.5 Air Quality Index Map
Gridded monitor data in Giovanni allows direct
comparisons with satellite observations
Giovanni PM2.5 Map (ug/m3)
Note This data set is Acquired by Giovanni from
DataFed via WCS.
22
Giovanni Surface Monitor-Satellite
ComparisonsJune 15- July 15th, 2006
MODIS/PM2.5 correlation
MODIS/ PM2.5 time series
Satellite measurements are a total column --- not
necessarily correlated with PM2.5 concentrations
!!!
23
IDEAInfusing Data into Environmental Applications
24
IDEA

• First tool designed for real time visualization
  of MODIS AOD and EPA PM2.5 (NASA, NOAA, EPA).
• Current Products
• MODIS RGBs and AOD real time map
• GASP AOD real time loop
• PM2.5 (time series and map)
• 850 mb wind vectors
• GOES derived wildfire locations
• AOD/PM2.5 Correlation statistics

25
http//www.star.nesdis.noaa.gov/smcd/spb/aq/
26
MODIS Aerosol Optical Depth (AOD) Image
Red high AOD Blue low AOD
Smoke
Haze
MODIS has 2 images per day Terra 1030
AM Aqua 130 PM
27
GOES Aerosol/Smoke Product (GASP)

• GASP is AOD derived from GOES
• Color scale is similar to MODIS AOD red is
  high AOD blue is low AOD white is clouds
• GASP has high temporal resolution (available
  every 30 minutes)

Haze
Haze
28
(No Transcript)
29
3-Day Composite History
30
IDEA Real Time co-location of MODIS and GASP
AOD with EPA PM2.5
31
IDEA Real Time co-location of MODIS and GASP
AOD with EPA PM2.5