100528 satellite obs_china_husar

Seminar at Fudan University, May 28, 2010, Shanghai, China

Rudolf B. HusarWith

Li Du, Erin RobinsonWashington University, St. Louis, MO, USA

Spatial and Temporal Pattern of Air Pollution over China Based on

Remote Sensing Observations

Atmospheric Aerosol Challenge:Characterization of Aerosols

• Aerosol complexity is due 7-dim. data space• The ‘aerosol dimensions’ D, C, S determine the

effects on health and climate

Dimension Abbr. Data SourcesSpatial dimensions X, Y Satellites, dense networks

Height Z Lidar, soundings

Time T Continuous monitoring

Particle size D Size-segregated sampling

Particle Composition C Speciated analysis

Particle Shape/Mixing S Microscopy, Source Type

Aerosol concentration: a (X, Y, Z, T, D, C, S)

Challenge: Vertical Distribution of Aerosols

Vertical Distribution:

• Layering• Size Distr.• Composition

Technical Challenge: Characterization• PM characterization requires many different instruments and analysis tools.

• Each sensor/network covers only a fraction of the 7-Dim PM data space.

Satellite-Integral

Satellites, integrate over height, size, composition, shape…dimensions

These data need de-convolution of the integral measures

Satellite Remote Sensing Since 1972

• First satellite aerosol paper, Francis Parmenter, 1972• Qualitative surface-satellite aerosol relationship shown, 1976• Focus on regional ‘hazy blobs’, sulfate pollution

Regional HazeLyons W.A., Husar R.B. Mon. Weather Rev. 1976

SMS GOES June 30 1975

Satellites show the synoptic aerosol patternand provide rich spatial context …

e.g. pollution in valleys.

Jan 10, 2003, SeaWiFS Dec 19, 2007, MODIS

The Perfect Dust Storm…Apr. 7, 2001 SaeWiFS

Asian Dust Cloud over N. America

On April 27, the dust cloud arrived in North America.

Regional average PM10 concentrations increased to 65 g/m3

Asian Dust

100 g/m3

Hourly PM10

In Washington State, PM10 concentrations exceeded 100 g/m3

Satellite(MODIS, OMI)

Visual Range(WMO)

Sun Photometer(Aeronet)

Challenge: Aerosol Retrieval Quality

MODIS vs. MISR: Poor AOT Correlation over Land

Land Ocean

MIS

R A

OT

MODIS-AOT MODIS-AOT

Aerosol AOT – MODISJanuary April

OctoberJuly

AERONET – MODIS AOT Comparison

AeronetMODIS

MODIS/Aeronet Ratio

Beijing

Hong Kong

Correlations good but systematic differences (slope)

MODIS4 AOT: Thursday

MODIS4 AOT: Sunday

MODIS4 AOT: Thursday

MODIS4 AOT: Sunday

MODIS Fire Pixels

OMI Absorbing Aerosol Index

January April

OctoberJuly

OMI CHCO Formaldehyde

January April

OctoberJuly

Population density

Emissions - NOx

Satellite Column Concentration: NO2, 2005

OMI Spectrometer Sensor

Satellite Column Concentration: NO2, 2009

OMI Spectrometer Sensor

OMI NO2 Day of Week: Thursday

OMI NO2 Day of Week: Sunday

Visibility is recorded at 7000+ stations hourly

Visual Range, Guilin 2010-05-23,24

Visual Range 16 km

Visual Range 9 km

Guilin

Surface Extinction Coefficient

Jun, Jul, Aug Dec, Jan, Feb

Sechuan Basin

Chengdu

Chengdu

Jun, Jul, Aug Chongqing

ChongqingDec, Jan, Feb

MODIS VISIBILITY

Xi’an

Xi’an

Xi’an

Jun, Jul, Aug Tianjin

TianjinDec, Jan, Feb

MODIS VISIBILITY

Summary

•Each data set has limitations, but gives self-consistent global-scale observations

• More detailed measurements are essential for the understanding

•There are still enormous challenges in integrating multi-sensory data for characterizing aerosols.

Combining global-scale remote sensing observations with detailed local observations and research conducted in China could yield faster

progress.

International Collaboration Opportunity:

Global Observing System of Systems (GEOSS)Pooling of Earth Observations nine Societal Benefit Areas

Any Dataset Serves Many Communities

Any Problem Requires Many

Datasets

New International Program - China is a Co-Chair of GEOSS EE