1 / 15

15 Years of Global PM 2.5 Estimates from Satellite

15 Years of Global PM 2.5 Estimates from Satellite. Aaron van Donkelaar with contributions from Randall Martin, and Brian Boys. Tuesday, June 10 th , 2014 Goldschmidt. PM 2.5 affects human health and longevity.

iram
Download Presentation

15 Years of Global PM 2.5 Estimates from Satellite

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. 15 Years of Global PM2.5 Estimates from Satellite Aaron van Donkelaar with contributions from Randall Martin, and Brian Boys Tuesday, June 10th, 2014 Goldschmidt

  2. PM2.5 affects human health and longevity • Life expectancy increases 7 months per 10 μg/m3 decrease in long-term exposure Extinction due to aerosol in the atmospheric column is called Aerosol Optical Depth (AOD) Previous Global Burden of Disease report for 2000 (GBD2000) limited by insufficient ground-level observations of fine Particulate Matter (PM2.5) MODIS Aqua February 5th, 2006 GBD2010 estimates PM2.5 in urban and rural areas causes: • 3.2 million deaths (3%) Makes use of satellite-derived data for complete global coverage 0 0.2 0.4 0.6 0.8 1.0 Aerosol Optical Depth Cohen et al., 2005

  3. AOD is related to PM2.5…but how to separate surface aerosol from column concentrations? • PM2.5 / AOD ratio is a function of: • vertical structure • aerosol type/hygroscopicity • meteorology Multiple approaches: • surface monitors calibration • empirical relations • model output 20 80 140 PM2.5 / AOD [μg/m3] GEOS-Chem v8-01-04

  4. Model output can in used in different ways • “Unconstrained”: • Combined MODIS, MISR AOD • Apply simulated AOD/PM2.5 • van Donkelaar et al., EHP, 2010 • MODIS/MISR, 2001-2006 • Good representation of magnitude (error = 1μg/m3+27%) • Unavailable after 2006 • Boys et al., submitted • SeaWiFS&MISR, 1998-2012 • Successful representation of trends observed TOA reflectance a priori AOD a posteriori AOD observational error a priori error CALIOP Space-borne LIDAR Chemical Transport Model • “Optimal Estimation”: • Error-constrained MODIS, simulated AOD • AERONET-based error estimates • Consistent optical properties • Local reflectance information • CALIOP-adjusted AOD/PM2.5 • van Donkelaar et al., JGR, 2013 • 2004-2010 • Good representation of magnitude (error = 1μg/m3+25%) • Unavailable before 2004 Optimal estimation constrains AOD retrieval by error: MODIS Imaging Spectroradiometer SeaWiFS Imaging Spectroradiometer MISR Imaging Spectroradiometer

  5. Combine PM2.5 to create a consistent 15 year dataset Unconstrained (UC) • 2001-2006 mass Optimal Estimation (OE) • 2004-2010 mass SeaWiFS&MISR • 1998-2012 annual variation/trend UC&OE • Weight by error over each land-cover type • 2001-2010 mass UC&OE + SeaWiFS&MISR • Apply relative variation from SeaWiFS/MISR • 1998-2012 annual mass Combined PM2.5 [μg/m3] PM2.5 (2001-2010) [μg/m3] In situ PM2.5 [μg/m3] van Donkelaar et al., submitted.

  6. Consistent Trends in Satellite-Derived and In Situ PM2.5 1999-2012 Eastern US PM2.5 Anomaly (μg m-3) SeaWiFS& MISR In Situ In Situ In Situ (1999-2012) 0.37 ± 0.06 μg m-3 yr-1 Satellite-Derived (1999-2012) 0.36 ± 0.13 μg m-3 yr-1 Satellite-Derived Boys et al., submitted

  7. Well correlated globally; potential bias at high PM2.5 200 150 100 50 0 1-σ error=1 μg/m3+47% y=0.68x+1.1 n=210 r=0.81 1-σ error=1 μg/m3+21% y=0.78x+1.7 n=512 r=0.73 40 30 20 10 0 Combined PM2.5 [μg/m3] Europe Global PM2.5 [μg/m3] 0 50 100 150 200 0 10 20 30 40 Ground Monitor PM2.5[μg/m3] van Donkelaar et al., submitted.

  8. Time series highlight global changes van Donkelaar et al., submitted.

  9. Significant global PM2.5 trends can be seen satellite @ in situ in situ PM2.5 [μg/m3] Boys et al., submitted.van Donkelaar et al., submitted.

  10. Changes in Long-term Population-Weighted Ambient PM2.5Clean Areas are Improving; High PM2.5 Areas are Degrading 100 80 60 40 20 0 WHO Guidelines and Interim Targets 100 80 60 40 20 0 Asia, East +1.6 North America, High Income -0.3 2012 (70%) 1998 AQG 1998 (51%) Year Exceedance of WHO AQG drops from 62% to 19% Exceedance of WHO IT1 increases from 51% to 70% 2012 2005 1998 2012 IT3 80 60 40 20 0 Asia, South +1.0 Europe, Western -0.3 IT2 Percent of Population [%] Percent of Population [%] IT1 80 60 40 20 0 Europe, Central -0.2 North Africa,Middle East +0.4 Global: +0.6 5 10 15 25 35 50 100 5 10 15 253550 100 5 10 15 253550 100 Annual Mean Exposure [μg/m3] Annual Mean Exposure [μg/m3] 1998-2012 exposure trend [μg/m3/yr] van Donkelaar et al., submitted.

  11. Questions?

  12. Global in situ monitors are sparse...and far from AERONET • PM2.5 estimates accuracy impacted by • AOD retrieval • Simulated aerosol vertical profile • Mass to extinction conversion • Sampling (clouds/snow) • Difficult to evaluate error sources • SPARTAN network collocates PM2.5 and AOD measurements • Satellite-derived PM2.5 evaluation, analysis, www.spartan-network.org Airphoton PM2.5 (Vanderlei Martins) Filter PM2.5 Filter PM10 3-λNephelometer AOD from CIMEL Sunphotometer (e.g. AERONET)

  13. Global impact of global data Global Burden of Disease 2010 • 488 authors from 303 institutions in 50 countries • Inclusion of rural populations • PM2.5 causal role in 3 million deaths per year Lim et al., Lancet, 2012 • Inform Epidemiological Studies: • Global childhood asthma (Anderson et al., 2012) • Lung cancer in Canada (Hystad et al., 2012) • Mortality in California (Jerrett et al., 2013) • Diabetes(Brook et al., 2013; Chen et al., 2013) • Global adverse birth outcomes (Fleisher et al., in press) • Hypertension(Chen et al., in press) • Low PM2.5 cardiovascular mortality (Crouse et al., 2012) Crouse et al., EHP, 2012

  14. Optimal estimation requires accurate error representation Error observed by AERONET Error is a function of species: • Emissions/chemistry • meteorology Estimate GC error: • Apply GC speciation • Inverse-distance / cross-correlation Observational error via brute force, extended via land cover van Donkelaar et al., JGR, 2013

  15. Satellite-Derived PM2.5 Trends Inferred from SeaWiFS& MISR AOD and GEOS-Chem AOD/PM2.5 MISR 2000 -2012 SeaWiFS 1998 -2010 PM2.5 Trend [μg m-3 yr-1] Boys et al., submitted

More Related