Conclusions & Future Work Transported smoke measured during four flights (Sept. 4, 6, 13 and 14) • No smoke measured at ground level • AOD-to-PM higher than expected • Aerosol was aged agricultural smoke from the Mississippi River Valley • Aerosol aging increases smoke hygroscopicity by 36% • f(RH) of 1.5 for aged smoke / 1.1 for fresh smoke • and thus increases AOD more than fresh smoke would Future work will look at AERONET and ground-based PM 2.5 measurements in order to study: • the frequency of these long-range transport events • the spatial extent of smoke transport in the southeast U.S. DISCOVER-AQ DISCOVER-AQ was funded by NASA’s Earth Venture-1 Program through the Earth System Science Pathfinder (ESSP) Program Office. We wish to thank the ESSP Program Office for the support, the NASA Wallops Flight Facility, the pilots, flight crew, and the entire DISCOVER-AQ Science Team. We also wish to thank the entire SEAC4RS science team. Long-Range Transport of Agricultural Smoke to Houston, TX: Effects on Aerosol Optical Depths A.J. Beyersdorf ([email protected]) 1 , S. Burton 1 , G. Chen 1 , J. Lin 2 , R. Martin 1 , R. Moore 1 , A. Nenes 2 , K.L. Thornhill 1 , E.L. Winstead 1 , L.D. Ziemba 1 & B.E. Anderson 1 1 NASA Langley Research Center, 2 Georgia Institute of Technology Vertical profiles of extinction during DISCOVER-AQ-Texas varied amongst flights • Boundary layer – low loadings of 30-70/Mm • 2,500-10,000 ft – transported smoke caused increased aerosol loadings for September 13-14 September 13 th • Smoke measured by both the HSRL and in situ measurements • Highest loadings in the north of the flight region • Back trajectory for layer aloft from a region of agricultural fires Differences in Back Trajectories • Boundary layer – low loadings of 30-70/Mm • 2,500-10,000 ft – transported smoke caused increased aerosol loadings for September 13-14 Smoke Transport 2.AQ.1 Measuring surface level particulate concentrations remains a challenge for Earth- observing satellites due to: 1) variability in aerosol vertical distribution, and 2) the effects of aerosol composition and hygroscopicity on optical properties. DISCOVER-AQ (Deriving Information on Surface conditions from Column and Vertically Resolved Observations Relevant to Air Quality) is a multi-year project aimed at understanding the variables that affect remote sensing measurements in U.S. urban areas. Four campaigns were performed in regions with differing aerosol composition and meteorology: Baltimore & Washington, DC, Summer 2011 San Joaquin Valley, CA, Winter 2012 Houston, TX, Summer 2012 Denver, CO, Summer 2013 Results from Previous Campaigns Maryland • aerosol was composed of a mixture of organics and ammonium sulfate • aerosol present in a well-mixed deep haze layer (~ 7500 ft) California • primarily ammonium nitrate aerosol • contained in a very shallow boundary layer (~ 2000 ft) except for the last two flights AOD-to-PM • The AOD-to-PM correlation is dependent on the height of the haze layer (boundary + residual) with Maryland having a higher ratio than measured during California. 0 0 0 0 0 0 0 Normalized Extinction RF01-09 RF11-12 15000 12500 10000 7500 5000 2500 0 Altitude (ft) 400 300 200 100 0 Extinction (Amb. RH at 532nm) 0 0 0 0 0 0 0 Normalized Extinction 15000 12500 10000 7500 5000 2500 0 Altitude (ft) 400 300 200 100 0 Extinction (Amb. RH at 532nm) Maryland California 9 Flights between Sept. 4 th and 26 th , 2013 • P-3B aircraft • 24 spirals (1,000-15,000 ft above ground levels) over 8 ground sites • In situ measurements of aerosols & trace gases • Aerosol number concentration, scattering, absorption, size & composition (by SP2 and PILS; particle into liquid sampler) • B-200 aircraft (30,000 ft) • High Resolution Spectral Lidar (HSRL) Sampling of agricultural fires during transit flight No correlation between ambient AOD and particulate mass during DAQ-TX • Ambient AOD (measured by the P3) varied between 0.06 (Sept 26 th ) & 0.32 (Sept. 14 th ) • Flight day particulate mass was less variable (6-11 mg/m 3 ) DISCOVER-AQ Texas Downtown Houston Galveston Conroe 0.6 0.4 0.2 0.0 AOD 40 30 20 10 0 PM(mg/m 3 ) Measured by PILS below 1000 ft 15000 12500 10000 7500 5000 2500 125 100 75 50 25 0 Extinction (Amb, 532nm) RF07 (9-24) RF08 (9-25) RF09 (9-26) 15000 12500 10000 125 100 75 50 25 0 Extinction (Amb, 532nm) RF03 (9-11) RF04 (9-12) RF05 (9-13) RF06 (9-14) 15000 12500 10000 7500 5000 2500 0 125 100 75 50 25 0 Extinction (Amb, 532nm) RF01 (9-4) RF02 (9-6) 15000 12500 10000 7500 5000 2500 0 Altitude (ft) 125 100 75 50 25 0 Extinction (Amb, 532nm) Normalized Extinction West Houston 80 60 40 20 0 LIDAR In Situ Measurements 4 0 Extinction 29.2 29.6 30.0 30.4 Latitude 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Pressure Altitude 80 60 40 20 0 Scattering & Absorption (/Mm, green) Aerosol Size – smoke was larger than fresher aerosol measured in the boundary layer Single Scattering Albedo – smoke had a higher SSA than the boundary layer and fresh agricultural fires indicating secondary aerosol formation Hygroscopicity – fresh smoke had an average f(RH) of 1.1 while the aged smoke averaged 1.5 Smoke Properties 4000 3000 2000 1000 0 Number (dNdlogDp) 10 2 4 6 8 100 2 4 6 8 1000 Diameter Frequency 2.8 2.6 2.4 2.2 2.0 Scat. Ang. Exp. Boundary Layer Elevated Layer Frequency 2.0 1.8 1.6 1.4 1.2 1.0 f(RH) Boundary Layer Elevated Layer Fresh. Ag. Fire Frequency 1.00 0.95 0.90 0.85 SSA Boundary Layer Elevated Layer Frequency 1.00 0.95 0.90 0.85 SSA Elevated Layer Ag. Fire Period 1 Period 2 Period 3 Aerosol Size Single Scattering Albedo Hygroscopicity 70 60 50 40 30 20 280 270 260 250 240 230 220 210 200 SEAC 4 RS & ARCTAS Biomass Burning SEAC 4 RS Fire Plumes • Preliminary identification by Bob Yokelson including western U.S. wildfires and agricultural fires in the southeast. • Additional identification to follow for more aged plumes downwind of the Rim fire. ARCTAS Fire Plumes • 495 plumes identified by Hecobian et al. (ACP, 2011) • Aged Asian wildfires sampled over Alaska • Fresh Canadian and California wildfires Aged Asian Wildfires (ARCTAS) Fresh Canadian Wildfires (ARCTAS) SEAC4RS Wildfires SEAC4RS Ag. Fires Fresh California Wildfires (ARCTAS) Agricultural Fires (in comparison to wildfires) • lower modified combustion efficiencies → smoldering fires • MCE = (ΔCO 2 )/(ΔCO+ΔCO 2 ) • high particulate emissions & lower single scattering albedo Western Wildfires (in comparison to ARCTAS fires) • highest single scattering albedos • f(RH) less than 1 → indicative of soot restructuring 2.8 2.6 2.4 2.2 2.0 1.8 Scat. Ang. Exponent 1.00 0.95 0.90 0.85 Modified Combustion Efficiency Agricultural Fires 4.0 3.5 3.0 2.5 2.0 1.5 Abs. Ang. Exponent 1.00 0.95 0.90 0.85 Modified Combustion Efficiency 1.00 0.95 0.90 0.85 Single Scat. Albedo 1.00 0.95 0.90 0.85 Modified Combustion Efficiency Changes in aerosol intensive properties with flame conditions (MCE) • Flaming conditions (high MCE) gives: • higher scattering angstrom exponent (smaller aerosol) • lower absorption angstrom exponent (less organic coating) • lower single scattering albedo (darker smoke) 0.5 0.4 0.3 0.2 0.1 0.0 AOD 20 15 10 5 0 PM from PILS (<1000 ft) All Profiles Flight Averages colored by altitude of max. extinction (black is highest) 0.6 0.4 0.2 0.0 AOD 40 30 20 10 0 PM(mg/m 3 ) Measured by PILS below 1000 ft Maryland Slope = 0.027 California Slope = 0.006 20 15 10 5 0 PM 30 25 20 15 10 5 0 Day in September PILS Mass (below 1000 ft) PM2.5 (9am - 5pm) Galveston Conroe Deer Park 0.4 0.3 0.2 0.1 0.0 P3-B AOD (Amb, 532nm) 30 25 20 15 10 5 0 Day in September