Investigating Secondary Aerosol Processes in the Amazon through Molecular-Level Characterization of Semi-Volatile Organics

Principal Investigator(s):
Allen Goldstein, University of California, Berkeley

Co-Investigator(s):
Nathan Kreisberg, Aerosol Dynamics Incorporated
Susanne Hering, Aerosol Dynamics Incorporated

Collaborator(s):
John Offenberg, US Environmental Protection Agency
Frank Keutsch, Harvard University
Jose-Luis Jimenez, University of Colorado, Boulder

The proposed research directly addresses DOE ASR Program Objectives to improve understanding of the generation, chemical transformation, and fate of primary and secondary aerosols in the tropics, associated with the GoAmazon ARM deployment. We aim to elucidate the effects of anthropogenic pollution on natural organic aerosol formation in this region through characterization and analysis of atmospheric organic chemical composition with unprecedented detail. Elucidating the human influence on natural atmospheric processes, especially in this region, is critical for understanding and modeling cloud formation, the hydrologic cycle, and the global radiative balance.

In areas where biogenic emissions are oxidized in the presence of anthropogenic pollutants such as SO2, NOx and black carbon, it has become increasingly apparent that secondary organic aerosol (SOA) formation from biogenic volatile organic compounds (BVOCs) is substantially enhanced. The Amazon forest is the dominant source of BVOCs globally, and the forest is rapidly being converted to urban and agricultural uses. Data collected downwind of Manaus during the GoAmazon2014/15 at the ARM facility at T3 afford the unparalleled opportunity to directly measure the influence of a highly polluted urban plume perturbing natural photochemical processing of BVOC. To understand the connection between primary BVOC emissions and their secondary products that form aerosols, we made hourly in-situ molecular level measurements using a Semi-Volatile Thermal desorption Aerosol Gas Chromatograph (SV-TAG) and collected quartz filter aerosol samples for later laboratory analysis during two intensive operating periods (IOPs) of the GoAmazon campaign (Jan-Mar & Aug-Oct, 2014) at T3.

Preliminary analyses have found several hundred VOC including over 20 sesquiterpenes, BVOC oxidation products of isoprene and monoterpenes, tracers of biomass burning and other sources. The vast majority of compounds in this hourly data set have yet to be identified. Almost none of the chemicals we measured have ever before been observed with hourly time resolution in the Amazon. Our focus for this proposal is to identify and quantify novel compounds observed in the atmosphere and use the temporal and chemical variability of all observed compounds to yield new insights into the atmospheric processing of volatile organic compounds. This information will be used by our group, and by multiple collaborating groups, to understand SOA formation, and as critical input for developing accurate models of background aerosol in order to understand the anthropogenic influences on SOA formation in the tropical systems that dominate global BVOC emissions.

We propose to complete five objectives.

  1. Conduct comprehensive analysis of SV-TAG data to identify
    "tracers of opportunity" that may be products of BVOC oxidation, including both vapor and particle phases.
  2. Confirm the identity of these compounds through GCxGC/VUV-HR-TOFMS and GCxGC/EI-HR-TOFMS analysis of filter samples, and quantify these compounds in the SV-TAG spectra through additional calibrations relative to internal standards.
  3. To the extent possible, identify oxidation products for compounds of interest found in GoAmazon, with confirming experiments through flow tube oxidation and/or smog chamber studies.
  4. Using results from Objectives 1, 2, and 3, quantify known and potential tracers for IOPs 1 and 2 with hourly resolution to better define primary and secondary aerosol sources through statistical analysis and comparisons between known and novel tracers, and to elucidate differences between periods with varying anthropogenic influence, especially as it relates to SOA formation pathways and partitioning.
  5. Develop a library of atmospheric organic compound mass spectra from the Amazon based on Objectives 2 and 3, and make it available to the scientific community.