AAAR 32nd Annual Conference
September 30 - October 4, 2013
Oregon Convention Center
Portland, Oregon, USA
Abstract View
Consistency of Long-term Black Carbon Trends from Thermal and Optical Measurements in the U.S. IMPROVE Network
L.-W. ANTONY CHEN, Judith Chow, John Watson, Bret Schichtel, Desert Research Institute
Abstract Number: 210 Working Group: Carbonaceous Aerosols in the Atmosphere
Abstract Decreasing trends of black carbon (BC), also termed elemental carbon (EC) as determined by thermal/optical analysis, have been reported at the U.S. IMPROVE (Interagency Monitoring of PROtected Visual Environ-ments) network from 1990 to 2004, consistent with the phase-in of cleaner engines, residential wood combustion technologies, and prescribed burning smoke mitigation practices. BC trends for the past decade (2000–2009) are examined in this study to address an upgrade of IMPROVE carbon instruments at 2005. The 2005 upgrade led to a transition from the IMPROVE to IMPROVE_A protocol. The new protocol does not change the temperatures plateaus but rather reflect the “actual” analysis temperatures that have been implemented since the inception of the IMPROVE network. It also enables simultaneous monitoring of filter reflectance and transmittance for charring correction. Filter attenuation determined from reflected light (T$_R) provides an independent measure of light absorption coefficient as well as BC on the filter. Digital thermograms of > 83 000 samples from 65 IMPROVE sites from 2000 through 2009 were reprocessed to obtain the T_R values. The changes in regression coefficients between BC and T$_R before and after 2005 were examined, and relationships between BC measured from the two periods (2000–2004 versus 2005–2009) were established. The variability is shown to be within ±10% in general with larger deviations occurring for high or low extreme BC levels. BC and T$_R trends assessed using a nonparametric test show universal decreasing trends across the U.S. The trends are well correlated, with national average downward rates (relative to the 2000–2004 baseline medians) of 4.5 %/yr for BC and 4.1 %/yr for T$_R. The consistency between BC and T$_R measurements adds to the weight of evidence that BC reductions are real rather than an artifact of changes to the measurement process.