Wednesday, 13 January 2016: 5:15 PM
Room 357 ( New Orleans Ernest N. Morial Convention Center)
The main goal of the GoAmazon 2014/15 experiment is to measure and understand the factors affecting aerosol particles over a tropical rain forest, especially the effects of anthropogenic pollution plume from large metropolitan areas as a perturbation to natural state of the pristine forest surrounding it. For helping the interpretation of data measured at the ground sites, we have performed HYSPLIT backward simulations, starting every 30-min from Jan-1st until Dec-31st 2014, for all the GoAmazon sites. Boundary conditions were taken every 3h at 1 x 1o resolution from the Global Data Assimilation System (GDAS) available from NOAA. We also experimented with the available 0.5 x 0.5o resolution, but no significant improvement was found, possibly due to the lack of vertical velocity fields, which hinders the model's ability to simulate the vertical mixing. An index for “in-plume” and “off-plume” events was built by selecting backward trajectories that started from T3 and passed over T2 / Manaus. We focus our analysis in the wet season to avoid having to disentangling the biomass burning transport. The transport time from T2 to T3 was found to be 7.5 ± 1.5 h on average. CN decreases from 5.0±2.5k to 2.3±1.3k # cm-3, while changes in black carbon (from 0.9±0.7 to 0.7±0.6 ug m-3), and CO (145±36 to 128±26 ppbv) are not as significant. At the same time, absorption decreases from 6 to 4 Mm-1 and the angstrom absorption exponent increases from 1.35 to 1.6, indicating a contamination by the brick factories between the two sites. On the other hand, scattering remains constant (8.5±1.5 Mm-1), while the scattering angstrom exponent slightly decreases from 1.2 to 1.1. We will discuss the possible physical mechanisms to explain these changes.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner