Lessons learned from simulating SOA formation in the Manaus plume with a detailed organic chemistry mechanism

Description

The GoAmazon field campaign was carried out in the vicinity of Manaus (Brazil) in 2014 and 2015. This metropolis of two million inhabitants is isolated in the middle of Amazonia and was identified as the ideal place to study interactions between pristine air masses and anthropogenic pollution. Using the Generator for Explicit Chemistry and Kinetics of Organics in the Atmosphere (GECKO-A), we ran a test case based on estimated Manaus and Amazon forest emissions. A boxmodel was used to simulate the fully explicit oxidation of 65 primary organic compounds with 15 million reactions involving 4.5 million species, of which 400,000 are able to condense into secondary organic aerosol. The model generally overestimated SOA mass by a factor of 5 to 10. Moreover, field measurements saw an increase in SOA mass by a factor of 2 when air masses were influenced by the city of Manaus. The model is unable to reproduce this trend, calculating an SOA mass decrease by 25% under high NOx conditions. These discrepancies must arise from processes not yet taken into account in the fully explicit chemical mechanism. In this work we therefore try to identify these processes that will need to be the subject of further research to allow their implementation in GECKO-A.