Free troposphere as the dominant source of CCN in the Equatorial Pacific boundary layer: long-range transport and teleconnections

Airborne aerosol measurements in the central equatorial Pacific during PASE (Pacific Atmospheric Sulfur Experiment) revealed that cloud condensation nuclei (CCN) activated in marine boundary layer (MBL) clouds were dominated by entrainment from the free troposphere (FT). About 65% entered at sizes effective as CCN in MBL clouds, while 25% entered the MBL too small to activate but subsequently grew via gas to particle conversion. The remaining 10% were inferred to be sea-salt aerosol; there was no discernable nucleation in the MBL. FT aerosols at low carbon monoxide (CO) mixing ratios (< 63 ppbv) were small and relatively volatile with a number mode around 30–40 nm dry diameter and tended to be associated with cloud outflow from distant deep convection (3000 km or more). Higher CO concentrations were commonly associated with trajectories from South America and the Amazon region (ca. 10 000 km away) and occurred in layers indicative of combustion sources partially scavenged by precipitation. These had number mode near 60–80 nm diameter with a large fraction already CCN.2 (those activated at 0.2% supersaturation and representative of MBL clouds) before entrainment into the MBL. Flight averaged concentrations of CCN.2 were similar for measurements near the surface, below the inversion and above the inversion, confirming that subsidence of FT aerosol dominated MBL CCN.2. Concurrent flight-to-flight variations of CCN.2 at all altitudes below 3 km imply MBL CCN.2 concentrations were in quasi-equilibrium with the FT over a 2–3 day time scale. This extended FT transport over thousands of kilometers indicates teleconnections between MBL CCN and cloud-scavenged sources of both natural and/or residual combustion origin. The low aerosol scattering and mass in such layers results in poor detection by satellite and this source of CCN is not represented in most current models. The measurements confirm nucleation in the MBL was not evident during PASE and argue against the CLAW hypothesis being effective in this region during PASE.