Date of this Version
Aerosol Science and Technology, 38:1080-1090, 2004
During a campaign to study ozone loss mechanisms in the Arctic stratosphere (SOLVE), several instruments on NASA's ER·2 aircraft obser ved a very low number density (0.1 I-I) of large, nitric-acid-containing particles that form the polar stratospheric clouds (PSCs). For effective physical and chemical characterization of these particles, the measurements from these instruments have to be intercompared and integrated. In particular, proper interpretation requires knowledge of the sampling characteristics of the particles into the instruments. Here, we present the calculation of the sampling characteristics of the one of the instruments on the ER-2, the NOAA NOy instrument. This instrument sampled ambient particles and gas from two forward-facing inlets located fore and aft on a particle-separation housing (the football ) and measured total NOy in the sample. In recent studies, ambient aero sol mass has been estimated by the difference of the measurements of the two inlets with the assumption that the rear inlet observation s represent the gas-phase NOy and small particles and the front inlet samples represent gas-phase NOy and all particle sizes with varied efficiency (anisokinetic sampling). This knowledge was derived largelyfrom semiempirical relations and potential flow studies of the housing. In our study, we usedCFD simulations to model the compressible flow conditions and considered noncontinuum effects in calculating particle trajectories. Our simulations show that the blunt body housing the inlets has a strong and complex interaction with the flow and particles sampled by the two inlets. The simulations show that the front inlet characteristics are influenced by the effect of the blunt body on the upstream pres sure field. The rear inlet sampling characteristics are influenced both by the shape and size of the inlet and its location on the blunt body.These interaction s result in calculated inlet characteristics that are significantly different from previously assumed values. Analysis of the SOLVE data, considering the ambient conditions and the calculated inlet sampling characteristics, in conjunction with thermodynamic growth modeling of super-cooled ternary solution (STS) particles, provides validation of the CFD results.