Sensitivities of Stratospheric Aerosol Dispersal to Variations in Location and Timing

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Tuesday, 6 January 2015: 4:30 PM
212A West Building (Phoenix Convention Center - West and North Buildings)
Aaron L. Match, Cornell University, Ithaca, NY; and J. John and L. Horowitz

Stratospheric aerosols warm the stratosphere and cool the troposphere. They are injected naturally by major volcanic eruptions and have been proposed as a form of geoengineering by solar radiation management. Transport of these stratospheric aerosols by the residual circulation helps determine their climate impacts. We use the Geophysical Fluid Dynamics Laboratory (GFDL) Atmospheric Model 3 (AM3) to simulate the dispersal of idealized stratospheric aerosol clouds initialized at ten locations and in four different seasons. The aerosols are transported as passive tracers with convection and wet/dry deposition. Tracers injected in the upper troposphere are rained out with an e-folding mass decay time of 1-2 months, while tracers injected in the stratosphere have e-folding mass decay times of 20-45 months. The upwelling branch of the meridional overturning circulation, calculated using the Transformed Eulerian Mean, causes tropical injections to persist 12-30 months longer than polar injections. Seasonal changes in circulation cause the meridional transport of aerosols to depend strongly on the season of injection, resulting in instantaneous total tracer mass varying up to 30% of the initial value. Varying injection longitude has negligible effect due to rapid zonal mixing on a monthly timescale. This work supports conclusions from prior studies of geoengineering strategies and volcanic cloud dispersal modeling. Future work includes adding radiative dynamical feedbacks, gravitational settling, and parameterized aerosol size distributions.