Seasonal impact of cloud nucleating aerosols on orographic snowfall

As part of an NSF-sponsored collaborative grant “Inhibition of Snowfall by Pollution Aerosols”, we have been performing simulations and observations to investigate the impacts of varying aerosol pollution amounts on precipitation. Sixty-day “seasonal” simulations were performed from January 1 – March 1 for the years of 2005, 2006, 2007, and 2008 with varying vertical profiles of CCN concentration. Maximum surface concentrations were varied from 100 to 800 to 1500 per cubic centimeter to represent a range from “clean” to “moderately polluted” to “highly polluted” environments, respectively. Simulation grids are nested down to 3km grid spacing to cover most of Colorado, with additional simulations at 1km grid spacing over the San Juan Mountain Range. The San Juan Range typically receives the greatest Colorado snowfall during a given winter season from particularly “wet” storms under southwesterly flow, and therefore, this area may be more susceptible to aerosol effects. Varying the model grid resolution and examining the differences in snowfall between simulations and observations aids in determining the optimal model configuration for resolving cloud scale aerosol effects and effectively representing observed accumulated snowfall.

The results show a significant response to varying aerosol amounts. Specifically, precipitation is found to decrease on the windward slopes and increase on the leeward slopes. This effect is most evident in the southern and western region of the San Juan Range where high moisture-laden storms are more prevalent. The aerosol-induced downwind precipitation shift over the Park Range is also present in each simulated season, but with lower amplitudes and slightly varying magnitudes among seasons. Seasons with greater overall snowfall exhibit a greater aerosol response in terms of both the magnitude of the change in total snow water equivalent as well as percentage change. The results are consistent with the findings of Saleeby et al. (2009) wherein higher CCN concentrations reduce ice particle riming rates and thus alter the seeder-feeder process. This effect requires the coexistence of a supercooled orographic cloud and snow falling from above and upwind. Snow crystals falling through a polluted cloud acquire very little rime, remain low density, have slower fall speeds, and thus have trajectories that more readily carry them toward lee slopes. Likewise, snow falling through a clean cloud may acquire substantial rime, increase in density, and transition into a graupel-like particle with higher fall speeds, reduced horizontal advection, and deposition on windward slopes. The shift in snow deposition due to aerosol loading can have serious hydrologic implications as snowpack is redistributed into neighboring river basins.

Decreasing the grid spacing from 3km to 1km in the simulations over the San Juan region leads to better resolution of the orographic cloud and topographic features. The 1km spacing simulations resulted in a 1.5% increase in resolved snowfall as well as a greater response to increases in CCN concentration. The 1km simulations experienced an increase in the areal impact and magnitude of the aerosol effect, likely due to a better resolved seeder-feeder effect along the major mountain ridges where orographic clouds are most prevalent.