Gravity wave events from mesoscale simulations, compared to polar stratospheric clouds observed from spaceborne lidar over the Antarctic Peninsula

摘要

We compare Gravity Waves (GW) and Polar Stratospheric Clouds (PSC) above the Antarctic Peninsula for winters (June to September) between 2006 and 2010. GW activity is inferred from stratospheric temperature and vertical winds from the Weather and Research Forecast mesoscale model (WRF), and documented as a function of time and geography for the studied period. Significant GW activity affects 36% of days and follows the Peninsula orography closely. Volumes of PSC, composed of ice and Nitric Acid Trihydrate (NAT), are retrieved using observations from the spaceborne lidar CALIOP (Cloud-Aerosol Lidar with Orthogonal Polarization). They are documented against GW activity as a function of time and longitude. Sixty-three percent of ice PSC are observed during GW events, when the average volume of PSC per profile doubles. Maximum ice PSC volumes are seen directly over the Peninsula (65°W), while maximum NAT PSC volumes appear downstream further East (～35°W). Effects of GW events on NAT PSC are felt as far East as 40°E. Our results support the importance of gravity waves as a major mechanism driving the evolution of ice PSC in the area, but the effects on NAT PSC are harder to detect. After a GW event ends, volumes of ice PSC get back to their usual levels in less than 24 h, while this process takes more than 48 h for NAT PSC. Daily profiles of H _2O and HNO _3 mixing ratios, retrieved from MLS observations, are used to derive ice and NAT frost points with altitude and time. Combining these frost points with modeled stratospheric temperatures, the volumes of air able to support ice and NAT crystals are quantified and compared with PSC volumes. Correlation is high for ice crystals, but not for NAT, consistent with their much slower nucleation mechanisms. Observations of ice PSC over the domain are followed by a strong increase (+50-100%) in NAT PSC formation efficiency 2 to 6 h later. This increase is followed by a steep drop (6-10 h later) and a longer period of slow decline (10-24 h later), at the end of which the NAT PSC formation efficiency is less than half its initial value. The fact that these effects tend to cancel each other out, coupled to the important lag in NAT PSC reaction to GW activity, suggest why it is especially difficult to quantify how GW activity impacts NAT PSC cover.