613 Seminar Series: Manisha Ganeshan

USRA/Climate & Radiation Laboratory

In the first part of this talk, the surface stability over Antarctica in the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA2) is evaluated against in-situ soundings. Recent high-resolution dropsonde observations from the 2010 Concordiasi field campaign in austral spring, reveal that wind-driven turbulence can erode surface-based inversions (SBIs) resulting in 33% and 18% of well-mixed boundary layers in West and East Antarctica, respectively.  Overall, MERRA2 has a good representation of surface stability although there are biases that contribute to regional differences. In MERRA2, development of unstable lapse rates through turbulence is favored during conditions of relatively weak surface radiative cooling (aided by shortwave and/or cloud-radiative forcing). As a result, the occurrence of mixed layers is frequently underpredicted in the cool and dry elevated regions of the East Antarctic plateau. A lower tropospheric cool bias exists in the model that tends to become neutralized (amplified) at lower levels during SBIs (mixed conditions) as a result of stronger (weaker) coupling with skin temperature. Even though the model does a better job in predicting mixed layers in West Antarctica, the repercussions of misrepresenting the surface stability are more severe in this region due to a stronger lower tropospheric cool bias. In particular, failure to predict a mixed layer can lead to substantial (greater than 15 °C) negative surface temperature errors in this region. Knowledge of the Antarctic atmospheric boundary layer structure thus remains critical for modeling the local energy budget and surface mass balance. 

 In the second part of this talk, the representation of an Antarctic polar low is investigated in high-resolution (quarter degree) GEOS analysis. Polar lows are convectively driven, warm-core, mesoscale cyclones that form in the poleward side of the jet stream, fueled by turbulent heat exchange between the ice-free ocean and cold atmosphere. Due to their rapid growth at sub-synoptic scales in regions with very few conventional observations, prediction of polar lows remains a challenge for Numerical Weather Prediction. Here, the representation of an Antarctic polar low during the austral spring of 2014, identified using AMSU-B brightness temperatures, is investigated in the analysis of two experiments viz., control using clear-sky AIRS radiances, and an observing system experiment (OSE) using adaptively thinned and cloud-cleared AIRS radiances. The representation of the system is substantially better in the OSE due to improved vertical alignment, vorticity, and structure of the warm-core.


Seminar Series Coordinators