Role of gravity waves in the Southern Hemispheric circulation and climate
WASCLIM - a ROMIC II Project (2019-2022)
WASCLIM is a project of the research initiative Role Of the Middle atmosphere In Climate (ROMIC-II)
ROMIC-II (project call in German) is the follow-up of the research initiative ROMIC. The two central research questions addressed in ROMIC-II are:
A. How does the variability of the sun induce changes in the tropospheric climate via changes in the circulation, physical processes, and chemical composition of the middle atmosphere?
B. How does tropospheric forcing, in particular anthropogenic influences, feed back to the troposphere and change the climate by its impact in the middle atmosphere?
Atmospheric Gravity Waves
Convection, fronts, flow over mountains, and spontaneous adjustments occurring at the tropospheric jet streams generate vertically propagating gravity waves (GWs) in the troposphere and
lower stratosphere (Smith 1979, Gill 1982, Baines 1995, Fritts and Alexander 2003, Sutherland 2010,
Nappo 2012, Plougonven and Zhang 2014). Generally, we differentiate between orographic GWs and
non-orographic GWs (OGWs and NOGWs). Through their far-field interactions, GWs constitute an
important coupling mechanism in the Earth’s atmosphere and the momentum deposition is described
as orographic and non-orographic GW drag (OGWD, NOGWD), respectively. The associated redistribution of momentum and energy controls the global middle atmospheric circulation (Dunkerton
1978, Lindzen 1981, Alexander et al. 2010).
The Problem: missing gravity wave drag at 60° S
Comparisons of stratospheric zonal-mean absolute GW momentum fluxes (GWMFs) from global circulation model (GCM) predictions and global satellite data reveal a large variability between currently used GW parametrizations. Some features are reproduced reasonably well in terms of magnitudes and the latitudinal
dependence, but there are still large deviations. Particular deviations are the "gap" in simulated
zonal mean GWMFs near 60° S in winter (Geller et al. 2013) and the large simulated
maxima above mountainous regions which are not observed by satellites such as SABER (Sounding
of the Atmosphere using Broadband Emission Radiometry) and HIRDLS (High Resolution Dynamics
Limb Sounder) (Geller et al. 2013). Most GCMs exhibit large systematic biases in
the simulation of the Southern Hemisphere (SH) polar stratospheric circulation. These biases are known as the "cold pole
bias" occurring, e.g., in the Canadian Middle Atmosphere Model (CMAM) or the Whole Atmosphere
Community Climate Model (WACCM) (SPARC CCMVal 2010, Butchart et al. 2011) as they underestimate the polar stratospheric temperatures. Due to isolation of the cold stratospheric air and
the weak planetary wave activity, there are significant late-biases in the simulated spring-time breakups of the
SH vortex. Most likely, the simplified representations of GWs is a major cause of the biases in the
simulation of SH stratospheric climate.
Summarizing and referring to a statement by T. G. Shepherd, the problem of parameterizing unresolved GWs and their sensitive interaction with large-scale dynamics is among the most uncertain
aspects of climate modeling (Shepherd 2014).
OGWs likely play an important role but also other processes could explain the "missing drag at
60° S" (de la Cámara et al. 2016).
Generally, the physical origin of the observed belt of enhanced GWMF around 60°
S can be explained
by different mechanisms: (i) The downwind advection and meridional refraction of
OGWs from the
Southern Andes and the Antarctic Peninsula into the PNJ (Sato et al. 2009). (ii) Unresolved OGWs
from small islands (Alexander et al. 2009, Alexander and Grimsdell 2013). (iii)
The generation of secondary GWs generated locally in the breaking region of these
primary OGWs (Hindley et al. 2015). (iv) NOGWs
from sources associated with winter storm tracks over the Southern Ocean (Hendricks et al. 2014).
(v) a zonally uniform distribution of small amplitude waves from non-orographic mechanisms
such as spontaneous adjustment and jet instability around the edge of the stratospheric PNJ (Hindley
et al. 2015).
The project WASCLIM investigates the impact of GWs on the SH
stratospheric circulation and tropospheric climate by means of field observations, high-resolution
numerical modelling, the analysis of meteorological reanalyses, and climate modelling. The
development and implementation of a simple parametrization for the horizontal propagation of
GWs, and, eventually, climate sensitivity simulations will quantify the resulting effects on
tropospheric climate and climate change. In the project, long-term ground-based measurements with autonomous temperature-lidar
CORAL (Compact Rayleigh Autonomous Lidar) at the southern tip of South America are continued. They are complemented by episodic
airborne measurements, amongst others, with the new ALIMA (Airborne Lidar for Studying the Middle Atmosphere) instrument and the infrared limb imager GLORIA (Gimballed Limb Observer for
Radiance Imaging of the Atmosphere) onboard the German research aircraft HALO (High Altitude and Long Range Research Aircraft) and ground-based meteor radar observations.
The region of the Andes and the Antarctic
Peninsula is a hotspot of stratospheric GW activity. A data set for different GW sources will be
assembled and high-resolution ICON (= global forecast model of the German Weather Service) simulations will complement observational data in a consistent way. This data set together with the drag
climatology derived from the reanalyses will be used to develop a source-transfer
parametrization mimicking the horizontal GW propagation. Moreover, ICON
simulations will be used to evaluate its performance in EMAC (= global chemistry
climate model of German research institutions) by means of nudged
EMAC simulations for the specific episode of the aircraft campaign. The
statistical mean representation of GWs in the modified EMAC model will be conducted based on
Latest news of the project (updated 2. Sept 2021)
♦ upcoming: A virtual scientific workshop on "gravity waves in the Southern Andes region" is scheduled for 28.-30. September 2021.
♦ The ROMIC-II Kick-off meeting was held on 26. February 2021 (virtual meeting).
♦ The virtual SouthTRAC-Gravity Waves meeting was held on 26./27./28. October 2020.
♦ The first SouthTRAC UTLS Chemistry and Transport Workshop was hosted virtually on 22./23. September 2020.
♦ The first data workshop for the gravity wave part of the SouthTRAC campaign took place on 20./21. January 2020 at DLR in Oberpfaffenhofen.
♦ Airborne observations were sucessfully completed in the framework of the SouthTRAC campaign in September to November 2019. For more details, visit the SouthTRAC website.
The project tasks are grouped into five scientific work packages and one workpackage coordinating their entity.
In this work package, ground-based and airborne observations are organized
and conducted in the region of the Southern Andes, the Drake Passage, and the Antarctic
Peninsula. The aim is the detailed probing and quantification of GW
activity and GWMFs in the troposphere and middle atmosphere. Ground-based remote sensing and
radiosonde observations at the southern tip of South America focus on OGWs and their temporal
development. Mesospheric wind measurements are relevant to quantify tidal motions and
Doppler shifting of ground-based GW measurements. Remote-sensing and in-situ measurements onboard HALO extend the observational volume
allowing the investigation of NOGW sources and the horizontal propagation of OGWs and NOGWs,
i.e., the spatial structure and extent of the GW field. The airborne measurements took place in the framework of the SouthTRAC campaign in
September to November 2019. Key instruments on HALO were the upward looking Rayleigh lidar ALIMA covering the middle atmosphere/lower
thermosphere, and the infrared limb imager GLORIA covering the upper troposphere and lower stratosphere region. This unique combination of two
cutting edge instruments offers new insights in the complex GW structures in the atmosphere.
WP2: Mesoscale modelling
In this work package, numerical simulations of orographic and non-orographic
GW events are performed using the global ICON model of DWD (Zängl et al. 2015). Locally enhanced
model resolution of around 1 km can be realized by nesting directly in ICON and allows
ICON to resolve the GW spectrum to a large part. Temporal evolution and spatial distribution of
GWMF can be computed from ICON simulations and justified against the observations and their analysis. This is done for the global ICON data which rely on orographic and non-orographic GW
parameterizations, and for the highly resolved nested data providing reliable 4D (time and space) GW
information. In this manner, the ICON data will help to establish a connection
of the tasks and results in the simplified model world (WP3, WP4 and WP5) to real world observations
WP3: GW fluxes and associated mean flow changes in reanalyses
The general goal of this work package is to provide a bridge between
the detailed GW analyses from the field campaign and the general representation of GWs
in climate models (WP5). Such a bridge is necessary to transform the knowledge gained by the
case study as part of the field campaign to a statistical characterisation needed to improve the GW
parameterisation in climate models. To do so, we will quantitatively assess GWMF and the associated
forcing of the mean flow in the SH polar stratosphere based on several ECMWF reanalysis products,
as well as operational analyses. Missing and/or misrepresented drag will be inferred both directly from
analysis increments due to data assimilation, as well as indirectly from the residual of the momentum
budget. The results are expected to both put the field observations into the perspective
of the past record, as well as provide a baseline for the improved GW parameterisation in EMAC
based on observationally-constrained global model runs.
WP4: Budgets of lateral propagation for GWs from different sources: development of a parametrization
In this work package, we seek an efficient way to represent horizontal GW propagation in a GCM
and thus reproducing the high GWMF in the SH PNJ by more realistic processes than in previous
studies. The GWMF contained in the PNJ is given by the strength of the source, the horizontal
and vertical propagation path and whether it reaches the PNJ, and the dissipation and filtering
which the GWs experiences. In a conventional parametrization implemented in a GCM, parametrized
GWs propagate only vertically. Ray-tracing handles horizontal propagation. However, in order to
significantly reduce the high computational cost that comes along with climate simulations, we follow
a different approach. Our approach is to perform ray-tracing offline,
evaluate the results statistically and compile look-up tables. For each orographic source, these will
provide a probability distribution µ that expresses statistically to which locations at an evaluation
height GWs propagate. In other words, we will spread GWs from a given source to all locations in
the lower stratosphere where µ is positive.
WP5: Global Modelling with EMAC
In this work package, the aim is to improve the representation of GWs in global models using observationally based results from the other work packages. GWMF and drag from observations, high-resolution
modelling, and reanalyses will serve as basis for the evaluation of GW activity in the global
model. While high-resolution model data can be compared to nudged EMAC simulations for the
specific episode of the campaign, the statistically mean representation of GWs in the model can only
be evaluated with long-term data of GW forcing as will be provided from reanalysis data.
The simplified horizontal propagation parameterization developed in WP4 will be implemented
in EMAC. The impact of GW modifications on the simulation of the climatological mean circulation
of the stratosphere, the response of stratospheric circulation to anthropogenic
forcing, and finally, the
dynamical downward coupling of stratospheric circulation change to the tropospheric climate will be
In particular, we will focus on the problem of the current "gap" in GWD around 60° S by means
of 1) comparison to observational data to identify the nature of the "gap" in more detail, and 2)
dedicated sensitivity studies with respect to modifications of the GW parametrizations. We aim
to evaluate the hypothesis that the missing GWD at 60° S in the stratosphere is due to neglecting
horizontal propagation of OGWs. As alternative hypothesis, we test whether additional sources of
GWs (orographic or non-orographic) around 60° S are needed to explain the missing GWD at 60° S.
Thereby, we will investigate whether similar effects on the mean stratospheric circulation are obtained
by altering GW propagation (re-distribution of GWD with same sources) or sources.