Aeolus Mission & Atmospheric Dynamics

In most cases, satellite validation is based on statistical comparison with reference data. However, satellite and reference measurements do neither exactly match in time and space (mistime and misdistance) nor address the same volume of air (misintegration). Therefore, the natural atmospheric variability leads to differences between both data sets. These differences must not be interpreted in terms of a satellite’s malfunction.

Based on ECMWF ERA-40 temperature data, the expected differences due to mistime and misdistance are quantified depending on location, height and season. The results are compared to satellite and radiosonde based results. Regions of lower atmospheric variability which are optimal for satellite validation are identified.  

The misintegration effect is shown for stratospheric gravity waves which are extracted from satellite and radiosonde based temperature measurements over Europe. We point out how this effect can be used for the extraction of additional information about the horizontal orientation of gravity waves in the future.

At least, an internet based tool is presented which allows the user to retrieve information about the mean differences due to mistime and misdistance for his specific location. This tool is integrated in the Alpine Data Analysis Centre (AlpenDAC) which is currently under construction and will be available via the environmental station Schneefernerhaus (UFS, www.schneefernerhaus.de), Germany.

Planetary waves (PW) are global scale waves in the atmosphere, which are well-known to considerably impact weather patterns in the middle latitudes. It is widely accepted that climate change leads to a change of the meridional temperature gradient. This should, in turn, change the planetary wave activity (PW-activity); the mid-latitudinal weather patterns are therefore expected to change. Moreover, planetary waves redistribute ozone in the stratosphere and wave breaking can lead to ozone streamer events. As such streamers are characterized by significantly reduced ozone column concentrations. They do not only impact the UV-radiation intensity which reaches the surface and which has consequences for human health. The reduced absorption of UV radiation also impacts the solar heating of the atmosphere which in turn has consequences for atmospheric dynamics.

In order to find out if the PW-activity has already changed and to observe streamer events, we use satellite based ozone total column measurements (GOME) and ERA–Interim temperature data (0 and 65 km height). We derive a measure for the PW-activity, the so-called dynamical activity index (DAI).

A general increase of the PW-activity in the stratosphere is evident from our analysis. Using the empirical mode decomposition (EMD) we are able to extract nonstationary signals of the time series. We find that longer term oscillations (QBO, ENSO, solar cycles) have a noticeable impact on the wave activity variability in all considered heights. This long-term modulation affects the overall trend of planetary wave activity. First preliminary results indicate that planetary waves favor breaking at the North Atlantic / European region. Accompanied, we observe a high frequency of ozone-streamer events in this area. As a consequence of the increase of PW-activity, the streamer frequency should also have changed.

Aeolus wind measurements will help to further study planetary wave breaking and characterize the impact on the European ozone concentration in unprecedented detail.

The Gimballed Limb Observer for Radiance Imaging of the Atmosphere (GLORIA) was deployed in January 2016 on board of the German research aircraft HALO to observe gravity waves.  Two research flights dedicated to investigate gravity waves were conducted in the frame of the ESA study GWEX (Gravity Wave EXperiment).  GLORIA observes thermal infrared emissions and has the unique capability to view the same volume of air under different angles. In this way, the 3D temperature structure of the atmosphere can be recovered using tomographic retrievals.  These 3D temperature fields are used to identify gravity waves inside the measurement volume and determine the 3D wave vector, infer gravity wave momentum flux and conduct propagtion studies. The first event investigated during GWEX was a complex superposition of mountain waves above Iceland, measured on 25-January-2016. Unexpected strong oblique propagation was observed. The second event occured on 28-January-2016 above Scandinavia and encompasses gravity waves from orography as well as from spontaneous imbalance. The waves propagated into the upper stratosphere, where they also could be observed by the AIRS nadir sounder, however with much reduced amplitudes and too long vertical wavelengths due to the observational filter of the nadir viewing geometry. For both events we compare the gravity wave signatures deduced from the temperatures with in situ wind measurements taken by the HALO on-board sensor system. In particular the second event is a text-book case for the polarization relations connecting the temperature and wind signatures in a gravity wave. Both cases, however, also highlight the challenges that are encountered in inferring gravity waves from wind observations. Temperature structures emphasize in particular mesoscale gravity waves, whereas vertical winds emphasize waves of very short horizotal wavelengths. In order to compare the two observations, vertical winds have to be filtered accordingly. The challenge for the horizontal wind is the separation between gravity wave signatures and the structure of the background flow in the UTLS region. The implications of these findings for analyzing global wind observations from Aeolus will be discussed.

The potential impact of the Aeolus horizontal line-of-sight (HLOS) winds has been extensively studied by using the global ECMWF variational data assimilation system. In preparation for the use of HLOS winds in mesoscale NWP systems in Europe, we have developed a mesoscale ensemble Kalman filter (EnKF) data assimilation system nested in the 50-member operational ECMWF ensemble prediction system (ENS). The applied Weather Research and Forecast model (WRF) and the ensemble adjustment Kalman filter (DART/WRF) over Europe and North Atlantic was used to carry out a series of the observing system simulation experiments (OSSEs) with the HLOS winds.

The impact of HLOS winds in the assimilation depends on the direction of sight with respect to the observed flow and on the assimilation methodology which spreads the impact of HLOS observations to the two wind components and other analyzed variables using the background-error covariances.

The comparison of OSSE experiments with a single wind component and full wind information and temperature data showed that the information content of HLOS winds is on average divided linearly between the zonal and meridional wind components depending on the observation azimuth. The assimilation of wind information in terms of the HLOS component may produce better analysis than the assimilation of the zonal and meridional winds in individual cases of interest for mesoscale prediction, as demonstrated in the case for fronts in the Northern Atlantic. Multivariate aspects of EnKF also provide significant analysis increments, especially if the HLOS observation is aligned along the front.  

Aeolus does not provide horizontally high resolution observations, but the processing software Level-2B allows tuning of the accumulation length which controls the accumulation of measurements at about 3 km resolution to a single HLOS profile observation. Extensive sensitivity experiments both the Rayleigh and Mie retrieval suggest a promising impact of the Mie retrieval with reduced accumulation length for the purpose of mesoscale model assimilation.  

Gravity waves are oscillations in wind velocity and temperature with buoyancy as a restoring force.  They transport energy and momentum through the atmosphere. In this way, they couple different compartments with each other and influence prominent circulation patterns in the stratosphere and mesosphere. Within the ESA study GWEX (Gravity Wave Experiment), an aircraft campaign has been performed in winter 2015/2016 to study the propagation of gravity waves. During this campaign, the first 3D tomographic measurements of gravity waves were taken with the infrared limb imager GLORIA (Gimballed Limb Observer for Radiance Imaging of the Atmosphere). The 3-D nature of the GLORIA measurements enables the determination of 3-D wave vectors, including the horizontal directions and orientation resolved momentum fluxes. This full wave characterization enables the use of ray-tracing models to identify the gravity wave sources and study their propagation.

A gravity wave event above southern Scandinavia has been probed with GLORIA on the 28 January 2016. This event is characterized by a complex situation where gravity waves from two different source mechanisms, orography and jet excitation, propagate through each other. This crossing of the waves is investigated using the GLORIA measurements, ray-tracing studies and ERA5 reanalysis data. The mountain waves propagate more or less straight upwards, whereas the jet-generated waves propagate more horizontal. The linear ray-tracing results show an excellent agreement with ERA5.

In the middle stratosphere, the results of both models are compared to measurements of the Atmospheric Infrared Sounder (AIRS) satellite instrument. Due to the relatively short vertical wavelengths of less than 15 km at 35 km altitude and the coarse vertical resolution of nadir viewing satellite instruments, the wave signatures are strongly suppressed in AIRS temperatures.  However, not only the amplitudes of the waves are underestimated, also the vertical wavelengths are overestimated. Only by including an extended observational filter for the AIRS instrument into the comparison good agreements to the models are achieved.

A growing concern about the global air quality, as well as the risks connected to the changing climate has led to a large increase in satellite remote sensing of atmospheric composition which provides observations of the  vertical aerosol profiles (in form of aerosol extinction coefficients). The amount of data will be further increased with the recent launch of the Aeolus satellite and the forthcoming launch of EarthCARE satellite. Meanwhile, the weather prediction problem has changed from forecasting  the basic meteorological variables to predicting also the atmospheric composition that requires description of aerosol dynamics in the NWP models. These two factors provide a strong motivation to explore the potential of tropospheric aerosols as the carriers of wind information.

We present a new model of intermediate complexity to simulate the aerosol, moisture, temperature and wind interactions  in 4D-Var. The model has been employed  for studying the wind tracing using aerosol data  using the OSSE type of  experiments.  We shall present numerical modelling results  that provide the upper bound of estimated usefulness of the aerosol observations in the wind tracing. It is shown that the wind tracing in 4D-Var strongly depends on the spatial density and accuracy of the tracer observations as well as the frequency of observation update and the assimilation window length.  The first two are needed to describe the spatial gradients of tracers and the last two provide information about the advection. In the case of linear flow, the observation spatial sampling is much more important than their update frequency. For the nonlinear flow, the opposite applies. The desired spatial resolution of observations is mainly a function of the flow properties; the faster the flow, the less dense tracer observations are needed to deduce the winds.

 An important goal of further research is to estimate the wind tracing potential in the case of unknown (unrepresented) sources. First results reveal a linear dependence of the wind analysis accuracy on the ratio of the local tracer source rate and the strength of the advection.

Furthermore, ongoing research deals with the wind tracing in a saturated atmosphere. We illustrate how nonlinearities of moist processes near saturation in the combination with the aerosol wet deposition and  couplings between the aerosol, moisture and dynamics have a detrimental impact on wind tracing from aerosol data.

With the successful launch of Aeolus, expectations are raised with respect to high-quality Aeolus products and their beneficial application in weather forecasting, climate research and atmospheric sciences. In the presentation the planned Aeolus product development and Cal/Val will be briefly described and put in the context of the global observing system and of atmospheric modelling capabilities, in particular with respect to weather, atmospheric circulation, transport and clouds.