An Update on the Impact of Aeolus HLOS Winds in Numerical Weather Prediction at ECMWF
Michael Rennie | European Centre for Medium-Range Weather Forecasts (ECMWF) | United Kingdom
Aeolus was launched over three and a half years ago and the L2B winds have been operationally assimilated at ECMWF for over two years.
The latest results from the assessment of the impact of L2B HLOS winds in ECMWF’s global NWP Prediction system will be presented. In particular, the impact with the second reprocessed Aeolus data from July 2019 - October 2020, the early part of which Aeolus winds had the largest signal-to-noise ratio and hence largest positive impact. This early FM-B laser period helps to give an impression of what could at least be achieved with a potential operational ESA/EUMETSAT Doppler Wind Lidar follow-on mission, in which significantly better SNR than Aeolus is sought.
In Observing System Experiments (OSEs), Aeolus provides statistically significant improvement of a good magnitude in short-range forecasts as verified by observations sensitive to temperature, wind and humidity, peaking at ~200 hPa in the extratropics and ~150 hPa in the tropics. Longer forecast range verification shows positive impact strongest at the 2-3 day forecast range, e.g. ~2% improvement in root mean square error for vector wind and temperature in the tropical upper troposphere and lower stratosphere and polar troposphere. Positive impact of up to ten days is found in the tropical lower stratosphere wind and temperature. This impact appears to be larger with the 2nd reprocessed versus the 1st and the NRT datasets; even some impact on the 500 hPa geopotential height in the N. Hemisphere is found. Experiments with variational QC modifications and a bias correction of the Rayleigh-clear winds as a function of temperature will be presented if time permits.
The operational Forecast Sensitivity Observation Impact (FSOI) metric shows that Aeolus is a useful contribution to the global observing system; with the Rayleigh-clear and Mie-cloudy winds providing similar overall short-range forecast impact in 2020-2021. Relative FSOI for the 2019 reprocessed dataset shows that Aeolus is amongst the most important satellite instruments; a good result for a demonstration mission. The relative FSOI is reduce to ~2% in late 2021 versus ~5% when the maximum atmospheric path signal was available in July 2019 (offline testing) due to SNR decreasing. Also, Aeolus winds have been absent for large fractions of 2020 and 2021 due to being flagged invalid (blocklisted) during instrument testing (aimed at understanding/mitigating the signal loss).
Tropical analysis and forecast uncertainties: a process-based evaluation of observing-system assimilation experiments with Aeolus winds
Prof. Dr. Nedjeljka Žagar | Universität Hamburg, Meteorological Institute, CEN | Germany
The tropics are the region with the largest uncertainties in the initial states for numerical weather prediction (analyses). Analysis uncertainties are largest in the tropical upper troposphere and the lower stratosphere (UTLS). One of the reasons is a lack of wind profiles which are more useful than temperature profiles in the tropics. This classical dynamical effect was described by J. Smagorinsky as “Not all data are equal in their information-yielding capacity. Some are more equal than others.”
Despite of their relatively small number and a relatively large random error in the Rayleigh channel, Aeolus wind profiles have a positive impact on the quality of global weather forecasts, with the maximal positive impact in the UTLS. Here we discuss one process contributing to the forecast improvements in the ECMWF model: the vertically-propagating Kelvin waves which are a major contributor to tropical variability.
Previous work showed that short-range forecast errors project on Kelvin waves significantly more in the easterly phase of the quasi-biennial oscillation (QBO) than in the westerly phase. Furthermore, it was shown that missing variance associated with Kelvin waves explains a large part of underdispersiveness of the ECMWF ensemble prediction system in the medium range in the tropics. It is unclear how well Kelvin wave dynamics is represented in global climate models, as they still poorly simulate the QBO and Madden-Julian oscillation, and their connections.
By filtering the Kelvin waves from the ECMWF analyses with and without Aeolus winds, we demonstrate that Aeolus alters the vertical structure of Kelvin waves in the layers with the strongest shear within UTLS. Changes in the Kelvin wave zonal wind within UTLS by Aeolus data can be up to +/-4 m/s. Similar to the Kelvin wave dynamics, the impact of assimilated Aeolus winds varies with periods 1-3 weeks. The coupling between the phase of QBO and the Aeolus winds is argued to happen through the Kelvin waves. The studied period May-September 2020 was characterised by a weakening easterly phase of the QBO. We suggest that a greater improvement of ECMWF forecasts in the tropical tropopause layer by Aeolus winds in May 2020 than later in summer 2020 was associated with the state of the QBO.
The impact of Aeolus wind observations on the predictability of extreme weather events
Dr. Giovanna De Chiara | ECMWF | United Kingdom
ESA’s Doppler wind Lidar mission, Aeolus, is an important new satellite observing system. It provides global coverage of wind profile information from the surface to 10 hPa. ECMWF was the first numerical weather prediction (NWP) centre to operationally assimilate the Aeolus horizonal line of site (HLOS) wind information on January 9, 2020. Impact studies have proven that the assimilation of Aeolus wind observations improves the quality of NWP forecasts of wind, temperature, and humidity, particularly in the upper troposphere and lower stratosphere, with the main improvements seen in the Tropics. This shows that Aeolus provides a valuable dataset that complements the existing Global Observing System (GOS).
The quality of the Aeolus HLOS wind information has evolved since launch. Reduction in instrument performance has caused degradations. This has been compensated by improvements in the Aeolus ground processing software and reprocessing activities. To achieve a long timeseries of high-quality Aeolus wind measurements, the Aeolus DISC (Data, Innovation, and Science Cluster) carried out a second reprocessing campaign that provided a new, high quality dataset covering the period from July 2019 to October 2020.
An ongoing ESA-funded project at ECMWF is investigating if the assimilation of Aeolus L2B winds improves the predictability of high impact and extreme weather events. The focus is on tropical cyclones and European extra-tropical storms. Also, the impact of Aeolus wind assimilation on the European forecast bust statistics is under evaluation. For this purpose, NWP observing system experiments with/without Aeolus L2B winds are being performed using the second reprocessed dataset and the operational data. A set of European and tropical severe weather events from July 2019 until the end of 2021 has been identified using various databases. The impact of Aeolus wind assimilation on these case studies and on the occurrence of forecast busts will be evaluated.
In this presentation the preliminary results will be presented and discussed.
Planetary wave dynamics with respect to streamer events: Lidar measurements to Identify Streamers and analyze Atmospheric waves (LISA) (Aeolus+Innovation)
Dr. Lisa Küchelbacher | German Aerospace Center
The jet stream represents a meridional barrier for air masses, but also energy fluxes. So-called "streamer events" in the upper troposphere / lower stratosphere are an example of how this barrier can be disrupted. During such events large-scale air masses from lower latitudes are irreversibly mixed into the circulation at higher latitudes with various consequences for atmospheric chemistry, energy and momentum balance. Streamers are the consequence of poleward planetary wave breaking, which modulate the jet stream and thus effecting the exchange of air masses and energy between the equator and the pole. Aeolus wind measurements allow the derivation of atmospheric wave structures on different temporal and spatial scales in particular above the oceans, where wind measurements from ground-based instruments are sparse and streamer events most likely occur.
We use Aeolus L2B wind measurements to derivate the planetary wave activity, so-called dynamical activity index (DAI). The DAI represents the mean amplitude of planetary waves up to wavenumber 10 in the mid-latitudes. A comparison of the DAI based on the Aeolus data and ERA-5 reanalysis data is presented. First case studies are shown addressing the structure of streamers and their relation to planetary wave breaking. Due to planetary wave breaking gravity waves might be excited at the flanks of streamers. The flanks of the streamers are characterized by comparatively strong wind shear. To identify the flanks of the streamer, and so possible source regions for streamers, we calculate the wind gradients along the track.
Gravity wave and PSC observations with ESA’s Aeolus satellite
Isabell Krisch | DLR - Deutsches Zentrum für Luft- und Raumfahrt e.V. | Germany
Since its launch in 2018, ESA’s Aeolus satellite provides global height resolved measurements of horizontal wind in the troposphere and lower stratosphere. It carries the world’s first spaceborne high spectral resolution wind lidar, the Atmospheric Laser Doppler Instrument (ALADIN). With its high-power laser, which operates at a wavelength of 354.8 nm, ALADIN can acquire measurements from roughly 30 km altitude down to either the ground or to the highest optically thick cloud layer. Besides the height resolved wind profiles, Aeolus also provides information on optical properties of clouds and aerosols.
The main objective of Aeolus is to improve numerical weather prediction (NWP). Multiple NWP centres have already shown the positive impact of Aeolus data and started its operational assimilation. However, detailed wind information is not only beneficial for NWP, but also for atmospheric dynamics research. Many dynamic features show characteristic wind patterns and/or are strongly influenced by the prevailing wind. Especially in the upper troposphere / lower stratosphere region, Aeolus measurements can provide valuable information for the investigation of dynamic features such as for example gravity waves.
In this study, we will highlight the use of Aeolus data to investigate the vertical change in gravity wave activity throughout the upper troposphere and lower stratosphere. This is of particular relevance for understanding the importance of oblique gravity wave propagation on a global scale. With its height resolved wind measurements, Aeolus provides a unique dataset for such investigations.
Additionally, we will show how the unique combination of wind measurements with optical properties provided by Aeolus can be used to gain a global picture of wave induced polar stratospheric cloud formation. Polar stratospheric clouds play a crucial role for the stratospheric ozone depletion above the poles. They form if the temperatures in the polar winter stratosphere fall below a certain temperature. The negative temperature perturbations of gravity waves can lead to the formation of polar stratospheric clouds even if the synoptic-scale temperature is still well above the formation threshold. Most of the current global chemistry-climate models do not yet contain this formation mechanism of polar stratospheric clouds through gravity waves. However, several modelling groups have already started to look into this issue. We will provide these modelling groups with a climatology and test data set of polar stratospheric clouds in the Artic/Antarctic with a special focus on wave induced polar stratospheric clouds. The Aeolus data set is predestined for creating such a climatology.
Exploring Aeolus’ capacities in resolving dynamical processes in the upper troposphere and lower stratosphere
Mathieu Ratynski | LATMOS (CNRS, Université Saint Quentin en Yvelines, Paris Sorbonne Université) | France
The European Space Agency's Aeolus satellite mission is designed to provide global information on the wind speed from the ground up to 30 km, which is highly demanded for weather forecasting. Aeolus satellite has been set into orbit in August 2018 and its payload consists of a sophisticated ALADIN lidar instrument measuring wind velocity by sensing Doppler spectral shift of the laser echo scattered by the different layers of the atmosphere.
Since the global atmospheric circulation is largely driven by middle atmosphere dynamics, it is essential that the climate models take a proper account for the dynamical processes. Small-scale atmospheric waves, called internal gravity waves (IGWs) pose a particular challenge for models, whereas inaccurate parameterization of IGWs can dramatically bias the predictions of future atmospheric circulation changes.
In this paper, we explore the capacities of Aeolus wind observations in capturing and resolving dynamical processes in the upper troposphere and lower stratosphere (UTLS) such as IGWs at various temporal and spatial scales. The perturbations in the vertical profiles of Rayleigh horizontal line-of-sight (HLOS) wind velocity, associated with IGW activity, are derived by subtracting the Aeolus-derived “background” wind profiles from the individual measurements. Then, the global distribution of the IGW kinetic energy in the UTLS and vertical wavelength is derived using Aeolus measurements over the entire mission lifespan. The derived evolution of IGW activity over the Aeolus mission lifetime is analyzed in consideration of the time-varying performance of ALADIN instrument. The latter is evaluated using two French ground-based Doppler wind lidars operating at a mid-latitude site (Observatoire de Haute-Provence) and at a southern tropical site (Maïdo Observatory at la Réunion island) as well as collocated meteorological radiosoundings.
The global spatiotemporal distribution of IGW from Aeolus observations is compared with that derived from global high-resolution temperature profiling data provided by GPS radio occultation (RO) GRAS (GPS Receiver for Atmospheric Sounding) instruments operating onboard MetOp satellites. The comparison of Aeolus and RO-derived global IGW distribution allows concluding on the capacities and limitations of Aeolus wind profiling for studying UTLS dynamics.