Validation and Assimilation of Aeolus Wind Observations

Table of Contents

1 Introduction

1.1 State of the art

1.1.1 Importance of atmospheric wind observations for NWP

1.1.2 Deficiencies in the global observing system

1.1.3 Wind profile observations from space - historical scientific and technological background

1.1.4 Aeolus mission objectives

1.2 Research goals and outline of the thesis

2 Basic principles

2.1 Aeolus observations

2.1.1 DWL ALADIN measurement concept

2.1.2 Level-2B wind processing

2.1.3 Overview: Satellite orbit and measurement geometry

2.1.4 Main changes in the Aeolus L2B data set during the mission

2.2 Data assimilation

2.2.1 Three-dimensional variational assimilation

2.2.2 Ensemble Kalman Filter

2.2.3 Global data assimilation system at DWD

2.2.4 Observation error components in data assimilation

3 Data and methodology

3.1 Data sets and methods for the Aeolus HLOS wind validation study

3.1.1 Evaluation period and region

3.1.2 Collocated radiosonde observations

3.1.3 DWD ICON and ECMWF IFS model equivalents

3.1.4 Quality control criteria for the Aeolus HLOS wind observations

3.1.5 Representativeness errors

3.1.6 Statistical metrics

3.2 Data sets and methods to assess the impact of Aeolus HLOS wind assimilation

3.2.1 Experimental set up

3.2.2 Verification data and methods

4 Results: Validation of Aeolus HLOS wind observations

4.1 Temporal evolution of systematic and random differences

4.2 Aeolus HLOS wind validation error estimates

4.2.1 Representativeness error

4.2.2 Model error and radiosonde wind observational error

4.2.3 Aeolus wind observational error

4.2.4 Comparison and classification of the validation error estimates

4.3 Rayleigh wind observation bias dependencies and correction approaches

4.3.1 Rayleigh wind bias dependence on latitude and orbit phase

4.3.2 Rayleigh wind bias dependence on latitude, longitude and orbit phase

4.4 Concluding remarks on the validation of Aeolus HLOS wind observations

5 Results: Impact of Aeolus HLOS wind assimilation in the global model ICON

5.1 A global statistical overview

5.1.1 Systematic changes in the analysis

5.1.2 Short-range forecast impact: Observation-based verification

5.1.3 Medium-range forecast impact: Analysis-based verification

5.2 Investigation of links between dynamical scenarios and particularly high impact of Aeolus on NWP forecasts

5.2.1 Impact on tropical stratospheric wind variations (QBO)

5.2.2 Impact on change in the ENSO state in the Eastern Pacific

5.2.3 Dynamical impact in the midlatitudes

5.3 Concluding remarks on the impact of Aeolus HLOS wind assimilation in the global model ICON

6 Conclusions and Outlook

6.1 Main conclusions

6.2 Outlook

Research Objectives and Themes

This thesis investigates the quality of Aeolus satellite wind observations and evaluates their impact on Numerical Weather Prediction (NWP) using the ICON data assimilation system. The primary research questions concern the characterization and correction of systematic errors in Aeolus data and the quantification of the relative benefit of these observations in global assimilation systems, particularly in relation to tropical and extratropical dynamical scenarios.

• Validation of Aeolus wind profile data against independent reference datasets.
• Estimation of observational and representativeness errors for data assimilation.
• Development and testing of bias correction approaches for Rayleigh wind observations.
• Analysis of forecast improvements from Aeolus data assimilation using Observing System Experiments (OSEs).
• Investigation of the dynamical influence of Aeolus on large-scale circulation systems like QBO, ENSO, and extratropical transition events.

Excerpt from the Book

Summary of Chapters

1 Introduction: Provides an overview of the role of wind observations in NWP, details the deficiencies in current observing systems, and outlines the research goals regarding the Aeolus mission.

2 Basic principles: Covers the physical fundamentals of Doppler Wind Lidars (DWL), the Aeolus ALADIN instrument design, and the theoretical framework of data assimilation methods relevant to this study.

3 Data and methodology: Describes the reference data sets used for validation, the quality control criteria applied to Aeolus data, and the setup of the Observing System Experiments (OSE) performed with the ICON model.

4 Results: Validation of Aeolus HLOS wind observations: Details the validation results for the systematic and random errors of Aeolus data, including the estimation of representativeness errors and various bias correction techniques.

5 Results: Impact of Aeolus HLOS wind assimilation in the global model ICON: Analyzes the impact of assimilating Aeolus data on analysis and forecast performance, with a specific focus on its influence during key dynamical scenarios like QBO, ENSO, and extratropical transition.

6 Conclusions and Outlook: Summarizes the main findings of the thesis and offers perspectives on future research and potential follow-on satellite missions.

Keywords

Aeolus, Doppler Wind Lidar, Data Assimilation, Numerical Weather Prediction, ICON, Validation, Systematic Errors, QBO, ENSO, Extratropical Transition, Representativeness Error, Rayleigh-Brillouin scattering, Observation Impact, Atmospheric Dynamics, Satellite Meteorology.

Frequently Asked Questions

What is the primary focus of this dissertation?

The dissertation focuses on the validation of Aeolus Doppler Wind Lidar (DWL) observations and the evaluation of their impact on numerical weather prediction within the ICON global model.

Which specific observation types does this study analyze?

The study analyzes Aeolus HLOS (Horizontal Line-of-Sight) wind observations, specifically separating them into Rayleigh-clear and Mie-cloudy wind products.

What is the goal of the validation part of this research?

The validation aims to characterize the systematic and random errors of Aeolus observations, determine the Aeolus observational error, and develop bias correction approaches.

How is the impact of Aeolus wind data assessed in the global model?

The impact is assessed through Observing System Experiments (OSEs), comparing a control run (without Aeolus) with an experimental run (with Aeolus) over a three-month period.

Which main dynamical systems are highlighted as areas of high impact?

The study highlights the impact of Aeolus observations on large-scale circulation systems, specifically the Quasi-Biennial Oscillation (QBO), the El Niño-Southern Oscillation (ENSO), and the extratropical transition of tropical cyclones.

What are the key statistical metrics used for validation?

The validation uses bias and random difference estimates, specifically employing the scaled Median Absolute Deviation (MAD) as a robust indicator of variance.

How does the representativeness error influence the study results?

Representativeness error arises due to the difference between line-based satellite measurements and model point-based representations, making it a critical component for correctly assigning observation error in data assimilation.

What is the conclusion regarding the benefit of Aeolus observations for NWP?

The research concludes that Aeolus provides a beneficial impact on NWP forecasts, particularly in data-sparse regions like the tropics and the Southern Hemisphere, and contributes significantly to the understanding of large-scale dynamical shifts.