Unterschiede

Hier werden die Unterschiede zwischen zwei Versionen angezeigt.

Link zu dieser Vergleichsansicht

Beide Seiten der vorigen RevisionVorhergehende Überarbeitung
Nächste Überarbeitung		Vorhergehende Überarbeitung
studium:themen_fuer_masterarbeiten_theo [2020/01/10 14:13] – keilstudium:themen_fuer_masterarbeiten_theo [2026/03/02 12:57] (aktuell) – george
Zeile 1: Zeile 1:
 +====== Master thesis topics ======
 +
 +===== External projects =====
 +  * **[[:studium:externe_arbeiten | Important information on external master thesis projects!]]**
 +
 +===== Form =====
 +  * {{:studium:Anleitung_Abschlussarbeiten_MIM.pdf|Allg. Hinweise zur Form der Abschlussarbeiten am MIM}}
 +
 +===== Topics for master theses in the theory group =====
 +
 +Bei Interesse an **Masterarbeiten** im Lehrstuhl für Theoretische Meteorologie bitte bei Prof. G. Craig, Prof. T. Birner, oder Dr. C. Keil nachfragen. Nachfolgend eine Auswahl derzeit offener Themen:
 +\\
 +
 +==== Atmospheric dynamics in AI weather models ==== 
 +AI weather prediction models such as GraphCast are trained to predict the change in the atmosphere over a time interval, typically 6 hours. The aim of this project is to examine these changes for typical weather systems such as cyclones and Rossby waves, and determine where they are produced in the network. Recent methods to visualize the internal state of the model will be used. Strong python coding skills will be needed, and experience using libraries such as PyTorch and JAX will be helpful, but not necessary.
 +
 +For more detailed information please contact George Craig. 
 +
 +==== Scale refinement at inference time in AI weather models ==== 
 +The aim of this project is to test the hypothesis that AI weather models work systematically from large spatial scales to small scales as they produce their forecasts. This may apply to both deterministic models with graph neural network or transformer architectures, and to generative diffusion model. To test the hypothesis, several different AI models will be run, and the latent space will be examined using Fourier transforms. Strong python coding skills will be needed, and experience using libraries such as PyTorch and JAX will be helpful, but not necessary.
 +
 +For more detailed information please contact George Craig. 
 +
 +==== Scale refinement during training in AI weather models ==== 
 +It is hypothesized that when AI weather models are trained, gradient descent causes the large spatial scales to be learned first. The goal of this project is to train a transformer-based AI weather model, ArchesWeather, and examine checkpoints through the training to see where different aspects of the atmosphere are learned. Strong python coding skills will be needed, and experience using libraries such as PyTorch and JAX will be helpful, but not necessary.
 +
 +For more detailed information please contact George Craig.
 +
 +==== Tracking forecast sensitivity through network layers ==== 
 +Backpropagation in an AI weather model can be used to  identify which regions of the atmosphere are relevant to the forecast of a particular weather event. With layer-wise relevance propagation, the method can be extended to show how information moves between features in the latent space of the model. This project will implement these methods and test them on some interesting weather systems. Strong python coding skills will be needed, and experience using libraries such as PyTorch and JAX will be helpful, but not necessary.
 +
 +For more detailed information please contact George Craig.
 +
 +==== Analyzing the “butterfly effect” in operational forecast data ====
 +The European Centre for medium-range weather forecasting (ECMWF) has accidentally run a one year-long butterfly experiment, with their deterministic and control forecast initially differing only by tiny truncation errors. This data set, together with the regular ensemble forecasts, provides a great opportunity to study uncertainty growth in weather prediction and to address related questions on forecast improvement potential and forecast busts. The thesis requires good python coding skills and interest in data analysis and visualization.
 +
 +For more detailed information please contact Tobias Selz or George Craig.
 +
 +==== Development of an AI-based algorithm for potential vorticity inversion ====
 +Piece-wise potential vorticity (PV) inversion is a method to understand spatial interactions in the atmosphere. It has been used to investigate the reasons for uncertainty growth or extreme weather. However, running the numerical inversion algorithm on forecast data is expensive and often leads to convergence issues. The goal of the thesis is to investigate, if AI-based methods could be used for the PV inversion instead. Therefore, a training data set has to be created first, applying the conventional method. After that, various neuronal network designs should be developed, trained and tested. The thesis required excellent python coding skills and working independently with neural networks.
 +
 +For more detailed information please contact Tobias Selz or George Craig.
 +
 +/**
 +==== Spatial variability budget of convective-scale precipitation forecasts ====
 +Short-range precipitation forecasts often exhibit overconfidence in ensemble forecasts, leading to discrepancies with observed precipitation extending beyond the predicted area. Incorporating model uncertainty schemes may mitigate this issue by introducing greater spatial variability.
 +Spatial variability can be assessed using spatial skill scores such as the Fractions Skill Score (FSS) and Structure-Amplitude-Location (SAL) score, applied to different combinations of ensemble members. Since these scores involve squared differences between ensemble members, they can be reformulated in a form of variance. Analyzing how this variance is modulated by including model uncertainty schemes can provide insights for improving ensemble forecasts.
 +This study examines how the spatial variability of precipitation forecasts is modified by a physically based stochastic perturbation scheme, with a particular focus on hourly precipitation during the summer over Germany. The variance budget framework proposed by Matsunobu et al. (2025) will be extended to incorporate FSS- and SAL-based spread metrics.
 +
 +For more detailed information please contact Takumi Matsunobu or Christian Keil.
 +**/
 +
 +====Dynamics-transport coupling in the stratosphere====
 +The stratospheric circulation is projected to change in response to a warming climate. A changing circulation can impact the transport of trace gases (ozone, water vapor and ozone destroying substances) in the middle and upper atmosphere. The project will focus on developing a better understanding of the dynamics-transport coupling in the stratosphere.
 +
 +Changes in the large-scale stratospheric dynamics and transport of trace gases will be investigated using a combination of theory, observations and numerical modeling. Key topics of investigation will be the subseasonal to climatological impact of stratospheric dynamical processes on the global distribution of tracers in the stratosphere. A hierarchy of climate models will be used.
 +
 +For more detailed information please contact Thomas Birner.
 +
 +/**
 +====Large-scale impacts of convective clouds====
 +
 +Although convective clouds are much smaller than large-scale weather systems, they can be a major source of error in
 +weather forecasts because errors grow rapidly. Studying this error growth requires large ensembles of forecasts to
 +accurately represent the many ways that the weather situation can develop.
 +The aim of this project is to develop a simple numerical model that can represent the error growth processes, but is
 +inexpensive to run in large ensembles. Different model formulations will be designed, programmed in Python and evaluated
 +using advanced verification measures.
 +
 +For more detailed information please contact George Craig.
 +**/
 +
 +====Stratosphere-troposphere and climate dynamics====
 +
 +Topics broadly in the area of stratosphere-troposphere and climate dynamics are available upon request. Recent research topics in our group include: variability and long-term trends in the width of the tropical belt, processes that govern the temperature structure of the tropical tropopause layer, the dynamics of sudden stratospheric warmings and their coupling to the troposphere, transport processes in the upper troposphere / lower stratosphere. Interested candidates are asked to look through research topics on our group's website, in particular our recent publications: //[[https://www.meteo.physik.uni-muenchen.de/~Thomas.Birner/pubs.html]]//.
 +
 +For more detailed information please contact Thomas Birner. 
 +
 +/**
 +
 +The stratospheric circulation in winter is dominated by a strong cyclonic vortex over the pole. This polar vortex is caused by radiative cooling due to polar night. However, upward propagating planetary waves frequently perturb the vortex in terms of both position and strength. For sufficiently strong planetary wave forcing, these disturbances can grow enough to destroy the polar vortex as a well-organized entity. These abrupt transitions in the hemispheric-scale stratospheric circulation are usually associated with a strong warming of the polar stratosphere and are thus often called //[[https://www.meteorologie.lmu.de/~Thomas.Birner/ssw.html | sudden stratospheric warmings (SSWs)]]//. Furthermore, SSWs tend to produce circulation anomalies at the surface.
 +
 +In this Master project we will explore the dynamical evolution leading up to and following SSW events based on meteorological reanalysis data and climate model output.
 +
 +**/
 +
 +====Assimilation of cloud information====
 +
 +The assimilation of cloud-related observations is challenging as traditional error metrics (e.g. RMS error) are not  really suitable for the evaluation of intermittent fields as clouds or precipitation (double penalty problem). To overcome this deficiency, various feature-based scores have been developed  for the verification of forecast precipitation fields, but these  scores have not yet been applied in the context of data assimilation.
 +
 +The goal of this thesis ist to test the use of feature-based metrics for the assimilation of cloud-affected satellite observations in the emsemble data assimilation system KENDA for the regional weather forecast model COSMO-DE. Different approaches shall be tested in an idealized setup of KENDA that is used by several people in the HErZ data assimilation group.
 +
 +Contact: Leonhard Scheck
 +
 +/**
 +====Tropospheric moisture variability and the development of tropical convection====
 +
 +Observations and high-resolution numerical simulations both show that tropospheric humidity affects the development of deep moist convection, with a drier troposphere typically limiting or delaying the apparition of the deepest clouds. This phenomenon is generally explained by the fact that a dry environment can quickly erode the cloud core through lateral mixing (called entrainment). Although convective cloud parameterizations  usually account for entrainment, the influence of environmental humidity is generally not well captured, in part because the host model (typically a weather prediction or climate model) cannot resolve all the small spatial humidity fluctuations.
 +
 +In this project, a Cloud Resolving Model (CRM) operating at about 100m resolution will be used to examine the connections between the development of deep clouds and small scale moisture fluctuations in the free troposphere. The analysis of high-resolution model data should constitute the first step towards a parameterization of convection that could explicitly account for unresolved moisture variability.
 +    
 +Contact: Julien Savre
 +**/
 +
 +==== Applicability of lossy compression methods to meteorological applications====
 +
 +The storage space requirements for output of numerical weather and climate prediction is growing faster than the cost of storage space is decreasing. Model resolution is continuously increased to overcome issues with parameterized physical processes like convection. At the same time the ensemble size (e.g. the number of model runs performed to create one forecast) is also increased to improve the assessment of uncertainty within the forecast. Both in combination results in really big data sets that are not only difficult to process but also very expensive to store.
 +
 +One way to reduce the amount of output is to rely more on online diagnostics and not to save large fractions of the output. While this approach appears promising in operational setups of weather services, it is only a partial solution in research. Visualization of arbitrary aspects of a model run would no longer be possible and experiments would have to be carried out again if changes are made to the online diagnostic.
 +
 +Another way is to store model output with reduced precision. File formats currently in use support only lossless compression or if lossy compression is possible only spatial correlation between neighboring data points is used (e.g., JPEG compression). The temporal correlation is ignored. In contrast, video compression algorithms are essentially based on the temporal correlation between successive time steps. Without reducing the quality, this results in a compression ratio that is by one order of magnitude higher than that of individual images. Central ideas of video compression should be directly transferable to the compression of model output. Examples are differential coding (only differences between time steps are stored) and motion compensation (for unchanged but moved parts of an image only a displacement vector is stored). On the other hand, assumptions about the perception by the human eye are not applicable (e.g., changes in brightness and color are not equally important).
 +
 +The following questions should be addressed in this thesis:\\
 +Which compression algorithms are best suited for meteorological model output? Candidates are video compression and general purpose algorithms. The plan is not to develop new algorithms, but to asses existing ones.\\
 +What are the characteristics of errors produced by the analyzed algorithms?\\
 +Which degree of compression is acceptable for a set of different meteorological applications?
 +
 +Contact: Robert Redl
 +
 +
 +==============================================
 +
 +Further themes are possible, please talk to Prof. G. Craig, Prof. T. Birner, or Dr. C. Keil.
 +\\
 + 
 +/**
 +
 +====Predictability of convection in a very big ensemble====
 +
 +Convection-permitting ensemble prediction systems (EPS) are nowadays 
 +operationally run at various Meteorological Services to represent
 +forecast uncertainty of local weather. However, the ensemble size of 
 +typical EPS ranges from 12 to 20 members due to computational constraints.
 +In cooperation with the RIKEN Center for Computational Science in Japan 
 +1000 member ensemble forecasts at kilometre-scale have been
 +performed for several high impact weather events across Germany.
 +
 +The aims of this Master project are to investigate the ensemble 
 +size necessary to capture forecast uncertainty realistically and to examine 
 +saturation limits of predictability using traditional and spatial methods.
 +Possible evaluation metrics contain the gridpoint based RMSE and its 
 +upscaled variant as well as the believable scale of the Fraction Skill 
 +Score FSS. The Gaussianity of variables will be assessed using histograms.
 +
 +A 1000 member ensemble forecast tailored to a weather event in Germany 
 +has not been performed so far and offers great new research opportunities.
 +A short-term research visit in Kobe (J) may be possible during the course
 +of this Master project.
 +
 +Contact: George Craig, Christian Keil
 +
 +
 +==== Assimilation of measurements from surface stations with KENDA-COSMO  ====
 +
 +
 +Surface observations represent an important source of information for 
 +numerical weather forecasts but their use has proven to be difficult and 
 +their full potential is currently not exploited.
 +
 +A dense network of surface stations provide measurements of surface 
 +pressure, 10 m wind, 2 m temperature and 2 m humidity over Germany. 
 +Using a kilometer-scale ensemble data assimilation (KENDA) system, these 
 +data can be assimilated into the COSMO-DE limited area model. However, 
 +currently only temperature and humidity are assimilated from all 
 +stations in Germany, while wind observations are only used in Northern 
 +Germany where the terrain is sufficiently flat, and surface pressure is 
 +not used at all.
 +
 +The aim of the project is to exploit the potential of these observations 
 +for convective-scale short-range forecasts using the KENDA-COSMO system. 
 +Error statistics of surface observations shall be investigated using 
 +already available COSMO-DE forecasts with a focus on the 
 +representativeness error of wind observations and height adjustment of 
 +pressure observations. These findings shall provide the basis for a 
 +series of data assimilation experiments computed with the local linux 
 +cluster.
 +
 +The project requires a basic knowledge of programming and numerical 
 +weather prediction. The project will be performed in the framework of 
 +the HErZ Data Assimilation Branch at MIM.
 +
 +Contact: Florian Harnisch, Martin Weissmann
 +
 +
 +==== Assimilation of humidity observations in COSMO-KENDA  ====
 +
 +In the framework of the [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] , a forward operator for visible MSG SEVIRI observations has been developed with the goal to assimilate these observations.
 +
 +This MSc project shall investigate the effect of assimilating complementary humidity observations (e.g. GPS delay or humidity-sensitive radiance channels) in the experimental KENDA esnsemble data assimilation system of DWD. 
 +
 +Research question:
 +- What is the effect of humidity- versus cloud-related observations?
 +
 +Contacts: Martin Weissmann
 +
 +
 +
 +==== Assimilation of Radar observations in COSMO-KENDA  ====
 +
 +
 +Using the COSMO model of the German Weather Service (DWD) several data assimilation experiments will be performed using radar observations. The KENDA-BACY System (Kilometre-scale ensemble data assimilation, Basic Cycling) provides a framework in which the LETKF can be used. 
 +
 +First, the general behaviour of the system will be analysed. Then the impact of various parameters like localization radius on the predictability of the system. 
 +Various fields can be analyzed and plotted to show the quality of the data assimilation and forecast. 
 +
 +Contact: Michael Würsch
 +
 +==== Interactive 3D Visualization of High-Resolution Numerical Weather Forecasts ====
 +
 +In collaboration between the Computer Graphics and Visualization Group at TU Munich and the Meteorological Institute at LMU Munich this interdisciplinary thesis project aims at visualizing numerical weather prediction data from the high-resolution COSMO model (http://www.cosmo-model.org/) in an interactive 3D system.
 +
 +The Computer Graphics and Visualization Group develops “Met.3D”, a 3D visualization system until now targeted at the visualization of weather predictions from the European Centre of Medium Range Weather Forecasts (ECMWF, http://www.ecmwf.int/). Your task will be the integration of COSMO forecasts into the system and the visualization and analysis of a case study. This will require the adaptation of the Met.3D visualization pipeline (data management, visualization algorithms) to the COSMO grid. Challenges will include the adaptation of visualization algorithms (e.g. raycaster) to the special vertical grid structure of COSMO, handling data fields larger than the available GPU memory, and handling the rotated coordinate system of the COSMO grid. Due to the high resolution of the COSMO model, you will create visualizations that will allow fascinating insights into the atmospheric structure over Europe.
 +
 +The thesis is suited for computer science students interested in meteorology as well as meteorology students interested in computer graphics. However, requirements for the thesis are good programming skills in C++, good knowledge of a graphics API (preferably OpenGL), and interest in the meteorological application. Knowledge of Linux and meteorological data formats (NetCDF and GRIB) would be a plus. In return, we offer an interesting thesis that will enable you to gain insight into the exciting world of computer science meeting atmospheric research.
 +
 +If you are interested, please contact:
 +
 +At TU Munich: Marc Rautenhaus, Email: marc.rautenhaus@tum.de\\
 +At LMU Munich: Tobias Selz, Email: tobias.selz@lmu.de
 +
 +==== Calculation and interpretation of the convective adjustment time-scale ==== 
 +
 +Many properties of convection and precipitation (like forecast quality, sources of uncertainty, predictability etc.) depend in some way on the prevailing meteorological weather regime. Thus it is desirable to objectively determine the dominant process in a given meteorological situation. Such a measure represents the convective time-scale τ that comprises a physically-based measure to distinguish equilibrium and non-equilibrium convection. Since it has been first suggested by Done et al. (2006) the time-scale has been applied in various applications.
 +
 +The aim of the project is to investigate several characteristics of the areal and temporal averaging as well as the optimal usage in the context of ensembles. Here, the convection-permitting ensemble prediction system COSMO-DE-EPS will be used. 
 +
 +The project requires basic knowledge of programming (F90, python) and numerical weather prediction models.
 +
 +Contact: Kirstin Kober, Christian Keil, George Craig
 +
 +
 +
 +==== Assimilation of lightning information in COSMO-DE model ==== 
 +
 +Lightning is closely related to convective precipitation and can serve as indicator for the location and intensity of convective systems. For this reason, lightning observations are commonly used for nowcasting purposes at many weather centres. In numerical weather prediction models in contrast, this potentially very valuable information on the initial state of convective activity is not presently exploited. This is partly due to the complexity of existing lightning parameterizations and partly to insufficient knowledge of the underlying processes of cloud electrification and discharge. 
 +
 +The aim of the project is the optimisation of a recently developed lightning parameterization (Dahl et al. 2011) and its implementation as a forward operator in the convective-scale data assimilation system. This study will be performed under the umbrella of the [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] at MIM, that closely collaborates with the Kilometer-scale Ensemble Data Assimilation (KENDA) project at DWD to develop an ensemble based assimilation algorithm for COSMO-DE.
 +
 +The project requires basic knowledge of programming (F90, python) and numerical weather prediction models.
 +
 +Contact: Christian Keil, Martin Weissmann, George Craig
 +
 +==== Investigation of precipitation amounts and further development of the Plant-Craig Convection Scheme  ==== 
 +
 +In global or coarse grid limited area models convective processes have to be parametrized in order to avoid the onset of unrealistic grid-scale convection. Numerous convection schemes have been developed in the past 40 years. Most of them are deterministic. The recently proposed Plant-Craig scheme is a stochastic parametrization scheme of the convection. Clouds are drawn randomly from a distribution and the closure assumption is only fulfilled on average. The scheme has been implemented into the COSMO model and tested in several case studies at a 7km resolution. These studies have revealed that the amount of precipitation produced by the scheme is in general too low.
 +
 +The goal of this project is to find out reasons and possible solutions to this problem. An other part of the project will be to further develop the Plant-Craig-scheme, for example to account for a displacement of the clouds by the flow.
 +
 +The project requires basic knowledge in programming (F90, python) and numerical weather prediction. There is also a related HiWi position.
 +
 +Contact: Tobias Selz, George Craig
 +
 +==== The impact of KENDA initial conditions on COSMO-DE ensemble forecasts ==== 
 +
 +The convection-permitting ensemble prediction system COSMO-DE-EPS is currently in its pre-operational phase at DWD.
 +The different sources of uncertainty are introduced by perturbing the initial and boundary conditions using an ad-hoc
 +multi-model approach, while the model error is introduced by varying different parameters in the physical parametrization schemes. Within the Kilometer-scale Ensemble Data Assimilation (KENDA) project at DWD in conjunction with the [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] at MIM, an ensemble based assimilation algorithm is developed for COSMO-DE.
 +
 +The goal of this project is to examine the impact of KENDA generated initial conditions on COSMO-DE-EPS forecasts, for instance, in terms of spread in precipitation and its growth with forecast lead time. Another subject of investigation is the horizontal and vertical structure of KENDA perturbations. This will be studied by conducting a series of COSMO-DE-EPS forecast experiments based on KENDA generated initial conditions for a period in summer 2011 and its comparison with already existing, postprocessed forecasts.
 +
 +The project requires basic knowledge of programming (F90, python) and numerical weather prediction models.
 +
 +Contact: Christian Keil, Martin Weissmann
 +
 +==== Examination of the performance of a stochastic convective parametrization during different flow conditions==== 
 +
 +In numerical weather prediction (NWP) models atmospheric processes not resolved by the model grid need to be parameterized.
 +At a horizontal resolution of about 10 km convection comprises such a process. The recently proposed stochastic convective
 +parametrization of Plant and Craig (PCCS, 2008) is implemented in the research version of the COSMO-EU model of DWD.
 +One question is the performance of PCCS during different flow conditions, i.e. compare precipitation epsisodes
 +characterized by either strong or weak synoptic-scale forcing.
 +
 +The convective timescale can serve as a predictor of the flow regime and has already been computed for the summer 2009.
 +Based on this available classification a handful of case studies ought to be performed with COSMO-EU including PCCS and the
 +model quality in forecasting summertime precipitation shall be examined.
 +The computation will be executed on the local linux-cluster at MIM.
 +
 +The project requires basic knowledge of programming (F90, python) and numerical weather prediction models.
 +
 +Contact: Christian Keil, George Craig
 +
 +==== Hybrid data assimilation methods==== 
 +
 +Data assimilation algorithms primarily in use today are based either on  variational or ensemble techniques.  Recently the algorithms that combine these two methods have been developed and applied for operational use in numerical weather prediction.      In this project,  the benefits of the combined approach on toy problems will be investigated.  Three toy problems will be chosen to mimic different dynamical regimes.  Different variants of hybrid algorithms will be evaluated  on accuracy as well as ability to preserve physical properties of the system.   
 +
 +The project requires basic knowledge of programming.  The project is within  [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] at MIM.
 +
 +Contact: Tijana Janjic Pfander
 +
 +==== Model vs. observation space localization for ensemble Kalman filter==== 
 +
 +In order to apply ensemble based Kalman filter methods in practice, localization is additionally applied as part of ensemble Kalaman filter algorithms.  Two methods for localization, one in observation space and one in model space are in use.  In the work by Campbell et al.  2009 it was shown that the radiance space localization is inferior to the model space localization.  Considering only the localization in the vertical, we will explore localization techniques and weighthing functions for satellite data assimilation on similar test problem as described in Campbell et al.  2009.  
 +
 +The project requires basic knowledge of programming (F90).  The project is within  [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] at MIM.
 +
 +Contact: Tijana Janjic Pfander
 +
 +==== Ensemble Generation==== 
 +
 +In order to improve the forecast of numerical weather prediction models, the proper descriptions of different sources of forecast error is necessary.   The goal of this master project is to investigate the impact of combining models with different numerical cores using toy problems.  For example, the impact of different advection schemes of each ensemble member on the properties of statistics calculated from ensemble are of interest.  Further, the impact of a specific scheme's numerical dissipation, both for space and time discretization on spread will be investigated.  For this master thesis, already existing programmes in F90 will need to be modified (moderately).
 +
 +The project requires basic knowledge of programming (F90).  The project is within  [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] at MIM.
 +
 +Contact: Tijana Janjic Pfander, Matthias Sommer, Christian Keil
 +
 +==== Evaluating observation impact calculations==== 
 +
 +The knowledge of the impact of different observations on the accuracy of NWP analyses and forecasts is crucial to optimize observing and data assimilation systems. Traditional methods as data denial experiments (i.e. rerunning the model for every possible configuration) can only be conducted occasionally due to their computational costs. For this reason, a new ensemble-based method to estimate the impact of observations following Liu and Kalnay (2008) has been implemented in the pre-operational DWD KENDA-COSMO ensemble data assimilation system in the framework of the [[http://www.meteo.physik.uni-muenchen.de/dokuwiki_en/doku.php?id=lscraig:herz | HErZ Data Assimilation Branch]] at MIM. 
 +
 +The task of the master thesis is to evaluate the accuracy of the method for different configurations (different lead time, localization, ensemble size) by conducting experiments with the KENDA-COSMO system. The goal is to find the optimal configuration for such a calculation in the future operational convectional data assimilation system of DWD. 
 +
 +The project requires basic knowledge of programming.
 +
 +Contact: Martin Weissmann and Matthias Sommer
 +
 +
 +\\
 +
 +
 +==== Ensemble Vorhersagen, Vorhersagbarkeit ====
 +(Bearbeiter: Florian Heinlein)
 +
 +Mit Ensemble Vorhersagen läßt sich die Vorhersagbarkeit quantifizieren. Beim DWD wird gegenwärtig ein hochauflösendes Ensemble basierend auf dem COSMO-DE mit 20 Membern entwickelt (COSMO-DE-EPS). Mit diesem Datenschatz lassen sich wichtige Fragen zur Vorhersagbarkeit untersuchen. In dieser Arbeit werden die 20 Member als deterministische Lösungen betrachtet. Basierend auf der diagnostischen Modellgröße der konvektiven Zeitskala kann das vorherrschende synoptische Regime identifiziert werden. Die Einteilung der Wetterlagen in die Regimes soll mit einer klassischen Wetterlagenklassifikation verglichen werden.
 +
 +Methoden und Vorgehen:
 +  - Einarbeitung in Thematik und den Umgang mit Daten des COSMO-DE-EPS
 +  - Berechnung der konvektiven Zeitskala und Einteilung in Regimes (lokal, frontal, stratiform)
 +  - Vergleich mit  klassischen Wetterlagenklassifikation
 +  - Auswertung und Diskussion der Ergebnisse
 +
 +Kontakt und Betreuung: Christian Keil
 +
 +
 +==== Vorhersagbarkeit durch Validierung von Ensemble Vorhersagen ====
 +
 +Die Qualität der Vorhersagen ist durch Verifikationsmaße quantifizierbar. Dafür existieren diverse Verfahren. In dieser Arbeit soll das räumliche Verifikationsmaß //Displacement and Amplitude Score// (DAS) und traditionelle Verfahren auf Niederschlagsvorhersagen individueller Members des Ensembles angewendet werden. Ziel ist die Überprüfung der Hypothese, das einzelne Ensemble Member für einen bestimmten Zeitraum (Regime) kontinuierlich besser sind als die Übrigen. 
 +
 +Methoden und Vorgehen:
 +  - Einarbeitung in Verifikationsmaße (konventionelle Maße und DAS) und den Umgang mit Modelldaten (time-lagged COSMO-DE)
 +  - Anwendung verschiedener Verifikationsmaße auf Ensemble Modellvorhersagen
 +  - Untersuchung der Güte in Abhängigkeit der meteorologischen Situation
 +  - Auswertung und Diskussion der Ergebnisse
 +
 +
 +Kontakt und Betreuung: Christian Keil, Kirstin Kober
 +
 +
 +==== Probabilistische Verifikation der Niederschlagsvorhersagen des COSMO-DE-EPS ====
 +
 +Das COSMO-DE-EPS ist das konvektions-auflösende Ensemble Vorhersagesystem des Deutschen Wetterdiensts (DWD). Dieses Ensemble soll für den Sommer 2009 mit probabilistischen Verifikationsmaßen analysiert werden. Ziel ist die Quantifizierung der Qualität des Ensembles unter Berücksichtigung der inherenten Unsicherheit.
 +
 +Methoden und Vorgehen:
 +  - Einarbeitung und Erzeugung von probabilistischen Vorhersagen
 +  - Einarbeitung in probabilistische Verifikationsmaße 
 +  - Kalibrierung
 +  - Anwendung dieser Verifikationsmaße auf die probabilistischen Vorhersagen
 +  - Auswertung und Diskussion der Ergebnisse
 +
 +
 +Kontakt und Betreuung: Kirstin Kober, Christian Keil
 +
 +
 +====Topics related to Tropical Meteorology====
 +
 +  * Observational analyses of tropical cyclones
 +
 +  * The thermodynamic structure of the tropical atmosphere
 +
 +  * The structure and dynamics of heat lows
 +
 +  * Gulf of Carpentaria convergence lines
 +
 +These projects would involve analyses of gridded analysis data and/or observational data and their interpretation.
 +
 +Contact: Prof. Roger Smith (Room 230)\\
 +Email: roger.smith@lmu.de
 +
 +
 +
 +Diplom-/Masterarbeit Meteorologie am Institut Physik der Atmosphäre am DLR Oberpfaffenhofen:
 +
 +
 +
 +==== Auswertung von Niederschlagsvorhersagen konvektiver Systeme im Bereich des Flughafen München ====
 +
 +COSMOAirport ist eine für Flughafenumgebungen angepasste Version des numerischen
 +Wettervorhersagemodells COSMO-DE des DWD. Das Modell wird stündlich gestartet und liefert u.a.
 +Niederschlagsprognosen bis zu 6 Stunden mit einer zeitlichen Auflösung von 15 Minuten für ein 360
 +km x 360 km großes Gebiet um den Flughafen München.
 +Während einer 2-monatigen Messkamapagne im Sommer 2010 am Flughafen München (MUC)
 +wurden kontinuierlich Prognosen erstellt (COSMO-MUC). Besonderes Interesse galt dabei Gewittern,
 +die den Flughafen München beeinflussten und zum Teil zu Behinderungen führten. Es wurden 21 Fälle
 +beobachtet, die dazu verwendet werden sollen, die Vorhersagegüte des COSMO-MUC mit der des
 +operationellen COSMO-DE des DWD (3-stündige Aktualisierung) zu vergleichen. Dabei kommen
 +verschiedene Verifikationsmaße zum Einsatz, der „Displacement and Amplitude Score“ (DAS), „False
 +Alarm Ratio“ (FAR) und der „Frequency Bias“ (FBI). Nach erfolgreicher Einarbeitung in die Thematik
 +sollen folgende Fragestellungen bearbeitet werden:
 +  * Abhängigkeit von DAS von Suchradius und Niederschlagsintensität
 +  * Abhängigkeit der Verifikationsmaße von der Größe des Berechnungsgebiet und der zeitlichen Auflösung
 +  * Vergleich mit Vorhersagen des operationellen COSMO-DE
 +Ziel der Arbeit ist es zu einer Aussage zu kommen in wie weit COSMO-MUC Prognosen von Gewittern
 +mit der stündlichen Aktualisierung zu einer Verbesserung führen verglichen mit den Prognosen des
 +operationellen COSMO-DE. Ferner soll eine Aussage über die Eignung der angewendeten
 +Verifikationsmaße für die Vorhersage konvektiver Systeme erfolgen.
 +
 +Anforderungsprofil: Studium der Meteorologie, Programmierkenntnisse erforderlich
 +Kontakt und Betreuung: Klaus.Dengler@dlr.de, Christian Keil
 +
 +
 +==== Numerische Simulation von latenten Wärmeflüssen über der Barentssee und Vergleich mit LIDAR-Messungen während IPY-THORPEX ====
 +
 +Während der Flugzeugmesskampagne IPY-THORPEX 
 +(www.pa.op.dlr.de/ipy_thorpex) wurden mit der DLR Falcon Lidar-Messungen 
 +des vertikalen Windes und der Wasserdampfkonzentration in der Grenzschicht 
 +über der Barentssee vorgenommen. Die Kombination dieser Lidar-Messungen 
 +erlaubten eine Berechnung des latenten Wärmeflusses mit Hilfe der 
 +Eddy-Kovarianz-Methode. Diese Messungen sollen nun mit numerischen 
 +Simulationen der Konvektion über Meerwasser (das "warme" Wasser des 
 +aufsteigenden Golfstromes und die kalte arktische Luft sind die treibenden 
 +Elemente der Konvektion) ergänzt und verglichen werden. Dazu sind 
 +sogenannte Grobstruktursimulationen (engl. large-eddy simulations) 
 +durchzuführen, die mit vorhandenen meteorologischen Daten des 
 +amerikanischen Weather and Research Modells (WRF) angetrieben werden. Die 
 +Masterarbeit ist für alle diejenigen geeignet, die sich für atmosphärische 
 +Dynamik, numerische Modellierung und Visualisierung von Daten 
 +interessieren. 
 +
 +Grundkenntnisse in einer Programmiersprache und in Linux 
 +sind durchaus vorteilhaft.
 +
 +
 +Kontakt und Betreuung: Andreas.Doernbrack@dlr.de
 +
 +==== Untersuchung von langlebigen Kondensstreifen mithilfe eines wolkenauflösenden numerischen Modells ==== 
 +
 +Ziel der Arbeit: Verbessertes Verständnis der Kondensstreifenentwicklung und des Übergangs in Kondensstreifen-Zirren. 
 +
 +Kondensstreifen haben einen nicht vernachlässigbaren Einfluss auf das Klima und stellen einen Forschungsschwerpunkt am Institut für Physik der Atmosphäre dar. Dabei werden Kondensstreifen mit unterschiedlichsten Methoden wie Satellitenbeobachtung, in-situ Messungen, Fernerkundung und numerischen Modellen untersucht. Die Eigenschaften junger Kondensstreifen, wie sie oft am Himmel zu beobachten sind, sind gut bekannt, da sie leicht zu beobachten und erkunden sind. Mit der Zeit verbreitern sich Kondensstreifen, verlieren ihre linienhafte Struktur und sind meist nicht von natürlich entstandenen Eiswolken zu unterscheiden. Diese langlebigen Kondensstreifen, auch Kondensstreifen-Zirrus genannt, sind bisher nur unzureichend untersucht. Mit dem wolkenauflösenden Modell EULAG-LCM soll daher die Entwicklung von Kondensstreifen über mehrere Stunden und deren Übergang in Kondensstreifen-Zirren untersucht werden. In einem ersten Schritt sollen Simulationen mit dem Modell durchgeführt werden. Anschließend sollen die geometrischen, optischen und mikrophysikalischen Eigenschaften in Abhängigkeit von den meteorologischen Umgebungsbedingungen analysiert werden. 
 +
 +Die Ausschreibung wendet sich an Studenten eines mathematisch-naturwissenschaftlichen Studiengangs. Erfahrungen im Umgang mit Fortran oder einer anderen Programmiersprache sind notwendig.
 +
 +Kontakt:
 +Dr. Simon Unterstraßer 
 +Tel: 08153/282553 
 +simon.unterstrasser@dlr.de
 +
 +==== Untersuchung charakteristischer Lebenszyklen von Kondensstreifen ==== 
 +
 +Ziel der Arbeit: Verbessertes Verständnis des Lebenszyklus von Kondensstreifen-Zirren. 
 +
 +Kondensstreifen haben einen nicht vernachlässigbaren Einfluss auf das Klima und stellen einen Forschungsschwerpunkt am Institut für Physik der Atmosphäre dar.
 +
 +
 +Die Eigenschaften junger Kondensstreifen, wie sie oft am Himmel zu beobachten sind, sind gut bekannt, da sie leicht zu beobachten und erkunden sind.
 +
 +Mit der Zeit verbreitern sich Kondensstreifen, verlieren ihre linienhafte Struktur und sind meist nicht von natürlich entstandenen Eiswolken zu unterscheiden. Diese langlebigen Kondensstreifen, auch Kondensstreifen-Zirrus genannt, sind bisher nur unzureichend untersucht. In dieser Arbeit sollen deren charakteristische Lebenszyklen untersucht werden, wobei dies stark an die Entwicklung sogenannter eisübersättigter Regionen in der oberen Troposphäre gekoppelt ist.
 +
 +Zuerst sollen daher Wetterdaten des COSMO-EU-Modells mit Trajektorienrechnungen analysiert werden, um typische Entwicklungen der relativen Feuchte innerhalb einer Luftmasse zu bestimmen.
 +Die gewonnen Daten fließen dann in das Modell EULAG ein, mit welchem Kondensstreifen numerisch simuliert werden. Durch die Berücksichtigung der Synoptik können detaillierte Aussagen über den Lebenszyklus von Kondensstreifen gemacht werden.
 +
 +
 +Die Ausschreibung wendet sich an Studenten eines mathematisch-naturwissenschaftlichen Studiengangs (bevorzugt mit meteorologischen Kenntnissen). Erfahrungen im Umgang mit Fortran oder einer anderen Programmiersprache sind notwendig. Unix/Linux-Kenntnisse sind wünschenswert.
 +
 +Kontakt:
 +Dr. Simon Unterstraßer 
 +Tel: 08153/282553 
 +simon.unterstrasser@dlr.de
 +
 +
 +==== Numerischen Simulation des Windeinflusses auf Flugzeug Wirbelschleppen landender Flugzeuge ====
 +
 +Als unvermeidbare Folge des Auftriebs erzeugen Flugzeuge ein Paar gegensinnig rotierender langlebiger Wirbel, die eine potenzielle Gefährdung nachfolgender Flugzeuge darstellen. Die daher zwischen aufeinanderfolgenden Flugzeugen vorgeschriebenen Staffelungsabstände tragen maßgeblich zu Kapazitätsengpässen großer Flughäfen bei. Der Endanflug ist unmittelbar vor der Landung besonders kritisch, da die Wirbelschleppen dann nicht mehr unter den Flugkorridor absinken, sondern im Gegenteil durch die Interaktion mit dem Boden auch wieder aufsteigen können. Zusätzlich sind in der geringen Flughöhe die Möglichkeiten des Piloten begrenzt, kritische Rollmomente zu kompensieren. 
 +Die Masterarbeit richtet sich vorrangig an Studenten eines naturwissenschaftlichen oder ingenieurwissenschaftlichen Studiengangs. Der Masterstudent soll mit der Large Eddy Simulation (LES) den Einfluss des Windes auf den Zerfall und Transport der Wirbelschleppen untersuchen. 
 +Für diese Untersuchungen soll der LES Code MGLET eingesetzt werden. Bestehende Auswertungsalgorithmen müssen angepasst werden. 
 +
 +Es sind folgende Arbeiten durchzuführen:
 +  * Simulation turbulenter Grenzschichtprofile mit Querwind, Gegenwind und Diagonalwind
 +  * Simulation des Wirbelverhaltens in Bodennähe unter Variation des Windes
 +  * Vergleich mit Messungen im Wasserschleppkanal bzw. am Flughafen
 +  * Wissenschaftliche Interpretation und Dokumentation der Arbeiten 
 +
 +Gewünschte Qualifikation:
 +  * Kenntnisse der Fluiddynamik
 +  * Erfahrung mit numerischen Simulationen   
 +  * Programmierkenntnisse in FORTRAN sowie Erfahrung mit UNIX
 +  * Besonderes Engagement und Ausdrucksfähigkeit
 +
 +
 +Kontakt: Anton Stephan, Tel.: +49 (0)8153 28-2566, Email: Anton.Stephan@dlr.de
 +Dr. Frank Holzäpfel, Tel.: +49 (0)8153 28-2529, Email: Frank.Holzaepfel@dlr.de
 +
 +==== Welchen Einfluss hat der Flugzeugtyp auf die Kondensstreifenentwicklung? Numerische Untersuchung mit EULAG.==== 
 +
 +Ziel der Arbeit: Untersuche, inwieweit die Eigenschaften von Kondensstreifen vom Typ des Flugzeugs abhängen, das den Kondensstreifen anfänglich erzeugt hat.
 +
 +Kondensstreifen haben einen nicht vernachlässigbaren Einfluss auf das Klima und stellen einen Forschungsschwerpunkt am Institut für Physik der Atmosphäre dar.
 +Dabei werden Kondensstreifen mit unterschiedlichsten Methoden wie Satellitenbeobachtung, in-situ Messungen, Fernerkundung und numerischen Modellen untersucht.
 +Mit dem numerischen Modell EULAG-LCM wurde in der Vergangenheit die Entwicklung von Kondensstreifen über mehrere Stunden und deren Übergang in sogenannte Kondensstreifen-Zirren untersucht. Der Aspekt verschiedener Flugzeugtypen wurde bisher noch nicht beleuchtet.
 +In dieser Arbeit soll daher untersucht werden, ob die Eigenschaften von Kondensstreifen-Zirren neben bekannten meteorologischen Parametern (Feuchte, Temperatur) auch von Flugzeugeigenschaften (Gewicht, Spannweite, Kerosinverbrauch) abhängen.
 +
 +
 +Die Ausschreibung wendet sich an Studenten eines mathematisch-naturwissenschaftlichen Studiengangs. Erfahrungen im Umgang mit Fortran oder einer anderen Programmiersprache sind notwendig.
 +
 +Kontakt:
 +Dr. Simon Unterstraßer 
 +Tel: 08153/282553 
 +simon.unterstrasser@dlr.de
 +
 +==== Investigation of tropical predictability in simulations using a stochastic convection scheme====
 +
 +In a series of model experiments (ICON) a stochastic convection scheme has been used to investigate the limits of atmospheric predictability that originate from the fast error growth that happens in moist convection and the spreading of this error upscale. However, although global simulations were performed this analysis has been restricted to the midlatitudes.
 +
 +The main idea behind this master thesis is to first re-apply the diagnostics that were used before to the tropics in addition with a basic evaluation of model performance and biases. In a second step diagnostics and phenomena that are specific to the tropics (eg. Kelvin waves, MJO) should be considered and their role in the error growth process could be investigated.
 +
 +This project requires interest in dynamical meteorology and modeling as well as a substantial amount of data analysis using python.
 +
 +Further reading:\\
 +Selz, 2019: Estimating the intrinsic limit of predictability using a stochastic convection scheme. (dataset description, midlatitude analysis)\\
 +Judth, 2020: Atmospheric Predictability of the Tropics, Middle Latitudes, and Polar Regions Explored through Global Storm-Resolving Simulations (Investigation of tropical predictability with a high-res model)
 +
 +Contact: Tobias Selz or George Craig
 +
 +**/