Chairpersons: Chiara Marsigli, Daniel Reinert
Within the project ICON-Seamless a new Earth System Model is developed for weather forecasts, seasonal and decadal climate predictions, as well as climate projections. In doing so, we use the expertise from the numerical weather prediction (NWP), which operates and maintains ICON-NWP, as well as the experience with the first ICON-Earth System Model version based on the physics of the MPI-M atmosphere model (ECHAM). The goal is to use common components for all time scales. As a first step we develop a model for seasonal and decadal predictions.
ICON-Seamless builds on the coupling of the atmosphere (ICON-NWP) and ocean (ICON-O) components via the coupling software YAC. Sea ice as a further important component is thereby included. Furthermore, to have a closed hydrological cycle and to represent the carbon and other biogeochemical cycles comprehensively, a suitable soil model based on the ICON-Land framework as well as the TERRA and JSBACH/QUINCY land models, are or will be coupled to ICON-NWP. In addition, transient external fields for aerosol, greenhouse gases, ozone and solar irradiance are implemented in ICON-NWP to be able to simulate historical time periods and scenarios of the future. In parallel, the ART modules (Aerosol and Reactive Trace gases), which allow a dynamic treatment of gases and aerosol, are adapted to the modified model physics. Intensive model evaluation supports the tuning. For future use in the field of (weather and) climate predictions, coupled data assimilation is being developed as well.
We give an overview of the current state of the development, experiments and potential areas of application.
How to cite: Früh, B., Potthast, R., Müller, W., Korn, P., Brienen, S., Fröhlich, K., Helmert, J., Köhler, M., Lorenz, S., Pham, T. V., Pohlmann, H., Schlemmer, L., Schnur, R., Schulz, J.-P., Sgoff, C., Vogel, B., Wirth, R., and Zängl, G.: ICON-Seamless, the development of a novel Earth System Model based on ICON for time scales from weather to climate, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-292, https://doi.org/10.5194/ems2022-292, 2022.
We describe the implementation of grid refinement in the atmosphere component of the ICOsahedral Nonhydrostatic (ICON) modeling system. It basically follows the classical two-way-nesting approach known from widely used mesoscale models like MM5 or WRF, but differs in the way how feedback from fine grids to coarser grids is applied. Moreover, the ICON implementation supports vertical nesting in the sense that the upper boundary of a nested domain may be lower than that of its parent domain. Compared to the well-established implementations on quadrilateral grids, new methods had to be developed for interpolating the lateral boundary conditions from the parent domain to the child domain(s) on triangular grids. These are based on radial basis functions (RBFs) and partly apply direct reconstruction of the prognostic variables at the required grid points, whereas gradient-based extrapolation from parent to child grid points is used in other cases. The technical implementation on the unstructured ICON grid is based upon sorting the boundary interpolation points at the beginning of the grid-point index vector, so that computations on boundary interpolation points can be excluded by appropriate start indices without the need of IF masks. The run-time flow control is written such that limited-area domains can be processed identically to nested domains except for the lateral boundary data supply.
To demonstrate the functionality and quality of the grid nesting in ICON, we present idealized tests based on the Schär mountain wave test case (Schär et al., 2002) and the Jablonowski-Williamson test case (Jablonowski and Williamson, 2006). The results show that the numerical disturbances induced at the nest boundaries are small enough to be negligible for real applications. This is confirmed by experiments closely following the configuration used for operational numerical weather prediction at DWD, which demonstrate that a regional refinement over Europe has a significant positive impact on the forecast quality in the northern hemisphere.
References
Schär, C. and D. Leuenberger and O. Fuhrer and D. Lüthi and C. Girard: A new terrain-following vertical coordinate formulation for atmospheric prediction models, Mon. Weather Rev., 130 , 2459-2480, 2002.
Jablonowski, C. and D.L. Williamson: A baroclinic instability test case for atmospheric model dynamical cores, Q. J. Roy. Meteor. Soc., 132, 2943-2975, 2006.
How to cite: Zängl, G., Reinert, D., and Prill, F.: Grid Refinement in ICON, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-172, https://doi.org/10.5194/ems2022-172, 2022.
Weather prediction centers are always looking for the best computational performance for their numerical weather prediction (NWP) model, given their financial budget. Over the last decades, most centers relied on computer systems with scalar x86 architectures. This, however, might not be the best choice for the mid-term future, as the development of CPUs with ever-increasing performance and memory bandwidth is slowing down.
Nowadays, hardware manufacturers advertise massively multiprocessing GPUs as one future pathway. Unfortunately, GPUs have their own programming paradigms, as they are still requiring a CPU as their driver and bring their own memory. This necessitates significant adaptions of existing codes. Porting a large and continuously developed community code, such as ICON, to emerging hardware architectures poses its own special challenges.
Many parts of the ICON framework have been made ready for GPU systems in a multi-institute effort over the past years. MeteoSwiss plans to use GPU ICON operationally for limited area forecasts in 2023. Current development activities also make ICON-GPU ready to support the enhanced feature set used operationally by the DWD (such as grid nesting and parametrizations for global simulations). It was decided to port ICON by introducing OpenACC compiler directives to the FORTRAN code. This iterative development model makes it possible to merge ported code back directly into the main code repository and to stay up-to-date with other developments like in model physics. A tolerance based testing suite was deployed to make sure that ported features remain functional also when non-GPU-related changes are introduced to already ported code sections.
We present the general porting strategy and the current state of the port. We discuss specific optimizations and the lessons learned while porting an actively developed code. Finally, we present the performance on current GPU and CPU machines and compare them to the currently operational setup on the DWD vector supercomputer.
How to cite: Jacob, M., Alexeev, D., Dietlicher, R., Cherkas, V., Germann, E., Gessler, F., Hupp, D., Jocksch, A., Lapillonne, X., Müller, C., Osuna, C., Reinert, D., Sawyer, W., Schättler, U., and Zängl, G.: ICON NWP on GPUs, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-138, https://doi.org/10.5194/ems2022-138, 2022.
This talk introduces the initial actions within the EUMETNET/C-SRNWP project ‘Evaluation and updates of ESA-CCI global land cover map for NWP needs’. We present the coordination activities of the working group and the first steps verifying and updating the latest ESA-CCI land cover map of 2020. Hereby, we started a survey in June 2021 with all members of the C-SRNWP Surface Expert team to evaluate their needs and possible future plans related to the usage of the ESA-CCI land cover product. The results show a clear need for accurate water information and the distinction between fresh and salt water. In addition, the distinction between lakes and rivers would be a benefit. Hence, we tried to find an up-to-date and accurate global water mask as reference to replace existing water bodies in the ESA-CCI land cover with new classes for fresh and salt water. In this manner, we use the Open Street Map (OSM), which is open source with a high level of detail and is updated continually by a community of more than eight million members. In a second step, we tried to find an adequate platform where scientists can report their founded issues on the ESA-CCI land cover map. We consider the open source end-to-end software development platform Gitlab as a suitable environment where scientists can interact also through programming codes. As an outlook regarding the other outcome of the survey, the level of detail of the urban class requires improvements and has to be divided in more subclasses using as e.g. Local Climate Zones.
How to cite: Oswald, S., Samuelsson, P., Kurzeneva, E., and Palmason, B.: Verification and updates of the ESA-CCI land cover product, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-31, https://doi.org/10.5194/ems2022-31, 2022.
The ICON (ICOsahedral Nonhydrostatic) is a joint project between the Deutscher Wetterdienst (DWD) and the Max-Planck-Institute for Meteorology (MPI-M) for the development of a unified global numerical weather prediction system. In 2018, COSMO (COnsortium for Small-scale Modeling) started the migration from the COSMO-LM to the ICON-LAM (ICON Limited Area Model) as the operational model. The main aim of this work is the presentation of a sensitivity study performed over a domain located in southern Italy (including the northern part of the Campania region and the southern part of Lazio) aimed to provide a contribution to the definition of a model configuration suitable for accurate weather forecasts over this area. Following the work performed by the authors with COSMO on a similar domain, the week 19-25 November 2018 has been selected as test case, when a low-pressure system coming from Western Mediterranean determined intense storms and gusts.
A computational grid R2B11, characterized by a very high resolution (about 1.2 km), has been adopted. Initial and boundary conditions are provided by the ECMWF IFS model at a spatial resolution of about 8.5 km. The sensitivity was carried out starting from a model configuration optimized by the authors (in a joint effort with the CMCC Foundation, Italy) over the whole Italian area, employing a grid with different resolution (R2B10, about 2.5 km). The reference configuration assumes that the shallow convection parameterization is active whereas the parts treating deep and mid-level convection are switched off. Moreover, a single moment cloud microphysics scheme and a diagnostic Kohler cloud cover scheme are employed.
A first sensitivity was performed with respect to the domain size, considering a reference domain (11.38° – 15.38° E; 40.25° - 42.25° N) and two additional domains respectively larger (in both directions) than 50% and 100% with respect to the original one. Then, a sensitivity to the numerical parameters, which have been shown to play a significant role in determining model response, has been carried out, e.g tkhmin (minimal diffusion coefficient for heat), rlam_heat (factor for laminar resistance for heat) and v0snow (factor for vertical velocity of snow).
Model evaluation has been conducted against ground observation data provided by CIRA instrumentation and by the SCIA system developed by ISPRA (Italy). Moreover, a comparison with forecasts provided by the COSMO model at 0.009° (about 1 km resolution) forced by the same driving data has been performed, in order to highlight the differences between the performances of the two models.
How to cite: Bucchignani, E., Cinquegrana, D., Montesarchio, M., and Zollo, A. L.: A sensitivity study on high resolution ICON-LAM and comparison with COSMO over Southern Italy , EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-57, https://doi.org/10.5194/ems2022-57, 2022.
The structural uncertainty of a climate model is defined as the range of outcomes that can be obtained through different representations of physical processes of the climate system, the selection of different unconstrained parameter values and different choices for the numerical solution of underlying fundamental equations. Exploring the range of these outcomes with the goal of determining a model configuration that produces results in closer agreement with observational data is defined as model calibration or tuning.
In this study, the preliminary results of the Coordinated Parameter Testing 2 (COPAT2) initiative of the CLM-Community are presented. In COPAT2, volunteer members of the community join forces together, with the objective of testing and providing recommended configurations for the new and final version of COSMO-CLM (6.0), as well as for the newly released regional climate model ICON-CLM, for climate modeling applications over the European CORDEX domain.
A series of sensitivity tests is performed in which various configurations of the models are explored. The aspects that are tested have been carefully selected, based on expert judgment. In the case of COSMO-CLM 6.0, the primary focus is on newly introduced and recently updated parameterizations and physical schemes. For ICON-CLM, these tests are the first ever conducted with the climate version of the model and are based on the operational configuration and on information of experiments performed for the development of the NWP mode of ICON.
The simulations are conducted at a horizontal resolution of approximately 12 km over Europe, using ERA5 reanalysis data as boundary conditions. In a first step, a series of relatively short tests is conducted over a 7-year period, from 1979 to 1985. Successively, depending on the sensitivity of the model to the applied changes in its configuration, a sub-set of simulations is extended over a total period of 12 years. The results are systematically analyzed with an evaluation suite that has been further developed and extended for COPAT2. The standardized analysis and condensation of results in very few indices summarizing the models performance allow for an easy and fast comparison of the quality of the different simulations.
Beside introducing preliminary results of the conducted sensitivity tests, an overview of the calibration strategy followed in COPAT2 will be presented, including information on the selected metrics, employed observational data sets and further details inherent to the ranking of the different experiments.
How to cite: Russo, E., Steger, C., Geyer, B., Petrik, R., Keuler, K., Rockel, B., Görgen, K., Ludwig, P., Feldmann, H., Sulis, M., Fallah, B., Truhetz, H., Hagemann, H., Schulz, J.-P., and Pothapakula, P.: The COPAT2 initiative of the CLM-Community: towards a recommended configuration of COSMO-CLM and ICON-CLM new model versions, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-643, https://doi.org/10.5194/ems2022-643, 2022.
Chairpersons: Estíbaliz Gascón, Balázs Szintai
AROME-Arctic (AA) is a regional high-resolution Arctic NWP model system. We will present recent results from the operational implementation, monitoring, and research and development performed at MET Norway towards further improving the short-range forecast capabilities in the European Arctic.
AA is the core basis for several important operational weather-related services. It is the source of weather forecasts on Svalbard and surrounding areas (e.g. spot forecasts presented on Yr.no, automated text forecasts for coastal waters and weather warnings) and various other downstream products in the Barents region, such as polar low and vessel icing forecasts, and serves as upper boundary conditions for ocean models. A configuration of AA is also used in the Copernicus Arctic Regional ReAnalysis.
With its unique location in the Arctic, special emphasis is needed in development and setup of the model in comparison to other model domains which largely cover land and mid-latitude regions. There are not as many conventional observations available in the region, thus remote sensing data is relatively more important. The presence of fast ice, snow on ice and moving sea ice presents unique challenges with large impacts on near surface temperature and humidity in the region.
Various new observation types and techniques are tested and evaluated for consideration in AA data assimilation, i.e. all-sky assimilation for microwave radiances, radar reflectivity assimilation, and ASCAT supermodding.
The AA model produces short-range forecasts (66 hours) every three hours. Downstream production includes lagged EPS products as a first step towards a full probabilistic forecast system with different perturbation strategies, which serves as a source to calibrated probability products, polar low tracking and polar low strike probability calculations.
Challenging aspects of Arctic weather prediction, like fog, are strongly related to the stable boundary layer and its interaction with the surface. A realistic description of the surface, soil and snowpack forms the basis for improvements. Improved surface schemes for soil diffusion and snow, as well as an improved land cover dataset (ECOCLIMAP-SG) and physical soil property data (SOILGRID) are important parts of the ongoing development. A more realistic depiction of the ocean boundary is investigated by experiments with two-way wave coupling of AA in an operational setup.
Additionally, the feasibility of applying AA wind speed forecasts in the pseudo-dynamic sea ice drift setup, and correcting the sea-ice state within the model domain based on the external data will be presented.
Process-based evaluation and verification of AA with observations from selected key locations and periods in the Arctic is performed in addition to routine verification. Further tests include the verification of snowfall by using active remote sensing observations.
How to cite: Grote, R., Støylen, E., and Ødegaard Køltzow, M.: AROME-Arctic - Recent developments of a regional high-resolution Arctic forecasting system., EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-249, https://doi.org/10.5194/ems2022-249, 2022.
The accurate and precise weather information in polar region have proved to be essential for the global weather and climate research. However, even though we leave in the ‘golden age’ of earth observations, there is a lack of in-situ observation coverage in the polar regions. Additionally, the use of radiances measured from polar-orbiting satellites is of limited use due to the difficulties using those kind of data over ice and snow. The major research expedition in Arctic, MOSAiC (Multidisciplinary drifting Observatory of Arctic Climate), managed to shed light on polar conditions with the collection of a huge variety of highly resolved data.
The SynopSys Project (Synoptic events during MOSAiC and their Forecast Reliability in the Troposhere-Stratosphere System) is a collaboration of the German Weather Service (DWD) with the Alfred-Wegener Institute (AWI) and the University of Bremen. The project aims to combine the state-of-the-art weather observations from the MOSAiC Expedition together with remote sensing products and meteorological forecast in order to identify and study synoptic events in the Arctic.
The current work focuses on the evaluation and improvement of the weather forecasting capabilities of ICON-NWP model in Arctic. In this frame, the latest version of the global model ICON is employed to assimilate the different kind MOSAiC data – from synoptic station data to ascending and descending radiosondes. On the one hand, a series of sensitivity studies has taken place to evaluate the different observation systems and identify the ones with the highest influence on the arctic model forecast. The improvement of the weather forecast itself and the weather analysis because of the assimilation of the project data is studied on the other hand, as well as their influence on the weather forecast of the mid-latitudes. The experiment period covers March and April 2020, which is of high meteorological interest, due to the observed day-to-day variability – a cold period at the beginning of the month was followed by strong warm air intrusion, challenging the model forecast and analysis performance.
How to cite: Paschalidi, Z., Cress, A., Jaiser, R., Rinke, A., You, C., Handorf, D., Monsees, F., Weber, M., and Rozanov, A.: Evaluation and Improvement of Arctic Forecast: Data Assimilation of MOSAiC Expedition Data for SynopSys Project, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-387, https://doi.org/10.5194/ems2022-387, 2022.
Numerical weather prediction is expected to profit considerably of an improved knowledge of the still underdetermined state of the atmospheric boundary layer. As of late, the spatially and temporally sparse existing measurements of e.g. radiosondes can be complemented with wind, temperature, and humidity profiles of ground-based remote-sensing instruments. The DWD evaluates several of those instruments for operational deployment in the framework of the project “Pilotstation”. Here, we will present the results of assimilating observations of the most mature of those systems, i.e. microwave radiometer (MWR) and Doppler lidar, into the ICON/KENDA assimilation system of the DWD.
The MWR measures brightness temperatures and thus, the profiles provided by the ICON model have to be transformed to observation space using the forward operator RTTOV-gb. We ran several assimilation experiments, especially with regard to the vertical localisation of the MWR channels. We will demonstrate how this localisation, together with the proper handling of interchannel cross-correlations, was key for obtaining a positive impact on the upper-air forecast statistics.
The Doppler lidar provides horizontal wind measurements, which exhibit a similar quality as the existing radar-wind profiler (RWP) observations and which can be assimilated directly. We will present the results of different assimilation experiments and discuss the impact in comparison with the RWP.
How to cite: Vural, J., Merker, C., Löffler, M., Knist, C., Kayser, M., and Schomburg, A.: Ground-based remote sensing of the atmospheric boundary layer: Assimilating microwave-radiometer and Doppler-lidar observations into the ICON/KENDA system, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-244, https://doi.org/10.5194/ems2022-244, 2022.
Conventional atmospheric measurement systems fail to provide observations of the essential variables characterizing the planetary boundary layer (PBL) with satisfactory spatial and temporal resolutions. Moreover, the available profile observations are very sparsely distributed. Due to this observational gap the thermodynamic structure of the PBL in the initial conditions for NWP is prone to errors, just as the representation of winds in the lower atmosphere. This affects the accuracy of forecasts of high-impact phenomena such as convective storms or winter fog and low stratus, and is relevant for the quality of downstream applications, warnings, and emergency responses.
Humidity and wind exhibit a very high variability in space and time and, together with temperature, determines the atmospheric stability. It is therefore of major interest to investigate the potential benefit of assimilating additional profile observations of humidity, temperature, and wind into the NWP system. In this contribution, we give an overview of our efforts to include novel, ground-based remote sensing profiler observations into the 1km mesh-size, LETKF-based, ensemble data assimilation system COSMO/KENDA-1.
We present the assimilation of brightness temperatures from three microwave radiometers installed on the Swiss Plateau using the RTTOV-gb forward operator, as well as experiments assimilating water vapor mixing ratio and temperature profiles from a Raman lidar located in Payerne. Additionally, we show an assimilation case study with wind lidars focusing on a local wind system in Basel.
The assimilation of Raman lidar observations leads to improved humidity analyses and precipitation forecasts, particularly for high intensities. It is further shown that state-dependent observation errors lead to more skillful results than constant observation errors.
How to cite: Crezee, B., Merker, C., Regenass, D., Leuenberger, D., Vural, J., Haefele, A., Hervo, M., Martucci, G., Bättig, P., and Arpagaus, M.: Assimilation of ground-based remote sensing profiler data at MeteoSwiss, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-568, https://doi.org/10.5194/ems2022-568, 2022.
MetCoOp is a co-operation for operational short range numerical weather forecasting formed by the Swedish Meteorological and Hydrological Institute (SMHI), Norwegian Meteorological Institute (MET), Finnish Meteorological Institute (FMI) and Estonian Environment Agency (ESTEA). MetCoOp forms a part of the United Weather Centre (UWC) initiative, aiming for common NWP-production by ten European countries in 2027. MetCoOp delivers the convection permitting LAM ensemble MEPS with 5 new members generated every hour, and the hourly updated deterministic very-short-range suite MNWC.
The interaction between forecasters and developers has been an important part of MetCoOp since its initiation as a bilateral co-operation between SMHI and MET in 2010. To this end, users and developers at the participating institutes come together weekly at regular video conferences and twice every year for a three-day face-to-face working meeting. Recommendations of the users play an important part in the procedure for introducing new developments into operational use. Typically, forecasters from each institute give their recommendations concerning forecast suites undergoing pre-operational testing based on their own experiences and assessment reports provided by the developers. Furthermore, duty forecasters are engaged in compiling monthly and seasonal verification reports showing the long-term evolution of MetCoOp forecasts.
At the regular video conferences incidents, updates, developments, meteorological performance, and other current events are discussed. Reviews of the meteorological quality of operational and pre-operational forecast suites are a fixed item in these "thursday-meetings", and often trigger vigorous discussion. Developers and users of MetCoOp forecasts take turns delivering these reviews, presenting deterministic and probabilistic verification scores and diagnostics, or cases of particular interest, and everybody is encouraged to sign up for giving a presentation. At the face-to-face meetings MetCoOp staff exchange experiences, work together, and make plans for future developments. At these meetings half a day is devoted to presenting and responding to forecaster feedback from each institute. Summary notes of the video conferences and face-to-face meetings, including slides, are available for future reference on the internal MetCoOp wiki-pages.
How to cite: Fortelius, C., Noer, G., Saarikalle, E., Sild, K., Wettergren, A., Berggren, L., and Eresmaa, R.: Interaction between forecasters and model developers within MetCoOp, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-190, https://doi.org/10.5194/ems2022-190, 2022.
Long-range scanning Doppler lidars can provide the vertical profile of the horizontal wind within the boundary layer. We used Doppler lidar wind profiles from six locations around the globe to verify the wind profile forecasts generated by ECMWF from the operational IFS model. The six locations selected cover a variety of surface types (rural, marine, mountainous urban, coastal urban).
We first validated the Doppler lidar observations at four locations by comparison with collocated radiosonde profiles to ensure that the Doppler lidar observations were of sufficient quality. The two observation types agree well, with the mean absolute error (MAE) in wind speed almost always less than 1 m s-1. Large deviations in the wind direction were usually seen only for low wind speeds. This is due to the ambiguity in wind direction at low wind speed and that the relative uncertainty in the wind direction measurement increases as the wind speeds decrease.
The Doppler lidar observations are at sufficient temporal resolution for us to generate time-height composites of the wind verification with one-hour resolution so that we can investigate the diurnal cycle. Verification of the model winds showed that the IFS model performs best over marine (ocean) locations. Larger errors were seen in locations where the surface was more complex, especially in the wind direction. For example, in Granada, which is near a high mountain range, the IFS model failed to capture a commonly occurring mountain breeze, a feature which is highly dependent on the orography at sub grid-scale.
At one location, we conditionally performed the wind verification based on the presence or absence of a low-level-jet diagnosed from the Doppler lidar observations. The IFS model was able to reproduce the presence of the low-level-jet but the wind speed maximum was about 2 m s-1 lower than observed. This is attributed to the effective vertical resolution of the model being too coarse to create the strong gradients in wind speed observed.
How to cite: Pentikäinen, P. and O'Connor, E.: Using Doppler lidar observations to verify wind profile forecasts from the ECMWF IFS model, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-229, https://doi.org/10.5194/ems2022-229, 2022.
We present a detailed evaluation of the turbulence forecast product eddy dissipation parameter (EDP) used at the Deutscher Wetterdienst (DWD). It is based on the turbulence parameterization scheme TURBDIFF, which is operational within the Icosahedral Nonhydrostatic (ICON) numerical weather prediction model used operationally by DWD. For aviation purposes, the procedure provides the cubic root of the eddy dissipation rate ε1/3 as an overall turbulence index. This quantity is a widely used measure for turbulence intensity as experienced by aircraft. The scheme includes additional sources of turbulent kinetic energy with particular relevance to aviation, which are briefly introduced. These sources describe turbulence generation by the subgrid-scale action of wake eddies, mountain waves, and convection, as well as horizontal shear as found close to fronts or the jet stream. Furthermore, we introduce a postprocessing calibration to an empirical EDR distribution, and we demonstrate the potential as well as limitations of the final EDP-based turbulence forecast by considering several case studies of typical turbulence events. Finally, we reveal the forecasting capability of this product by verifying the model results against one year of aircraft in situ EDR measurements from commercial aircraft. We find that the forecasted EDP performs favorably when compared to the Ellrod index. In particular, the turbulence signal from deep convection, which is accounted for in the EDP product, is advantageous when spatial nonlocality is allowed in the verification procedure.
We further compare against data from the SOUTHTRAC campaign that took place in 2019 over the Andes.
In particular we compare high quality turbulence data from the HALO aircragt against ICON.
The model runs in global and also local mode. Since ICON is not used so far for aviation turbulence
forecasting at convectionn permitting scales we expect insights for the development
of future turbulence products. The focus is on turbulence generated by
mountain waves and jet stream dynamics.
How to cite: Goecke, T., Machulskaya, E., Geldenhuys, M., Ungermann, J., Dörnbrack, A., and Schumann, U.: Aviation Turbulence Forecasting at DWD with ICON: Methodology, Case Studies, and Verification, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-696, https://doi.org/10.5194/ems2022-696, 2022.
To mitigate the impacts associated with adverse weather conditions, meteorological services issue weather warnings to the general public. These warning systems rely heavily on forecasts issued by underlying prediction systems. When deciding which prediction system(s) to utilise when constructing warnings, it is important to compare systems in their ability to forecast the occurrence and severity of extreme weather events: if a warning system has access to more accurate forecasts for extreme weather, then it has the potential to generate more useful warnings. However, evaluating forecasts for extreme events is known to be a challenging task. This is exacerbated further by the fact that high-impact weather often manifests as a result of several confounding features, a realisation that has led to considerable recent research on so-called compound weather events. Both univariate and multivariate methods are therefore required to evaluate forecasts for high-impact weather. In this work, we review weighted verification tools, which allow particular outcomes to be emphasised during forecast evaluation, and demonstrate how these can be used to verify forecasts for compound weather events. We compare different approaches to construct weighted scoring rules, and exploit recent developments to apply these scores in a multivariate setting. Additionally, we leverage existing results on weighted scores to introduce weighted probability integral transform (PIT) histograms, allowing forecast calibration to be assessed conditionally on particular outcomes having occurred. To illustrate the practical benefit afforded by these weighted verification tools, they are applied to forecasts for extreme heat events issued by the Swiss Federal Office of Meteorology and Climatology (MeteoSwiss). Extreme heat events are defined in terms of MeteoSwiss heat warning criteria, and these verification methods therefore assess the forecasts' potential to generate useful weather warnings.
How to cite: Allen, S., Bhend, J., Martius, O., and Ziegel, J.: Evaluating the potential of MeteoSwiss forecast strategies when constructing weather warnings, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-458, https://doi.org/10.5194/ems2022-458, 2022.
Evaluation of Columnar Database Tables for Producing Climate Aggregates on Demand
Dáire Casey, Conor Lally, Dára Elliot
Met Éireann, Maynooth University
Modern automated observing networks generally produce observations at a time resolution of one minute or less. For climatological and other purposes, longer time resolutions are required such as ten minute, hourly, daily, and monthly. Using traditional relational databases, this necessitates the creation of derived values, i.e. aggregates computed from a set of individual observations such as accumulation, averages and extremes.
With the potential for quality control processes to subsequently modify observation values, ensuring consistency between the different time granularities is exceptionally difficult. This can lead to undesirable variations between related original values, derived, and aggregates.
New columnar databases and related analytical query engines provide massive performance enhancements for certain data-types. This introduces the possibility of determining climatological derived values and aggregates on demand as opposed to pre-calculating them. This eliminates the potential for inconsistency between time granularities. But considering the huge size of many observations and climate datasets, is the performance of columnar databases good enough to meet the operational requirements of a National Meteorological Organisation? This will be determined by querying a columnar database and an equivalent row-oriented database with a set of varied queries and comparing the speed at which results are returned. This research will have important implications for observations and climate data storage systems used by National Meteorological Organisations such as Ireland’s Met Éireann. Moreover, this approach may prove advantageous to those wishing to use data science applications on climate and weather data sets on the cloud.
How to cite: Casey, D., Lally, C., and Elliott, D.: Evaluation of Columnar Database Tables for Producing Climate Aggregates on Demand, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-255, https://doi.org/10.5194/ems2022-255, 2022.
Objective forecast verification provides the basis to motivate changes to the forecast system. At MeteoSwiss, we are introducing statistical ensemble postprocessing into our automated forecast production. These automated forecasts are accessed by the Swiss general population through the MeteoSwiss website and mobile app and they will form the basis for a range of derived products. Therefore, it is crucial to evaluate the new forecasts broadly, i.e. on the diverse aspects of forecast quality relevant for the variety of products used across all of Switzerland.
The core component of this evaluation system consists of a web portal for interactive visualization of verification measures. This web portal provides the means to compare forecast meteograms at stations with the corresponding observations for a quick visual inspection akin to the information available on the mobile app. In addition, forecast quality of the most recent forecasts is monitored using various verification measures. Finally, re-forecasting with experimental configurations is integrated to produce in-depth reporting on the effect of novel forecasts to support decisions on the development of the forecasting system. To facilitate near real-time analyses, atomic verification scores per meteorological parameter, forecast source, forecast issue time, station, and time are pre-computed and stored in a partitioned database. The partitioning allows for rapid multi-threaded access at analysis time from the interactive web portal. The objective verification is complemented with feedback by forecasters on duty on individual cases during the pre-operational phase of new forecast developments. Here, we will showcase how each of these parts is used to assess release candidates for the automatic forecast production.
How to cite: Bhend, J., Spirig, C., Moret, L., and Liniger, M. A.: Objective verification for development and monitoring of automated weather forecasts, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-332, https://doi.org/10.5194/ems2022-332, 2022.
Verification of precipitation forecasts are rarely published in scientific literature. This study deals with a verification of the 24-hour-precipitation forecast of the global numerical weather prediction models GFS, ICON, ARPEGE and NEMS, as well as the reanalysis model ERA5.
For model comparison more than 10’000 precipitation measurements worldwide from the measurement network METAR were used in hourly resolution. Annual, monthly, and daily precipitation sums of the global prediction models were compared with quality-controlled measurements for the year 2020. Continuous error metrics (e.g., mean absolute errors (MAE), mean bias errors (MBE)) as well as categorical error metrics (e.g., probability of detection, false alarm rate, Heidke Skill Score) are calculated for each measurement location separately, as well as averaged by a 2 x 2 degrees grid to account for the unequal global distribution of the measurements.
In general, the model errors are large in equatorial regions, areas close to the coast and in mountainous areas, where the annual precipitation amount is typically large. The best model performance was analysed for the reanalysis model ERA5 with a mean absolute error of 234 mm and a small mean bias error of -6 mm. ICON showed the lowest MAE of the operational weather forecast models (MAE = 253 mm) followed by GFS (MAE = 270 mm), ARPEGE (MAE = 281 mm) and NEMS (MAE = 296 mm). NEMS and ICON tend to underestimate annual precipitation amounts, whereas GFS and ARPEGE tend to overestimate annual precipitation amounts. Daily precipitation events larger than 1 mm were detected most accurate with ICON (HSS = 0.46) and ERA5 (HSS = 0.45), followed by GFS, ARPEGE and NEMS.
For each country, the model with the highest accuracy was determined based on the MAE of annual precipitation sums and based on the accuracy of daily precipitation events. For more than 80 % of all countries worldwide ERA5 or ICON showed the best model performance.
How to cite: Schlögl, S., Huonder, U., and Müller, M.: Global verification of the 24-hour precipitation forecast of numerical weather prediction models, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-383, https://doi.org/10.5194/ems2022-383, 2022.
This study introduces the remapping process of the state-of-the-art Numerical Weather Prediction (NWP) model, the Korean Integrated Model (KIM) system, developed by the Korea Institute of Atmospheric Prediction Systems (KIAPS). The KIM employs the quasi-uniform and pole singularity-free spherical geometry, a cubed-sphere grid, essentially requiring the grid transformation in the pre- and post-process. The remapping process of the KIM consists of two parts: (1) figuring out reference points from source grid points and weights at each reference point and (2) calculating the values at destination points by multiplying the values at reference points and appropriate weights. The KIM uses the linked-cell mapping algorithm to search the reference points around the destination point. The KIM provides various interpolation methods including the alternative-bilinear algorithm (Kim et al., 2019), which can obtain the weights from irregular grid points without assuming a rectangular shape. We are developing an advanced pre- and post-process that downscales cubed-sphere grids to a higher resolution using neural network techniques as an ongoing project. We also plan to share our results that improved the efficiency of the remapping process, which requires a massive resource compared to the simplicity of its calculation. For example, only the address and weights of reference grid points required for the interpolation are stored in a remap matrix (pruning) as an efficient storage managing strategy. Moreover, the advantage of parallelization, evaluated in the Korean 5th super-computer, will also be presented. Through parallelizing the remapping process, the speed improves up to about twelve times, and a more stable calculation speed is expected.
How to cite: Kim, S.-W., Shim, T., and Kim, J.: The State-of-the-art Remapping Process for the Korean Integrated Model (KIM), EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-337, https://doi.org/10.5194/ems2022-337, 2022.
