Convection-permitting Ensemble Prediction System Research Articles

Impacts of Multiscale Components of Initial Perturbations on Error Growth Characteristics and Ensemble Forecasting Skill

Abstract To compare the roles of two kinds of initial perturbations in a convection-permitting ensemble prediction system (CPEPS) and reveal the effects of the differences in large-scale/small-scale perturbation components on the CPEPS, three initial perturbation schemes are introduced, including a dynamical downscaling (DOWN) scheme originating from a coarse-resolution model, a multiscale ensemble transform Kalman filter (ETKF) scheme, and a filtered ETKF (ETKF_LARGE) scheme. First, the comparisons between the DOWN and ETKF schemes reveal that they behave differently in many ways. Specifically, the ensemble spread and forecast error for precipitation in the DOWN scheme are larger than those in the ETKF; the probabilistic forecasting skill for precipitation in the DOWN scheme is better than that in the ETKF at small neighborhood radii, whereas the advantages of the ETKF begin to appear as the neighborhood radius increases; DOWN possesses better spread–skill relationships than ETKF and has comparable probabilistic forecasting skills for nonprecipitation. Second, the comparisons between DOWN and ETKF_LARGE indicate that the differences in the large-scale initial perturbation components are key to the differences between DOWN and ETKF. Third, the comparisons between ETKF and ETKF_LARGE demonstrate that the small-scale initial perturbations are important since they can increase the precipitation spread in the early times and decrease the forecast errors while simultaneously improving the probabilistic forecasting skill for precipitation. Given the advantages of the DOWN and ETKF schemes and the importance of both large-scale and small-scale initial perturbations, multiscale initial perturbations should be constructed in future research.

A Comparison of Sea Surface Temperature Perturbation Methods for a Convection Permitting Ensemble Prediction System Over the European Arctic

A Comparison of Sea Surface Temperature Perturbation Methods for a Convection Permitting Ensemble Prediction System Over the European Arctic

Machine Learning Methods for Postprocessing Ensemble Forecasts of Wind Gusts: A Systematic Comparison

Abstract Postprocessing ensemble weather predictions to correct systematic errors has become a standard practice in research and operations. However, only a few recent studies have focused on ensemble postprocessing of wind gust forecasts, despite its importance for severe weather warnings. Here, we provide a comprehensive review and systematic comparison of eight statistical and machine learning methods for probabilistic wind gust forecasting via ensemble postprocessing that can be divided in three groups: state-of-the-art postprocessing techniques from statistics [ensemble model output statistics (EMOS), member-by-member postprocessing, isotonic distributional regression], established machine learning methods (gradient-boosting extended EMOS, quantile regression forests), and neural network–based approaches (distributional regression network, Bernstein quantile network, histogram estimation network). The methods are systematically compared using 6 years of data from a high-resolution, convection-permitting ensemble prediction system that was run operationally at the German weather service, and hourly observations at 175 surface weather stations in Germany. While all postprocessing methods yield calibrated forecasts and are able to correct the systematic errors of the raw ensemble predictions, incorporating information from additional meteorological predictor variables beyond wind gusts leads to significant improvements in forecast skill. In particular, we propose a flexible framework of locally adaptive neural networks with different probabilistic forecast types as output, which not only significantly outperform all benchmark postprocessing methods but also learn physically consistent relations associated with the diurnal cycle, especially the evening transition of the planetary boundary layer.

Impacts of different perturbation methods on multiscale interactions between multisource perturbations for convection‐permitting ensemble forecasting during SCMREX

AbstractThis study investigates the impacts of different perturbation methods on the multiscale interactions between different‐source perturbations and precipitation forecasting in convection‐permitting short‐term forecasts. Initial Condition (IC) and MOdel physics (MO) perturbations are applied to a convection‐permitting ensemble prediction system (CPEPS) in southern China for the Southern China Monsoon Rainfall Experiment (SCMREX) in May 2014. The analysis presented in this study is based on 32 12‐hr forecasts and builds upon perturbation methods explored in the previous work. Different perturbation methods show different multiscale characteristics especially for four factors: the meso‐β‐scale saturation rates and saturation values for MO perturbations and the forecast‐perturbation magnitudes and changes in convective activities relative to the control member for IC perturbations. All of these factors are closely associated with the occurrence of dispersion reduction due to adding MO perturbations to IC perturbations. Such factors can be changed to different degrees by modifying the initial magnitudes of IC perturbations and by combining various types of perturbation methods. The faster meso‐β‐scale saturations and smaller meso‐β‐scale saturation values for MO perturbations, as well as the larger forecast‐perturbation magnitudes and larger enhancements of convective activities relative to the control member for IC perturbations boost the dispersion reduction. To improve precipitation forecasting, it is instructive to apply dispersion reduction to the design of perturbation methods for CPEPSs, where forecast errors have been overestimated by IC perturbations especially at smaller scales.

Accounting for skew when post-processing MOGREPS-UK temperature forecast fields

AbstractWhen statistically post-processing temperature forecasts, it is almost always assumed that the future temperature follows a Gaussian distribution conditional on the output of an ensemble prediction system. Recent studies, however, have demonstrated that it can at times be beneficial to employ alternative parametric families when post-processing temperature forecasts, that are either asymmetric or heavier-tailed than the normal distribution. In this article, we compare choices of the parametric distribution used within the Ensemble Model Output Statistics (EMOS) framework to statistically post-process 2m temperature forecast fields generated by the Met Office’s regional, convection-permitting ensemble prediction system, MOGREPS-UK. Specifically, we study the normal, logistic and skew-logistic distributions. A flexible alternative is also introduced that first applies a Yeo-Johnson transformation to the temperature forecasts prior to post-processing, so that they more readily conform to the assumptions made by established post-processing methods. It is found that accounting for the skewness of temperature when post-processing can enhance the performance of the resulting forecast field, particularly during summer and winter and in mountainous regions.

Case Dependence of Multiscale Interactions between Multisource Perturbations for Convection-Permitting Ensemble Forecasting during SCMREX

AbstractThis study examines the case dependence of the multiscale characteristics of initial condition (IC) and model physics (MO) perturbations and their interactions in a convection-permitting ensemble prediction system (CPEPS), focusing on the 12-h forecasts of precipitation perturbation energy. The case dependence of forecast performances of various ensemble configurations is also examined to gain guidance for CPEPS design. Heavy-rainfall cases over Southern China during the Southern China Monsoon Rainfall Experiment (SCMREX) in May 2014 were discriminated between the strongly and weakly forced events in terms of synoptic-scale forcing, each of which included 10 cases. In the cases with weaker forcing, MO perturbations showed larger influences while the enhancements of convective activities relative to the control member due to IC perturbations were less evident, leading to smaller dispersion reduction due to adding MO perturbations to IC perturbations. Such dispersion reduction was more sensitive to IC and MO perturbation methods in the weakly and strongly forced cases, respectively. The dispersion reduction improved the probabilistic forecasts of precipitation, with more evident improvements in the cases with weaker forcing. To improve the benefits of dispersion reduction in forecasts, it is instructive to elaborately consider the case dependence of dispersion reduction, especially the various sensitivities of dispersion reduction to different-source perturbation methods in various cases, in CPEPS design.

Ensemble approach to weather forecasting

Numerical simulations of the Earth System evolution are the basis of weather forecasts. Due to chaotic nature of the atmosphere, an inevitable inaccuracy of initial state information, and models imperfection their results are always uncertain. To be correctly interpreted by a user, each weather forecast should be supplemented by an estimate of its uncertainty. An ensemble approach is the only practically affordable method to solve the problem. With this approach, it is possible not only to get an a priori estimate of forecast uncertainty, but also to improve a mean forecast skill and to present alternative scenarios of future weather evolution. Ensemble prediction systems are highly time-consuming and can be implemented only on powerful supercomputers. A global ensemble medium-range prediction system has been running operationally at the Hydrometcenter of Russia since 2015. A short-range limited-area convection-permitting ensemble prediction system was developed for meteorological support of the Sochi Olympic Games 2014 and is now applied for research. Characteristics of the systems, peculiarities of their computer implementation and verification results are discussed.

HarmonEPS—The HARMONIE Ensemble Prediction System

Abstract HarmonEPS is the limited-area, short-range, convection-permitting ensemble prediction system developed and maintained by the HIRLAM consortium as part of the shared ALADIN–HIRLAM system. HarmonEPS is the ensemble realization of HARMONIE–AROME, used for operational short-range forecasting in HIRLAM countries. HarmonEPS contains a range of perturbation methodologies to account for uncertainties in the initial conditions, forecast model, surface, and lateral boundary conditions. This paper describes the state of the system at the version labeled cycle 40 and highlights some directions for further development. The different perturbation methods available are evaluated and compared where appropriate. Several institutes have operational or preoperational implementations of HarmonEPS, such as MEPS (Finland, Norway, and Sweden), COMEPS (Denmark), IREPS (Ireland), KEPS (the Netherlands), AEMET-γSREPS (Spain), and RMI-EPS (Belgium), and these systems are briefly described and compared with the ensemble prediction system (IFS ENS) from the European Centre for Medium-Range Weather Forecasts (ECMWF).

Multiscale Characteristics of Different-Source Perturbations and Their Interactions for Convection-Permitting Ensemble Forecasting during SCMREX

AbstractThe multiscale characteristics of initial conditions (ICs), model physics (MO), and lateral boundary conditions (LBCs) perturbations, as well as their interactions, are investigated for an experimental convection-permitting ensemble prediction system used in southern China, with a focus on 32 12-h forecasts collected over a half-month period of the Southern China Monsoon Rainfall Experiment (SCMREX) in May 2014. Forecast perturbations differed from each other in terms of magnitude, evolution, and vertical distribution between components at different scales. The meso-β-scale MO perturbations show faster growth and saturation, as well as larger magnitude than the meso-α-scale ones, especially in the presence of moist convection. For IC perturbations, damping was present for nonprecipitation variables, while rapid growth and saturation occurred for precipitation. Adding LBCs perturbations to IC or MO perturbations caused linear impacts, which caused consistent, although small, perturbation increments. Nonlinear impacts of adding MO perturbations to IC perturbations were evident and were intrinsic to the predictability limits of IC and MO perturbations, which were closely related to moist dynamics. These nonlinear impacts had the most significant effects on meso-β-scale precipitation perturbations and can effectively improve precipitation prediction. These results highlight the importance of developing multiscale and multisource perturbation methods that reasonably include the interactions among different-scale perturbations from different sources, especially the nonlinear impacts of adding MO perturbations to IC perturbations.

MOGREPS-UK Convection-Permitting Ensemble Products for Surface Water Flood Forecasting: Rationale and First Results

ABSTRACT Flooding is one of the costliest hazards in the United Kingdom. A large part of the annual flood damage is caused by surface water flooding that is a direct result of intense rainfall. Traditional catchment-based approaches to flood prediction are not applicable for surface water floods. However, given sufficiently accurate forecasts of rainfall intensity, with sufficient lead time, actions can be taken to reduce their impact. These actions require reliable information about severity and areas at risk that is clear and easily interpreted. The accuracy requirements, in particular, are very challenging, as they relate to prediction of intensities that occur only infrequently and that typically affect only small areas. In this paper, forecasts of intense rainfall from a new convection-permitting ensemble prediction system are evaluated using radar observations of intense rain and surface water flooding reports. An urban flooding case that occurred in Edinburgh in 2011 is first investigated and then a broader look is taken at performance through a 3-month period during the London Olympic and Paralympic Games in 2012. Conclusions are drawn about the value of the ensemble and the particular means of presenting the forecasts, and areas requiring further work are highlighted.

Bayesian Model Verification of NWP Ensemble Forecasts

Abstract Forecasts of convective precipitation have large uncertainties. To consider the forecast uncertainties of convection-permitting models, a convection-permitting ensemble prediction system (EPS) based on the Consortium for Small-scale Modeling (COSMO) model with a horizontal resolution of 2.8 km covering all of Germany is being developed by the Deutscher Wetterdienst (DWD). The deterministic model is named COSMO-DE. Vertical structures of temperature and humidity affect the potential for convective instability. For verification of vertical model profiles, radiosonde data are used. However, the observed state is uncertain by itself because of the well-known limits in observing the atmosphere. In this work the authors use a probabilistic approach, which considers the observation error as well as the model uncertainty to validate multidimensional state vectors (e.g., temperature profiles) of the COSMO-DE-EPS and of two mesoscale ensembles with horizontal resolution of 10 km and parameterized convection. The mesoscale ensembles are the COSMO short-range EPS (COSMO-SREPS) and the COSMO limited-area EPS (COSMO-LEPS). The approach is based on Bayesian statistics and allows for both verification and comparison of ensembles. The investigation period comprises August 2007 for a comparison of the COSMO-DE-EPS with the COSMO-SREPS. A period of 5 days in July 2007 is used to demonstrate the potential of the Bayesian approach for verification by evaluating the COSMO-SREPS and COSMO-LEPS against COSMO-EU analyses. Based on the Bayesian approach, it is shown that the temperature profiles modeled by the COSMO-DE-EPS are more consistent with the observed profiles than those of COSMO-SREPS.

Mismatching Perturbations at the Lateral Boundaries in Limited-Area Ensemble Forecasting: A Case Study

AbstractAn experimental convection-permitting ensemble prediction system (EPS) has recently been developed at the Met Office where the analysis uncertainty is estimated by means of an ensemble transform Kalman filter (ETKF). In this paper, the author reports on a case study where mismatches between the analysis perturbations and the perturbations coming from the lateral boundaries lead to the generation of significant spurious perturbations in the ensemble forecasts of the surface pressure. To alleviate ensemble perturbation mismatches originating from the ensemble technique, he tests a so-called scale-selective ETKF where a revised transform matrix is applied only to the small-scale component of the high-resolution forecasts, while the large-scale component of the analysis perturbations is taken from the driving EPS. Results show that the new approach successfully removes the spurious perturbations in the surface pressure fields and also provides some benefits in the precipitation forecasts for the case studied. An examination of ensemble-derived forecast error covariances reveals that ensemble perturbation mismatches at the lateral boundaries tend to decrease the degree of balance between the mass field and the rotational wind field and to produce more compact horizontal and vertical correlations. Finally, the limitations of the scale-selective approach and future directions are discussed.

Impact of Stochastic Physics in a Convection-Permitting Ensemble

A stochastic physics scheme is tested in the Application of Research to Operations at Mesoscale (AROME) short-range convection-permitting ensemble prediction system. It is an adaptation of ECMWF’s stochastic perturbation of physics tendencies (SPPT) scheme. The probabilistic performance of the AROME model ensemble is found to be significantly improved, when verified against observations over two 2-week periods. The main improvement lies in the ensemble reliability and the spread–skill consistency. Probabilistic scores for several weather parameters are improved. The tendency perturbations have zero mean, but the stochastic perturbations have systematic effects on the model output, which explains much of the score improvement. Ensemble spread is an increasing function of the SPPT space and time correlations. A case study reveals that stochastic physics do not simply increase ensemble spread, they also tend to smooth out high-spread areas over wider geographical areas. Although the ensemble design lacks surface perturbations, there is a significant end impact of SPPT on low-level fields through physical interactions in the atmospheric model.

Hydro-meteorological evaluation of a convection-permitting ensemble prediction system for Mediterranean heavy precipitating events

Abstract. An assessment of the performance of different convection-permitting ensemble prediction systems (EPSs) is performed, with a focus on Heavy Precipitating Events (HPEs). The convective-scale EPS configuration includes perturbations of lateral boundary conditions (LBCs) by using a global ensemble to provide LBCs, initial conditions (ICs) through an ensemble data assimilation technique and perturbations of microphysical parameterisations to account for part of model errors. A probabilistic evaluation is conducted over an 18-day period. A clear improvement is found when uncertainties on LBCs and ICs are considered together, but the chosen microphysical perturbations have no significant impact on probabilistic scores. Innovative evaluation processes for three HPE case studies are implemented. First, maxima diagrams provide a multi-scale analysis of intense rainfall. Second, an hydrological evaluation is performed through the computation of discharge forecasts using hourly ensemble precipitation forecasts as an input. All ensembles behave similarly, but differences are found highlighting the impact of microphysical perturbations on HPEs forecasts, especially for cases involving complex small-scale processes.

Ensemble prediction for nowcasting with a convection-permitting model—I: description of the system and the impact of radar-derived surface precipitation rates

A key strategy to improve the skill of quantitative predictions of precipitation, as well as hazardous weather such as severe thunderstorms and flash floods is to exploit the use of observations of convective activity (e.g. from radar). In this paper, a convection-permitting ensemble prediction system (EPS) aimed at addressing the problems of forecasting localized weather events with relatively short predictability time scale and based on a 1.5 km grid-length version of the Met Office Unified Model is presented. Particular attention is given to the impact of using predicted observations of radar-derived precipitation intensity in the ensemble transform Kalman filter (ETKF) used within the EPS. Our initial results based on the use of a 24-member ensemble of forecasts for two summer case studies show that the convectivescale EPS produces fairly reliable forecasts of temperature, horizontal winds and relative humidity at 1 h lead time, as evident from the inspection of rank histograms. On the other hand, the rank histograms seem also to show that the EPS generates too much spread for forecasts of (i) surface pressure and (ii) surface precipitation intensity. These may indicate that for (i) the value of surface pressure observation error standard deviation used to generate surface pressure rank histograms is too large and for (ii) may be the result of non-Gaussian precipitation observation errors. However, further investigations are needed to better understand these findings. Finally, the inclusion of predicted observations of precipitation from radar in the 24-member EPS considered in this paper does not seem to improve the 1-h lead time forecast skill.

Uncertainties in COSMO-DE precipitation forecasts introduced by model perturbations and variation of lateral boundaries

As a first step towards a convection-permitting ensemble prediction system (EPS), this study explores the use of perturbation methods within the numerical weather prediction (NWP) model COSMO-DE. The study isolates uncertainty sources so that their impact can be separately studied. The focus is set on uncertainties in model physics and lateral boundary conditions which are represented by a multi-parameter and a multi-boundary approach. Experimental ensemble forecasts of precipitation with a lead time of 24 h are generated. Three ensemble setups are constructed: one with model perturbations, one with variations of boundaries and one with combined perturbations. The investigation period comprises 15 days in summer 2007. Deterministic verification shows that each individual member leads to quantitative precipitation forecasts (QPFs) within a reasonable quality range. Verification shows that the probabilistic precipitation forecasts of the experimental ensembles are superior to the deterministic forecasts. Measures of ensemble dispersion show that the impact of the perturbations on the forecast strongly varies with lead time, with model perturbations always dominating the first few hours and variations of lateral boundaries often dominating the following forecast hours. The study concludes that the applied perturbation methods lead to potentially useful probabilistic precipitation forecasts and should be considered as part of a future EPS design.