Forecasts Ensemble Prediction System Research Articles

Regional Solar Irradiance Forecast for Kanto Region by Support Vector Regression Using Forecast of Meso-Ensemble Prediction System

From the perspective of stable operation of the power transmission system, the transmission system operators (TSO) needs to procure reserve adjustment power at the stage of the previous day based on solar power forecast information from global horizontal irradiance (GHI). Because the reserve adjustment power is determined based on information on major outliers in past forecasts, reducing the maximum forecast error in addition to improving the average forecast accuracy is extremely important from the perspective of grid operation. In the past, researchers have proposed various methods combining the numerical weather prediction (NWP) and machine learning techniques for the one day-ahead solar power forecasting, but the accuracy of NWP has been a bottleneck issue. In recent years, the development of the ensemble prediction system (EPS) forecasts based on probabilistic approaches has been promoted to improve the accuracy of NWP, and in Japan, EPS forecasts in the mesoscale domain, called mesoscale ensemble prediction system (MEPS), have been distributed by the Japan Meteorological Agency (JMA). The use of EPS as a machine learning model is expected to improve the maximum forecast error, as well as the accuracy, since the predictor can utilize various weather scenarios as information. The purpose of this study is to examine the effect of EPS on the GHI prediction and the structure of the machine learning model that can effectively use EPS. In this study, we constructed the support vector regression (SVR)-based predictors with multiple network configurations using MEPS as input and evaluated the forecast error of the Kanto region GHI by each model. Through the comparison of the prediction results, it was shown that the machine learning model can achieve average accuracy improvement while reducing the maximum prediction error by MEPS, and knowledge was obtained on how to effectively provide EPS information to the predictor. In addition, machine learning was found to be useful in improving the systematic error of MEPS.

Robust flight planning impact assessment considering convective phenomena

Thunderstorms are one of the leading causes of Air Traffic Management delays. In this paper, we assess how incorporating convective information into flight planning algorithms can lead to reductions in reroutings due to storm encounters during the execution of the flight. We use robust open-loop optimal control methodology at the flight planning level and incorporate meteorological uncertainties based on Ensemble Prediction System forecasts. Convective risk areas can be derived from the latter to be included in the objective function. At the execution level, the planned trajectories are included in an air traffic simulator (NAVSIM) under observed weather (wind and storms). In this simulation process, track modifications might be triggered in case of encountering an observed thunderstorm. A tool termed DIVMET based on pathfinding algorithms has been integrated into NAVSIM is considered to that end. Results show that planning robust trajectories (avoiding thus convective areas) reduces the number of storms encounters and increases predictability. This increase in predictability is at a cost in terms of fuel and time, also quantified.

Dependence of skill and spread of the ensemble forecasts on the type of perturbation and its relationship with long-term norms of precipitation and temperature

A new computing cluster has been operating since 2016 at the Institute of Meteorology and Water Management – National Research Institute. Increasing computing power enabled the implementation of ensemble prediction system forecasts in the operational mode and the use of a new computer for research purposes. As part of the priority project on “Study of Disturbances in the Representation of Modeling Uncertainty in Ensemble Development” and the earlier project entitled “COSMO Towards Ensemble in Km in Our Countries), implemented in the Working Group 7 (Predictability and Ensemble Methods) as part of the COSMO modeling consortium, specific studies were carried out to test ensemble forecasts. This research concerned the impact of variability of physical fields characterizing the soil surface (a selected parameter determining evaporation from the soil surface and soil surface temperature) using various methods of perturbation. Numerical experiments were completed for the warm period (from June to September) 2013.

Postprocessing ensemble forecasts of vertical temperature profiles

Abstract. Weather forecasts from ensemble prediction systems (EPS) are improved by statistical models trained on past EPS forecasts and their atmospheric observations. Recently these corrections have moved from being univariate to multivariate. The focus has been on (quasi-)horizontal atmospheric variables. This paper extends the correction methods to EPS forecasts of vertical profiles in two steps. First univariate distributional regression methods correct the probability distributions separately at each vertical level. In the second step copula coupling re-installs the dependence among neighboring levels by using the rank order structure of the EPS forecasts. The method is applied to EPS data from the European Centre for Medium-Range Weather Forecasts (ECMWF) at model levels interpolated to four locations in Germany, from which radiosondes are released to measure profiles of temperature and other variables four times a day. A winter case study and a summer case study, respectively, exemplify that univariate postprocessing fails to preserve stable layers, which are crucial for many atmospheric processes. Quantile resampling and a resampling that preserves the relative distance between individual EPS members improve the calibration of the raw forecasts of the temperature profiles as shown by rank histograms. They also improve the multivariate metrics of energy score and variogram score and retain the stable layers. Improvements take place over all times of the day and all seasons. They are largest within the atmospheric boundary layer and for shorter lead times.

Comparative Study on Probabilistic Forecasts of Heavy Rainfall in Mountainous Areas of the Wujiang River Basin in China Based on TIGGE Data

Based on the ensemble precipitation forecast data in the summers of 2014–2018 from the Observing System Research and Predictability Experiment (THORPEX) Interactive Grand Global Ensemble (TIGGE), a comparative study of two multi-model ensemble methods, the Bayesian model average (BMA) and the logistic regression (LR), was conducted. Meanwhile, forecasts of heavy precipitation from the two models over the Wujiang River Basin in China for the summer of 2018 were compared to verify their performances. The training period sensitivity test results show that a training period of 2 years was the best for BMA probability forecast model. Compared with the BMA method, the LR model required more statistical samples and its optimal length of the training period was 5 years. According to the Brier score (BS), for precipitation events exceeding 10 mm with lead times of 1–7 days, the BMA outperformed the LR and the raw ensemble prediction system forecasts (RAW) except for forecasts with a lead time of 1 day. Furthermore, for heavy rainfall events exceeding 25 and 50 mm, the RAW and the BMA performed much the same in terms of prediction. The reliability diagram of the two multi-model ensembles (i.e., BMA and LR) was more reliable than the RAW for heavy and moderate rainfall forecasts, and the BMA model had the best performance. The BMA probabilistic forecast can produce a highly concentrated probability density function (PDF) curve and can also provide deterministic forecasts through analyzing percentile forecast results. With regard to the heavy rainfall forecast in mountainous areas, it is recommended to refer to the forecast with a larger percentile between the 75th and 90th percentiles. Nevertheless, extreme events with low probability forecasts may occur and cannot be ignored.

Characterisation and predictability of a strong and a weak forcing severe convective event – a multi-data approach

Two severe summer-time convective events in Germany are investigated which can be classified by the prevailing synoptic conditions into a strong and a weak forcing case. The strong forcing case exhibits a larger scale precipitation pattern caused by frontal ascent whereas scattered convection is dominating the convective activity in the weak forcing case. Other distinguished differences between the cases are faster movement of convective cells and larger regions with significant loss mainly due to severe gusts in the strong forcing case. A comprehensive set of various observations is used to characterise the two different events. The observations include measurements from a lightning detection network, precipitation radar, geostationary satellite and weather stations, as well as information from an automated cell detection algorithm based on radar reflectivity which is combined with severe weather reports, and damage data from insurances. Forecast performance at various time scales is analysed ranging from nowcasting and warning to short-range forecasting. Various methods and models are examined, including human warnings, observation-based nowcasting algorithms and high-resolution ensemble prediction systems. The analysis shows the advantages of a multi-sensor and multi-source approach in characterising convective events and their impacts. Using data from various sources allows to combine the different strengths of observational data sets, especially in terms of spatial coverage or data accuracy, e.g. damage data from insurances provide good spatial coverage with little meteorological information while measurements at weather stations provide accurate but pointwise observations. Furthermore, using data from multiple sources allow for a better understanding of the convective life cycle. Several parameters from different instruments are shown to have a predictive skill for convective development, these include satellite-based cloud-top cooling rates as measure for intensive convective growth, 3D-radar reflectivity, mesocyclone detection from doppler radar, overshooting top detection or lightning jumps to evaluate storm intensification and formation of severe weather. This synergetic approach can help to improve nowcasting algorihtms and thus the warning process. The predictability of the analysed severe convective events differs with different types of forcing which is reflected in both, convective-scale ensemble prediction system forecasts and human weather warnings. Human warnings show larger false alarm rates in the weak forcing case. Ensemble predictions are able to capture the characteristics of the convective precipitation. The forecast skill is connected strongly to the synoptic situation and the presence of large-scale forcing increases the forecast skill. This has to be considered for potential future warn-on-forecast strategies.

High resolution forecasting for wind energy applications using Bayesian model averaging

Two methods of post-processing the uncalibrated wind speed forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF) ensemble prediction system (EPS) are presented here. Both methods involve statistically post-processing the EPS or a downscaled version of it with Bayesian model averaging (BMA). The first method applies BMA directly to the EPS data. The second method involves clustering the EPS to eight representative members (RMs) and downscaling the data through two limited area models at two resolutions. Four weighted ensemble mean forecasts are produced and used as input to the BMA method. Both methods are tested against 13 meteorological stations around Ireland with 1 yr of forecast/observation data. Results show calibration and accuracy improvements using both methods, with the best results stemming from Method 2, which has comparatively low mean absolute error and continuous ranked probability scores.

Verification of TIGGE Multimodel and ECMWF Reforecast-Calibrated Probabilistic Precipitation Forecasts over the Contiguous United States*

Abstract Probabilistic quantitative precipitation forecasts (PQPFs) were generated from The Observing System Research and Predictability Experiment (THORPEX) Interactive Grand Global Ensemble (TIGGE) database from July to October 2010 using data from Europe (ECMWF), the United Kingdom [Met Office (UKMO)], the United States (NCEP), and Canada [Canadian Meteorological Centre (CMC)]. Forecasts of 24-h accumulated precipitation were evaluated at 1° grid spacing within the contiguous United States against analysis data based on gauges and bias-corrected radar data. PQPFs from ECMWF’s ensembles generally had the highest skill of the raw ensemble forecasts, followed by CMC. Those of UKMO and NCEP were less skillful. PQPFs from CMC forecasts were the most reliable but the least sharp, and PQPFs from NCEP and UKMO ensembles were the least reliable but sharper. Multimodel PQPFs were more reliable and skillful than individual ensemble prediction system forecasts. The improvement was larger for heavier precipitation events [e.g., >10 mm (24 h)−1] than for smaller events [e.g., >1 mm (24 h)−1]. ECMWF ensembles were statistically postprocessed using extended logistic regression and the five-member weekly reforecasts for the June–November period of 2002–09, the period where precipitation analyses were also available. Multimodel ensembles were also postprocessed using logistic regression and the last 30 days of prior forecasts and analyses. The reforecast-calibrated ECMWF PQPFs were much more skillful and reliable for the heavier precipitation events than ECMWF raw forecasts but much less sharp. Raw multimodel PQPFs were generally more skillful than reforecast-calibrated ECMWF PQPFs for the light precipitation events but had about the same skill for the higher-precipitation events; also, they were sharper but somewhat less reliable than ECMWF reforecast-based PQPFs. Postprocessed multimodel PQPFs did not provide as much improvement to the raw multimodel PQPF as the reforecast-based processing did to the ECMWF forecast. The evidence presented here suggests that all operational centers, even ECMWF, would benefit from the open, real-time sharing of precipitation forecast data and the use of reforecasts.

Singular Vectors for Tropical Cyclone–Like Vortices in a Nondivergent Barotropic Framework

Abstract In this study, singular vectors (SVs) are calculated for tropical cyclone (TC)–like vortices on an f plane and β plane using a barotropic model, and the structure and time evolution of the SVs are investigated. In the f-plane study, SVs are calculated for TC-like vortices that do and do not satisfy a necessary condition of barotropic instability of normal modes, in which the vorticity gradient changes sign. It is found that, in the case where the initial vortices do not meet the condition, 1) the SVs are tilted against the shear of the background angular velocity as found earlier by Nolan and Farrell, indicating the growth of SVs through the Orr mechanism; 2) the leading singular value increases with the maximum tangential wind speed Vmax and decreases with the radius of the maximum wind (RMW); and 3) the locations of SVs move outward with increasing RMW, Vmax, and the optimization time. In the case where the initial vortex allows for barotropic instability, the SV is initially tilted against the background shear and exhibits transient growth for a limited period. At a certain time during the initial growth, the SV “locks in” to a normal mode structure and remains in that structure so that it may grow exponentially with time. In contrast to the SVs on an f plane, the azimuthal distribution of the SVs on a β plane becomes more asymmetric, and the extent of the asymmetry increases as the strength of the beta gyres increases. On the β plane, all first and second SVs calculated in this study have an azimuthal wavenumber-1 structure at the optimization time, regardless of whether the vorticity gradient of initial TC-like vortices changes sign and the TC-like vortices include the beta gyres at initial time. It is found that when the first and second SVs are used as ensemble initial perturbations, the linear combination of the initial first and second SVs can shift the vortex toward any direction at the optimization time. This is true even when SVs with a low horizontal resolution are used as initial perturbations, as in the European Centre for Medium-Range Weather Forecasts (ECMWF) and Japan Meteorological Agency (JMA) ensemble prediction system. Such wavenumber-1 perturbations could be useful for generating sufficient spread among the tropical cyclone tracks in ensemble forecasts.

Novel Metrics for Evaluation of Ensemble Forecasts of Tropical Cyclone Structure

Abstract Predicting extratropical transition (ET) of a tropical cyclone poses a significant challenge to numerical forecast models because the storm evolution depends on both the timing of the phasing between the tropical cyclone and midlatitude weather systems and the structures of each system. Ensemble prediction systems offer the potential for assessing confidence in numerical guidance during ET cases. Thus, forecasts of storm structure changes during ET from the European Centre for Medium-Range Weather Forecasts (ECMWF) ensemble prediction system (EPS) are explored using two novel validation approaches. The evolution of the (initially tropical) storm structure is characterized in the framework of the cyclone phase space (CPS) and the validation metrics are based on separation between the EPS forecasts and verifying analyses in the CPS. The first validation approach utilizes two metrics and most closely resembles traditional forecast validation techniques. The second approach involves clustering the ensemble member initializations and operational analyses during the life cycles of each tropical cyclone to provide a reference structure evolution against which to evaluate the EPS forecasts. Application of these metrics is demonstrated for two case studies of ET in the western North Pacific: Typhoons Tokage (2004) and Maemi (2003). Both validation approaches identify a decline in EPS structure forecast accuracy for all valid times coinciding with ET onset and beyond, as well as during a weakening tropical stage prior to ET for Tokage. While track forecast errors contribute to structure errors in the EPS forecasts, they are not an overwhelming factor. The two validation approaches highlight the inability of ensemble member forecasts to appropriately weaken the warm core prior to and during ET, and the effects this has on forecasts of ET timing. The analyses adopted in this study provide a basis for future assessments of ensemble forecast skill of cyclone structure during ET.

Regional Differences in the Prediction of Extratropical Cyclones by the ECMWF Ensemble Prediction System

Abstract A regional study of the prediction of extratropical cyclones by the European Centre for Medium-Range Weather Forecasts (ECMWF) Ensemble Prediction System (EPS) has been performed. An objective feature-tracking method has been used to identify and track the cyclones along the forecast trajectories. Forecast error statistics have then been produced for the position, intensity, and propagation speed of the storms. In previous work, data limitations meant it was only possible to present the diagnostics for the entire Northern Hemisphere (NH) or Southern Hemisphere. A larger data sample has allowed the diagnostics to be computed separately for smaller regions around the globe and has made it possible to explore the regional differences in the prediction of storms by the EPS. Results show that in the NH there is a larger ensemble mean error in the position of storms over the Atlantic Ocean. Further analysis revealed that this is mainly due to errors in the prediction of storm propagation speed rather than in direction. Forecast storms propagate too slowly in all regions, but the bias is about 2 times as large in the NH Atlantic region. The results show that storm intensity is generally overpredicted over the ocean and underpredicted over the land and that the absolute error in intensity is larger over the ocean than over the land. In the NH, large errors occur in the prediction of the intensity of storms that originate as tropical cyclones but then move into the extratropics. The ensemble is underdispersive for the intensity of cyclones (i.e., the spread is smaller than the mean error) in all regions. The spatial patterns of the ensemble mean error and ensemble spread are very different for the intensity of cyclones. Spatial distributions of the ensemble mean error suggest that large errors occur during the growth phase of storm development, but this is not indicated by the spatial distributions of the ensemble spread. In the NH there are further differences. First, the large errors in the prediction of the intensity of cyclones that originate in the tropics are not indicated by the spread. Second, the ensemble mean error is larger over the Pacific Ocean than over the Atlantic, whereas the opposite is true for the spread. The use of a storm-tracking approach, to both weather forecasters and developers of forecast systems, is also discussed.

Verification of the ECMWF Ensemble Prediction System Forecasts: A Study of Large-Scale Patterns

Abstract The aim of this work is the evaluation of some basic measures of skill of the Ensemble Prediction System used operationally at the European Centre for Medium-Range Weather Forecasts. Probabilistic forecasts for the occurrence of predefined flow patterns are analyzed and validated throughout the 10-day forecast range. The weather regimes used as fixed patterns are those defined by Vautard for the 700-hPa geopotential height and used during the field phase of the Fronts and Atlantic Storm-Track Experiment as a component in the planning of the intense observing periods. The evaluation of the ensemble forecasts is performed for three cold seasons, from 1996 to 1999, using the Brier score and the related skill score. In this case the sample climate and a simple probabilistic forecast (poor man's ensemble) have been used as reference forecasts. The results show a positive skill with respect to the reference forecasts up to the 240-h time step.

Ensemble prediction of tropical cyclones using targeted diabatic singular vectors

The feasibility of using the European Centre for Medium-Range Weather Forecasts's ensemble prediction system for tropical-cyclone prediction is studied in terms of the spread in cyclone tracks and intensities. It is shown that significant spread in the tracks is obtained through the use of initial perturbations based on targeted singular vectors derived using linearized physics. Inclusion of stochastic physics in the forecast model leads to larger spread in the central pressures. The spread in the tracks shows a high level of sensitivity to the background state used in deriving the singular vectors.

Ensemble prediction of tropical cyclones using targeted diabatic singular vectors

AbstractThe feasibility of using the European Centre for Medium‐Range Weather Forecasts's ensemble prediction system for tropical‐cyclone prediction is studied in terms of the spread in cyclone tracks and intensities. It is shown that significant spread in the tracks is obtained through the use of initial perturbations based on targeted singular vectors derived using linearized physics. Inclusion of stochastic physics in the forecast model leads to larger spread in the central pressures. The spread in the tracks shows a high level of sensitivity to the background state used in deriving the singular vectors.