scholarly journals Addressing up-scaling methodologies for convection-permitting EPSs using statistical and machine learning tools

2021 ◽  
Vol 18 ◽  
pp. 145-156
Author(s):  
Tiziana Comito ◽  
Colm Clancy ◽  
Conor Daly ◽  
Alan Hally
Keyword(s):  
Machine Learning ◽  
Weather Forecasting ◽  
Forecast Accuracy ◽  
Ensemble Prediction ◽  
Learning Tools ◽  
Ensemble Prediction System ◽  
Rainfall Events ◽  
Levels Of Detail ◽  
Scaling Algorithms ◽  
Up Scaling

Abstract. Convection-permitting weather forecasting models allow for prediction of rainfall events with increasing levels of detail. However, the high resolutions used can create problems and introduce the so-called “double penalty” problem when attempting to verify the forecast accuracy. Post-processing within an ensemble prediction system can help to overcome these issues. In this paper, two new up-scaling algorithms based on Machine Learning and Statistical approaches are proposed and tested. The aim of these tools is to enhance the skill and value of the forecasts and to provide a better tool for forecasters to predict severe weather.

scholarly journals Three-dimensional visualization of ensemble weather forecasts – Part 2: Forecasting warm conveyor belt situations for aircraft-based field campaigns

2015 ◽  
Vol 8 (7) ◽  
pp. 2355-2377 ◽  
Author(s):  
M. Rautenhaus ◽  
C. M. Grams ◽  
A. Schäfler ◽  
R. Westermann
Keyword(s):  
System Integration ◽  
Visual Analysis ◽  
Weather Forecasting ◽  
Three Dimensional ◽  
Conveyor Belt ◽  
Ensemble Prediction ◽  
Wind Fields ◽  
Uncertainty Visualization ◽  
Ensemble Prediction System ◽  
Weather Forecasts

Abstract. We present the application of interactive three-dimensional (3-D) visualization of ensemble weather predictions to forecasting warm conveyor belt situations during aircraft-based atmospheric research campaigns. Motivated by forecast requirements of the T-NAWDEX-Falcon 2012 (THORPEX – North Atlantic Waveguide and Downstream Impact Experiment) campaign, a method to predict 3-D probabilities of the spatial occurrence of warm conveyor belts (WCBs) has been developed. Probabilities are derived from Lagrangian particle trajectories computed on the forecast wind fields of the European Centre for Medium Range Weather Forecasts (ECMWF) ensemble prediction system. Integration of the method into the 3-D ensemble visualization tool Met.3D, introduced in the first part of this study, facilitates interactive visualization of WCB features and derived probabilities in the context of the ECMWF ensemble forecast. We investigate the sensitivity of the method with respect to trajectory seeding and grid spacing of the forecast wind field. Furthermore, we propose a visual analysis method to quantitatively analyse the contribution of ensemble members to a probability region and, thus, to assist the forecaster in interpreting the obtained probabilities. A case study, revisiting a forecast case from T-NAWDEX-Falcon, illustrates the practical application of Met.3D and demonstrates the use of 3-D and uncertainty visualization for weather forecasting and for planning flight routes in the medium forecast range (3 to 7 days before take-off).

scholarly journals Initial perturbations based on Ensemble Transfrom Kalman Filter with rescaling method for ensemble forecast

2021 ◽  
Author(s):  
Jingzhuo Wang ◽  
Jing Chen ◽  
Hanbin Zhang ◽  
Hua Tian ◽  
Yining Shi
Keyword(s):  
Growth Rate ◽  
Kalman Filter ◽  
Weather Forecasting ◽  
Initial Perturbation ◽  
Ensemble Forecast ◽  
Ensemble Prediction ◽  
Forecast Errors ◽  
Model Uncertainties ◽  
Ensemble Prediction System ◽  
Probabilistic Forecasts

AbstractEnsemble forecast is a method to faithfully describe initial and model uncertainties in a weather forecasting system. Initial uncertainties are much more important than model uncertainties in the short-range numerical prediction. Currently, initial uncertainties are described by Ensemble Transform Kalman Filter (ETKF) initial perturbation method in Global and Regional Assimilation and Prediction Enhanced System-Regional Ensemble Prediction System (GRAPES-REPS). However, an initial perturbation distribution similar to the analysis error cannot be yielded in the ETKF method of the GRAPES-REPS. To improve the method, we introduce a regional rescaling factor into the ETKF method (we call it ETKF_R). We also compare the results between the ETKF and ETKF_R methods and further demonstrate how rescaling can affect the initial perturbation characteristics as well as the ensemble forecast skills. The characteristics of the initial ensemble perturbation improve after applying the ETKF_R method. For example, the initial perturbation structures become more reasonable, the perturbations are better able to explain the forecast errors at short lead times, and the lower kinetic energy spectrum as well as perturbation energy at the initial forecast times can lead to a higher growth rate of themselves. Additionally, the ensemble forecast verification results suggest that the ETKF_R method has a better spread-skill relationship, a faster ensemble spread growth rate and a more reasonable rank histogram distribution than ETKF. Furthermore, the rescaling has only a minor impact on the assessment of the sharpness of probabilistic forecasts. The above results all suggest that ETKF_R can be effectively applied to the operational GRAPES-REPS.

scholarly journals From research to applications – examples of operational ensemble post-processing in France using machine learning

2020 ◽  
Vol 27 (2) ◽  
pp. 329-347 ◽  
Author(s):  
Maxime Taillardat ◽  
Olivier Mestre
Keyword(s):  
Machine Learning ◽  
Heat Waves ◽  
Computation Time ◽  
Ensemble Prediction ◽  
Post Processing ◽  
Ensemble Forecasts ◽  
Ensemble Prediction System ◽  
Practical Applications ◽  
Medium Resolution ◽  
Hourly Rainfall

Abstract. Statistical post-processing of ensemble forecasts, from simple linear regressions to more sophisticated techniques, is now a well-known procedure for correcting biased and poorly dispersed ensemble weather predictions. However, practical applications in national weather services are still in their infancy compared to deterministic post-processing. This paper presents two different applications of ensemble post-processing using machine learning at an industrial scale. The first is a station-based post-processing of surface temperature and subsequent interpolation to a grid in a medium-resolution ensemble system. The second is a gridded post-processing of hourly rainfall amounts in a high-resolution ensemble prediction system. The techniques used rely on quantile regression forests (QRFs) and ensemble copula coupling (ECC), chosen for their robustness and simplicity of training regardless of the variable subject to calibration. Moreover, some variants of classical techniques used, such as QRF and ECC, were developed in order to adjust to operational constraints. A forecast anomaly-based QRF is used for temperature for a better prediction of cold and heat waves. A variant of ECC for hourly rainfall was built, accounting for more realistic longer rainfall accumulations. We show that both forecast quality and forecast value are improved compared to the raw ensemble. Finally, comments about model size and computation time are made.

scholarly journals Novel Metrics for Evaluation of Ensemble Forecasts of Tropical Cyclone Structure

2009 ◽  
Vol 137 (9) ◽  
pp. 2830-2850 ◽  
Author(s):  
Daniel Veren ◽  
Jenni L. Evans ◽  
Sarah Jones ◽  
Francesca Chiaromonte
Keyword(s):  
Tropical Cyclone ◽  
Forecast Accuracy ◽  
Ensemble Member ◽  
Life Cycles ◽  
Ensemble Prediction ◽  
Structure Evolution ◽  
Track Forecast ◽  
Forecast Errors ◽  
Reference Structure ◽  
Ensemble Prediction System

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.

scholarly journals Application of Multimodel Superensemble Technique on the TIGGE Suite of Operational Models

Geomatics ◽  
2021 ◽  
Vol 1 (1) ◽  
pp. 81-91
Author(s):  
Amit Bhardwaj ◽  
Vinay Kumar ◽  
Anjali Sharma ◽  
Tushar Sinha ◽  
Surendra Pratap Singh
Keyword(s):  
Real Time ◽  
Weather Forecasting ◽  
Weather Prediction ◽  
Daily Rainfall ◽  
Ensemble Prediction ◽  
Individual Member ◽  
Ensemble Prediction System ◽  
Consensus Forecasts ◽  
European Centre ◽  
Medium Range

One widely recognized portal which provides numerical weather prediction forecasts is “The Observing System Research and Predictability Experiment” (THORPEX) Interactive Grand Global Ensemble (TIGGE), an initiative of WMO project. This data portal provides forecasts from 1 to 16 days (2 weeks in advance) for many variables such as rainfall, winds, geopotential height, temperature, and relative humidity. These weather forecasting centers have delivered near-real-time (with a delay of 48 hours) ensemble prediction system data to three TIGGE data archives since October 2006. This study is based on six years (2008–2013) of daily rainfall data by utilizing output from six centers, namely the European Centre for Medium-Range Weather Forecasts, the National Centre for Environmental Prediction, the Center for Weather Forecast and Climatic Studies, the China Meteorological Agency, the Canadian Meteorological Centre, and the United Kingdom Meteorological Office, and make consensus forecasts of up to 10 days lead time by utilizing the multimodal multilinear regression technique. The prediction is made over the Indian subcontinent, including the Indian Ocean. TRMM3B42 daily rainfall is used as the benchmark to construct the multimodel superensemble (SE) rainfall forecasts. Based on statistical ability ratings, the SE offers a better near-real-time forecast than any single model. On the one hand, the model from the European Centre for Medium-Range Weather Forecasting and the UK Met Office does this more reliably over the Indian domain. In a case of Indian monsoon onset, 05 June 2014, SE carries the lowest RMSE of 8.5 mm and highest correlation of 0.49 among six member models. Overall, the performance of SE remains better than any individual member model from day 1 to day 10.

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

Energies ◽  
2021 ◽  
Vol 14 (11) ◽  
pp. 3245
Author(s):  
Takahiro Takamatsu ◽  
Hideaki Ohtake ◽  
Takashi Oozeki ◽  
Tosiyuki Nakaegawa ◽  
Yuki Honda ◽  
...  
Keyword(s):  
Machine Learning ◽  
Support Vector Regression ◽  
Forecast Error ◽  
Learning Model ◽  
Ensemble Prediction ◽  
Support Vector ◽  
Prediction System ◽  
Ensemble Prediction System ◽  
Machine Learning Model ◽  
Kanto Region

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.

scholarly journals From research to applications – Examples of operational ensemble post-processing in France using machine learning

2020 ◽  
Author(s):  
Maxime Taillardat ◽  
Olivier Mestre
Keyword(s):  
Machine Learning ◽  
Heat Waves ◽  
Computation Time ◽  
Ensemble Prediction ◽  
Post Processing ◽  
Ensemble Forecasts ◽  
Ensemble Prediction System ◽  
Practical Applications ◽  
Medium Resolution ◽  
Hourly Rainfall

Abstract. Statistical post-processing of ensemble forecasts, from simple linear regressions to more sophisticated techniques, is now a well-known procedure in order to correct biased and misdispersed ensemble weather predictions. However, practical applications in National Weather Services is still in its infancy compared to deterministic post-processing. This paper presents two different applications of ensemble post-processing using machine learning at an industrial scale. The first is a station-based post-processing of surface temperature in a medium resolution ensemble system. The second is a gridded post-processing of hourly rainfall amounts in a high resolution ensemble prediction system. The techniques used rely on quantile regression forests (QRF) and ensemble copula coupling (ECC), chosen for their robustness and simplicity of training whatever the variable subject to calibration. Moreover, some variants of classical techniques used such as QRF or ECC have been developed in order to adjust to operational constraints. A forecast anomaly-based QRF is used for temperature for a better prediction of cold and heat waves. A variant of ECC for hourly rainfall is built, accounting for more realistic longer rainfall accumulations. It is shown that forecast quality as well as forecast value is improved compared to the raw ensemble. At last, comments about model size and computation time are made.

Object-based aviaton convection forecasts from global ensemble model

2020 ◽  
Author(s):  
Adrien Warnan
Keyword(s):  
Situational Awareness ◽  
Forecast Accuracy ◽  
Meteorological Conditions ◽  
Air Traffic ◽  
Global Model ◽  
Ensemble Prediction ◽  
Strong Impact ◽  
Ensemble Prediction System ◽  
Weather Events ◽  
Cloud Top Pressure

<div>In the framework of a single european sky and to improve Air Traffic Flow and Capacity</div><div>Management, involved actors need to share a common situational awareness of the airspace</div><div>conditions. These include the meteorological conditions with a focus on weather phenomena with a</div><div>strong impact on air traffic and network operations such as convection.</div><div>Within this framework, Météo-France is working to provide a new product of convection forecasts</div><div>based on ensemble prediction system (EPS) and global model, and is developing a similarity-based</div><div>method (Rottner<em> et al.,</em> 2019) at Centre National de la Recherche Météorologique (CNRM).</div><div>The method aims to detect areas where meteorological conditions are homogeneous which are</div><div>called called objects. The latter are defined by a reference histogram representing the</div><div>meteorological phenomena to be detected and so are physically consistent. The similarity-based</div><div>method can be applied to each member of a EPS. The results can be summarized by a map that</div><div>contains the information predicted by all members of the ensemble. Thus, it provides spatialization</div><div>for local weather events. This method is currently tested to detect rainfall objects, however we can</div><div>apply this method to detect other event type, like convection. To discriminate the convection</div><div>characteristic within these rainfall objects, we use a diagnostic of cloud top pressure extracted from</div><div>global model outputs. Furthermore, to improve the convection forecast accuracy, the similarity-</div><div>based method can also be applied to several models to create a composite of convection forecast.</div><div>Météo-France will soon deliver a convection potential product to aeronautical users.</div>

scholarly journals 3-D visualization of ensemble weather forecasts – Part 2: Forecasting warm conveyor belt situations for aircraft-based field campaigns

2015 ◽  
Vol 8 (2) ◽  
pp. 2161-2212 ◽  
Author(s):  
M. Rautenhaus ◽  
C. M. Grams ◽  
A. Schäfler ◽  
R. Westermann
Keyword(s):  
System Integration ◽  
Visual Analysis ◽  
Weather Forecasting ◽  
Conveyor Belt ◽  
Ensemble Prediction ◽  
Wind Fields ◽  
Uncertainty Visualization ◽  
Ensemble Prediction System ◽  
Conveyor Belts ◽  
Warm Conveyor Belts

Abstract. We present the application of interactive 3-D visualization of ensemble weather predictions to forecasting warm conveyor belt situations during aircraft-based atmospheric research campaigns. Motivated by forecast requirements of the T-NAWDEX-Falcon 2012 campaign, a method to predict 3-D probabilities of the spatial occurrence of warm conveyor belts has been developed. Probabilities are derived from Lagrangian particle trajectories computed on the forecast wind fields of the ECMWF ensemble prediction system. Integration of the method into the 3-D ensemble visualization tool Met.3D, introduced in the first part of this study, facilitates interactive visualization of WCB features and derived probabilities in the context of the ECMWF ensemble forecast. We investigate the sensitivity of the method with respect to trajectory seeding and forecast wind field resolution. Furthermore, we propose a visual analysis method to quantitatively analyse the contribution of ensemble members to a probability region and, thus, to assist the forecaster in interpreting the obtained probabilities. A case study, revisiting a forecast case from T-NAWDEX-Falcon, illustrates the practical application of Met.3D and demonstrates the use of 3-D and uncertainty visualization for weather forecasting and for planning flight routes in the medium forecast range (three to seven days before take-off).

scholarly journals The influence of the new ECMWF Ensemble Prediction System resolution on wind power forecast accuracy and uncertainty estimation

2012 ◽  
Vol 8 (1) ◽  
pp. 143-147 ◽  
Author(s):  
S. Alessandrini ◽  
S. Sperati ◽  
P. Pinson
Keyword(s):  
Wind Power ◽  
Prediction Accuracy ◽  
Wind Farm ◽  
Forecast Accuracy ◽  
Horizontal Resolution ◽  
Ensemble Prediction ◽  
Prediction System ◽  
Meteorological Model ◽  
Ensemble Prediction System ◽  
Medium Range Weather Forecast

Abstract. The importance of wind power forecasting (WPF) is nowadays commonly recognized because it represents a useful tool to reduce problems of grid integration and to facilitate energy trading. If on one side the prediction accuracy is fundamental to these scopes, on the other it has become also clear that a reliable estimation about their uncertainty is paramount. In fact prediction accuracy is unfortunately not constant and can depend on the location of a particular wind farm, on the forecast time and on the atmospheric situation. Previous studies indicated that the spread of power forecasts derived from the Ensemble Prediction System (EPS) in use at the European Centre for Medium-Range Weather Forecast (ECMWF) could be used as indicator of a three-hourly, three days ahead, wind power forecast's accuracy. In this paper a new application of the EPS, whose horizontal resolution was increased on January 2010 from T399/T255 (60 km) to T639/T319 (32 km), shows an improvement in the results implying that the power spread has actually enough correlation with the error calculated on the deterministic forecast in order to be used as an accuracy predictor. The periods for this comparison are from January 2008 until October 2008 (T399/T255) and from January 2011 until October 2011 (T639/T319). Moreover we have focused our attention on the influence of the new EPS configuration on the performance of a deterministic WPF conducted with the ensemble mean: the results show that increasing the EPS resolution yields a single-valued WPF whose performance is comparable with that of the new ECMWF deterministic high-resolution meteorological model, whose spatial resolution increased from T799 (25 km) to T1279 (15 km).

Sign in / Sign up

Export Citation Format

Share Document