Probabilistic Forecasting of Thunderstorms in the Eastern Alps

Simon, Thorsten; Fabsic, Peter; Mayr, Georg J.; Umlauf, Nikolaus; Zeileis, Achim

doi:10.1175/mwr-d-17-0366.1

Cited by 17 publications

(25 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is one more important difference between the present study and the work by Thomas et al (2018), namely the way in which the final model is estimated. After the selection procedure, their final model is fitted by gradient boosting.…”

Section: Connection To Other Studiesmentioning

confidence: 87%

“…We discuss the relation of the present work to two other studies: firstly, a work with meteorological background on the prediction of thunderstorm occurrence in the Eastern Alps (Simon et al, 2018). Secondly, a work from the statistical literature which focuses on gradient boosting for distributional regression and which presents a case study with a count data variable as response (Thomas et al, 2018). Simon et al (2018) postprocess the deterministic highresolution (HRES) ECMWF forecast from 2010 to 2015 to predict the probability of the occurrence of thunderstorms.…”

Section: Connection To Other Studiesmentioning

confidence: 99%

“…This finding suggests than the short training period of the present study covers a sufficient variety of atmospheric processes leading to lightning, which enables the statistical model to learn these processes. Thomas et al (2018) apply a hurdle model with a zerotruncated negative binomial in their study about abundance of wintering sea ducks. The abundance of sea ducks is quantified on a grid which leads to a response quantity with similar properties to the present lightning counts: 75 % zeros and overdispersion.…”

Section: Connection To Other Studiesmentioning

confidence: 99%

“…They also separate the hurdle model for the selection of terms by gradient boosting with stability selection. However, in Thomas et al (2018)'s study, terms for the dispersion parameter θ have also been selected, which could also be a consequence of less regularization within the individual boosting runs.…”

Section: Connection To Other Studiesmentioning

confidence: 99%

“…This concept has been used successfully in several studies (e.g. Simon et al, 2018;Thomas et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

NWP-based lightning prediction using flexible count data regression

Simon

Mayr

Umlauf

et al. 2019

Adv. Stat. Clim. Meteorol. Oceanogr.

Self Cite

View full text Add to dashboard Cite

A method to predict lightning by postprocessing numerical weather prediction (NWP) output is developed for the region of the European Eastern Alps. Cloud-to-ground (CG) flashes -detected by the groundbased Austrian Lightning Detection & Information System (ALDIS) network -are counted on the 18 × 18 km 2 grid of the 51-member NWP ensemble of the European Centre for Medium-Range Weather Forecasts (ECMWF). These counts serve as the target quantity in count data regression models for the occurrence of lightning events and flash counts of CG. The probability of lightning occurrence is modelled by a Bernoulli distribution. The flash counts are modelled with a hurdle approach where the Bernoulli distribution is combined with a zero-truncated negative binomial. In the statistical models the parameters of the distributions are described by additive predictors, which are assembled using potentially nonlinear functions of NWP covariates. Measures of location and spread of 100 direct and derived NWP covariates provide a pool of candidates for the nonlinear terms. A combination of stability selection and gradient boosting identifies the nine (three) most influential terms for the parameters of the Bernoulli (zero-truncated negative binomial) distribution, most of which turn out to be associated with either convective available potential energy (CAPE) or convective precipitation. Markov chain Monte Carlo (MCMC) sampling estimates the final model to provide credible inference of effects, scores, and predictions. The selection of terms and MCMC sampling are applied for data of the year 2016, and out-of-sample performance is evaluated for 2017. The occurrence model outperforms a reference climatology -based on 7 years of data -up to a forecast horizon of 5 days. The flash count model is calibrated and also outperforms climatology for exceedance probabilities, quantiles, and full predictive distributions.

show abstract

Section: Connection To Other Studiesmentioning

confidence: 87%

Section: Connection To Other Studiesmentioning

confidence: 99%

Section: Connection To Other Studiesmentioning

confidence: 99%

Section: Connection To Other Studiesmentioning

confidence: 99%

“…This concept has been used successfully in several studies (e.g. Simon et al, 2018;Thomas et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

NWP-based lightning prediction using flexible count data regression

Simon

Mayr

Umlauf

et al. 2019

Adv. Stat. Clim. Meteorol. Oceanogr.

Self Cite

View full text Add to dashboard Cite

show abstract

A deep‐learning model to predict thunderstorms within 400 km² South Texas domains

Kamangir

Collins

Tissot

et al. 2020

Meteorological Applications

View full text Add to dashboard Cite

A deep-learning neural network (DLNN) model was developed to predict thunderstorm occurrence within 400 km 2 of south Texas (USA) domains for up to 15 hr (±2 hr accuracy) in advance. The input features were chosen primarily from numerical weather prediction model output parameters/variables; cloud-to-ground lightning served as the target. The deep-learning technique used was the stacked deionizing autoencoder (SDAE) in order to create a higher order representation of the features. Logistic regression was then applied to the SDAE output to train the predictive model. An iterative technique was used to determine the optimal SDAE architecture. The performance of the optimized DLNN classifiers exceeded that of the corresponding shallow neural network models, a classifier via a combination of principal component analysis and logistic regression, and operational weather forecasters, based on the same data set. K E Y W O R D Sdeep learning, numerical weather prediction, stacked deionizing autoencoder, thunderstorm prediction

show abstract

Pre‐tactical convection prediction for air traffic flow management using LSTM neural network

Jardines,

Soler,

García‐Heras

et al. 2024

Meteorological Applications

View full text Add to dashboard Cite

This paper aims to explore machine learning techniques for post‐processing high‐resolution Numerical Weather Prediction (NWP) products for the early detection of convection. Data from the Arome Ensemble Prediction System and satellite observations from the Rapidly Developing Thunderstorm (RDT) product by Météo‐France are used to train a recurrent neural network model to predict areas of total convection and moderate convection. The learning task is formulated as a binary classification problem using a long short‐term memory (LSTM) network architecture. Results from the LSTM model are compared with an object‐based probabilistic approach to forecast convection using metrics such as a receiver operating characteristics (ROC) curve, the Brier score and reliability. Results indicate that the LSTM model performs similarly to the object‐based probabilistic benchmark when classifying moderate convection areas and shows improved skill when classifying areas of total convective. Finally, the LSTM model results are presented within an air traffic management context to showcase the potential use of machine learning models within an operational application.

show abstract

Probabilistic Forecasting of Thunderstorms in the Eastern Alps

Cited by 17 publications

References 27 publications

NWP-based lightning prediction using flexible count data regression

NWP-based lightning prediction using flexible count data regression

A deep‐learning model to predict thunderstorms within 400 km² South Texas domains

Pre‐tactical convection prediction for air traffic flow management using LSTM neural network

Contact Info

Product

Resources

About

Probabilistic Forecasting of Thunderstorms in the Eastern Alps

Cited by 17 publications

References 27 publications

NWP-based lightning prediction using flexible count data regression

NWP-based lightning prediction using flexible count data regression

A deep‐learning model to predict thunderstorms within 400 km2 South Texas domains

Pre‐tactical convection prediction for air traffic flow management using LSTM neural network

Contact Info

Product

Resources

About

A deep‐learning model to predict thunderstorms within 400 km² South Texas domains