Bow echo detection and rainfall scenarios : two ways to extract relevant information from convective-scale ensembles

The use of ensemble prediction systems (EPS) enables the quantification of forecast uncertainty. However, the use of EPS is challenging due to the large amount of information it provides. Forecasts from EPS are typically summarised using statistical measures (such as quantiles maps). Although this mathematical representation is effective in capturing the ensemble distribution, it lacks physical consistency, which raises issues for many applications of EPS in an operational context. We propose two different ways for providing physically consistent synthesis of French convection-permitting AROME-EPS forecasts.

The first way is dedicated to a specific type of Mesoscale Convective System (MCS) with a bow shape, called a bow echo. To summarise the risk of bow echoes in AROME-EPS forecasts, a convolutional neural network has been trained to automatically detect these MCSs in AROME-EPS members. Different synthesis plots are based on these detections. They have been designed in collaboration with forecasters and improved, based on feedback since 2021. A case study will be presented to better understand the usefulness of these synthesis plots.

The second way is a rainfall scenario synthesis. The aim is to automatically generate a few scenarios that are representative of the different possible outcomes. Each scenario is a reduced set of EPS members. To design a scenario synthesis, the procedure can be divided into two parts. The first step aims to extract relevant features from each EPS member to reduce the problem dimensionality. Then, clustering is done based on these features. The originality of our work is to leverage the capabilities of deep learning for feature extraction. For this purpose, we use a convolutional autoencoder (CAE) to learn an optimal low-dimensional representation (also called latent space representation) of the input forecast field. In this work, the algorithm is developed to work on 1-hour accumulated rainfall from AROME-EPS. To visualise these scenario synthesis, an interactive plot has been developed. This interactive plot summarises information concerning scenario size, members included in a scenario or trajectory across lead times for a specific member. This plot will be presented for a case study in this presentation.

The two methods proposed are shown to provide an additional and complementary information, useful for facilitating the human expertise. In addition, their design is generic enough to be applied to other events and variables.