EuLerian Identification of ascending AirStreams (ELIAS 2.0) in numerical weather prediction and climate models – Part 1: Development of deep learning model

Abstract

Abstract. Physical processes on the synoptic scale are important modulators of the large-scale extratropical circulation. In particular, rapidly ascending airstreams in extratropical cyclones, so-called warm conveyor belts (WCBs), modulate the upper-tropospheric Rossby wave pattern and are sources and magnifiers of forecast uncertainty. Thus, from a process-oriented perspective, numerical weather prediction (NWP) and climate models should adequately represent WCBs. The identification of WCBs usually involves Lagrangian air parcel trajectories that ascend from the lower to the upper troposphere within 2 d. This requires expensive computations and numerical data with high spatial and temporal resolution, which are often not available from standard output. This study introduces a novel framework that aims to predict the footprints of the WCB inflow, ascent, and outflow stages over the Northern Hemisphere from instantaneous gridded fields using convolutional neural networks (CNNs). With its comparably low computational costs and relying on standard model output alone, the new diagnostic enables the systematic investigation of WCBs in large data sets such as ensemble reforecast or climate model projections, which are mostly not suited for trajectory calculations. Building on the insights from a logistic regression approach of a previous study, the CNNs are trained using a combination of meteorological parameters as predictors and trajectory-based WCB footprints as predictands. Validation of the networks against the trajectory-based data set confirms that the CNN models reliably replicate the climatological frequency of WCBs as well as their footprints at instantaneous time steps. The CNN models significantly outperform previously developed logistic regression models. Including time-lagged information on the occurrence of WCB ascent as a predictor for the inflow and outflow stages further improves the models' skill considerably. A companion study demonstrates versatile applications of the CNNs in different data sets including the verification of WCBs in ensemble forecasts. Overall, the diagnostic demonstrates how deep learning methods may be used to investigate the representation of weather systems and their related processes in NWP and climate models in order to shed light on forecast uncertainty and systematic biases from a process-oriented perspective.

Everything Hits at Once: How Remote Rainfall Matters for the Prediction of the 2021 North American Heat Wave

“…Forecasts with the 30% lowest and highest RMSE are grouped into the “good” and “bad” categories, respectively, with overall 230 individual forecasts in each group. Within these subgroups, imprints of WCBs are detected by using a novel technique based on convolutional neural networks (ELIAS2.0; Quinting & Grams, 2022; Quinting et al., 2022; Supplementary Methods in Supporting Information S1).…”

“…The LAGRANTO documentation and information on how to access the source code are provided in Sprenger and Wernli (2015). Information and the source code for the convolutional neural networks model ELIAS 2.0 are available from Quinting and Grams (2022a, 2022b) and Quinting et al. (2022).…”

Warm conveyor belt activity over the Pacific: Modulation by the Madden-Julian Oscillation and impact on tropical-extratropical teleconnections

“…In addition to the trajectory-based data set, we employ a deep learning-based WCB data set, which is derived with the methodology introduced by Quinting and Grams (2022). The intention is to provide a baseline for addressing the overarching questions https://doi.org/10.5194/egusphere-2023-783 Preprint.…”

“…In short, Quinting and Grams (2022) developed separate convolutional neural network (CNN) models with variants of the UNet architecture (Ronneberger et al, 2015) for each of the three WCB stages. The CNN models take five atmospheric variables as predictors and provide conditional probabilities of occurrence for each WCB stage as output.…”

Towards a diagnostic framework unifying different perspectives on blocking dynamics: insight into a major blocking in the North Atlantic-European region

“…In addition, we create an additional set of trajectories to detect WCBs as trajectories that ascend by at least 600 hPa in 48 hours based on a similar methodology as Madonna et al (2014). Analogous to Quinting and Grams (2022), we distinguish different stages of the WCB and assign all WCB trajectory parcels that are located above 400 hPa to the WCB outflow stage. For this purpose, two-day backward trajectories are started 3-hourly in the northern hemisphere at an equidistant grid of ∆x = 100 km and at 13 equidistant vertical levels between 400 and 100 hPa.…”