MULTIVARIATE MULTI-STEP CONVECTION NOWCASTING WITH DEEP NEURAL NETWORKS: THE NOVARA CASE STUDY

被引：3

作者：

Biondi, Riccardo ^{[1
]}

Chkeir, Sandy ^{[1
]}

Anesiadou, Aikaterini ^{[1
]}

Mascitelli, Alessandra ^{[1
]}

Realini, Eugenio ^{[2
]}

Nisi, Luca ^{[3
]}

Cimarelli, Corrado ^{[4
]}

机构：

[1] Univ Padua, Dipartimento Geosci, Padua, Italy

[2] Geomat Res & Dev GReD Srl, Via Cavour 2, I-22074 Lomazzo, Italy

[3] MeteoSwiss, Via Monti 146, CH-6605 Locarno, Switzerland

[4] Ludwig Maximilian Univ Munich LMU, Munich, Germany

来源：

2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022) | 2022年

关键词：

Nowcasting; Deep Learning; LSTM;

D O I：

10.1109/IGARSS46834.2022.9883665

中图分类号：

P [天文学、地球科学];

学科分类号：

07 ;

摘要：

Severe weather events are constantly increasing over northern Italy impacting the air traffic of one of the major airports of Europe: Milano Malpensa. Monitoring and predicting extreme convection is very challenging especially when it develops locally in a short time range. This work is performed within two projects funded by the H2020 SESAR Programme, with the objective of nowcasting with high accuracy the strong weather events affecting the airport. We collected different types of data from 10 locations around the airport and developed an end-to-end nowcasting deep neural networks based model for each of these stations. We show in this paper the results that we obtained for Novara, the only station for which we have available weather stations, radar, GNSS and lightning.

引用

页码：6598 / 6601

页数：4

共 13 条

[1] A Deep Learning Approach to Predict Weather Data Using Cascaded LSTM Network [J].