No abstract

Tropical cyclones (TCs) are the most destructive weather systems that form over the tropical oceans, with 90 storms forming globally every year. The timely detection and tracking of TCs are important for advanced warning to the affected regions. As these storms form over the open oceans far from the continents, remote sensing plays a crucial role in detecting them. Here we present an automated TC detection from satellite images based on a novel deep learning technique. In this study, we propose a multi-staged deep learning framework for the detection of TCs, including, (i) a detector -Mask Region-Convolutional Neural Network (R-CNN), (ii) a wind speed filter, and (iii) a classifier -CNN. The hyperparameters of the entire pipeline is optimized to showcase the best performance using Bayesian optimization. Results indicate that the proposed approach yields high precision (97.10%), specificity (97.59%), and accuracy (86.55%) for test images.

The monsoon low-pressure systems (LPSs) are synoptic scale vortices embedded in the large-scale monsoon circulation. Although the LPS form in all monsoon regions, they are most prominent over the Indian monsoon region, with an average of 12 ( E 2) storm genesis during June to September (JJAS) every year . The LPS are the main rain-bearing systems that help distribute precipitation over deep interior parts of the continental India (Hunt et al., 2016;Krishnamurthy & Ajayamohan, 2010;Krishnan et al., 2011;Sikka, 1977). The current generation coupled climate models participating in the Coupled Model Intercomparison Project Phase 5 (CMIP5) have low skill in simulating the LPS activity over India (Praveen et al., 2015). They have also found that the LPS activity contributes as much as 60% of the total JJAS rainfall received over Central India and that, the models that simulate LPS-related rainfall correctly were able to simulate the total rainfall reliably. Similar results were also obtained by Hunt and Fletcher (2019).A dry bias over continental India is a common problem across the CMIP5 models (

No abstract

<div><div><div><p>The synoptic-scale (3 - 7 days) variability is a dominant contributor to the Indian summer monsoon (ISM) seasonal precipitation. An accurate prediction of ISM precipitation by dynamical or statistical models remains a challenge. Here we show that the sea level pressure (SLP) can be used as a proxy to predict the active-break cycle as well as the genesis of low- pressure-systems (LPS), using a deep learning model, namely, convolutional long short-term memory (ConvLSTM) networks. The deep learning model is able to reliably predict the daily SLP anomalies over Central India and the Bay of Bengal at a lead time of 7 days. As the fluctuations in SLP drive the changes in the strength of the atmospheric circulation, the prediction of SLP anomalies is useful in predicting the intensity of ISM. It is demonstrated that the ConvLSTM possesses better prediction skill compared to a conventional numerical weather prediction model, indicating the usefulness of a physics guided deep learning model in medium range weather forecasting.</p></div></div></div>