Tropical Cyclone Intensity Estimation Using Multi-Dimensional Convolutional Neural Networks from Geostationary Satellite Data

• Published in: Remote Sensing Vol. 12; no. 1; p. 108
• Main Authors: Lee, Juhyun, Im, Jungho, Cha, Dong-Hyun, Park, Haemi, Sim, Seongmun
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 28.12.2019
• Subjects: 2d/3d convolutional neural networks; Algorithms; Altitude; Artificial neural networks; Atmosphere; Atmospheric models; Climate change; Cyclones; Deep learning; Earth atmosphere; heat; hurricanes; Lower atmosphere; Machine learning; Mathematical models; Meteorological satellites; Mimicry; multispectral imaging; Neural networks; Object recognition; Pattern recognition; Q; Remote sensing; Root-mean-square errors; Satellite imagery; satellites; Science; Sea level; Sensors; Synchronous satellites; Tropical cyclones; tropical cyclones; multispectral imaging; 2D/3D convolutional neural networks
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs12010108

For a long time, researchers have tried to find a way to analyze tropical cyclone (TC) intensity in real-time. Since there is no standardized method for estimating TC intensity and the most widely used method is a manual algorithm using satellite-based cloud images, there is a bias that varies depending on the TC center and shape. In this study, we adopted convolutional neural networks (CNNs) which are part of a state-of-art approach that analyzes image patterns to estimate TC intensity by mimicking human cloud pattern recognition. Both two dimensional-CNN (2D-CNN) and three-dimensional-CNN (3D-CNN) were used to analyze the relationship between multi-spectral geostationary satellite images and TC intensity. Our best-optimized model produced a root mean squared error (RMSE) of 8.32 kts, resulting in better performance (~35%) than the existing model using the CNN-based approach with a single channel image. Moreover, we analyzed the characteristics of multi-spectral satellite-based TC images according to intensity using a heat map, which is one of the visualization means of CNNs. It shows that the stronger the intensity of the TC, the greater the influence of the TC center in the lower atmosphere. This is consistent with the results from the existing TC initialization method with numerical simulations based on dynamical TC models. Our study suggests the possibility that a deep learning approach can be used to interpret the behavior characteristics of TCs.