Short-term prediction of the intensity and track of tropical cyclone via ConvLSTM model

Tong, Biao; Wang, X.; Fu, Jiyang; Chan, P. W.; He, Yuncheng

doi:10.1016/j.jweia.2022.105026

Cited by 32 publications

(18 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To investigate the importance of temporal information in TC intensity estimation tasks and the impact of the temporal sequence length on the estimation results, we conducted experiments. Previous studies have shown that an excessively long temporal length can interfere with current intensity estimation 39 . Therefore, we chose to consider only the historical information within the past 24 h for the intensity estimation.…”

Section: Resultsmentioning

confidence: 99%

“…Previous studies have shown that an excessively long temporal length can interfere with current intensity estimation. 39 Therefore, we chose to consider only the historical information within the past 24 h for the intensity estimation. Since the imaging frequency of the TCIR dataset is every 3 h, we conducted experiments with different temporal sequence lengths T ranging from 1 to 8.…”

Section: Setting the Temporal Lengthmentioning

confidence: 99%

See 1 more Smart Citation

Spatiotemporal fusion convolutional neural network: tropical cyclone intensity estimation from multisource remote sensing images

Fu,

Hu,

et al. 2024

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Utilizing multisource remote sensing images to accurately estimate tropical cyclone (TC) intensity is crucial and challenging. Traditional approaches rely on a single image for intensity estimation and lack the capability to perceive dynamic spatiotemporal information. Meanwhile, many existing deep learning methods sample from a time series of fixed length and depend on computation-intensive 3D feature extraction modules, limiting the model's flexibility and scalability. By organically linking the genesis and dissipation mechanisms of a TC with computer vision techniques, we introduce a spatiotemporal fusion convolutional neural network that integrates three distinct improvement approaches. First, an a priori aware nonparametric fusion module is introduced to effectively fuse key features from multisource remote sensing data. Second, we design a scale-aware contraction-expansion module. This module effectively captures detailed features of the TC by connecting information from different scales through a weighted and up-sampling method. Finally, we propose a 1D-2D conditional sampling training method that balances single-step regression (for short sequences) and latent-variable-based temporal modeling (for long sequences) to achieve flexible spatiotemporal feature perception, thereby avoiding the data scale constraint imposed by fixed sequence lengths. Through qualitative and quantitative experimental comparisons, the proposed spatiotemporal fusion convolutional neural network achieved a root-mean-square error of 8.89 kt, marking a 29.7% improvement over the advanced Dvorak technique, and its efficacy in actual TC case analyses indicates its practical viability and potential for broader applications.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Setting the Temporal Lengthmentioning

confidence: 99%

Spatiotemporal fusion convolutional neural network: tropical cyclone intensity estimation from multisource remote sensing images

Fu,

Hu,

et al. 2024

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

show abstract

“…The simulations calculate the wind speed/grade prediction accuracy for ten instances. From the instances, nine correctly predicted the wind speed/grade using the LEGEMP method, eight correctly predicted using Tong et al [1], and seven instances were correctly predicted using Na et al [1]. As a result, the overall wind speed/grade prediction accuracy using the three methods was observed to be 90%, 80%, and 70%, respectively.…”

Section: Wind Speed/grade Prediction Accuracymentioning

confidence: 92%

Enhancing Cyclone Intensity Prediction for Smart Cities Using a Deep-Learning Approach for Accurate Prediction

Jayaraman,

Venkatachalam,

Eid

et al. 2023

Atmosphere

View full text Add to dashboard Cite

Accurate cyclone intensity prediction is crucial for smart cities to effectively prepare and mitigate the potential devastation caused by these extreme weather events. Traditional meteorological models often face challenges in accurately forecasting cyclone intensity due to cyclonic systems’ complex and dynamic nature. Predicting the intensity of cyclones is a challenging task in meteorological research, as it requires expertise in extracting spatio-temporal features. To address this challenge, a new technique, called linear support vector regressive gradient descent Jaccardized deep multilayer perceptive classifier (LEGEMP), has been proposed to improve the accuracy of cyclone intensity prediction. This technique utilizes a dataset that contains various attributes. It employs the Herfindahl correlative linear support vector regression feature selection to identify the most important characteristics for enhancing cyclone intensity forecasting accuracy. The selected features are then used in conjunction with the Nesterov gradient descent jeopardized deep multilayer perceptive classifier to predict the intensity classes of cyclones, including depression, deep depression, cyclone, severe cyclone, very severe cyclone, and extremely severe cyclone. Experimental results have demonstrated that LEGEMP outperforms conventional methods in terms of cyclone intensity prediction accuracy, requiring minimum time, error rate, and memory consumption. By leveraging advanced techniques and feature selection, LEGEMP provides more reliable and precise predictions for cyclone intensity, enabling better preparedness and response strategies to mitigate the impact of these destructive storms. The LEGEMP technique offers an improved approach to cyclone intensity prediction, leveraging advanced classifiers and feature selection methods to enhance accuracy and reduce error rates. We demonstrate the effectiveness of our approach through rigorous evaluation and comparison with conventional prediction methods, showcasing significant improvements in prediction accuracy. Integrating our enhanced prediction model into smart city disaster management systems can substantially enhance preparedness and response strategies, ultimately contributing to the safety and resilience of communities in cyclone-prone regions.

show abstract

“…With the increased adoption and development of artificial intelligence (AI), new deep learning based statistical prediction methods now outperform the general statistical methods in terms of predicting marine dynamics. At present, various deep learning neural networks have emerged, such as convolutional neural networks (CNNs) (Karpathy et al, 2014;Kim, 2014;LeCun et al, 2015;Shelhamer et al, 2017;Kohler and Langer, 2020), long short term memory networks (LSTMs) (Hochreiter and Schmidhuber, 1997;Chen et al, 2022), convolutional long short term memory networks (ConvLSTM) (Shi et al, 2015;Tong et al, 2022), and transformers (Vaswani et al, 2017;. As one of the most popular models, CNNs have the advantages of offering 1) powerful selflearning ability, 2) high processing efficiency for multipledimensional data, and 3) self-adaptability (Krizhevsky et al, 2012;Oquab et al, 2014;LeCun et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

Monthly extended ocean predictions based on a convolutional neural network via the transfer learning method

Miao

Zhang

et al. 2023

Front. Mar. Sci.

View full text Add to dashboard Cite

Sea surface temperature anomalies (SSTAs) and sea surface height anomalies (SSHAs) are indispensable parts of scientific research, such as mesoscale eddy, current, ocean-atmosphere interaction and so on. Nowadays, extended-range predictions of ocean dynamics, especially in SSTA and SSHA, can provide daily prediction services in the range of 30 days, which bridges the gap between synoptic-scale weather forecasts and monthly average scale climate predictions. However, the forecast efficiency of extended range remains problematic. With the development of ocean reanalysis and satellite remote sensing products, large amounts datasets provide an unprecedented opportunity to use big data for the extended range prediction of ocean dynamics. In this study, a hybrid model, combing convolutional neural network (CNN) model with transfer learning (TL), was established to predict SSTA and SSHA at monthly scales, which makes full use of these data resources that arise from delayed gridding reanalysis products and real-time satellite remote sensing observations. The proposed model, where both ocean and atmosphere reanalysis datasets serve as the pretraining dataset and the satellite remote sensing observations are employed for fine-tuning based on the transfer learning (TL) method, can effectively capture the evolving spatial characteristics of SSTAs and SSHAs with low prediction errors over the 30 days range. When the forecast lead time is 30 days, the root means square errors for the SSTAs and SSHAs model results are 0.32°C and 0.027 m in the South China Sea, respectively, indicating that this model has not only satisfactory prediction performance but also offers great potential for practical operational applications in improving the skill of extended-range predictions.

show abstract

Short-term prediction of the intensity and track of tropical cyclone via ConvLSTM model

Cited by 32 publications

References 33 publications

Spatiotemporal fusion convolutional neural network: tropical cyclone intensity estimation from multisource remote sensing images

Spatiotemporal fusion convolutional neural network: tropical cyclone intensity estimation from multisource remote sensing images

Enhancing Cyclone Intensity Prediction for Smart Cities Using a Deep-Learning Approach for Accurate Prediction

Monthly extended ocean predictions based on a convolutional neural network via the transfer learning method

Contact Info

Product

Resources

About