Distributed Typhoon Track Prediction Based on Complex Features and Multitask Learning

Typhoons are common natural phenomena that often have disastrous aftermaths, particularly in coastal areas. Consequently, typhoon track prediction has always been an important research topic. It chiefly involves predicting the movement of a typhoon according to its history. However, the formation and movement of typhoons is a complex process, which in turn makes accurate prediction more complicated; the potential location of typhoons is related to both historical and future factors. Existing works do not fully consider these factors; thus, there is significant room for improving the accuracy of predictions. To this end, we presented a novel typhoon track prediction framework comprising complex historical features—climatic, geographical, and physical features—as well as a deep-learning network based on multitask learning. We implemented the framework in a distributed system, thereby improving the training efficiency of the network. We verified the efficiency of the proposed framework on real datasets.

[1]  John A. Knaff,et al.  Further improvements to the Statistical Hurricane Intensity Prediction Scheme (SHIPS) , 2005 .

[2]  Quanan Zheng,et al.  Spatiotemporal Distributions of Ocean Color Elements in Response to Tropical Cyclone: A Case Study of Typhoon Mangkhut (2018) Past over the Northern South China Sea , 2021, Remote. Sens..

[3]  Yuping Zhang,et al.  A Novel Data-Driven Tropical Cyclone Track Prediction Model Based on CNN and GRU With Multi-Dimensional Feature Selection , 2020, IEEE Access.

[4]  James S. Goerss,et al.  Tropical Cyclone Track Forecasts Using an Ensemble of Dynamical Models , 2000 .

[5]  Chi-Keong Goh,et al.  Co-evolutionary multi-task learning for dynamic time series prediction , 2017, Appl. Soft Comput..

[6]  Guillaume Charpiat,et al.  Tropical Cyclone Track Forecasting Using Fused Deep Learning From Aligned Reanalysis Data , 2020, Frontiers in Big Data.

[7]  C. M. Kishtawal,et al.  Multimodel Ensemble Forecasts for Weather and Seasonal Climate , 2000 .

[8]  Abdollah Homaifar,et al.  A Sparse Recurrent Neural Network for Trajectory Prediction of Atlantic Hurricanes , 2016, GECCO.

[9]  Rohitash Chandra,et al.  Application of cooperative neuro-evolution of Elman recurrent networks for a two-dimensional cyclone track prediction for the south pacific region , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[10]  Peng Zhao,et al.  A nowcasting model for the prediction of typhoon tracks based on a long short term memory neural network , 2018, Acta Oceanologica Sinica.

[11]  Kim-Han Thung,et al.  A brief review on multi-task learning , 2018, Multimedia Tools and Applications.

[12]  Dianhui Wang,et al.  Effective Deep Attributed Network Representation Learning With Topology Adapted Smoothing , 2021, IEEE Transactions on Cybernetics.

[13]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Fumio Yamazaki,et al.  Inundation Assessment of the 2019 Typhoon Hagibis in Japan Using Multi-Temporal Sentinel-1 Intensity Images , 2021, Remote. Sens..

[15]  Xiaofeng Li,et al.  CNN-Based Tropical Cyclone Track Forecasting from Satellite Infrared Images , 2020, IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium.

[16]  Yan Zhang,et al.  Diagnosis and Classification of Typhoon-Associated Low-Altitude Turbulence Using HKO-TDWR Radar Observations and Machine Learning , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[17]  Rohitash Chandra,et al.  Dynamic Cyclone Wind-Intensity Prediction Using Co-Evolutionary Multi-task Learning , 2017, ICONIP.

[18]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[19]  Sam Ganzfried,et al.  Predicting Hurricane Trajectories using a Recurrent Neural Network , 2018, AAAI.

[20]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[21]  Hyojin Kim,et al.  Deep-Hurricane-Tracker: Tracking and Forecasting Extreme Climate Events , 2019, 2019 IEEE Winter Conference on Applications of Computer Vision (WACV).

[22]  Donghyun You,et al.  Prediction of a typhoon track using a generative adversarial network and satellite images , 2019, Scientific Reports.

[23]  Rohitash Chandra,et al.  Cooperative neuro-evolution of Elman recurrent networks for tropical cyclone wind-intensity prediction in the South Pacific region , 2015, 2015 IEEE Congress on Evolutionary Computation (CEC).

[24]  Harry C. Weber.  Hurricane track prediction using a statistical ensemble of numerical models , 2003 .

[25]  Abhijit Mukherjee,et al.  Joint Learning for Cyclone Track Nowcasting , 2020, MACLEAN@PKDD/ECML.