| Prediction of visibility in the Arctic based on dynamic Bayesian network analysis |
| |
| 作者姓名： | Shijun Zhao Yulong Shan Ismail Gultepe |
| |
| 作者单位： | 1. College of Meteorology and Oceanography, National University of Defense Technology |
| |
| 摘 要： | With the accelerated warming of the world, the safety and use of Arctic passages is receiving more attention.Predicting visibility in the Arctic has been a hot topic in recent years because of navigation risks and opening of ice-free northern passages. Numerical weather prediction and statistical prediction are two methods for predicting visibility. As microphysical parameterization schemes for visibility are so sophisticated, visibility prediction using numerical weather prediction models inclu...
|
Prediction of visibility in the Arctic based on dynamic Bayesian network analysis |
| |
| Shijun Zhao,Yulong Shan,Ismail Gultepe.Prediction of visibility in the Arctic based on dynamic Bayesian network analysis[J].Acta Oceanologica Sinica,2022,41(4):57-67. |
| |
| Authors: | Shijun Zhao Yulong Shan Ismail Gultepe |
| |
| Institution: | 1.College of Meteorology and Oceanography, National University of Defense Technology, Nanjing 211101, China2.Cloud Physics and Severe Weather Research, Environment and Climate Change Canada, Toronto M4Y1M7, Canada |
| |
| Abstract: | With the accelerated warming of the world, the safety and use of Arctic passages is receiving more attention. Predicting visibility in the Arctic has been a hot topic in recent years because of navigation risks and opening of ice-free northern passages. Numerical weather prediction and statistical prediction are two methods for predicting visibility. As microphysical parameterization schemes for visibility are so sophisticated, visibility prediction using numerical weather prediction models includes large uncertainties. With the development of artificial intelligence, statistical prediction methods have received increasing attention. In this study, we constructed a statistical model with a physical basis, to predict visibility in the Arctic based on a dynamic Bayesian network, and tested visibility prediction over a 1°×1° grid area averaged daily. The results show that the mean relative error of the predicted visibility from the dynamic Bayesian network is approximately 14.6% compared with the inferred visibility from the artificial neural network. However, dynamic Bayesian network can predict visibility for only 3 days. Moreover, with an increase in predicted area and period, the uncertainty of the predicted visibility becomes larger. At the same time, the accuracy of the predicted visibility is positively correlated with the time period of the input evidence data. It is concluded that using a dynamic Bayesian network to predict visibility can be useful over Arctic regions for projected climatic changes. |
| |
| Keywords: | Arctic visibility prediction artificial neural network dynamic Bayesian network |
|
| 点击此处可从《海洋学报(英文版)》浏览原始摘要信息 |
| 点击此处可从《海洋学报(英文版)》下载免费的PDF全文 |
