課程資訊
課程名稱
氣候變遷與極端事件:深度學習的應用
Climate Change and Extreme Events - Deep Learning Applications 
開課學期
112-2 
授課對象
理學院  大氣科學研究所  
授課教師
梁禹喬 
課號
AtmSci7113 
課程識別碼
229 M8690 
班次
 
學分
2.0 
全/半年
半年 
必/選修
選修 
上課時間
星期四8,9(15:30~17:20) 
上課地點
大氣系A100 
備註
與羅敏輝合授
總人數上限:47人 
 
課程簡介影片
 
核心能力關聯
核心能力與課程規劃關聯圖
課程大綱
為確保您我的權利,請尊重智慧財產權及不得非法影印
課程概述

Recent progress in artificial intelligence has spurred considerable interest in its applications for earth science problems. To expand our knowledge of this up-rising research area, this course overviews and discusses recent progress on the deep learning applications on studying climate change and extreme events. The topics cover applying deep learning technique to detect and quantify climate change signals, improve seasonal prediction skill, and enhance our understanding of various components in the earth system. Neural networks, particularly salient in processing image, are the major target tool in this class, as the spatial information is crucial for our physical understanding in climate change and extreme events. Students are expected to lead paper discussion and complete their deep learning projects with underlying physics or mechanism as well as interpretability addressed.  

課程目標
This course aims to enhance and update our knowledge of recent progress on deep learning applications on studying climate change and extreme events. 
課程要求
Basic statistics, calculus, linear algebra, and Python programming skills are preferred.  
預期每週課後學習時數
 
Office Hours
 
指定閱讀
 
參考書目
Attribution and detection
Bone et al. (2023): https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2022MS003475
Ham et al. (2023): https://www.nature.com/articles/s41586-023-06474-x

Internal variability and forced response
Barnes et al. (2020): https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2020MS002195
Bone et al. (2023): https://essopenarchive.org/users/653004/articles/660203-separation-of-internal-and-forced-variability-of-climate-using-a-u-net

Climate prediction
ENSO, Ham et al. (2021): https://www.nature.com/articles/s41586-019-1559-7

Extreme events
Heatwave: https://www.frontiersin.org/articles/10.3389/fclim.2022.789641/full
Flood: https://iopscience.iop.org/article/10.1088/1755-1315/1086/1/012036/meta

Atmosphere - jet stream
jet shift, Connolly et al. (2023): https://journals.ametsoc.org/view/journals/aies/2/2/AIES-D-22-0094.1.xml

Ocean - sea surface height
https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023MS003709
 
評量方式
(僅供參考)
   
課程進度
週次
日期
單元主題
Week 1
2/22  introduction 
Week 2
2/29  internal variability and forced response
Bone et al. (2023)
https://essopenarchive.org/users/653004/articles/660203-separation-of-internal-and-forced-variability-of-climate-using-a-u-net 
Week 3
3/7  internal variability and forced response
Barnes et al. (2020)
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020MS002195 
Week 4
3/14  invited talk (Dr. Buo-Fu Chen)  
Week 5
3/21  attribution and detection
Bone et al. (2023)
https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2022MS003475 
Week 6
3/28  attribution and detection
Ham et al. (2023)
https://www.nature.com/articles/s41586-023-06474-x 
Week 7
4/4  no class 
Week 8
4/11  midterm presentation1 
Week 9
4/18  climate prediction
Ham et al. (2021)
https://www.nature.com/articles/s41586-019-1559-7 
Week 10
4/25  midterm presentation2 
Week 11
5/2  extreme events: floods
Mistry and Parekh (2022)
https://iopscience.iop.org/article/10.1088/1755-1315/1086/1/012036/meta 
Week 12
5/9  extreme events: heatwaves
Jacques-Dumas et al. (2022)
https://www.frontiersin.org/articles/10.3389/fclim.2022.789641/full 
Week 13
5/16  buffer week 
Week 14
5/23  final presentation 
Week 15
5/30  final presentation 
Week 16
6/6  no class