Collaborative Research: ANSWERS: Prediction of Geoeffective Solar Eruptions, Geomagnetic Indices, and Thermospheric Density Using Machine Learning Methods

合作研究：答案：使用机器学习方法预测地球有效太阳喷发、地磁指数和热层密度

• 批准号: 2149747
• 负责人: Xiaoli Bai
• 金额: $ 54万
• 依托单位: Rutgers University New Brunswick
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-05-01 至 2025-04-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2149747&HistoricalAwards=false
• 关键词: Collaborative; Research; ANSWERS; Prediction; Geoeffective

Understanding and predicting eruptions on the Sun and their terrestrial impacts are a research as well as strategic national priority, as such space weather affects our electronic communication, electric power supply, satellite infrastructure, national defense, and more. This project is a collaboration among Rutgers University, New Jersey Institute of Technology, West Virginia University, and Montclair State University that will improve our ability to predict several linked space weather components: geoeffective solar eruptions, the global magnetic response of Earth to these eruptions, as well as variation of neutral density in the Earth’s thermosphere and its effect on satellite drag. The work covers many aspects of geospace science, solar physics, and data science including machine learning. The innovative machine learning tools developed from the project will be applicable for analyzing disparate data sets in astronomy and other areas of science. Faculty members, early career researchers including a postdoctoral fellow and graduate students will collaborate on the project, creating a multidisciplinary training program for future generations of scientists. The project will emphasize diversity and the participation of underrepresented minorities through both the research efforts and education activities such as K-12 teacher workshops.The two key science questions are: What are the physical mechanisms for the onset of geoeffective solar eruptions? And what are the effects of solar eruptions on neutral density in the thermosphere? Specifically, the project will create synthetic vector magnetograms using ground- and space-based data for solar cycles 23 and 24; develop machine learning (ML) tools to predict solar flares and associated geoeffective coronal mass ejections (CMEs) based on magnetogram parameters; predict geomagnetic indices from derived magnetic properties of solar active regions and CMEs, solar wind parameters and solar images; and predict neutral density in the thermosphere using ML approaches that integrate satellite data, observed and predicted geomagnetic indices, and empirical neutral density models. Most of the funding will be used to support three graduate students (one at WVU and two at NJIT) and a postdoc at Rutgers. K-12 teacher workshops will be organized by Montclair State University. ANSWERS projects advance the nation’s STEM expertise and societal resilience to space weather hazards by filling key knowledge gaps regarding the coupled Sun-Earth system.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

了解和预测太阳上的火山爆发及其对地球的影响是一项研究，也是国家的战略优先事项，因为这种空间天气影响我们的电子通信，电力供应，卫星基础设施，国防等。该项目是罗格斯大学、新泽西理工学院、西弗吉尼亚大学和蒙特克莱尔州立大学之间的一个合作项目，将提高我们预测若干相互关联的空间天气组成部分的能力：地球效应太阳爆发、地球对这些爆发的全球磁响应以及地球热层中性密度的变化及其对卫星阻力的影响。这项工作涵盖了地球空间科学，太阳物理学和数据科学的许多方面，包括机器学习。该项目开发的创新机器学习工具将适用于分析天文学和其他科学领域的不同数据集。教职员工，包括博士后研究员和研究生在内的早期职业研究人员将在该项目上合作，为未来几代科学家创建多学科培训计划。该项目将通过研究工作和教育活动，如K-12教师讲习班，强调多样性和代表性不足的少数群体的参与。两个关键的科学问题是：什么是地球效应太阳爆发开始的物理机制？太阳爆发对热层中的中性密度有什么影响？具体而言，该项目将利用第23和24个太阳活动周期的地面和空间数据创建合成矢量磁图;开发机器学习工具，以根据磁图参数预测太阳耀斑和相关的地球效应日冕物质抛射;根据太阳活动区和日冕物质抛射的衍生磁性、太阳风参数和太阳图像预测地磁指数;并使用ML方法预测热层中的中性密度，该方法集成了卫星数据、观测和预测的地磁指数以及经验中性密度模型。大部分资金将用于支持三名研究生（一名在西弗吉尼亚大学，两名在NJIT）和罗格斯大学的博士后。K-12教师讲习班将由蒙特克莱尔州立大学组织。ANSWERS项目通过填补有关太阳-地球耦合系统的关键知识空白，推进国家的STEM专业知识和社会对空间天气灾害的复原力。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

MSIS‐UQ: Calibrated and Enhanced NRLMSIS 2.0 Model With Uncertainty Quantification

• DOI: 10.1029/2022sw003267
• 发表时间: 2022-08
• 期刊: Space Weather
• 作者: R. Licata;P. Mehta;D. Weimer;W. Tobiska;J. Yoshii

Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling

• DOI: 10.1029/2022sw003189
• 发表时间: 2022-06
• 期刊: Space Weather
• 作者: R. Licata;P. Mehta;D. Weimer;D. Drob;W. Tobiska;J. Yoshii

Thermospheric density predictions during quiet time and geomagnetic storm using a deep evidential model-based framework

• DOI: 10.1016/j.actaastro.2023.06.023
• 发表时间: 2023-10
• 期刊: Acta Astronautica
• 影响因子: 3.5
• 作者: Yiran Wang;X. Bai

Global Thermospheric Density Prediction Model Based on Deep Evidential Framework

基于深度证据框架的全球热层密度预测模型

• 发表时间: 2023
• 期刊: 2023 AAS/AIAA Astrodynamics Specialist Conference
• 作者: Yiran Wang, Xiaoli Bai

Probabilistic Solar Proxy Forecasting With Neural Network Ensembles

• DOI: 10.1029/2023sw003675
• 发表时间: 2023-06
• 期刊: Space Weather
• 作者: Joshua D. Daniell;P. Mehta