Article(id=1311998697550995584, tenantId=1146029695717560320, journalId=1311385791232049272, issueId=1311998371833934610, articleNumber=null, orderNo=null, doi=10.3878/j.issn.1006-9895.2510.25091, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1748188800000, receivedDateStr=2025-05-26, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1790734465253, onlineDateStr=2026-09-30, pubDate=1784044800000, pubDateStr=2026-07-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1790734465253, onlineIssueDateStr=2026-09-30, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1790734465253, creator=13701087609, updateTime=1790734465253, updator=13701087609, issue=Issue{id=1311998371833934610, tenantId=1146029695717560320, journalId=1311385791232049272, year='2026', volume='50', issue='4', pageStart='1017', pageEnd='1326', issueExtLink='null', onlineDate='null', pubDate='1784044800000', pubDateStr='2026-07-15', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1790734387596, creator='13701087609', updateTime=1790739443587, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1312019578272702984, tenantId=1146029695717560320, journalId=1311385791232049272, issueId=1311998371833934610, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1312019578272702985, tenantId=1146029695717560320, journalId=1311385791232049272, issueId=1311998371833934610, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=1295, endPage=1308, ext={EN=ArticleExt(id=1311998698394050689, articleId=1311998697550995584, tenantId=1146029695717560320, journalId=1311385791232049272, language=EN, title=Automatic Identification of Cold and Warm Fronts in Extratropical Cyclones over the Eurasian Continent Based on the DETR Model, columnId=1311998374337934099, journalTitle=Chinese Journal of Atmospheric Sciences, columnName=Articles, runingTitle=null, highlight=null, articleAbstract=

An extratropical cyclone is an important weather system in mid-latitude regions and has a profound impact on global weather and climate. The frontal system is a key component of the cyclone, and its accurate identification is crucial for improving the accuracy of weather forecasts. In recent years, deep learning methods have been applied to automatic frontal recognition, but research on frontal identification over the Eurasian continent has been limited, with a focus on detecting single-type fronts. Achieving simultaneous recognition and classification of cold and warm fronts remains a major challenge. Therefore, this paper proposes a method for the synchronous recognition of cold and warm fronts based on the Detection Transformer (DETR) model called DETR-FRONT (DETR-FRO). This method constructs RGB feature images using meteorological variables, such as sea level pressure, temperature advection, temperature, and relative humidity at 850 hPa. The DETR model is trained using an artificially created cold and warm front label set, enabling automatic synchronous recognition and classification of both types of fronts. DETR-FRO improves the original DETR model by enhancing its transmission channel and incorporating multiple meteorological variables to construct RGB feature images. This modification strengthens the ability of the model to capture and express frontal features, leading to improved recognition performance for cold and warm fronts. Multidimensional validation results show that DETR-FRO can simultaneously identify cold and warm fronts in extratropical cyclones over the Eurasian continent with high accuracy and stability. In a typical snowstorm case study, the DETR-FRO model accurately reconstructs the cold and warm fronts and their evolution processes. The recognition results are consistent with dynamic and thermodynamic fields, further confirming the model’s generalization ability and meteorological interpretability. This improvement provides new technical support for weather forecasting and the study of extratropical cyclones. In addition, the long-term frontal dataset created using DETR-FRO, covering the period from 1995 to 2024, reveals the seasonal distribution characteristics of cold and warm front frequencies in extratropical cyclones over the Eurasian continent at the climate scale. This information contributes to a deeper understanding of the climatic characteristics of cold and warm fronts in extratropical cyclones.

, authors=Danfeng SONG1, 2, Yujing QIN1, 2, Jiachen WANG3, Qian LIU1, 2, Xiang WANG1, 2, authorsList=Danfeng SONG, Yujing QIN, Jiachen WANG, Qian LIU, Xiang WANG, authorCompany=null, correspAuthors=Yujing QIN, authorNote=null, correspAuthorsNote=null, copyrightStatement=Copyright © 2026 Chinese Journal of Atmospheric Sciences. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1311998712260419757, articleId=1311998697550995584, tenantId=1146029695717560320, journalId=1311385791232049272, language=CN, title=基于DETR模型的欧亚大陆温带气旋冷暖锋自动识别研究, columnId=1311998374501511957, journalTitle=大气科学, columnName=论文, runingTitle=null, highlight=null, articleAbstract=

温带气旋是中纬度地区重要的天气系统,对全球天气及气候均具有深远影响。锋面系统是温带气旋的关键结构,其精准识别对于提高天气预报的准确性至关重要。近两年来,深度学习方法开始在锋面自动识别领域得到应用,然而针对欧亚大陆的锋面自动识别研究较少,并且多集中于单类锋面的检测,实现冷锋和暖锋的同步识别与分类仍然存在较大困难。因此本文提出了一种基于DETR(DEtection TRansformer)模型的冷暖锋同步识别方法DETR-FRO(DETR-FRONT)。该方法利用海平面气压、850 hPa的温度平流、温度及相对湿度等气象变量构建RGB特征图像,并结合人工冷暖锋标签集训练DETR模型,实现了冷暖锋自动同步识别与分类。DETR-FRO改进了原DETR模型的输送通道,引入多个气象变量构建RGB特征图像以提升锋面特征表达能力,使得冷暖锋均有较好的识别效果。多维度验证结果表明,该方法能够同步识别欧亚大陆温带气旋的冷暖锋,并且具有较高的准确性和稳定性。在典型暴雪个例的应用中,DETR-FRO模型能够准确还原冷锋、暖锋及其演变过程,且识别结果与动力、热力要素场有较好的匹配,进一步验证了该模型的泛化能力与气象可解释性。这一改进也为天气预报和温带气旋的研究提供了新的技术支持。此外,基于DETR-FRO构建的1995~2024年长时间序列锋面数据集,从气候尺度揭示了欧亚大陆温带气旋冷暖锋频率的季节分布特征,有利于加深温带气旋冷暖锋气候特征研究的理解。

, authors=宋丹凤1, 2, 秦育婧1, 2, 王佳晨3, 刘倩1, 2, 王向1, 2, authorsList=宋丹凤, 秦育婧, 王佳晨, 刘倩, 王向, authorCompany=null, correspAuthors=秦育婧, authorNote=

宋丹凤,女,硕士研究生,主要从事人工智能与气象应用方面研究。E-mail:

, correspAuthorsNote=
秦育婧,女,副教授,主要从事极端天气气候、人工智能与气象应用、大气环流异常及短期临近预报研究。E-mail:
, copyrightStatement=版权所有©《大气科学》编辑部 2026, copyrightOwner=北京中科期刊出版有限公司, extLink=null, articleAbsUrl=null, sourceXml=3mfySxoioDVSMLxNVs/9Yw==, magXml=4xqXT3k7lIzLNxJ0Lhx6SQ==, pdfUrl=null, pdf=v3ifUC23k0fmW6xTAbXgcQ==, pdfFileSize=11348506, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=sOGRq0SSklV1O087ojTfAQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=QNQzPl7rnhMaj6h6dqkX0w==, mapNumber=null, fund=null)}, authors=[Author(id=1311998714735059130, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2720769486@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998715246764221, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998714735059130, language=EN, stringName=Danfeng SONG, firstName=Danfeng, middleName=null, lastName=SONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998715993350334, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998714735059130, language=CN, stringName=宋丹凤, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio={"content":"

宋丹凤,女,硕士研究生,主要从事人工智能与气象应用方面研究。E-mail:

"}, bioImg=null, bioContent=

宋丹凤,女,硕士研究生,主要从事人工智能与气象应用方面研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])]), Author(id=1311998716416975040, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=qinyujing@nuist.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1311998716836405444, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998716416975040, language=EN, stringName=Yujing QIN, firstName=Yujing, middleName=null, lastName=QIN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998718505738439, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998716416975040, language=CN, stringName=秦育婧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])]), Author(id=1311998718623178953, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998718954528971, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998718623178953, language=EN, stringName=Jiachen WANG, firstName=Jiachen, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3Nanjing Meteorological Observation Center, Nanjing 210019, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998719030026445, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998718623178953, language=CN, stringName=王佳晨, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3南京市气象探测中心, 南京210019, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998714143662262, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=3, ext=[AuthorCompanyExt(id=1311998714307240119, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714143662262, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Nanjing Meteorological Observation Center, Nanjing 210019), AuthorCompanyExt(id=1311998714319823032, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714143662262, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3南京市气象探测中心, 南京210019)])]), Author(id=1311998719118106831, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998719403319506, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719118106831, language=EN, stringName=Qian LIU, firstName=Qian, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998719491399891, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719118106831, language=CN, stringName=刘倩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])]), Author(id=1311998719755641045, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998719894053080, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719755641045, language=EN, stringName=Xiang WANG, firstName=Xiang, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998719965356249, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719755641045, language=CN, stringName=王向, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])])], keywords=[Keyword(id=1311998720334455002, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, orderNo=1, keyword=Extratropical Cyclone), Keyword(id=1311998720657416412, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, orderNo=2, keyword=Front Identification), Keyword(id=1311998720766468317, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, orderNo=3, keyword=Deep Learning), Keyword(id=1311998721039098078, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, orderNo=4, keyword=DEtection TRansformer (DETR) model), Keyword(id=1311998721437556959, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, orderNo=1, keyword=温带气旋), Keyword(id=1311998723132055776, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, orderNo=2, keyword=锋面识别), Keyword(id=1311998723249496289, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, orderNo=3, keyword=深度学习), Keyword(id=1311998723538903266, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, orderNo=4, keyword=DETR模型)], refs=[Reference(id=1311998733248717061, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=Berry G, Reeder M J, Jakob C. 2011. A global climatology of atmospheric fronts [J]. Geophys. Res. Lett., 38(4): L04809. doi:10.1029/2010GL046451, articleTitle=null, refAbstract=null), Reference(id=1311998733328408838, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Biard J C, Kunkel K E. 2019. Automated detection of weather fronts using a deep learning neural network [J]. Adv. Stat. Climate Meteorol. Oceanogr., 5(2): 147−160. doi:10.5194/ascmo-5-147-2019, articleTitle=null, refAbstract=null), Reference(id=1311998733617815815, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=Bueh C, Xie Z W. 2015. An objective technique for detecting large-scale tilted ridges and troughs and its application to an East Asian cold event [J]. Mon. Wea. Rev., 143(12): 4765−4783. doi:10.1175/MWR-D-14-00238.1, articleTitle=null, refAbstract=null), Reference(id=1311998733689118984, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Carion N, Massa F, Synnaeve G, et al. 2020. End-to-end object detection with transformers [C]//16th European Conference on Computer Vision. Glasgow: Springer, 213–229. doi:10.1007/978-3-030-58452-8_13, articleTitle=null, refAbstract=null), Reference(id=1311998734037246217, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Catto J L, Pfahl S. 2013. The importance of fronts for extreme precipitation [J]. J. Geophys. Res., 118(19): 10791−10801. doi:10.1002/jgrd.50852, articleTitle=null, refAbstract=null), Reference(id=1311998734108549386, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=Catto J L, Jakob C, Berry G, et al. 2012. Relating global precipitation to atmospheric fronts [J]. Geophys. Res. Lett., 39(10): L10805. doi:10.1029/2012GL051736, articleTitle=null, refAbstract=null), Reference(id=1311998734473453836, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=杜晓丹. 2024. 锢囚型温带气旋暴雪过程统计分析及个例研究 [D]. 南京信息工程大学硕士学位论文., articleTitle=null, refAbstract=null), Reference(id=1311998734544757005, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Du Xiaodan. 2024. Statistical analysis and case study of the blocking type extratropical cyclone snowstorm process [D]. M. S. thesis (in Chinese), Nanjing University of Information Science & Technology. doi:10.27248/d.cnki.gnjqc.2024.000120, articleTitle=null, refAbstract=null), Reference(id=1311998734876107022, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Feng M R, Qin Y J, Lu C H. 2021. An objective identification method for wintertime cold fronts in Eurasia [J]. Adv. Atmos. Sci., 38(10): 1695−1705. doi:10.1007/s00376-021-0315-8, articleTitle=null, refAbstract=null), Reference(id=1311998736549634319, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Hénin R, Ramos A M, Schemm S, et al. 2019. Assigning precipitation to mid-latitudes fronts on sub-daily scales in the North Atlantic and European sector: Climatology and trends [J]. Int. J. Climatol., 39(1): 317−330. doi:10.1002/joc.5808, articleTitle=null, refAbstract=null), Reference(id=1311998736641909008, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Hewson T D. 1998. Objective fronts [J]. Meteor. Appl., 5(1): 37−65. doi:10.1017/S1350482798000553, articleTitle=null, refAbstract=null), Reference(id=1311998737031979281, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Hobbs P V, Locatelli J D, Martin J E. 1990. Cold fronts aloft and the forecasting of precipitation and severe weather east of the Rocky Mountains [J]. Wea. Forecasting, 5(4): 613−626. doi:10.1175/1520-0434(1990)005<0613:CFAATF>2.0.CO;2, articleTitle=null, refAbstract=null), Reference(id=1311998737367523602, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Justin A D, Willingham C, McGovern A, et al. 2023. Toward operational real-time identification of frontal boundaries using machine learning [J]. Artif. Intell. Earth Syst., 2(3): e220052. doi:10.1175/AIES-D-22-0052.1, articleTitle=null, refAbstract=null), Reference(id=1311998737480769811, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Justin A D, McGovern A, Allen J T. 2025. FrontFinder AI: Efficient identification of frontal boundaries over the continental United States and NOAA’ s unified surface analysis domain using the UNET3+ model architecture [J]. Artif. Intell. Earth Syst., 4(1): e240043. doi:10.1175/AIES-D-24-0043.1, articleTitle=null, refAbstract=null), Reference(id=1311998737849868564, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Lagerquist R, McGovern A, Gagne II D J. 2019. Deep learning for spatially explicit prediction of synoptic-scale fronts [J]. Wea. Forecasting, 34(4): 1137−1160. doi:10.1175/WAF-D-18-0183.1, articleTitle=null, refAbstract=null), Reference(id=1311998738210578710, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=刘蕊. 2024-11-15(003). 由一个深厚温带气旋带来的气候启示 [N]. 中国气象报., articleTitle=null, refAbstract=null), Reference(id=1311998738319630615, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Rui. 2024-11-15(003). Climate insights from a deep extratropical cyclone (in Chinese) [N]. China Meteorological News. doi:10.28122/n.cnki.ncqxb.2024.002708, articleTitle=null, refAbstract=null), Reference(id=1311998738390933784, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=McCann D W, Whistler J P. 2001. Problems and solutions for drawing fronts objectively [J]. Meteor. Appl., 8(2): 195−203. doi:10.1017/S1350482701002079, articleTitle=null, refAbstract=null), Reference(id=1311998738709700889, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Niebler S, Miltenberger A, Schmidt B, et al. 2022. Automated detection and classification of synoptic-scale fronts from atmospheric data grids [J]. Wea. Climate Dyn., 3(1): 113−137. doi:10.5194/wcd-3-113-2022, articleTitle=null, refAbstract=null), Reference(id=1311998738814558490, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Parfitt R, Czaja A, Seo H. 2017. A simple diagnostic for the detection of atmospheric fronts [J]. Geophys. Res. Lett., 44(9): 4351−4358. doi:10.1002/2017GL073662, articleTitle=null, refAbstract=null), Reference(id=1311998739116548379, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Qin Y J, He S Y, Lu C H, et al. 2024a. Application of the mask R-CNN model to cold front identification in Eurasia [J]. Int. J. Climatol., 44(11): 3766−3777. doi:10.1002/joc.8549, articleTitle=null, refAbstract=null), Reference(id=1311998739582116125, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Qin Y J, Liu Q, Lu C H. 2024b. Cold front identification using the DETR model with satellite cloud imagery [J]. Remote Sens., 17(1): 36. doi:10.3390/rs17010036, articleTitle=null, refAbstract=null), Reference(id=1311998741154980127, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Qin Y J, Wang J C, Lu C H, et al. 2025. A method for identifying warm fronts in Eurasia [J]. Int. J. Climatol., 45(1): e8701. doi:10.1002/joc.8701, articleTitle=null, refAbstract=null), Reference(id=1311998741276614945, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Renard R J, Clarke L C. 1965. Experiments in numerical objective frontal analysis [J]. Mon. Wea. Rev., 93(9): 547−556. doi:10.1175/1520-0493(1965)093<0547:EINOFA>2.3.CO;2, articleTitle=null, refAbstract=null), Reference(id=1311998741683462434, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Schultz D M, Vaughan G. 2011. Occluded fronts and the occlusion process: A fresh look at conventional wisdom [J]. Bull. Amer. Meteor. Soc., 92(4): 443−466. doi:10.1175/2010BAMS3057.1, articleTitle=null, refAbstract=null), Reference(id=1311998742010618147, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Simmonds I, Keay K, Tristram Bye J A. 2012. Identification and climatology of Southern Hemisphere mobile fronts in a modern reanalysis [J]. J. Climate, 25(6): 1945−1962. doi:10.1175/JCLI-D-11-00100.1, articleTitle=null, refAbstract=null), Reference(id=1311998742090309924, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Smith R K, Reeder M J. 1988. On the movement and low-level structure of cold fronts [J]. Mon. Wea. Rev., 116(10): 1927−1944. doi:10.1175/1520-0493(1988)116<1927:OTMALL>2.0.CO;2, articleTitle=null, refAbstract=null), Reference(id=1311998742417465637, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=孙淑清, 高守亭. 2005. 现代天气学概论[M]. 北京: 气象出版社, 49., articleTitle=null, refAbstract=null), Reference(id=1311998742484574503, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Sun Shuqing, Gao Shouting. 2005. Introduction to Modern Meteorology (in Chinese) [M]. Beijing: China Meteorological Press, 49., articleTitle=null, refAbstract=null), Reference(id=1311998742849478952, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=王新敏, 邹旭恺, 翟盘茂. 2007. 北半球温带气旋的变化 [J]. 气候变化研究进展, 3(3): 154−157., articleTitle=null, refAbstract=null), Reference(id=1311998742937559337, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Xinmin, Zou Xukai, Zhai Panmao. 2007. Researches on extratropical cyclone variability in the Northern Hemisphere [J]. Adv. Climate Change Res. (in Chinese), 3(3): 154−157. doi:10.3969/j.issn.1673-1719.2007.03.006, articleTitle=null, refAbstract=null), Reference(id=1311998743289880874, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=曾康, 闵锦忠, 庄潇然, 等. 2024. 基于生成对抗网络的强对流临近预报方法及其在中国东部地区的应用评估 [J]. 大气科学, 48(6): 2316−2328., articleTitle=null, refAbstract=null), Reference(id=1311998743759642923, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Zeng Kang, Min Jinzhong, Zhuang Xiaoran, et al. 2024. Severe convection nowcasting method based on a generative adversarial network and its application evaluation in eastern China [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 48(6): 2316−2328. doi:10.3878/j.issn.1006-9895.2310.23094, articleTitle=null, refAbstract=null), Reference(id=1311998745361867052, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang J L, Liu P, Zhang F, et al. 2018. CloudNet: Ground-based cloud classification with deep convolutional neural network [J]. Geophys. Res. Lett., 45(16): 8665−8672. doi:10.1029/2018GL077787, articleTitle=null, refAbstract=null), Reference(id=1311998745793880365, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=张颖娴, 丁一汇, 李巧萍. 2012. ERA40再分析资料揭示的北半球和东亚地区温带气旋生成频率变化 [J]. 气象, 38(6): 646−656., articleTitle=null, refAbstract=null), Reference(id=1311998746636935470, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Yingxian, Ding Yihui, Li Qiaoping. 2012. Cyclogenesis frequency changes of extratropical cyclones in the Northern Hemisphere and East Asia revealed by ERA40 reanalysis data [J]. Meteorological Monthly (in Chinese), 38(6): 646−656. doi:10.7519/j.issn.1000-0526.2012.06.002, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)]), AuthorCompany(id=1311998714143662262, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=3, ext=[AuthorCompanyExt(id=1311998714307240119, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714143662262, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Nanjing Meteorological Observation Center, Nanjing 210019), AuthorCompanyExt(id=1311998714319823032, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714143662262, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3南京市气象探测中心, 南京210019)])], figs=[ArticleFig(id=1311998724042219748, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.1, caption=Workflow of the DETR-FRO model for the simultaneous identification of cold and warm fronts involves several steps. The purple dashed box outlines the process of constructing the cold and warm front label set and the RGB feature image training set. The black dashed box represents the simplified network structure of the DETR model. The red dashed box indicates the input examples from the cold and warm front dataset, whereas the blue dashed box shows the output recognition results., figureFileSmall=a+slu8cUxvHlPZLRCpBHXA==, figureFileBig=sOGRq0SSklV1O087ojTfAQ==, tableContent=null), ArticleFig(id=1311998724390346981, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图1, caption=DETR-FRO模型同步识别冷暖锋方法的工作流程(紫色虚线框表示构建冷暖锋标签集与RGB特征图像训练集的过程;黑色虚线框表示DETR模型的简化网络结构;红色虚线框为输入的冷暖锋数据集示例;蓝色虚线框为输出的识别结果)。, figureFileSmall=a+slu8cUxvHlPZLRCpBHXA==, figureFileBig=sOGRq0SSklV1O087ojTfAQ==, tableContent=null), ArticleFig(id=1311998724918829286, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.2, caption=(a) Manual analysis of cold and warm fronts at 1200 UTC on August 25, 2008; corresponding binary label for the (b) cold front and (c) warm front. In Fig. (a), 850 hPa temperature advection (shading, units: 10−4 K s−1), temperature (red contours, units: °C), sea level pressure (black contours, units: hPa), and the 10 m wind field., figureFileSmall=Hm8hhG6+tCsk+b7aT7dPdQ==, figureFileBig=BdoYcyaU8nzkwq0t14oGNQ==, tableContent=null), ArticleFig(id=1311998725652832487, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图2, caption=2008年8月25日12时(协调世界时,下同)(a)人工分析冷暖锋结果、(b)转化为冷锋标签后的黑白二值图像、(c)转化为暖锋标签后的黑白二值图像,其中(a)中包含850 hPa温度平流(填色,单位:10−4 K s−1)、温度(红色等值线,单位:°C)、海平面气压(黑色等值线,单位:hPa)和10 m风场。, figureFileSmall=Hm8hhG6+tCsk+b7aT7dPdQ==, figureFileBig=BdoYcyaU8nzkwq0t14oGNQ==, tableContent=null), ArticleFig(id=1311998726076457192, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.3, caption=Schematic of the generation of training images from meteorological variables at 1200 UTC on August 25, 2008. (a–d) Fields of 850 hPa temperature advection (units: 10−4 K s−1), 850 hPa temperature (units: °C), sea level pressure (units: hPa), and 850 hPa relative humidity; (e–g) assignment of the four variables to the R, G, and B channels; (h) resulting RGB feature image., figureFileSmall=uQbKmPaNfj0g2SqLHh67JA==, figureFileBig=F04pDUvVoG6pai43J+Mm5A==, tableContent=null), ArticleFig(id=1311998727536074986, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图3, caption=2008年8月25日12时的气象要素生成训练图像示意图:(a–d)850 hPa温度平流(单位:10−4 K s−1)、850 hPa温度(单位:°C)、地面海平面气压(单位:hPa)、850 hPa相对湿度分布情况;(e–g)四种要素放置在R、G、B通道的分布情况;(h)合成的RGB特征图像。, figureFileSmall=uQbKmPaNfj0g2SqLHh67JA==, figureFileBig=F04pDUvVoG6pai43J+Mm5A==, tableContent=null), ArticleFig(id=1311998727632543980, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.4, caption=Segmentation results directly identified from feature images by the DETR-FRO model at (a) 1200 UTC on August 25, 2008 and (b) 0000 UTC on October 6, 2014. In both panels, 850 hPa temperature advection (shading, units: 10−4 K s−1), sea level pressure (black contours, units: hPa), and 10 m winds are overlaid., figureFileSmall=kXbhicPx8AE2h1pxNhSuNg==, figureFileBig=2e8RBuTZjf0vj9hUed9FNA==, tableContent=null), ArticleFig(id=1311998727796121838, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图4, caption=(a)2008年8月25日12时与(b)2014年10月6日00时DETR-FRO模型由特征图像直接识别的分割结果,其中,叠加了850 hPa温度平流(填色,单位:10−4 K s−1)、海平面气压(黑色等值线,hPa)和10 m水平风场。, figureFileSmall=kXbhicPx8AE2h1pxNhSuNg==, figureFileBig=2e8RBuTZjf0vj9hUed9FNA==, tableContent=null), ArticleFig(id=1311998728144249071, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.5, caption=DETR-FRO identification results of cold and warm fronts at (a–c) 0000 UTC on August 5, 2024 and (d–f) 0000 UTC on September 5, 2024 under three combinations of input variables: (a, d) The advt-T850-RH combination; (b, e) the advt-T850-SLP combination; (c, f) the advt-T850-SLP-RH combination. In each panel, 850 hPa temperature advection (shading, units: 10−4 K s−1), sea level pressure (black contours, units: hPa), and 10 m winds are overlaid., figureFileSmall=VUr+R9IKuszZSucEs4VrHA==, figureFileBig=xiofD+S/wAOUV2hdCtxFmg==, tableContent=null), ArticleFig(id=1311998728215552240, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图5, caption=三种变量组合下2024年(a–c)8月5日00时与(d–f)9月5日00时DETR-FRO冷暖锋识别结果:(a,d)advt-T850-RH组合;(b,e)advt-T850-SLP组合;(c,f)advt-T850-SLP-RH组合。图中叠加了850 hPa温度平流(填色,单位:10−4 K s−1)、海平面气压(黑色等值线,单位:hPa)和10 m风场。, figureFileSmall=VUr+R9IKuszZSucEs4VrHA==, figureFileBig=xiofD+S/wAOUV2hdCtxFmg==, tableContent=null), ArticleFig(id=1311998728563679473, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.6, caption=Surface cold (blue lines) and warm fronts (red lines) from 0000 UTC on September 10 to 0000 UTC on September 11, 2023: (a–c) Sea level pressure field (black contours, units: hPa) and NOAA longwave infrared brightness temperature field (shading, units: K); (d–f) 850 hPa temperature advection field (shading, units: 10−4 K s−1), temperature field (red contours, units: °C), horizontal wind field (vectors), and geopotential height field (black contours, units: dagpm)., figureFileSmall=dD96I9jGIYlcWK0p1obDTA==, figureFileBig=UWmXtfZoULRqg8fb9XeJcw==, tableContent=null), ArticleFig(id=1311998728651759860, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图6, caption=2023年9月10日00时至11日00时(a–c)地面冷(蓝线)、暖锋(红线)锋线、海平面气压场(黑色等值线,单位:hPa)和NOAA长波红外通道亮度温度(填色,单位:K),(d–f)地面冷(蓝线)、暖锋(红线)锋线、850 hPa温度平流(填色,单位:10−4 K s−1)、温度场(红色等值线,单位:°C)、水平风场(矢量箭头)和位势高度场(黑色等值线,单位:dagpm)。, figureFileSmall=dD96I9jGIYlcWK0p1obDTA==, figureFileBig=UWmXtfZoULRqg8fb9XeJcw==, tableContent=null), ArticleFig(id=1311998729046024438, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.7, caption=Evolution of meteorological variables at 12-h intervals from 1200 UTC on November 7 to 1200 UTC on November 8, 2021: (a–c) distributions of surface meteorological variables, including 12-h accumulated precipitation (shading, units: mm), sea level pressure (black contours, units: hPa), surface winds (vectors), and surface cold (blue lines) and warm fronts (red lines), snowfall is indicated by black dots, with dot sizes proportional to snowfall intensity; (d–f) 850 hPa fields showing temperature advection (shading, units: 10−4 K s−1), temperature (red contours, units: °C), and surface fronts, green dots mark regions with relative humidity greater than 90%., figureFileSmall=Mh3Aqh9sc+OqJQb+n3TiOg==, figureFileBig=39bxh4ZdBDa9avwb4BC6zg==, tableContent=null), ArticleFig(id=1311998729440289015, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图7, caption=2021年11月7日12时至8日12时逐12 h气象要素演变:(a–c)地面气象要素分布,包括逐12 h累积降水量(填色,单位:mm)、海平面气压(黑色等值线,单位:hPa)、地面风(矢量)及地面锋线(冷锋:蓝线;暖锋:红线,下同),降雪以黑色点表示,点的大小与降雪强度成正比;(d–f)850 hPa气象要素场,包括温度平流(填色,单位:10−4 K s−1)、温度(红色等值线,单位:°C)及地面锋线,图中相对湿度大于90%的区域以绿色点表示。, figureFileSmall=Mh3Aqh9sc+OqJQb+n3TiOg==, figureFileBig=39bxh4ZdBDa9avwb4BC6zg==, tableContent=null), ArticleFig(id=1311998729536758008, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.8, caption=NOAA longwave infrared satellite images at 12 h intervals from 1200 UTC on November 7 to 1200 UTC on November 8, 2021, and vertical cross sections of meteorological variables at the corresponding cold and warm front locations: (a–c) NOAA longwave infrared brightness temperature (shading, units: K), overlaid with surface fronts (cold front: blue solid line; warm front: red solid line); (d–f) meridional–vertical cross sections along 120°E showing potential temperature (brown contours, units: K) and horizontal winds (barbs), with the latitude of the cold front indicated by blue dashed lines; (g–i) meridional–vertical cross sections along 126°E, 129°E, and 136°E showing specific humidity (shading, units: g kg−1), potential temperature (brown contours, units: K), and horizontal wind (barbs), with the latitude of the warm front indicated by red dashed lines., figureFileSmall=MjhLxHbvs/E26t2jcIMOjQ==, figureFileBig=mtgq/DOhk8MCFdHUTL8hNg==, tableContent=null), ArticleFig(id=1311998729838747897, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图8, caption=2021年11月7日12时~8日12时逐12 h NOAA长波红外云图及对应冷、暖锋位置处的气象要素垂直剖面:(a–c)NOAA长波红外通道亮度温度(填色,单位:K),叠加地面锋线位置(冷锋:蓝色实线;暖锋:红色实线);(d–f)沿120°E的经向垂直剖面,显示位温(棕色等值线,单位:K)和水平风(风羽),并标示冷锋对应的纬度位置(蓝色虚线);(g–i)沿126°E、129°E和136°E的经向—垂直剖面,显示比湿(填色,单位:g kg−1)、位温(棕色等值线,单位:K)和水平风(风羽),暖锋对应的纬度位置以红色虚线表示。, figureFileSmall=MjhLxHbvs/E26t2jcIMOjQ==, figureFileBig=mtgq/DOhk8MCFdHUTL8hNg==, tableContent=null), ArticleFig(id=1311998729951994106, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.9, caption=Proportion of extratropical cyclone cold fronts from March 1995 to March 2024 in (a) winter, (b) spring, (c) summer, and (d) autumn (shaded) and sea level pressure field (contours, units: hPa), figureFileSmall=F74KclIjDdmI8ML2iUN4Kw==, figureFileBig=EV7CnEPADMMuspsnchEm3g==, tableContent=null), ArticleFig(id=1311998730262372603, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图9, caption=1995年3月至2024年3月(a)冬季、(b)春季、(c)夏季和(d)秋季温带气旋冷锋发生比例(阴影)及海平面气压场(等值线,hPa), figureFileSmall=F74KclIjDdmI8ML2iUN4Kw==, figureFileBig=EV7CnEPADMMuspsnchEm3g==, tableContent=null), ArticleFig(id=1311998730354647292, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Fig.10, caption=Proportion of extratropical cyclone warm fronts from March 1995 to March 2024 in (a) winter, (b) spring, (c) summer, and (d) autumn (shaded) and sea level pressure field (contours, units: hPa)., figureFileSmall=kCdLmCCmDC33rAfszXjEpw==, figureFileBig=Slmd6RIh3cz5fAxyqoT1QA==, tableContent=null), ArticleFig(id=1311998731956871421, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=图10, caption=1995年3月至2024年3月(a)冬季、(b)春季、(c)夏季和(d)秋季温带气旋暖锋发生比例(阴影)和海平面气压场(等值线,单位:hPa)。, figureFileSmall=kCdLmCCmDC33rAfszXjEpw==, figureFileBig=Slmd6RIh3cz5fAxyqoT1QA==, tableContent=null), ArticleFig(id=1311998732032368895, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Tab.1, caption=

Hyperparameter settings in the DETR-FRO model

, figureFileSmall=null, figureFileBig=null, tableContent=
任务超参数设置
目标检测批量大小(batch size)12
训练轮次(training epochs)200
学习率(learning rate)10−4
权重衰减(weight_decay)10−3
学习率衰减(lr_drop)10
目标分割批量大小(batch size)4
训练轮次(training epochs)200
学习率(learning rate)10−4
学习率衰减(lr_drop)35
), ArticleFig(id=1311998732154003712, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=表1, caption=

DETR-FRO模型中超参数的设置

, figureFileSmall=null, figureFileBig=null, tableContent=
任务超参数设置
目标检测批量大小(batch size)12
训练轮次(training epochs)200
学习率(learning rate)10−4
权重衰减(weight_decay)10−3
学习率衰减(lr_drop)10
目标分割批量大小(batch size)4
训练轮次(training epochs)200
学习率(learning rate)10−4
学习率衰减(lr_drop)35
), ArticleFig(id=1311998732380496129, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Tab.2, caption=

Comparison of cold and warm fronts identified by the DETR-FRO model and manual cold and warm fronts in 2015 for each season

, figureFileSmall=null, figureFileBig=null, tableContent=
冷锋暖锋
OAMROAMR
注:*为全年最大值,#为全年最小值
全年0.760.210.770.23
冬季0.750.25*0.760.27*
春季0.79*0.240.760.24
夏季0.72#0.200.74#0.22
秋季0.770.18#0.81*0.20#
), ArticleFig(id=1311998732472770818, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=表2, caption=

2015年各季节DETR-FRO模型识别的冷暖锋与人工冷暖锋的比较结果

, figureFileSmall=null, figureFileBig=null, tableContent=
冷锋暖锋
OAMROAMR
注:*为全年最大值,#为全年最小值
全年0.760.210.770.23
冬季0.750.25*0.760.27*
春季0.79*0.240.760.24
夏季0.72#0.200.74#0.22
秋季0.770.18#0.81*0.20#
), ArticleFig(id=1311998732850258179, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=EN, label=Tab.3, caption=

Comparison of cold and warm front identification results in 2015 using three meteorological element combinations with manual cold and warm front data

, figureFileSmall=null, figureFileBig=null, tableContent=
冷锋暖锋
OAMROAMR
注:表中OA与MR值均为2015年逐12小时的全年平均值;其中*为同类指标最大值,#为最小值
advt-T850-RH0.71#0.28*0.750.30*
advt-T850-SLP0.720.270.72#0.27
advt-T850-SLP-RH0.76*0.21#0.77*0.23#
), ArticleFig(id=1311998732934144260, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, language=CN, label=表3, caption=

3种气象要素组合的2015年冷暖锋识别结果与人工的冷暖锋比较

, figureFileSmall=null, figureFileBig=null, tableContent=
冷锋暖锋
OAMROAMR
注:表中OA与MR值均为2015年逐12小时的全年平均值;其中*为同类指标最大值,#为最小值
advt-T850-RH0.71#0.28*0.750.30*
advt-T850-SLP0.720.270.72#0.27
advt-T850-SLP-RH0.76*0.21#0.77*0.23#
)], attaches=null, journal=Journal(id=1311385095715782771, delFlag=0, nameCn=大气科学, nameEn=Chinese Journal of Atmospheric Sciences, nameHistory1=null, nameHistory2=null, issn=1006-9895, eissn=null, cn=11-1768/O4, coden=null, periodic=1, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=LP50vgRD4hM0fI04zyL4ig==, journalPrice=null, startedYear=null, abbrevIsoEn=Chinese Journal of Atmospheric Sciences, journalRemark=null, publicationField=null, createdTime=1790588171177, updatedTime=1790588695815, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=D, firstLetterEn=D, subjectCode=Science, subjectName=null, subjectCodeEn=Science, subjectNameEn=null, picCn=LP50vgRD4hM0fI04zyL4ig==, picEn=ET4pX8RqS33ayXuWESJ8vw==, jcr=null, cjcr=null, exts=[JournalExt(id=1311387296293810494, language=CN, name=大气科学, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1790588695834, updatedTime=1790588695834, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.dqkxqk.ac.cn/dqkx/dqkx/author/login, submissionEditorUrl=http://www.dqkxqk.ac.cn/dqkx/dqkx/editor/login, submissionReviewUrl=http://www.dqkxqk.ac.cn/dqkx/dqkx/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1311387296352530751, language=EN, name=Chinese Journal of Atmospheric Sciences, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1790588695848, updatedTime=1790588695848, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.dqkxqk.ac.cn/dqkx/dqkx/author/login, submissionEditorUrl=http://www.dqkxqk.ac.cn/dqkx/dqkx/editor/login, submissionReviewUrl=http://www.dqkxqk.ac.cn/dqkx/dqkx/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1311385791232049272, websiteList=[Website(id=1311386135251747142, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1311385791232049272, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/dqkx/CN, language=CN, createTime=1790588419020, createBy=18614031015, updateTime=1790590177432, updateBy=18614031015, name=大气科学-中文, tplId=1146099689490845704, title=大气科学, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1311393672596381721, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=articleTextType, value=kx, createTime=1790590216063, updateTime=1790590216063, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672554438678, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=banner, value=null, createTime=1790590216053, updateTime=1790590216053, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672613158940, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=grayFlag, value=0, createTime=1790590216067, updateTime=1790590216067, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672537661461, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=logo, value=https://castjournals.cast.org.cn/joweb/dqkx/CN/file/pic?fileId=y/JfrvKousZ4v9l7+IrWug==, createTime=1790590216049, updateTime=1790590216049, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672629936158, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=minRunFlag, value=0, createTime=1790590216071, updateTime=1790590216071, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672583798808, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dqkx/CN/file/pic, createTime=1790590216060, updateTime=1790590216060, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672621547549, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=silenceFlag, value=0, createTime=1790590216069, updateTime=1790590216069, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672579604503, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1790590216059, updateTime=1790590216059, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672600576026, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=themeColor, value=null, createTime=1790590216064, updateTime=1790590216064, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393672608964635, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135251747142, code=themeStyle, value=null, createTime=1790590216066, updateTime=1790590216066, creator=18614031015, updator=18614031015)]), Website(id=1311386135402742107, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1311385791232049272, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/dqkx/EN, language=EN, createTime=1790588419057, createBy=18614031015, updateTime=1790590174040, updateBy=18614031015, name=大气科学-英文, tplId=1146101810881728533, title=Chinese Journal of Atmospheric Sciences, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1311393636084965391, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=articleTextType, value=kx, createTime=1790590207358, updateTime=1790590207358, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636055605260, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=banner, value=null, createTime=1790590207351, updateTime=1790590207351, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636114325522, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=grayFlag, value=0, createTime=1790590207365, updateTime=1790590207365, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636043022347, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=logo, value=https://castjournals.cast.org.cn/joweb/dqkx/EN/file/pic?fileId=y/JfrvKousZ4v9l7+IrWug==, createTime=1790590207348, updateTime=1790590207348, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636135297044, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=minRunFlag, value=0, createTime=1790590207370, updateTime=1790590207370, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636076576782, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dqkx/EN/file/pic, createTime=1790590207356, updateTime=1790590207356, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636122714131, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=silenceFlag, value=0, createTime=1790590207368, updateTime=1790590207368, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636063993869, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1790590207354, updateTime=1790590207354, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636089159696, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=themeColor, value=null, createTime=1790590207359, updateTime=1790590207359, creator=18614031015, updator=18614031015), WebsiteProps(id=1311393636101742609, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1311386135402742107, code=themeStyle, value=null, createTime=1790590207362, updateTime=1790590207362, creator=18614031015, updator=18614031015)])], journalTitle=大气科学, weixinUrl=null, journalUrl=https://www.iapjournals.ac.cn/dqkx/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Chinese Journal of Atmospheric Sciences, journalPhotoCn=LP50vgRD4hM0fI04zyL4ig==, journalPhotoEn=ET4pX8RqS33ayXuWESJ8vw==, journalFirstLetter=D, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/dqkx/CN/10.3878/j.issn.1006-9895.2510.25091, detailUrlEn=https://castjournals.cast.org.cn/joweb/dqkx/EN/10.3878/j.issn.1006-9895.2510.25091, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dqkx/CN/PDF/10.3878/j.issn.1006-9895.2510.25091, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dqkx/EN/PDF/10.3878/j.issn.1006-9895.2510.25091, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1784044800000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于DETR模型的欧亚大陆温带气旋冷暖锋自动识别研究
收藏切换
PDF下载
宋丹凤 1, 2 , 秦育婧 1, 2 , 王佳晨 3 , 刘倩 1, 2 , 王向 1, 2
大气科学 | 论文 2026,50(4): 1295-1308
收起
收藏切换
大气科学 |论文 2026 , 50 (4) : 1295 -1308
基于DETR模型的欧亚大陆温带气旋冷暖锋自动识别研究
全屏
[Author(id=1311998714735059130, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2720769486@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998715246764221, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998714735059130, language=EN, stringName=Danfeng SONG, firstName=Danfeng, middleName=null, lastName=SONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998715993350334, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998714735059130, language=CN, stringName=宋丹凤, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio={"content":"

宋丹凤,女,硕士研究生,主要从事人工智能与气象应用方面研究。E-mail:

"}, bioImg=null, bioContent=

宋丹凤,女,硕士研究生,主要从事人工智能与气象应用方面研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])]), Author(id=1311998716416975040, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=qinyujing@nuist.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1311998716836405444, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998716416975040, language=EN, stringName=Yujing QIN, firstName=Yujing, middleName=null, lastName=QIN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998718505738439, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998716416975040, language=CN, stringName=秦育婧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])]), Author(id=1311998718623178953, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998718954528971, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998718623178953, language=EN, stringName=Jiachen WANG, firstName=Jiachen, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3Nanjing Meteorological Observation Center, Nanjing 210019, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998719030026445, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998718623178953, language=CN, stringName=王佳晨, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3南京市气象探测中心, 南京210019, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998714143662262, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=3, ext=[AuthorCompanyExt(id=1311998714307240119, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714143662262, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Nanjing Meteorological Observation Center, Nanjing 210019), AuthorCompanyExt(id=1311998714319823032, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714143662262, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3南京市气象探测中心, 南京210019)])]), Author(id=1311998719118106831, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998719403319506, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719118106831, language=EN, stringName=Qian LIU, firstName=Qian, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998719491399891, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719118106831, language=CN, stringName=刘倩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])]), Author(id=1311998719755641045, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1311998719894053080, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719755641045, language=EN, stringName=Xiang WANG, firstName=Xiang, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1311998719965356249, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, authorId=1311998719755641045, language=CN, stringName=王向, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
2南京信息工程大学大气科学学院, 南京210044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1311998713141223598, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=1, ext=[AuthorCompanyExt(id=1311998713178972335, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998713459990704, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998713141223598, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044)]), AuthorCompany(id=1311998714030416051, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, xref=2, ext=[AuthorCompanyExt(id=1311998714042998964, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1311998714051387573, tenantId=1146029695717560320, journalId=1311385791232049272, articleId=1311998697550995584, companyId=1311998714030416051, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学大气科学学院, 南京210044)])])]
宋丹凤1, 2 , 秦育婧1, 2 , 王佳晨3, 刘倩1, 2, 王向1, 2
作者信息
  • 1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
  • 2南京信息工程大学大气科学学院, 南京210044
  • 3南京市气象探测中心, 南京210019
通讯作者:
秦育婧,女,副教授,主要从事极端天气气候、人工智能与气象应用、大气环流异常及短期临近预报研究。E-mail:
作者简介:

宋丹凤,女,硕士研究生,主要从事人工智能与气象应用方面研究。E-mail:

Automatic Identification of Cold and Warm Fronts in Extratropical Cyclones over the Eurasian Continent Based on the DETR Model
Danfeng SONG1, 2 , Yujing QIN1, 2 , Jiachen WANG3, Qian LIU1, 2, Xiang WANG1, 2
Affiliations
  • 1State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster, Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
  • 2School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing 210044
  • 3Nanjing Meteorological Observation Center, Nanjing 210019
出版时间: 2026-07-15 doi: 10.3878/j.issn.1006-9895.2510.25091
文章导航
收藏切换

温带气旋是中纬度地区重要的天气系统,对全球天气及气候均具有深远影响。锋面系统是温带气旋的关键结构,其精准识别对于提高天气预报的准确性至关重要。近两年来,深度学习方法开始在锋面自动识别领域得到应用,然而针对欧亚大陆的锋面自动识别研究较少,并且多集中于单类锋面的检测,实现冷锋和暖锋的同步识别与分类仍然存在较大困难。因此本文提出了一种基于DETR(DEtection TRansformer)模型的冷暖锋同步识别方法DETR-FRO(DETR-FRONT)。该方法利用海平面气压、850 hPa的温度平流、温度及相对湿度等气象变量构建RGB特征图像,并结合人工冷暖锋标签集训练DETR模型,实现了冷暖锋自动同步识别与分类。DETR-FRO改进了原DETR模型的输送通道,引入多个气象变量构建RGB特征图像以提升锋面特征表达能力,使得冷暖锋均有较好的识别效果。多维度验证结果表明,该方法能够同步识别欧亚大陆温带气旋的冷暖锋,并且具有较高的准确性和稳定性。在典型暴雪个例的应用中,DETR-FRO模型能够准确还原冷锋、暖锋及其演变过程,且识别结果与动力、热力要素场有较好的匹配,进一步验证了该模型的泛化能力与气象可解释性。这一改进也为天气预报和温带气旋的研究提供了新的技术支持。此外,基于DETR-FRO构建的1995~2024年长时间序列锋面数据集,从气候尺度揭示了欧亚大陆温带气旋冷暖锋频率的季节分布特征,有利于加深温带气旋冷暖锋气候特征研究的理解。

温带气旋  /  锋面识别  /  深度学习  /  DETR模型

An extratropical cyclone is an important weather system in mid-latitude regions and has a profound impact on global weather and climate. The frontal system is a key component of the cyclone, and its accurate identification is crucial for improving the accuracy of weather forecasts. In recent years, deep learning methods have been applied to automatic frontal recognition, but research on frontal identification over the Eurasian continent has been limited, with a focus on detecting single-type fronts. Achieving simultaneous recognition and classification of cold and warm fronts remains a major challenge. Therefore, this paper proposes a method for the synchronous recognition of cold and warm fronts based on the Detection Transformer (DETR) model called DETR-FRONT (DETR-FRO). This method constructs RGB feature images using meteorological variables, such as sea level pressure, temperature advection, temperature, and relative humidity at 850 hPa. The DETR model is trained using an artificially created cold and warm front label set, enabling automatic synchronous recognition and classification of both types of fronts. DETR-FRO improves the original DETR model by enhancing its transmission channel and incorporating multiple meteorological variables to construct RGB feature images. This modification strengthens the ability of the model to capture and express frontal features, leading to improved recognition performance for cold and warm fronts. Multidimensional validation results show that DETR-FRO can simultaneously identify cold and warm fronts in extratropical cyclones over the Eurasian continent with high accuracy and stability. In a typical snowstorm case study, the DETR-FRO model accurately reconstructs the cold and warm fronts and their evolution processes. The recognition results are consistent with dynamic and thermodynamic fields, further confirming the model’s generalization ability and meteorological interpretability. This improvement provides new technical support for weather forecasting and the study of extratropical cyclones. In addition, the long-term frontal dataset created using DETR-FRO, covering the period from 1995 to 2024, reveals the seasonal distribution characteristics of cold and warm front frequencies in extratropical cyclones over the Eurasian continent at the climate scale. This information contributes to a deeper understanding of the climatic characteristics of cold and warm fronts in extratropical cyclones.

Extratropical Cyclone  /  Front Identification  /  Deep Learning  /  DEtection TRansformer (DETR) model
宋丹凤, 秦育婧, 王佳晨, 刘倩, 王向. 基于DETR模型的欧亚大陆温带气旋冷暖锋自动识别研究. 大气科学, 2026 , 50 (4) : 1295 -1308 . DOI: 10.3878/j.issn.1006-9895.2510.25091
Danfeng SONG, Yujing QIN, Jiachen WANG, Qian LIU, Xiang WANG. Automatic Identification of Cold and Warm Fronts in Extratropical Cyclones over the Eurasian Continent Based on the DETR Model[J]. Chinese Journal of Atmospheric Sciences, 2026 , 50 (4) : 1295 -1308 . DOI: 10.3878/j.issn.1006-9895.2510.25091
温带气旋是中纬度地区典型的斜压性低压系统,在全球大气环流中发挥着关键作用(张颖娴等, 2012),其发展演变不仅主导着日常天气的变化,还可引发包括强风、暴雨暴雪、风暴潮在内的多种灾害性天气(王新敏等, 2007; 刘蕊, 2024; 杜晓丹, 2024),对社会生产和人类活动造成重大影响。与热带气旋不同,温带气旋常伴有锋面系统,而锋面系统是温带气旋的核心特征,也是驱动天气演变的关键动力结构(Schultz and Vaughan, 2011)。不同类型的锋面可引发不同的天气现象和灾害,例如,冷锋的推进容易触发强对流降水和剧烈天气过程(Smith and Reeder, 1988; Hobbs et al., 1990; Catto and Pfahl, 2013)。暖锋则主要带来持续性降水(孙淑清和高守亭, 2005),而锢囚锋阶段往往对应气旋成熟期,此时降水最强,天气影响最显著(Schultz and Vaughan, 2011)。因此,锋面在温带气旋的发生发展中起着至关重要的作用,其演变特征直接影响天气过程的强度和持续时间。为了提高中高纬度地区的天气预报能力,减轻气象灾害影响,深入研究温带气旋中的锋面系统具有重要的科学价值和现实意义。
目前,天气预报业务中的锋面分析主要依靠人工分析,并且依赖于预报员的主观经验积累(McCann and Whistler, 2001)。因此,研究人员提出了多种锋面自动识别技术。这些方法目前主要分为两大类(Qin et al., 2024b; Justin et al., 2025):一类是传统的数值锋面分析方法(Numerical Frontal Analysis methods,NFA),它是基于各种气象要素的客观识别算法;另一类则是基于深度学习(Machine Learning/Deep Learning,DL/ML)的方法,主要通过神经网络模型实现锋面自动识别。
NFA方法最早是由Renard and Clarke(1965)提出的,通过定义一种基于网格数据的热锋面参数(Thermal Frontal Parameter,TFP),使用TFP的最大值区域来定位锋面。但由于当时湿度场分析不可靠等原因,所得结果并不十分理想。Hewson(1998)提出了基于热锋面参数(TFP)和两个遮掩变量的锋面定位方法。McCann and Whistler(2001)在Hewson(1998)的基础上,分析了该方法在锋线识别中的局限性,并提出了优化方案:采用700~1000 hPa的厚度作为锋面定位计算的变量,以减少地形的影响,并通过锋面平流算法计算锋面速度来判断锋面的性质。此外,不少研究基于不同要素的分布特征,提出了多种改进的锋面识别方法。Berry et al.(2011)和Catto et al.(2012)使用两个阈值速度将锋面分为三种类型;Simmonds et al.(2012)提出了基于风场矢量特征的锋面识别方法;Parfitt et al.(2017)采用等压相对涡度和水平温度梯度归一化乘积超过阈值的区域进行锋面定位;Hénin et al.(2019)基于温度平流区分了暖锋和冷锋;Feng et al.(2021)提出了一种冷锋自动识别方案,结合TFP和温度平流来识别冷锋区域,并提取暖界作为冷锋。NFA方法能够快速地给出锋面点所在的位置,但大多数方法在将这些点连成锋面线条时还存在困难,且阈值的选取仍有一定的主观性。
近年来,得益于大量高质量的气象数据,深度学习方法在气象领域有了很大的发展(曾康等, 2024)。基于深度学习的锋面自动识别技术也开始得到发展。Biard and Kunkel(2019)以温度、比湿、气压以及纬向风和经向风速场作为输入训练了一个二维卷积神经网络,在北美洲2003~2015年期间成功检测出高达90%的分析锋面。Lagerquist et al.(2019)采用卷积神经网络(Convolutional Neural Network,CNN)识别网格数据中的锋面,并提出了一种后处理方法,通过将概率网格转换为具体的锋面对象,提高了检测结果的可解释性。Niebler et al.(2022)在九个气压层上使用五个输入变量进行训练,在一定程度上识别了美国和德国区域内的冷锋、暖锋、准静止锋和锢囚锋,与Biard and Kunkel(2019)相比,他们将气压变量替换为垂直速度,以增强对锋面结构的捕捉能力。Justin et al.(2023)则采用了UNet 3+网络架构,并基于ERA5再分析数据进行分组训练,其研究与Lagerquist et al.(2019)相比,有效提升了模型对美国区域内冷锋、暖锋和锢囚锋的识别能力。此外,Qin et al.(2024a)针对欧亚大陆提出了一种基于Mask R-CNN的冷锋识别方法,在连续时间序列和极端降水事件中展现了较高的准确性和有效性。近年来,基于Transformer架构的DETR模型在目标检测与分割任务中表现优异(Carion et al., 2020)。在此基础上,Qin et al.(2024b)将卫星云图数据引入DETR模型中,为准确确定冷锋的位置和形态提供了新见解。与一些依赖拟合的NFA方法相比,基于深度学习的图像识别方法能够直接提取锋线,并减少人为阈值设定带来的局限性。
上述研究工作推动了锋面识别的自动化,同时也有利于锋面活动长期特征的研究。然而目前针对欧亚大陆的锋面识别研究仍相对较少,且大多数研究仅关注单类锋面(如冷锋或暖锋)的研究。现在常用的模型DETR在输入方面也存在一定局限:原始DETR模型以RGB三通道图像作为输入,这在一定程度上限制了气象特征的表达能力。在以往的锋面识别研究中(Qin et al., 2024b, 2025),冷锋识别通常依赖于850 hPa的温度平流(Temperature Advection,advt)、温度(T850)和海平面气压(Sea Level Pressure,SLP)等要素,而暖锋识别更侧重于advt、T850以及850 hPa的相对湿度(Relative Humidity,RH)等气象要素。仅依赖RGB三通道输入的原始DETR模型,难以兼顾多种关键气象要素,从而在冷暖锋的同步识别中表现受限。此外,Qin et al.(2025)研究发现,与冷锋相比,暖锋及其锋区的特征通常更弱,导致暖锋的识别难度显著高于冷锋。基于此,本文将提出一种基于DETR模型的欧亚大陆温带气旋中冷暖锋同步识别方法DETR-FRO,创新性地引入了四通道气象要素输入,突破原始模型的输入要素种类的限制,克服了冷暖锋同步识别中暖锋识别率较低的问题。该模型输出的温带气旋冷、暖锋的长期数据集,将有利于加深温带气旋发生发展机制的理解。
本文使用的主要资料:
(1)1995~2024年欧洲中期天气预报中心(European Centre for Medium-Range Weather Forecasts,ECMWF)提供的EAR-5逐小时再分析资料,水平分辨率:0.25°×0.25°,气象要素包括:1000~500 hPa层次的温度、u、v风场和比湿,850 hPa的相对湿度,海平面气压场,10 m风场以及逐小时累计降水量和降雪量。
(2)美国国家海洋和大气管理局(National Oceanic and Atmospheric Administration,NOAA)提供的逐3小时地球静止红外通道亮度温度,空间分辨率为0.07°×0.07°。
如无特别说明,本文所用时间均为协调世界时。
图1展示了本文提出的基于DETR模型的欧亚大陆温带气旋中的冷暖锋识别方法DETR-FRO的整体工作流程。包含以下三部分:(1)通过人工分析冷暖锋的位置构建用于深度学习的标签集;利用SLP、advt、T850及RH等四个关键气象要素,通过1×1卷积将其整合为RGB三通道伪彩色图像,生成RGB特征图像训练集(构建过程如图1紫色虚线框所示);(2)将构建好的RGB特征图像训练集及其对应的锋面标签(如图1红色虚线框所示)一同输入DETR模型网络(如图1中黑色虚线框所示)进行训练,得到DETR-FRO冷暖锋识别模型;(3)将待识别的RGB图像输入训练好的DETR-FRO模型,获得对应的冷暖锋识别结果(如图1蓝色虚线框所示)。为了进一步说明该方法的实际应用效果,下面以2008年8月25日12时的气象数据为例,详细介绍该识别方法的具体实现过程。
首先构建人工冷暖锋标签集,具体步骤如下:(1)随机选取2008年、2009年和2014年三年的数据作为训练样本,2015年全年数据作为测试集;(2)针对欧亚大陆区域(10°~70°N,0°~155°E)绘制4年逐12小时的850 hPa与地面的气象要素图像;(3)根据等温线分布特征、冷暖平流和闭合低压等天气系统特征,逐一对每个时次的气象图像进行人工冷暖锋分析。如图2所示,2008年8月25日12时在黑海北部(50°~60°N,30°~50°E),蒙古高原(40°~50°N,60°~80°E),贝加尔湖南侧(45°~50°N,110°~120°E)均存在闭合低压中心,并且等温线密集,伴有明显的冷暖平流。因此我们在这3处地区分别分析冷锋(绿色粗实线)、暖锋(黄色粗实线)。进一步提取出冷锋与暖锋的位置,并将其转化为黑白二值图像,如图2b、c所示。按照上述方法处理本文选取的4年数据,获得模型训练和测试所需的所有冷暖锋标签集。
图3展示了2008年8月25日12时的气象要素转换为训练图像的过程。我们在DETR-FRO模型中创新性地引入了四通道气象要素输入,通过一个1×1卷积网络对SLP、advt、T850及RH等四个关键气象要素进行特征加权融合,最终生成可被DETR模型处理的RGB彩色图像。这一改进不仅突破了原始模型的输入要素种类的限制,也有效提升了模型在冷暖锋同步识别中的表现能力(相关结果详见第3.2节)。其中数据转换的具体做法如下:(1)利用850 hPa上u、v风场和温度T850计算出advt;(2)随后将欧亚大陆范围内的advt(图3a)、T850(图3b)、SLP(图3c)、RH(图3d)四个要素分别进行归一化处理,将物理量数值映射至0~255灰度范围;(3)定义一个1×1卷积模型,利用权重向量ω对四维气象场(通道1:advt,通道2:T850,通道3:SLP,通道4:RH)进行特征加权,其中advt和T850分别映射至R和G通道,SLP和RH共同映射至B通道(图3e–g),最后融合生成RGB图像(图3h)。权重向量如下,映射的具体方法如公式(1)
$ \omega =\left(\begin{array}{cccc}\begin{array}{c}1\\0\\0\end{array} & \begin{array}{c}0\\1\\0\end{array} & \begin{array}{c}0\\0\\0.5\end{array} & \begin{array}{c}0\\0\\0.5\end{array}\end{array}\right) $
$ x_{i,j}^*=\frac{x_{i,j}-x_{\mathrm{min}}}{x_{\mathrm{max}}-x_{\mathrm{min}}}\times255, $
其中,$ x_{i,j}^{*} $是处理后的变量,xmin是变量的最小值,xmax是变量的最大值。
对所有时次的气象数据均按照上述方法生成对应的RGB特征图像。随后,将2008、2009、2014年的特征图像按照8∶2的比例随机划分成训练集与验证集,并与相应的冷暖锋标签一起组成DETR-FRO冷暖锋数据集。
本文采用基于COCO格式数据集的DETR模型作为锋面识别框架,并引入ResNet-101作为主干特征提取网络(backbone)。具体配置如下:模型的分类类别设定为2+1类,即冷锋、暖锋两类以及背景一类;训练集中包含1754张样本图像,共标注冷锋个例6048个、暖锋个例3059个;验证集中包含438张图像,对应冷锋个例1513个、暖锋个例773个;测试集包含2015年全年数据730张图像,对应冷锋个例1996个、暖锋个例1443个;所有图像的分辨率统一为621×241像素。在单块GPU(显存41 GB)的硬件环境下,依据表1中设定的超参数,对模型进行目标检测与目标分割任务的联合训练。
在第73轮训练后,模型在验证集上的总损失降至最低,为9.23,并在此后训练过程中保持稳定波动,表明模型已达到最优性能。以该轮模型参数为基础,构建最终的DETR-FRO冷暖锋同步识别模型。在推理阶段采用置信度阈值为0.8的筛选策略,以输出高置信度的冷锋与暖锋识别结果。图4a给出了2008年8月25日12时DETR-FRO模型识别出的冷、暖锋结果。可以看出,模型识别出的锋面空间位置与人工分析结果(图2a)具有一致性,表明DETR-FRO具有良好的锋面结构提取能力。进一步分析可知,识别出的3组冷、暖锋位置均与海平面气压场、风场及温度平流场有着较好的匹配,进一步验证了DETR-FRO识别结果的合理性。图4b展示了2014年10月6日00时(验证集样本)的DETR-FRO识别结果。模型在日本列岛附近识别到1 条冷锋与1 条暖锋。识别出的冷暖锋均起始于低压中心,其分布符合锋面的天气学特征:冷锋沿着低压等值线的曲率最大处并大致平行于高压等值线的方向;暖锋沿着低压槽区展开,并伴随着气旋性风向切变。850 hPa锋区内存在明显的冷暖平流与地面冷暖锋对应。此外,模型在贝加尔湖南侧还识别到1条冷锋,该冷锋自风场辐合中心向东南方向伸展,并大致平行于外围高压等值线。这进一步说明模型提取的锋面特征与动力、热力结构具有较好的一致性。
参考Bueh and Xie(2015)以及Zhang et al.(2018)提出总体精度(Overall Accuracy,OA)和漏报率(Miss Rate,MR)两个指标,从长时间尺度评估DETR-FRO方法识别冷暖锋的效果。这两个指标的定义如下:
$ \mathrm{OA}=\frac{M}{I}, $
$ \mathrm{MR}=\frac{O-M}{O}, $
其中,O为人工识别的冷(暖)锋总数,I为DETR-FRO识别的冷(暖)锋总数,M为二者匹配的冷(暖)锋数量,O–M则表示人工识别但模型未能识别的冷(暖)锋数量,即模型的漏报数量。具体的匹配标准为,在同一时次中DETR-FRO方法识别的冷(暖)锋与人工冷(暖)锋之间的距离小于5°,则认为两者匹配。OA值越高,越接近1,表明DETR-FRO识别的冷(暖)锋与人工识别的冷(暖)锋匹配程度越高。相反,MR值越低,越接近0,说明DETR-FRO模型漏报的概率越小。当OA和MR都在合理范围内时,可以认为DETR-FRO模型在温带气旋中冷暖锋识别方面是可靠的。
冷暖锋的活动频率计算公式如下:
$ F_{^i}=\frac{m_i}{t}\text{,} $
其中,Fi表示第i个格点上冷(暖)锋活动频率;mi表示第i 个格点上冷(暖)锋出现的次数;t表示总时次数。
为了直观评估DETR-FRO在冷暖锋识别中的性能,我们将2015年测试集上DETR-FRO的识别结果与人工标注的冷暖锋进行了对比。表2给出了测试集2015年各季节DETR-FRO识别的冷暖锋与人工识别结果的对比情况。结果显示,冷锋识别的OA值均超过0.72,MR值均低于0.25,暖锋识别的OA值均超过0.74,MR值低于0.27,这表明DETR-FRO识别的冷、暖锋与人工识别的冷、暖锋整体上匹配度较高,漏报率较低。其中,春季冷锋识别效果最佳(OA达0.79),秋季暖锋识别效果最佳(OA达0.81)。夏季冷、暖锋的匹配程度略低于其他季节,这可能与夏季中高纬大气斜压性较弱有关。冬季冷暖锋漏报率略高,秋季最低。总体来看,冷暖锋全年平均OA均高于0.75,MR低于0.23,说明DETR-FRO模型在全年温带气旋冷暖锋识别中均表现出较好的可靠性和稳定性。
为了讨论本文所选气象要素的合理性以及四通道气象要素输入的优势,我们选择了advt-T850-RH、advt-T850-SLP、advt-T850-SLP-RH三种气象要素组合分别训练模型,并对其结果进行对比分析及评估。表3给出了三种气象要素组合下,模型在测试集2015年冷暖锋识别结果与人工冷暖锋的对比情况。结果表明,三种组合中,advt-T850-SLP-RH组合在冷锋和暖锋的OA指标上均达到最高值(分别为0.76和0.77),同时在MR指标上也达到最小值(分别为0.21和0.23)。
图5进一步给出了2024年8月5日00时与9月5日00时三种气象要素组合对应的冷暖锋识别结果。结果表明,在8月5日00时(图5a–c),advt-T850-SLP-RH四要素组合在东北与内蒙交界处识别到气旋中的冷、暖锋,而advt-T850-RH与advt-T850-SLP两个三要素的组合仅识别出冷锋。结合该气旋前侧存在等温线较密集、暖平流较强和风向气旋性切变的特征,可以判断该区域应存在暖锋结构,说明四要素组合在暖锋识别方面要优于三要素组合。在9月5日00时(图5d–f),三种气象要素组合均能够识别到西西伯利亚平原和鄂霍茨克海附近气旋系统中的冷暖锋。而从冷暖锋的连续性、结构完整性与气象要素的对应关系来看,advt-T850-SLP-RH四要素组合表现最优。
综上,advt-T850-SLP-RH四要素组合在全年测试集与个例中均展现出最佳的识别性能,充分验证了四通道气象要素输入方案在冷暖锋同步识别中的优越性。因此,本文最终选取advt-T850-SLP-RH组合作为DETR-FRO模型的输入特征图像,以提升模型在实际应用中的识别准确性和可靠性。
为了进一步验证DETR-FRO方法识别温带气旋中冷暖锋的效果,选取了一个连续时间段进行分析。图6给出了2023年9月10日00时至11日00时期间,贝加尔湖南侧温带气旋的发展演变过程。图中展示了冷暖锋的识别结果、850 hPa锋区特征、地面系统演变以及长波红外云图。
9月10日00时(图6a,d)模型在贝加尔湖南侧识别到了一条从低压中心延伸出的冷锋,该冷锋锋区附近等温线密集,冷平流显著。此时模型没有识别到暖锋,从图中也可以看出暖平流较弱,暖锋云系也不明显,说明此时暖锋尚未形成。12 h后(图6e),伴随贝加尔湖南侧的低压系统东移增强,冷锋也随之东移并增强。同时模型识别到暖锋生成,对应图6b、e中显示锋区内暖平流增强并出现强的暖式风向切变,对应长波红外云图可以看到,暖锋锋前上空覆盖大片卷云区,其下方出现积状云,且亮度温度低于240 K,表明暖锋云系活跃。这表明,DETR-FRO模型成功捕捉到了暖锋的生成过程,说明其能够反映锋面的时间演变特征。至9月11日00时(图6f),锋面系统进一步东移,冷锋开始减弱,锋区温度梯度有所减小,暖锋持续维持,锋区内仍存在典型的暖锋云系特征,云带内可见反气旋曲率的云系纹线。这进一步验证了DETR-FRO模型识别锋面系统的能力,尤其是对于锋面的移动、发展和衰减过程,均能够较好地反映其与动力、热力要素之间的关系。
从以上分析可以看出,DETR-FRO模型识别出的冷暖锋与实际气象要素场高度吻合,其识别结果在各时次均与850 hPa锋区、地面系统、云图特征相匹配。这表明,模型能够同时学习到冷锋和暖锋的关键特征,并实现对锋面结构的精准定位,进一步验证了其对锋面识别的适用性和泛化能力。
2021年11月7~9日,受黄淮气旋影响我国东北和华北等地出现了一次暴雪过程。此次过程具有降雪强度大、影响范围广的特点,内蒙古东南至吉林西部出现罕见大暴雪,伴随极端强降雪、降温和大风天气,对公众出行、生产生活造成严重影响(杜晓丹, 2024)。以下将通过分析DETR-FRO模型识别出的冷、暖锋在此次暴雪过程中的演变特征,验证其与关键气象要素的匹配情况,及DETR-FRO模型冷、暖锋对系统演变的表达能力。
图7给出了11月7日12时至8日12时逐12 h地面和850 hPa气象要素场。可以看出,11月7日12时(图7a),蒙古高压控制着东亚大陆大部分地区,气旋中心位于渤海湾附近。受气旋环流影响,该区域风向辐合。DETR-FRO模型识别到此时气旋冷暖锋已经形成,锋线自气旋中心穿出,与冷高压前沿等压线平行,降水主要集中在冷锋附近及暖锋前,而降雪则集中在气旋头部位置。结合850 hPa(图7d),冷、暖锋锋区特征明显,等温线密集并存在明显的温度平流,地面冷、暖锋均位于相应锋区的偏暖空气一侧,且锋面走向与等温线走向基本平行。同时,在锋区附近可见相对湿度大于90%的区域呈带状分布,且暖锋锋前相对湿度较大,锋后相对湿度较小,与降水区域也有较好的对应关系。由以上分析可知,DETR-FRO模型确定的气旋冷暖锋与地面气压系统、降水场、850 hPa的冷暖平流分布和相对湿度场均有较好的匹配。
12 h后,气旋东移至朝鲜半岛附近,中心气压加深至1010 hPa,锋面系统随之东移入海,气旋进一步发展成熟。此时暖锋前降水范围扩大,降水量有所增加(图7b)。850 hPa上可见锋区的温度梯度加大,冷暖空气的交汇更加剧烈(图7e)。24 h后(图7c、f),气旋继续向东北方向推进至日本海附近,中心气压进一步加深至1005 hPa,冷、暖锋也进一步东移,降水分布随之东移并仍然与锋面位置匹配。从降水场及风场的分布形态来看,此时锋面气旋已展现出锢囚特征,但因DETR-FRO模型并未识别锢囚锋,此处不再进一步讨论。通过以上分析可知,从时间演变特征来看,DETR-FRO模型识别的冷、暖锋对锋面气旋的发展演变过程有较好的表现能力。
图8a–c给出了11月7日12时至8日12时逐12 h红外云图。分析可知,7日12时(图8a),气旋呈现出明显的斜压叶状云系特征,模型识别到的冷锋位于云系下凹区,暖锋则位于斜压叶云带凸起区下方,符合斜压叶状云系与锋面结构在空间上的对应关系。8日00时(图8b),云系演变为钩状云系,冷锋对应于云界清晰边缘,暖锋位于云带反气旋方向的后缘,其位置变化与云系形态的演变相吻合。8日12时(图8c),云系已进一步发展为典型的锢囚锋云系。此时冷锋位于云系南侧整齐边界处,暖锋位于气旋中心西南侧的弯曲云带位置,符合典型锢囚锋云系的空间结构特征。可见,DETR-FRO模型识别的冷暖锋与锋面气旋演变过程中的云系形态具有较好的对应关系。
图8d–f与图8g–i分别给出了模型识别的冷锋与暖锋位置处气象要素的垂直剖面图。7日12时,在30°N附近(图8d),位温等值线密集并向冷区倾斜,锋区处1000~700 hPa风随高度逆转,说明有明显的冷平流,对应模型识别出的冷锋位置;在42°N暖锋附近(图8g),同样可见位温等值线密集并向冷区倾斜,锋区处1000~700 hPa风随高度顺转,说明有明显的暖平流,暖锋后暖区的湿度明显大于暖锋前,表现出典型的暖锋垂直结构特征。8日00时至12时(图8e–f、h–i),冷锋和暖锋位置处的气象要素剖面也都表现出典型的锋面垂直结构特征。因此,DETR-FRO模型识别的锋面位置与实际气象要素的垂直结构具有较好的一致性。
综上,在本次气旋暴雪过程中,DETR-FRO模型识别的锋面准确地表达了冷、暖锋随气旋的演变过程。锋面系统与降水落区、水汽输送及热力动力要素场保持时空一致性,进一步验证了DETR-FRO方法在温带气旋锋面识别中的适用性。
将1995~2024年间逐12 h的RGB特征图像输入至训练好的DETR-FRO模型,得到了长时间序列的温带气旋冷暖锋数据集。图9给出了1995年3月至2024年3月温带气旋冷锋在四季的发生频率分布。可以看出,冷锋频率存在明显的季节变化,冬季(图9a)的发生频率最高,春季(图9b)和秋季(图9d)次之,夏季(图10c)最低。与前人的研究一致(Qin et al., 2024b),冬季冷锋频率高值区主要位于东北平原、鄂霍次克海西侧、日本海北部及日本南部的西北太平洋区域,发生频率超过9.5%,与北太平洋的温带风暴路径位置一致。其中,东北平原和日本海北部附近的冷锋频率最高可达15.5%,而蒙古地区的冷锋频率最低。春季,高值区主要位于波斯湾附近、西西伯利亚平原和东北平原。秋季,冷锋频率的分布形势与春季基本一致,但东北平原和日本海附近的高值区频率相对增加,波斯湾附近的高值区频率相对降低。夏季冷锋活动整体偏弱,仅在伊朗高原、巴尔喀什湖和蒙古高原以东地区存在相对高值中心。四季温带气旋冷锋的高值区多位于低压控制区域,表明DETR-FRO识别出的冷锋在天气动力学上具有较好的合理性。
图10进一步展示了温带气旋暖锋的四季发生频率分布。与冷锋相比,暖锋频率整体偏低,但同样表现出明显的季节性差异。冬季(图10a)暖锋频率最高,西西伯利亚平原和西太平洋地区为主要高值区,发生频率超过4.5%。春季(图10b)和秋季(图10d)的暖锋频率分布形势与冬季基本一致,但东北平原地区的暖锋频率相对增加。夏季(图10c)暖锋活动明显减弱,仅在西西伯利亚平原和西太平洋地区有相对高值区。与前人研究(Qin et al., 2024b, 2025)相比,DETR-FRO识别的温带气旋冷暖锋发生频率在四季的空间分布形势基本一致,但数值整体偏低,这是因为本文的研究对象仅包含温带气旋中的冷暖锋。
本文提出了一种基于DETR模型的冷暖锋同步识别方法DETR-FRO。该方法利用海平面气压、850 hPa的温度平流、温度及相对湿度等气象变量构建RGB特征图像,并结合人工冷暖锋标签集训练DETR模型,实现了冷暖锋自动同步识别与分类。通过将DETR-FRO的锋面识别结果与人工识别的冷暖锋进行对比,并结合其在连续时间段内的识别表现,进一步验证了该方法在欧亚大陆温带气旋系统中同步识别冷锋与暖锋具有较高的准确性与稳定性。在典型暴雪个例的应用中,DETR-FRO模型能够准确还原冷锋、暖锋及其演变过程,且识别结果与动力、热力要素场有较好的匹配,进一步验证了该模型的泛化能力与气象可解释性。此外,基于模型识别结果构建的1995~2024年长时间序列锋面数据集,从气候尺度揭示了欧亚大陆温带气旋冷暖锋频率的季节分布特征。
与原始DETR模型相比,DETR-FRO改进了原模型的输送通道,使其可以输入多个气象要素场,而不是局限于原始模型的RGB三个通道。引入多个通道的好处是,可以利用更多的气象要素特征提升了锋面特征表达能力,使得冷暖锋均有较好的识别效果。参考之前的锋面客观识别方法(Qin et al., 2024a, 2024b, 2025)所建议的效果较好的气象要素,本文采用了海平面气压、850 hPa的温度平流、温度及相对湿度四通道输入,经过与三通道气象要素输入结果的比较,增加通道数后,冷暖锋的识别效果均得到了一定程度的改善。这种方法也能为其他复杂天气系统的自动分析提供新的技术途径。目前,本模型只针对气旋中的冷、暖锋识别,而锢囚锋的识别仍存在较大挑战,针对锢囚锋与冷、暖锋在气象特征上的差异,后续研究可以考虑引入条件机制(Conditional Mechanism),设计局部查询,针对不同锋面系统进行更精细化的特征学习。该模块的引进需要大量的测试工作,可能需要先考虑获得更多冷、暖锋及锢囚锋标签的可能途径。

参考文献 引证文献
排序方式:
1
Berry G, Reeder M J, Jakob C. 2011. A global climatology of atmospheric fronts [J]. Geophys. Res. Lett., 38(4): L04809. doi:10.1029/2010GL046451
2
Biard J C, Kunkel K E. 2019. Automated detection of weather fronts using a deep learning neural network [J]. Adv. Stat. Climate Meteorol. Oceanogr., 5(2): 147−160. doi:10.5194/ascmo-5-147-2019
3
Bueh C, Xie Z W. 2015. An objective technique for detecting large-scale tilted ridges and troughs and its application to an East Asian cold event [J]. Mon. Wea. Rev., 143(12): 4765−4783. doi:10.1175/MWR-D-14-00238.1
4
Carion N, Massa F, Synnaeve G, et al. 2020. End-to-end object detection with transformers [C]//16th European Conference on Computer Vision. Glasgow: Springer, 213–229. doi:10.1007/978-3-030-58452-8_13
5
Catto J L, Pfahl S. 2013. The importance of fronts for extreme precipitation [J]. J. Geophys. Res., 118(19): 10791−10801. doi:10.1002/jgrd.50852
6
Catto J L, Jakob C, Berry G, et al. 2012. Relating global precipitation to atmospheric fronts [J]. Geophys. Res. Lett., 39(10): L10805. doi:10.1029/2012GL051736
7
杜晓丹. 2024. 锢囚型温带气旋暴雪过程统计分析及个例研究 [D]. 南京信息工程大学硕士学位论文.
Du Xiaodan. 2024. Statistical analysis and case study of the blocking type extratropical cyclone snowstorm process [D]. M. S. thesis (in Chinese), Nanjing University of Information Science & Technology. doi:10.27248/d.cnki.gnjqc.2024.000120
8
Feng M R, Qin Y J, Lu C H. 2021. An objective identification method for wintertime cold fronts in Eurasia [J]. Adv. Atmos. Sci., 38(10): 1695−1705. doi:10.1007/s00376-021-0315-8
9
Hénin R, Ramos A M, Schemm S, et al. 2019. Assigning precipitation to mid-latitudes fronts on sub-daily scales in the North Atlantic and European sector: Climatology and trends [J]. Int. J. Climatol., 39(1): 317−330. doi:10.1002/joc.5808
10
Hewson T D. 1998. Objective fronts [J]. Meteor. Appl., 5(1): 37−65. doi:10.1017/S1350482798000553
11
Hobbs P V, Locatelli J D, Martin J E. 1990. Cold fronts aloft and the forecasting of precipitation and severe weather east of the Rocky Mountains [J]. Wea. Forecasting, 5(4): 613−626. doi:10.1175/1520-0434(1990)005<0613:CFAATF>2.0.CO;2
12
Justin A D, Willingham C, McGovern A, et al. 2023. Toward operational real-time identification of frontal boundaries using machine learning [J]. Artif. Intell. Earth Syst., 2(3): e220052. doi:10.1175/AIES-D-22-0052.1
13
Justin A D, McGovern A, Allen J T. 2025. FrontFinder AI: Efficient identification of frontal boundaries over the continental United States and NOAA’ s unified surface analysis domain using the UNET3+ model architecture [J]. Artif. Intell. Earth Syst., 4(1): e240043. doi:10.1175/AIES-D-24-0043.1
14
Lagerquist R, McGovern A, Gagne II D J. 2019. Deep learning for spatially explicit prediction of synoptic-scale fronts [J]. Wea. Forecasting, 34(4): 1137−1160. doi:10.1175/WAF-D-18-0183.1
15
刘蕊. 2024-11-15(003). 由一个深厚温带气旋带来的气候启示 [N]. 中国气象报.
Liu Rui. 2024-11-15(003). Climate insights from a deep extratropical cyclone (in Chinese) [N]. China Meteorological News. doi:10.28122/n.cnki.ncqxb.2024.002708
16
McCann D W, Whistler J P. 2001. Problems and solutions for drawing fronts objectively [J]. Meteor. Appl., 8(2): 195−203. doi:10.1017/S1350482701002079
17
Niebler S, Miltenberger A, Schmidt B, et al. 2022. Automated detection and classification of synoptic-scale fronts from atmospheric data grids [J]. Wea. Climate Dyn., 3(1): 113−137. doi:10.5194/wcd-3-113-2022
18
Parfitt R, Czaja A, Seo H. 2017. A simple diagnostic for the detection of atmospheric fronts [J]. Geophys. Res. Lett., 44(9): 4351−4358. doi:10.1002/2017GL073662
19
Qin Y J, He S Y, Lu C H, et al. 2024a. Application of the mask R-CNN model to cold front identification in Eurasia [J]. Int. J. Climatol., 44(11): 3766−3777. doi:10.1002/joc.8549
20
Qin Y J, Liu Q, Lu C H. 2024b. Cold front identification using the DETR model with satellite cloud imagery [J]. Remote Sens., 17(1): 36. doi:10.3390/rs17010036
21
Qin Y J, Wang J C, Lu C H, et al. 2025. A method for identifying warm fronts in Eurasia [J]. Int. J. Climatol., 45(1): e8701. doi:10.1002/joc.8701
22
Renard R J, Clarke L C. 1965. Experiments in numerical objective frontal analysis [J]. Mon. Wea. Rev., 93(9): 547−556. doi:10.1175/1520-0493(1965)093<0547:EINOFA>2.3.CO;2
23
Schultz D M, Vaughan G. 2011. Occluded fronts and the occlusion process: A fresh look at conventional wisdom [J]. Bull. Amer. Meteor. Soc., 92(4): 443−466. doi:10.1175/2010BAMS3057.1
24
Simmonds I, Keay K, Tristram Bye J A. 2012. Identification and climatology of Southern Hemisphere mobile fronts in a modern reanalysis [J]. J. Climate, 25(6): 1945−1962. doi:10.1175/JCLI-D-11-00100.1
25
Smith R K, Reeder M J. 1988. On the movement and low-level structure of cold fronts [J]. Mon. Wea. Rev., 116(10): 1927−1944. doi:10.1175/1520-0493(1988)116<1927:OTMALL>2.0.CO;2
26
孙淑清, 高守亭. 2005. 现代天气学概论[M]. 北京: 气象出版社, 49.
Sun Shuqing, Gao Shouting. 2005. Introduction to Modern Meteorology (in Chinese) [M]. Beijing: China Meteorological Press, 49.
27
王新敏, 邹旭恺, 翟盘茂. 2007. 北半球温带气旋的变化 [J]. 气候变化研究进展, 3(3): 154−157.
Wang Xinmin, Zou Xukai, Zhai Panmao. 2007. Researches on extratropical cyclone variability in the Northern Hemisphere [J]. Adv. Climate Change Res. (in Chinese), 3(3): 154−157. doi:10.3969/j.issn.1673-1719.2007.03.006
28
曾康, 闵锦忠, 庄潇然, 等. 2024. 基于生成对抗网络的强对流临近预报方法及其在中国东部地区的应用评估 [J]. 大气科学, 48(6): 2316−2328.
Zeng Kang, Min Jinzhong, Zhuang Xiaoran, et al. 2024. Severe convection nowcasting method based on a generative adversarial network and its application evaluation in eastern China [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 48(6): 2316−2328. doi:10.3878/j.issn.1006-9895.2310.23094
29
Zhang J L, Liu P, Zhang F, et al. 2018. CloudNet: Ground-based cloud classification with deep convolutional neural network [J]. Geophys. Res. Lett., 45(16): 8665−8672. doi:10.1029/2018GL077787
30
张颖娴, 丁一汇, 李巧萍. 2012. ERA40再分析资料揭示的北半球和东亚地区温带气旋生成频率变化 [J]. 气象, 38(6): 646−656.
Zhang Yingxian, Ding Yihui, Li Qiaoping. 2012. Cyclogenesis frequency changes of extratropical cyclones in the Northern Hemisphere and East Asia revealed by ERA40 reanalysis data [J]. Meteorological Monthly (in Chinese), 38(6): 646−656. doi:10.7519/j.issn.1000-0526.2012.06.002
2026年第50卷第4期
PDF下载
158
62
引用本文
BibTeX
文章信息
doi: 10.3878/j.issn.1006-9895.2510.25091
  • 接收时间:2025-05-26
  • 首发时间:2026-09-30
  • 出版时间:2026-07-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-05-26
基金
作者信息
    1南京信息工程大学气候系统预测与变化应对全国重点实验室/气象灾害教育部重点实验室/气象灾害预报预警与评估协同创新中心, 南京210044
    2南京信息工程大学大气科学学院, 南京210044
    3南京市气象探测中心, 南京210019

通讯作者:

秦育婧,女,副教授,主要从事极端天气气候、人工智能与气象应用、大气环流异常及短期临近预报研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/dqkx/CN/10.3878/j.issn.1006-9895.2510.25091
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏