使用全局数据集的LSTM网络预测流

IF 2.6 Q2 WATER RESOURCES Frontiers in Water Pub Date : 2023-06-05 DOI:10.3389/frwa.2023.1166124
Katharina Wilbrand, Riccardo Taormina, Marie-claire ten Veldhuis, M. Visser, M. Hrachowitz, J. Nuttall, R. Dahm
{"title":"使用全局数据集的LSTM网络预测流","authors":"Katharina Wilbrand, Riccardo Taormina, Marie-claire ten Veldhuis, M. Visser, M. Hrachowitz, J. Nuttall, R. Dahm","doi":"10.3389/frwa.2023.1166124","DOIUrl":null,"url":null,"abstract":"Streamflow predictions remain a challenge for poorly gauged and ungauged catchments. Recent research has shown that deep learning methods based on Long Short-Term Memory (LSTM) cells outperform process-based hydrological models for rainfall-runoff modeling, opening new possibilities for prediction in ungauged basins (PUB). These studies usually feature local datasets for model development, while predictions in ungauged basins at a global scale require training on global datasets. In this study, we develop LSTM models for over 500 catchments from the CAMELS-US data base using global ERA5 meteorological forcing and global catchment characteristics retrieved with the HydroMT tool. Comparison against an LSTM trained with local datasets shows that, while the latter generally yields superior performances due to the higher spatial resolution meteorological forcing (overall median daily NSE 0.54 vs. 0.71), training with ERA5 results in higher NSE in most catchments of Western and North-Western US (median daily NSE of 0.83 vs. 0.78). No significant changes in performance occur when substituting local with global data sources for deriving the catchment characteristics. These results encourage further research to develop LSTM models for worldwide predictions of streamflow in ungauged basins using available global datasets. Promising directions include training the models with streamflow data from different regions of the world and with higher quality meteorological forcing.","PeriodicalId":33801,"journal":{"name":"Frontiers in Water","volume":" ","pages":""},"PeriodicalIF":2.6000,"publicationDate":"2023-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Predicting streamflow with LSTM networks using global datasets\",\"authors\":\"Katharina Wilbrand, Riccardo Taormina, Marie-claire ten Veldhuis, M. Visser, M. Hrachowitz, J. Nuttall, R. Dahm\",\"doi\":\"10.3389/frwa.2023.1166124\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Streamflow predictions remain a challenge for poorly gauged and ungauged catchments. Recent research has shown that deep learning methods based on Long Short-Term Memory (LSTM) cells outperform process-based hydrological models for rainfall-runoff modeling, opening new possibilities for prediction in ungauged basins (PUB). These studies usually feature local datasets for model development, while predictions in ungauged basins at a global scale require training on global datasets. In this study, we develop LSTM models for over 500 catchments from the CAMELS-US data base using global ERA5 meteorological forcing and global catchment characteristics retrieved with the HydroMT tool. Comparison against an LSTM trained with local datasets shows that, while the latter generally yields superior performances due to the higher spatial resolution meteorological forcing (overall median daily NSE 0.54 vs. 0.71), training with ERA5 results in higher NSE in most catchments of Western and North-Western US (median daily NSE of 0.83 vs. 0.78). No significant changes in performance occur when substituting local with global data sources for deriving the catchment characteristics. These results encourage further research to develop LSTM models for worldwide predictions of streamflow in ungauged basins using available global datasets. Promising directions include training the models with streamflow data from different regions of the world and with higher quality meteorological forcing.\",\"PeriodicalId\":33801,\"journal\":{\"name\":\"Frontiers in Water\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":2.6000,\"publicationDate\":\"2023-06-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers in Water\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.3389/frwa.2023.1166124\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"WATER RESOURCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Water","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3389/frwa.2023.1166124","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"WATER RESOURCES","Score":null,"Total":0}
引用次数: 2

摘要

对于测量差和未测量的集水区,流量预测仍然是一个挑战。最近的研究表明,基于长短期记忆(LSTM)细胞的深度学习方法在降雨径流建模方面优于基于过程的水文模型,为未测量流域(PUB)的预测开辟了新的可能性。这些研究通常采用本地数据集来开发模型,而在全球尺度上对未测量的盆地进行预测则需要对全球数据集进行训练。在这项研究中,我们利用全球ERA5气象强迫和HydroMT工具检索的全球流域特征,从CAMELS-US数据库中开发了500多个流域的LSTM模型。与使用本地数据集训练的LSTM相比,后者通常由于更高的空间分辨率气象强迫(总体日NSE中位数0.54比0.71)而具有更优越的性能,而使用ERA5训练在美国西部和西北部的大多数集水区产生更高的NSE(日NSE中位数0.83比0.78)。当用全局数据源代替局部数据源来推导集水区特征时,不会出现显著的性能变化。这些结果鼓励进一步研究开发LSTM模型,利用可用的全球数据集对未测量流域的全球流量进行预测。有希望的方向包括用来自世界不同地区的流量数据和更高质量的气象强迫训练模型。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Predicting streamflow with LSTM networks using global datasets
Streamflow predictions remain a challenge for poorly gauged and ungauged catchments. Recent research has shown that deep learning methods based on Long Short-Term Memory (LSTM) cells outperform process-based hydrological models for rainfall-runoff modeling, opening new possibilities for prediction in ungauged basins (PUB). These studies usually feature local datasets for model development, while predictions in ungauged basins at a global scale require training on global datasets. In this study, we develop LSTM models for over 500 catchments from the CAMELS-US data base using global ERA5 meteorological forcing and global catchment characteristics retrieved with the HydroMT tool. Comparison against an LSTM trained with local datasets shows that, while the latter generally yields superior performances due to the higher spatial resolution meteorological forcing (overall median daily NSE 0.54 vs. 0.71), training with ERA5 results in higher NSE in most catchments of Western and North-Western US (median daily NSE of 0.83 vs. 0.78). No significant changes in performance occur when substituting local with global data sources for deriving the catchment characteristics. These results encourage further research to develop LSTM models for worldwide predictions of streamflow in ungauged basins using available global datasets. Promising directions include training the models with streamflow data from different regions of the world and with higher quality meteorological forcing.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Frontiers in Water
Frontiers in Water WATER RESOURCES-
CiteScore
4.00
自引率
6.90%
发文量
224
审稿时长
13 weeks
期刊最新文献
River-floodplain connectivity and residence times controlled by topographic bluffs along a backwater transition Rating curve development and uncertainty analysis in mountainous watersheds for informed hydrology and resource management Characterization of sewage quality and its spatiotemporal variations in a small town in Eastern Guangdong, China Model and remote-sensing-guided experimental design and hypothesis generation for monitoring snow-soil–plant interactions Precipitation fuels dissolved greenhouse gas (CO2, CH4, N2O) dynamics in a peatland-dominated headwater stream: results from a continuous monitoring setup
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1