利用机器学习算法和气候数据评估和预测气象干旱

IF 4.8 2区 环境科学与生态学 Q1 ENVIRONMENTAL SCIENCES
Khalid En-Nagre , Mourad Aqnouy , Ayoub Ouarka , Syed Ali Asad Naqvi , Ismail Bouizrou , Jamal Eddine Stitou El Messari , Aqil Tariq , Walid Soufan , Wenzhao Li , Hesham El-Askary
{"title":"利用机器学习算法和气候数据评估和预测气象干旱","authors":"Khalid En-Nagre ,&nbsp;Mourad Aqnouy ,&nbsp;Ayoub Ouarka ,&nbsp;Syed Ali Asad Naqvi ,&nbsp;Ismail Bouizrou ,&nbsp;Jamal Eddine Stitou El Messari ,&nbsp;Aqil Tariq ,&nbsp;Walid Soufan ,&nbsp;Wenzhao Li ,&nbsp;Hesham El-Askary","doi":"10.1016/j.crm.2024.100630","DOIUrl":null,"url":null,"abstract":"<div><p>Monitoring drought in semi-arid regions due to climate change is of paramount importance. This study, conducted in Morocco’s Upper Drâa Basin (UDB), analyzed data spanning from 1980 to 2019, focusing on the calculation of drought indices, specifically the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) at multiple timescales (1, 3, 9, 12 months). Trends were assessed using statistical methods such as the Mann-Kendall test and the Sen’s Slope estimator. Four significant machine learning (ML) algorithms, including Random Forest, Voting Regressor, AdaBoost Regressor, and K-Nearest Neighbors Regressor, were evaluated to predict the SPEI values for both three and 12-month periods. The algorithms’ performance was measured using statistical indices. The study revealed that drought distribution within the UDB is not uniform, with a discernible decreasing trend in SPEI values. Notably, the four ML algorithms effectively predicted SPEI values for the specified periods. Random Forest, Voting Regressor, and AdaBoost demonstrated the highest Nash-Sutcliffe Efficiency (NSE) values, ranging from 0.74 to 0.93. In contrast, the K-Nearest Neighbors algorithm produced values within the range of 0.44 to 0.84. These research findings have the potential to provide valuable insights for water resource management experts and policymakers. However, it is imperative to enhance data collection methodologies and expand the distribution of measurement sites to improve data representativeness and reduce errors associated with local variations.</p></div>","PeriodicalId":54226,"journal":{"name":"Climate Risk Management","volume":"45 ","pages":"Article 100630"},"PeriodicalIF":4.8000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2212096324000470/pdfft?md5=86c5c7e81b15816365b705592e8fb810&pid=1-s2.0-S2212096324000470-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Assessment and prediction of meteorological drought using machine learning algorithms and climate data\",\"authors\":\"Khalid En-Nagre ,&nbsp;Mourad Aqnouy ,&nbsp;Ayoub Ouarka ,&nbsp;Syed Ali Asad Naqvi ,&nbsp;Ismail Bouizrou ,&nbsp;Jamal Eddine Stitou El Messari ,&nbsp;Aqil Tariq ,&nbsp;Walid Soufan ,&nbsp;Wenzhao Li ,&nbsp;Hesham El-Askary\",\"doi\":\"10.1016/j.crm.2024.100630\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Monitoring drought in semi-arid regions due to climate change is of paramount importance. This study, conducted in Morocco’s Upper Drâa Basin (UDB), analyzed data spanning from 1980 to 2019, focusing on the calculation of drought indices, specifically the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) at multiple timescales (1, 3, 9, 12 months). Trends were assessed using statistical methods such as the Mann-Kendall test and the Sen’s Slope estimator. Four significant machine learning (ML) algorithms, including Random Forest, Voting Regressor, AdaBoost Regressor, and K-Nearest Neighbors Regressor, were evaluated to predict the SPEI values for both three and 12-month periods. The algorithms’ performance was measured using statistical indices. The study revealed that drought distribution within the UDB is not uniform, with a discernible decreasing trend in SPEI values. Notably, the four ML algorithms effectively predicted SPEI values for the specified periods. Random Forest, Voting Regressor, and AdaBoost demonstrated the highest Nash-Sutcliffe Efficiency (NSE) values, ranging from 0.74 to 0.93. In contrast, the K-Nearest Neighbors algorithm produced values within the range of 0.44 to 0.84. These research findings have the potential to provide valuable insights for water resource management experts and policymakers. However, it is imperative to enhance data collection methodologies and expand the distribution of measurement sites to improve data representativeness and reduce errors associated with local variations.</p></div>\",\"PeriodicalId\":54226,\"journal\":{\"name\":\"Climate Risk Management\",\"volume\":\"45 \",\"pages\":\"Article 100630\"},\"PeriodicalIF\":4.8000,\"publicationDate\":\"2024-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2212096324000470/pdfft?md5=86c5c7e81b15816365b705592e8fb810&pid=1-s2.0-S2212096324000470-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Climate Risk Management\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2212096324000470\",\"RegionNum\":2,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENVIRONMENTAL SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Climate Risk Management","FirstCategoryId":"93","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2212096324000470","RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0

摘要

监测气候变化导致的半干旱地区干旱至关重要。本研究在摩洛哥上德拉盆地(UDB)进行,分析了 1980 年至 2019 年的数据,重点是计算干旱指数,特别是多个时间尺度(1、3、9、12 个月)的标准化降水指数(SPI)和标准化降水蒸散指数(SPEI)。使用 Mann-Kendall 检验和 Sen's Slope 估计器等统计方法对趋势进行了评估。评估了四种重要的机器学习(ML)算法,包括随机森林算法、投票回归算法、AdaBoost 回归算法和 K-Nearest Neighbors 回归算法,以预测 3 个月和 12 个月期间的 SPEI 值。这些算法的性能使用统计指数进行衡量。研究结果表明,UDB 内的干旱分布并不均匀,SPEI 值呈明显的下降趋势。值得注意的是,四种 ML 算法都能有效预测特定时期的 SPEI 值。随机森林算法、投票回归算法和 AdaBoost 算法的纳什-萨特克利夫效率(NSE)值最高,从 0.74 到 0.93 不等。相比之下,K-近邻算法的效率值在 0.44 到 0.84 之间。这些研究成果有可能为水资源管理专家和政策制定者提供有价值的见解。不过,当务之急是改进数据收集方法,扩大测量点的分布范围,以提高数据的代表性,减少与地方差异相关的误差。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Assessment and prediction of meteorological drought using machine learning algorithms and climate data

Monitoring drought in semi-arid regions due to climate change is of paramount importance. This study, conducted in Morocco’s Upper Drâa Basin (UDB), analyzed data spanning from 1980 to 2019, focusing on the calculation of drought indices, specifically the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) at multiple timescales (1, 3, 9, 12 months). Trends were assessed using statistical methods such as the Mann-Kendall test and the Sen’s Slope estimator. Four significant machine learning (ML) algorithms, including Random Forest, Voting Regressor, AdaBoost Regressor, and K-Nearest Neighbors Regressor, were evaluated to predict the SPEI values for both three and 12-month periods. The algorithms’ performance was measured using statistical indices. The study revealed that drought distribution within the UDB is not uniform, with a discernible decreasing trend in SPEI values. Notably, the four ML algorithms effectively predicted SPEI values for the specified periods. Random Forest, Voting Regressor, and AdaBoost demonstrated the highest Nash-Sutcliffe Efficiency (NSE) values, ranging from 0.74 to 0.93. In contrast, the K-Nearest Neighbors algorithm produced values within the range of 0.44 to 0.84. These research findings have the potential to provide valuable insights for water resource management experts and policymakers. However, it is imperative to enhance data collection methodologies and expand the distribution of measurement sites to improve data representativeness and reduce errors associated with local variations.

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
Climate Risk Management
Climate Risk Management Earth and Planetary Sciences-Atmospheric Science
CiteScore
8.20
自引率
4.50%
发文量
76
审稿时长
30 weeks
期刊介绍: Climate Risk Management publishes original scientific contributions, state-of-the-art reviews and reports of practical experience on the use of knowledge and information regarding the consequences of climate variability and climate change in decision and policy making on climate change responses from the near- to long-term. The concept of climate risk management refers to activities and methods that are used by individuals, organizations, and institutions to facilitate climate-resilient decision-making. Its objective is to promote sustainable development by maximizing the beneficial impacts of climate change responses and minimizing negative impacts across the full spectrum of geographies and sectors that are potentially affected by the changing climate.
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信