Scholarlink

<p>Drought is widely regarded as one of the most devastating climate-related phenomena, which adversely affects agricultural production, water resources, ecosystems, and human livelihoods worldwide. To improve drought monitoring, assessment, mitigation and decision making, this study uses advanced technologies and drought indices, integrating meteorological, hydrological and agricultural data through geospatial tools. We analyze LANDSAT 8 satellite imagery and apply supervised machine learning algorithms, principally Support Vector Machines (SVM), using the Google Earth Engine (GEE) platform to examine land use patterns and classify land cover types. Using the Vegetation Health Index (VHI), this study detected agricultural drought events in the studied area for the years 2000, 2003, 2005, 2006, 2008, 2016, and 2017. The study establishes spatial correlations between seasonal VHI and other drought indices, including SPI, PAA, and NDWI, to elucidate spatial relationships between agricultural, meteorological, and hydrological droughts. A detailed seasonal drought profile is generated for El Tarf province in northeastern Algeria using Boolean spatial correlations, revealing significant drought periods from 2000 to 2020. The intersection of correlations between VHI and SPI-3, PAA, and NDWI revealed that the most affected regions were, respectively, spring (68.19%), summer (55.78%), winter (55.30%) and autumn (43.88%), throughout the province for 20 years. This research highlights how crucial it is to incorporate technological innovations into drought monitoring, offering a scientific foundation for policy development and risk reduction measures in the context of climate change.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Spatial and temporal trends in meteorological, hydrological and agricultural droughts in northeastern Algeria

  • 期刊论文

  • 发表日期：

    2026年5月27日

查看全文
  • Mohamed Fethi Hadjadj,
  • Rabia Malkia,
  • Haroun Chenchouni,
  • Djamal Bengusmia,
  • Ali Slimani

摘要

Drought is widely regarded as one of the most devastating climate-related phenomena, which adversely affects agricultural production, water resources, ecosystems, and human livelihoods worldwide. To improve drought monitoring, assessment, mitigation and decision making, this study uses advanced technologies and drought indices, integrating meteorological, hydrological and agricultural data through geospatial tools. We analyze LANDSAT 8 satellite imagery and apply supervised machine learning algorithms, principally Support Vector Machines (SVM), using the Google Earth Engine (GEE) platform to examine land use patterns and classify land cover types. Using the Vegetation Health Index (VHI), this study detected agricultural drought events in the studied area for the years 2000, 2003, 2005, 2006, 2008, 2016, and 2017. The study establishes spatial correlations between seasonal VHI and other drought indices, including SPI, PAA, and NDWI, to elucidate spatial relationships between agricultural, meteorological, and hydrological droughts. A detailed seasonal drought profile is generated for El Tarf province in northeastern Algeria using Boolean spatial correlations, revealing significant drought periods from 2000 to 2020. The intersection of correlations between VHI and SPI-3, PAA, and NDWI revealed that the most affected regions were, respectively, spring (68.19%), summer (55.78%), winter (55.30%) and autumn (43.88%), throughout the province for 20 years. This research highlights how crucial it is to incorporate technological innovations into drought monitoring, offering a scientific foundation for policy development and risk reduction measures in the context of climate change.