Scholarlink

<p>Agricultural drought assessment in the northwestern region of Bangladesh is essential for promoting sustainable farming practices, mitigating crop losses, and enhancing resilience to climate variability. This study offers a comprehensive remote sensing-based evaluation of agricultural drought from 2016 to 2024, utilizing Google Earth Engine (GEE) to analyze spatiotemporal variations in drought-related variables and indicators. The results reveal a significant upward trend in Land Surface Temperature (LST), peaking at 32.45&#xa0;°C in 2024, which heightens the risk of agricultural drought. Meanwhile, the Normalized Difference Vegetation Index (NDVI) remained relatively stable, with slight increases over time, indicating the resilience of irrigated crops. Three drought indices, derived from LST and NDVI—the Vegetation Condition Index (VCI), Temperature Condition Index (TCI), and Vegetation Health Index (VHI)—consistently showed that the region predominantly experienced ‘‘no drought’’ conditions, with the most significant area, 33,311 km2 (94%), classified as ‘‘no drought’’ in 2023 according to the VHI. Mild and moderate droughts were observed in smaller areas, with the most affected regions being 7,178 km2 (20%) for mild drought in 2024 and 4,593 km2 (13%) for moderate drought in 2017, as indicated by the TCI. The study also highlighted a notable decline in rainfall trends over the study period. Rainfall-based indices such as the Precipitation Condition Index (PCI) and Standardized Precipitation Index (SPI) showed that over 60% of the area experienced mild to moderate drought conditions, with the largest affected areas covering 31,490 km2 (91%) for ‘‘mild drought’’ and 31,853 km2 (92%) for ‘‘moderate drought,’’ while minimal areas experienced ‘‘no drought.’’ Comparative assessment of the five drought-related indicators identified VHI as the most suitable integrated index for agricultural drought monitoring, as it combines vegetation condition and temperature-related stress, whereas SPI and PCI primarily provide rainfall-based support. Severe and extreme droughts were rare, appearing in isolated pockets, particularly in the TCI index, with affected areas ranging from 568 km2 to 3,538 km2, potentially linked to the gradual rise in temperatures. Furthermore, the analysis of the 2024 Land Use/Land Cover (LULC) map revealed that croplands dominate the region (63.57%), making it particularly vulnerable to agricultural drought, with urbanization (20.99%) further exacerbating water stress.</p> Graphical Abstract <p></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

Remote sensing-based drought monitoring in northwestern Bangladesh: leveraging google earth engine for comprehensive assessment

  • 期刊论文

  • 发表日期：

    2026年6月22日

查看全文
  • Md. Shafayat Ullah,
  • Md Zidanur Rahman Zidan,
  • Sujit Kumar Roy,
  • Mohammad Mohinuzzaman,
  • Shahriar Md Arifur Rahman,
  • Md. Kamruzzaman Tusar,
  • Protima Sarker

摘要

Agricultural drought assessment in the northwestern region of Bangladesh is essential for promoting sustainable farming practices, mitigating crop losses, and enhancing resilience to climate variability. This study offers a comprehensive remote sensing-based evaluation of agricultural drought from 2016 to 2024, utilizing Google Earth Engine (GEE) to analyze spatiotemporal variations in drought-related variables and indicators. The results reveal a significant upward trend in Land Surface Temperature (LST), peaking at 32.45 °C in 2024, which heightens the risk of agricultural drought. Meanwhile, the Normalized Difference Vegetation Index (NDVI) remained relatively stable, with slight increases over time, indicating the resilience of irrigated crops. Three drought indices, derived from LST and NDVI—the Vegetation Condition Index (VCI), Temperature Condition Index (TCI), and Vegetation Health Index (VHI)—consistently showed that the region predominantly experienced ‘‘no drought’’ conditions, with the most significant area, 33,311 km2 (94%), classified as ‘‘no drought’’ in 2023 according to the VHI. Mild and moderate droughts were observed in smaller areas, with the most affected regions being 7,178 km2 (20%) for mild drought in 2024 and 4,593 km2 (13%) for moderate drought in 2017, as indicated by the TCI. The study also highlighted a notable decline in rainfall trends over the study period. Rainfall-based indices such as the Precipitation Condition Index (PCI) and Standardized Precipitation Index (SPI) showed that over 60% of the area experienced mild to moderate drought conditions, with the largest affected areas covering 31,490 km2 (91%) for ‘‘mild drought’’ and 31,853 km2 (92%) for ‘‘moderate drought,’’ while minimal areas experienced ‘‘no drought.’’ Comparative assessment of the five drought-related indicators identified VHI as the most suitable integrated index for agricultural drought monitoring, as it combines vegetation condition and temperature-related stress, whereas SPI and PCI primarily provide rainfall-based support. Severe and extreme droughts were rare, appearing in isolated pockets, particularly in the TCI index, with affected areas ranging from 568 km2 to 3,538 km2, potentially linked to the gradual rise in temperatures. Furthermore, the analysis of the 2024 Land Use/Land Cover (LULC) map revealed that croplands dominate the region (63.57%), making it particularly vulnerable to agricultural drought, with urbanization (20.99%) further exacerbating water stress.

Graphical Abstract