干旱区生态环境知识资源中心(Arid): Sentinel-2 accurately estimated wheat yield in a semi-arid region compared with Landsat 8

Arid

DOI	10.1080/01431161.2023.2232542
	Sentinel-2 accurately estimated wheat yield in a semi-arid region compared with Landsat 8
	Faqe Ibrahim, Gaylan Rasul; Rasul, Azad; Abdullah, Haidi
通讯作者	Ibrahim, GRF
来源期刊	INTERNATIONAL JOURNAL OF REMOTE SENSING
ISSN	0143-1161
EISSN	1366-5901
出版年	2023
卷号	44 期号:13 页码:4115-4136
英文摘要	Wheat and barley are crucial food resources for the global population, making their growth and monitoring essential to enhance food security worldwide. Effective observation of these crops is necessary to address production issues and mitigate the impacts of weather changes. Advancements in remote sensing technology have significantly improved the observation and estimation processes. In this study, various spectral vegetation indices were utilized, along with canopy biophysical properties (such as LAI) and biochemical properties (like chlorophyll). These properties were derived from satellite data, specifically Landsat 8 and Sentinel-2, using tools like Google Earth Engine (GEE) and the R Program. Samples of wheat and barley were collected before reaching their optimal harvest stage, and a correlation was established between the vegetation indices (e.g. NDVI, NDWI, EVI, SAVI, CMFI, SR, RVI, GRVI, and NDRI) and actual production data. Yield prediction algorithms were employed, and the results were used to generate prediction yield maps. The findings revealed a strong relationship between the vegetation indices derived from Sentinel-2 and Landsat images and the actual grain yield, with an R-2 of 0.77 and 0.71, respectively. Additionally, the study demonstrated that the most robust relationship was observed between the LAI data obtained from Sentinel-2 and cereal yield data, achieving an R-2 of 0.68. Among the indices derived from Landsat images, NDWI exhibited the highest correlation with an R-2 of 0.59. The root mean square error (RMSE) was found to be the lowest for Sentinel-2 (0.57) and Landsat 8 (1.54). Furthermore, the study indicated that the least significant relationship for grain yield prediction was observed between the NDRI index for Sentinel-2 (R-2 0.1) and the SAVI index for Landsat images (R-2 0.47).
英文关键词	GEE Landsat 8 OLI multi-linear regression remote sensing vegetation indices wheat and barley
类型	Article
语种	英语
开放获取类型	Green Published
收录类别	SCI-E
WOS记录号	WOS:001028210100001
WOS关键词	DIFFERENCE WATER INDEX ; VEGETATION INDEX ; CROP YIELD ; TIME ; NDWI
WOS类目	Remote Sensing ; Imaging Science & Photographic Technology
WOS研究方向	Remote Sensing ; Imaging Science & Photographic Technology
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/397039
推荐引用方式 GB/T 7714	Faqe Ibrahim, Gaylan Rasul,Rasul, Azad,Abdullah, Haidi. Sentinel-2 accurately estimated wheat yield in a semi-arid region compared with Landsat 8[J],2023,44(13):4115-4136.
APA	Faqe Ibrahim, Gaylan Rasul,Rasul, Azad,&Abdullah, Haidi.(2023).Sentinel-2 accurately estimated wheat yield in a semi-arid region compared with Landsat 8.INTERNATIONAL JOURNAL OF REMOTE SENSING,44(13),4115-4136.
MLA	Faqe Ibrahim, Gaylan Rasul,et al."Sentinel-2 accurately estimated wheat yield in a semi-arid region compared with Landsat 8".INTERNATIONAL JOURNAL OF REMOTE SENSING 44.13(2023):4115-4136.