干旱区生态环境知识资源中心(Arid): Delineation and optimization of cotton farmland management zone based on time series of soil-crop properties at landscape scale in south Xinjiang, China

Arid

DOI	10.1016/j.still.2023.105744
	Delineation and optimization of cotton farmland management zone based on time series of soil-crop properties at landscape scale in south Xinjiang, China
	Wang, Nan; Xu, Dongyun; Xue, Jie; Zhang, Xianglin; Hong, Yongsheng; Peng, Jie; Li, Hongyi; Mouazen, Abdul Mounem; He, Yong; Shi, Zhou
通讯作者	Shi, Z
来源期刊	SOIL & TILLAGE RESEARCH
ISSN	0167-1987
EISSN	1879-3444
出版年	2023
卷号	231
英文摘要	Traditional farmland management approaches treat farmland as a homogeneous unit, leading to uneven distribution and waste of resources, which is contrary to precision agriculture. Management zones (MZ) delineation is an essential method of site-specific management (SSM) for accomplishing variable rate input of resources in subregions by recognizing the heterogeneity of farmland soils. Aimed at the soil salinization problem in Xinjiang, this study estimates the salt content of soil profiles and essential nutrient content using time series Sentinel-2 images and machine learning methods, and describes crop growth conditions through dynamic time warping. Subsequently, geographically weighted principal component analysis and possibilistic fuzzy C-means algorithm were employed to outline MZ based on soil-crop characteristics. Three scales (10 m, 100 m, and field scales divided by ridges) were examined to obtain the optimal scale of delineation. Soil-crop properties were estimated at a spatial resolution of 10 m using machine learning methods with R2 > 0.65. In the zoning process, the field scale was selected with accuracy and practicality in agricultural practices, with comparatively higher homogeneity within zones and variation between zones. The soils in the study area were categorized into three zones at field scale with a fuzzy performance index of 0.68, a normalized classification entropy of 0.71, and fuzzy class of 0.30. Soil function is primarily restricted by salinity and fertility. Fertilizer recommendation strategies were given for the three zones. This study uses time-series remote sensing images to obtain a high-precision estimation of soil-crop characteristics at 10 m, and optimize the scale of farmland management zoning, which provides a valuable technology and scale selection for SSM for cotton management zone delineation in saline farmland in arid and semi-arid regions.
英文关键词	Management zones Remote sensing Dynamic time warping Multivariate spatial clustering Scales optimization
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:000999874500001
WOS关键词	VEGETATION INDEX ; ELECTRICAL-CONDUCTIVITY ; SPATIAL VARIABILITY ; WET SEASONS ; SALINITY ; YIELD ; REGION ; MODIS ; PHOSPHORUS ; PHENOLOGY
WOS类目	Soil Science
WOS研究方向	Agriculture
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/398665
推荐引用方式 GB/T 7714	Wang, Nan,Xu, Dongyun,Xue, Jie,et al. Delineation and optimization of cotton farmland management zone based on time series of soil-crop properties at landscape scale in south Xinjiang, China[J],2023,231.
APA	Wang, Nan.,Xu, Dongyun.,Xue, Jie.,Zhang, Xianglin.,Hong, Yongsheng.,...&Shi, Zhou.(2023).Delineation and optimization of cotton farmland management zone based on time series of soil-crop properties at landscape scale in south Xinjiang, China.SOIL & TILLAGE RESEARCH,231.
MLA	Wang, Nan,et al."Delineation and optimization of cotton farmland management zone based on time series of soil-crop properties at landscape scale in south Xinjiang, China".SOIL & TILLAGE RESEARCH 231(2023).