Smart agricultural (SA) technology has become a technological support for modern agriculture. By exploring the decision-making process and psychological motivation of farmers in adopting SA technology, it is conducive to achieving the popularisation of SA technology and promoting the modernisation of agriculture. Based on microscopic research data, a Structural Equation Model (SEM) is used to analyse the influencing factors and extent of cotton farmers’ adoption of SA technologies, using Deconstructive Theory of Planned Behavior (DTPB) as the analytical framework. This was combined with in-depth interviews to further reveal the motivations and influencing mechanisms of cotton farmers’ adoption of SA technologies. The results show that under the behavioural belief dimension, cotton farmers value the positive effect of perceived usefulness even though the risk of the technology itself has a dampening effect on adoption intentions. Under the normative belief dimension, superior influence influenced the willingness to adopt SA technologies to a greater extent than peer influence. Under the control belief dimension, factors such as self-efficacy and information channels influence willingness to adopt technology and behaviour. In addition, behavioural attitudes, subjective norms, and perceived behavioural control all contribute to cotton farmers’ willingness to adopt SA technologies, and can also influence behaviour directly or indirectly through willingness to adopt. Policy and technology satisfaction positively moderate the transition from willingness to behaviour. Therefore, preferential policies are proposed to reduce the cost of adopting SA technologies; to continuously improve the level of SA technologies; to establish SA technology test plots to provide a reference base; and to increase knowledge training on SA and expand access to information.

The coordinated development of farmland transfer (FT) and labor migration (LM) is of great efficiency significance to facilitate the development of rural economy and implement the rural revitalization strategy. The study used socioeconomic data from 30 provinces/autonomous regions/municipalities (hereafter referred to as provinces) in China to measure the coupling coordination degree (CCD) of FT and LM. It adopted the coupling coordination degree model (CCDM), exploratory spatial data analysis method (ESDA), and gray relational analysis model (GARM) to investigate the spatial differences in the CCD and its influencing factors. The results indicate the following: (1) Regional differences are evident despite the fact that the comprehensive evaluation level of FT and LM in the various provinces is relatively low and displaying a rising trend. (2) The CCD of FT and LM exhibits a fluctuating upward trend and is at the primary coupling coordination stage, with a significant difference in coupling coordination levels between regions, and a spatial distribution pattern of central region > eastern region > northeast region > western region. (3) The CCD shows a strong global spatial positive correlation with clear fluctuations, demonstrating the agglomeration dispersion development tendency over time; the local spatial agglomeration state emerges and stabilizes. According to the distribution pattern, the Western region exhibits weak agglomeration type, whereas the eastern and central regions exhibit strong agglomeration type. (4) There are significant variations between provinces in terms of the intensity of the CCD of FT and LM, as well as the level of concurrent employment business, the level of non-agricultural industry development, the level of urbanization, the level of agricultural equipment, and the land approval.

Agriculture big data can use the advantages of information technology to provide data support and technical support for improving the level of farmland transfer and labor transfer. And the coupling and coordinated development of the two is an essential way to ensure the effective promotion of the rural revitalization strategy and complete the building of a moderately prosperous society in an all-round path. First, aided by the coupling coordination model and exploratory spatial data analysis, a detailed analysis of the spatiotemporal patterns and dynamics of farmland transfer and labor transfer was conducted. The driving factors were quantitatively examined using the gray correlation model. The results indicate that the degree of coupling and coordination between Xinjiang’s farmland transfer and labor transfer ranges from 0.50 to 0.70, basically on the between of slight and primary coordination, and the degree of coordination among the regions varies greatly. The labor resource endowment, agricultural equipment level, farmland ownership confirmation, nonagricultural industry development level, and urbanization level are significantly impacting the degree of coupling and coordination between Xinjiang’s farmland transfer and labor transfer.