The non-farming use of cropland has led to food insecurity in China due to drastic land use (LU) changes under the stresses of ecological restoration and urbanization, particularly in non-major grain-producing areas. Questions were raised about spatiotemporal cropland losses/gains and their drivers in these areas in the future for sustainable development of the agriculture sector. However, the answers to these questions have not been well acknowledged. This study, therefore, presents analyses of cropland area change from 1990 to 2018 and from 2018 to 2051 in Shaanxi province based on the Future Land Use Simulation (FLUS) model that follows the integration of the Shared Socioeconomic Pathway 2 and the Representative Concentration Pathway 4.5 (SSP245) within the International Coupled Model Intercomparison Project 6 (CMIP6). The results highlight that ecological restoration and fast-paced urbanization mainly drove the alarming non-farming use of cropland. The per capita cropland area is projected to increase, but the cropland loss will still occur, which potentially causes food insecurity. Thus, food security will be a challenging issue in the near future. The quantitative findings call for careful designs of LU policies, taking into account cropland protection, socio-economic development, and ecological restoration.

Spatiotemporally mapping winter wheat is imperative for informing and shaping global food security policies. Traditional mapping methods heavily rely on sufficient and reliable samples obtained through labor-intensive fieldwork and manual sample collection. However, these methods are time-consuming, costly, and lack timely and continuous data collection. To address these challenges and fully leverage remote sensing big data and cloud computing platforms like Google Earth Engine (GEE), this paper developed an algorithm for Auto-Generating Winter Wheat Samples for mapping (AGWWS). The AGWWS utilizes historical samples to determine the optimal migration threshold by measuring Spectral Angle Distance (SAD), Euclidean Distance (ED), and Near-Infrared band Difference Index (NIRDI). This facilitates the auto-generation of winter wheat sample sets for the years 2000, 2005, 2010, 2015, and 2021. Approximately two-thirds of the samples were allocated for training, with the remaining one-third used for validating the mapping method, employing the One-Class Support Vector Machine (OCSVM). The Huang–Huai–Hai (HHH) Plain, a major winter wheat production region, was selected to perform the algorithm and subsequent analysis on. Different combinations of the hyper-parameters, gamma and nu, of the OCSVM based on the Gaussian Radial Basis Function Kernel were tested for each year. Following correlation analysis between the winter wheat area derived from the generated maps and the national statistical dataset at the city level, the map with the highest corresponding R² was chosen as the AGWWS map for each year (0.77, 0.77, 0.80, 0.86, and 0.87 for 2000, 2005, 2010, 2015, and 2021, respectively). The AGWWS maps ultimately achieved an average Overall Accuracy of 81.65%. The study then explores the Non-Grain Production of Winter Wheat (NGPOWW) by analyzing winter wheat change maps from 2000–2005, 2005–2010, 2005–2010, and 2015–2021 in the HHH Plain. Despite an overall increase in the total planted area of winter wheat, the NGPOWW phenomena has led to concerning winter wheat planting marginalization. Compensatory winter wheat areas are notably situated in mountainous and suburban cultivated lands with low qualities. Consequently, despite the apparent expansion in planted areas, winter wheat production is anticipated to be adversely affected. The findings highlight the necessity for improved cultivated land protection policies monitoring the land quality of the compensation and setting strict quota limits on occupations.