Cropping patterns describe how crops are arranged and sequenced through time, and they directly shape agricultural inputs, yields, and the associated regional climate and ecological processes. Optimizing the spatial distribution of cropping patterns, integrating agricultural intensification and sustainable development, is now a broadly endorsed objective. The Jianghan Plain is an important agricultural production base in central China in Hubei Province, characterized by intensive cultivation and diverse cropping systems. In recent years, non-grain and non-farming conversion has become increasingly prominent, accompanied by growing pressure on resources and the environment. Timely and accurate mapping of cropping patterns and their spatial optimization from food security and ecological security perspectives are essential for guiding sustainable agricultural restructuring on the Jianghan Plain. However, existing cropping pattern optimization studies have primarily focused on administrative-scale structural adjustments, leaving spatially explicit, plot-level approaches largely underexplored. This study integrated time-series remote sensing data and phenological knowledge to map the major cropping patterns on the Jianghan Plain, constructed a sustainable intensification evaluation framework from food security and ecological security perspectives, incorporating six indicators including output value per unit area, net primary productivity, crop water requirement, non-point source pollution load, and greenhouse gas emission intensity, to quantitatively evaluate the production benefits and ecological costs of cropping patterns from food security and ecological security perspectives. On this basis, through incorporating multi-dimensional indicators such as resources, environment, and socio-economic factors, a Bayesian Network for cropping patterns optimization was built to explore the cropping patterns optimization results under three scenarios: production, conservation, and trade-off. The results indicated that: (1) There were significant differences in the production benefits and ecological costs of cropping patterns. There were significant differences in the production benefits and ecological costs among cropping patterns. In terms of production benefits, double-cropping paddy rice achieved the highest score (0.91), followed by winter wheat-paddy rice (0.87) and winter rapeseed-paddy rice (0.84). Regarding ecological costs, winter wheat-soybean and winter rapeseed-soybean exhibited the lowest ecological pressure (both -0.37), while double-cropping paddy rice incurred the highest ecological cost. (2) The trade-off scenario, which balances food security with ecological conservation, was identified as the optimal solution. Winter wheat-paddy rice, winter rapeseed-paddy rice, winter wheat-soybean, and double cropping paddy rice, which achieved relatively high comprehensive expected effects, were identified as viable alternatives for cropping pattern optimization on the Jianghan Plain, with their combined proportion adjusted from the current 11.9% to 32.6%. Spatially, double-cropping paddy rice is best suited to the southeastern Jianghan Plain, winter wheat-soybean is recommended for the central region, winter wheat-paddy rice is optimal for the southwestern area and along the Han River, and winter rapeseed-paddy rice is most appropriate for higher-altitude areas distant from rivers. This optimization ensures agricultural output while maximizing regional ecological positive benefits, achieving sustainable intensive agriculture. The proposed Bayesian Network framework enables plot-level spatial optimization of cropping patterns under multivariate and multi-scenario conditions, offering a technically feasible approach to optimizing cropping patterns from the dual perspectives of food security and ecological security. This approach enhances cropland use efficiency and provides actionable insights to optimize cropping patterns in comparable regions, while charting a viable pathway toward green, sustainable agricultural development and national food security.

A spatial distribution map of crops is often required in a wide range of agricultural and environmental applications, including the land use patterns, cropping intensity, and grain yields in sustainable agriculture. Current crop mapping techniques can rely primarily on machine learning algorithms due to the training data and portability. Deep learning can also require large amounts of training data, with the data dependency. However, the reliable, accurate, sufficiently large, and labeled datasets have consistently been limited for the data-driven tasks in the large-scale or heterogeneous agricultural regions. The high-quality training data can be compounded by cloud contamination in the optical remote sensing, inconsistent field surveys, and phenological variability over the landscapes, leading to the persistent challenge of scalable crop mapping. Crops can often follow a relatively stable and biologically driven growth rhythm under environmental and agronomic practices. The growth cycle can be divided into characteristic phenological stages (e.g., emergence, vegetative growth, flowering, and maturity). The time-series remote sensing data can be expected to systematically capture and quantify the vegetation indices, such as the NDVI or EVI. These temporal dynamics can also reduce the dependence on the large training data. Expert phenological knowledge can be encoded into computational models. In this study, a cropping pattern mapping was introduced to incorporate the crop phenology knowledge into the classification framework using Bayesian Networks. Key phenological features were extracted at the critical growth stages, according to a small number of the representative training samples. Knowledge probabilistic encoding was also performed to define the conditional dependencies. A Bayesian Network was constructed to tailor to the phenology-driven crop type classification. Empirical experiments were conducted to validate the effectiveness of the model in a region with highly complex cropping patterns. The results demonstrate that: 1) The model parameters were established either without training data or with only a limited number of samples, particularly with the phenological knowledge as a guiding framework. Thus, the accuracy was maintained under conditions of sample scarcity, with an overall mapping precision of over 92%. The prior knowledge effectively served as a data surrogate in knowledge remote sensing; 2) The Bayesian Network classification framework exhibited a 'weak learning - strong inference' performance, whereby the overfitting was avoided for the limited samples rather than the domain-specific knowledge structures for the inference. The highly precise fitting was observed in the data-driven models during the inference phase. The dependency of the machine learning on data was reduced by 42%. As a result, the data dependency and overfitting were avoided to enhance the interpretability, transparency, and portability of the classification in the machine learning models. The strong robustness was also achieved to generalize under the different temporal and spatial contexts, especially with the sparse or costly training data. The domain knowledge was then integrated with the probabilistic graphical modeling. A promising pathway can represent crop mapping in data-constrained environments. The finding can offer a practical alternative to fully data-driven approaches in the classification tasks during remote sensing. A geospatial data layer can also support decision-making in agricultural planning and food security evaluation.