Climate is the most important environmental factor influencing yield during rice growth and development. To investigate the relationships between climate and yield under different crop rotation patterns and planting methods, three typical rotation patterns (vegetable–rice (V), rape–rice (R), and wheat–rice (W)) and two mechanical planting methods (mechanical transplanting (T1) and mechanical direct seeding (T2)) were established. The results showed that compared to the V rotation pattern, the average daily temperature (ADT) during the sowing to heading stage increased under both R and W rotation patterns, which significantly shortened the growth period. Thus, the effective accumulated temperature (EAT), photosynthetic capacity, effective panicle (EP), and spikelet per panicle (SP) under R and W rotation patterns significantly decreased, leading to reductions in grain yield (GY). VT2 had a higher ratio of productive tillers (RPT), relative chlorophyll content (SPAD), leaf area index (LAI), and net photosynthetic rate (Pn) than those of VT1, which significantly increased panicle dry matter accumulation (DMA), resulting in an increase in GY. Although RT2 and WT2 had a higher RPT than those of RT1 and WT1, the GY of RT1 and WT1 decreased due to the significant reductions in EAT and photosynthetic capacity. Principal component analysis (PCA) showed that the comprehensive score for different rotation patterns followed the order of V > R > T with VT2 ranking first. The structural equation model (SEM) showed that EAT and ADT were the most important climate factors affecting yield, with total effects of 0.520 and −0.446, respectively. In conclusion, mechanical direct seeding under vegetable–rice rotation pattern and mechanical transplanting under rape–rice or wheat–rice rotation pattern were the rice-planting methods that optimized the climate resources in southwest China.

A critical precondition for realizing mechanized transplantation in rice cultivation is the implementation of seedling tray techniques. To augment the efficacy of seeding, a precise evaluation of the quality of rice seedling cultivation in these trays is imperative. This research centers on the analysis of rice seedling tray images, employing deep learning as the foundational technology. The aim is to construct a computational model capable of autonomously evaluating seeding quality within the ambit of intelligent seedling cultivation processes. This study proposes a virtual grid-based image segmentation preprocessing method. It involves dividing the complete image of a rice seedling tray into several grid images. These grid images are then classified and marked using an improved ResNet50 model that integrates the SE attention mechanism with the Adam optimizer. Finally, the objective of detecting missing seeding areas is achieved by reassembling the marked grid images. The experimental results demonstrate that the improved ResNet50 model, integrating the SE attention mechanism and employing an initial learning rate of 0.01 over 50 iterations, attains a test set accuracy of 95.82%. This accuracy surpasses that of the AlexNet, DenseNet, and VGG16 models by respective margins of 4.55%, 2.07%, and 2.62%. This study introduces an innovative model for the automatic assessment of rice seeding quality. This model is capable of rapidly evaluating the seeding quality during the seedling phase; precisely identifying the locations of missing seeds in individual seedling trays; and effectively calculating the missing seed rate for each tray. Such precision in assessment is instrumental for optimizing seedling processes

Ratoon rice is the production of a second rice crop from the stubble left behind after the main crop has been harvested. To identify suitable ratoon rice cultivars for a dual-purpose grain and forage production system, 25 widely cultivated rice cultivars in southwestern China were systematically evaluated based on ratooning ability, forage productivity, ratoon rice yield, and Cd accumulation characteristics. Significant variations were observed among the tested cultivars, indicating substantial differences in their suitability for this production system. Principal component analysis identified three cultivars, CKY637, CKYSM, and FY498, as promising candidates because of their favorable combination of ratooning ability (1.33–1.61), silage rice yield (8.85–11.27 t ha⁻¹), ratoon rice yield (7.83–9.65 t ha⁻¹), and low grain Cd accumulation (0.0430–0.0457 mg kg⁻¹). Among them, CKY637 achieved the highest comprehensive evaluation score, whereas CKYSM exhibited particularly stable performance in maintaining low Cd accumulation, and FY498 was well balanced between yield and quality. These results suggest that cultivar suitability for the dual-purpose system depends on the coordinated performance of multiple traits, particularly strong ratooning ability, high forage and ratoon rice productivity, short total growth duration, and low Cd accumulation in ratoon rice grains. Overall, CKY637, CKYSM, and FY498 represent candidate cultivar resources for the first-season forage + second-season ratoon rice system and provide practical materials for its application in subtropical rice-growing regions.