Soil organic carbon (SOC) and iron (Fe) oxides are known to affect the formation and stability of soil aggregates. However, the effects of SOC and Fe oxides on soil aggregates stability under straw returning and potassium (K) fertilizer application in paddy–upland rotation systems are less well-studied. This study primarily investigated soil aggregates dynamics and their stability indices (mean weight diameter, MWD; geometric mean diameter, GMD), and soil binders (SOC and iron oxides) after rice and rape harvests under four treatments: F1，mineral nitrogen (N) and phosphorus (P) fertilizer; F2, mineral NPK fertilizer; F3, mineral NP fertilizer with straw returning; F4, mineral NPK fertilizer with straw returning in rice–rape cropping system. Straw returning treatments (F3 and F4) significantly (P <0.05) increased MWD and GMD, but the effect of K is not obvious. The soil aggregates stability was higher after the rape harvest than rice harvest, but SOC content was the opposite. Straw input can increase the contents of SOC, alkane-C and aromatic-C concentrations, especially in >0.25 mm aggregates. Long-term straw incorporation significantly increased the amorphous (Feo) and complex iron oxides (Fep) concentrations. SOC and Fep in bulk soil and >5 mm aggregates were significantly related with MWD, and significant relationship was observed between MWD and Feo in <5 mm fractions. Thus, the high levels of SOC, alkane-C, Feo and Fep in soil after straw returning were responsible for the aggregate stability, but the effect of potassium application is not obvious in a rice–rape cropping system.

The Yangtze River basin is distributed across subtropical monsoon climate regions, and has four seasons, including a hot rainy season. These climatic conditions provide favorable conditions for paddy-upland rotation. This paper summarizes the spatiotemporal changes in soil potassium (K) and K cycles in soil-plant systems, as well as environmental impacts on K changes, and provides information for optimal K management. During the past 30 years, soil available K increased by -7.1% to 103.4%. The increase was lower in Hunan, Guizhou, Zhejiang, and Jiangsu provinces (<10%) and higher in Anhui, Jiangxi, Henan, and Chongqing provinces (>30%), demonstrating that soil K pools were enhanced. Farm manure was gradually replaced by synthetic K sources, such as straw and mineral fertilizers, which contributed to an increase in crop yields and soil available K. The meta-analysis results showed that comprehensive K management strategies increased crop yield and soil available K by 11.0% and 44.3%, respectively, on average. Other factors such as balanced fertilization, recycling of straw, increase in atmospheric deposition, decrease in leaching, runoff, and soil K fixation also greatly influenced soil K changes, leading to improvements in crop yields, soil structure, soil fertility, and nutrient availability. Positive K cycles and appropriate K fertilizer use will facilitate proper K management, including cycling of straw, improving machinery and equipment, and estimating the optimal K fertilizer dose after straw. Future studies should focus on tradeoffs between different forms of K under various environmental conditions and accurate estimates of reductions in mineral-K fertilizer requirements following straw return.