A field experiment was conducted to explore how long-term integrated organic or inorganic fertilization influences soil bacterial communities in dryland farming systems. High-throughput amplicon sequencing was applied to profile bacterial composition and diversity under six treatments—no fertilizer (T0), chemical fertilizer alone (T1), and chemical fertilizer combined with cattle manure (T2), sheep manure (T3), bio-organic fertilizer (T4), or potassium fulvate (T5)—and to pinpoint the major environmental determinants. The results showed that combined fertilization markedly elevated soil nutrient levels, with cattle manure (T2) and sheep manure (T3) exhibiting the strongest effects. The bacterial community was dominated by Actinomycetota, Pseudomonadota, and Chloroflexota. Relative abundances of Pseudomonadota and Bacteroidota increased under combined fertilization, whereas Actinomycetota declined—particularly in T2 and T3. At the genus level, Nocardioides, KD4-96, and Subgroup 6 were predominant. Moreover, T2 and T3 significantly raised the Simpson index relative to T1. Redundancy analysis revealed that soil organic matter (p = 0.002) and pH (p = 0.03) were strongly associated with variation in bacterial community composition. Collectively, long-term combined organic-inorganic fertilization with different organic amendments effectively enhances soil fertility and bacterial community structure, offering a promising approach for nutrient management in the dryland farming regions of the Ningxia Hui Autonomous Region, northwestern China.

Identifying soil nutrient limiting factors and fertilization effects in the irrigated silty soil region of Ningxia is key to improving wheat (Triticum aestivum L.) quality and yield. A field experiment was conducted with five treatments: conventional fertilization (TF), recommended fertilization (RF), nitrogen deficiency (RF-N), phosphorus deficiency (RF-P), and potassium deficiency (RF-K). The results showed that under RF, soil nutrients remained at relatively high levels, with no significant differences compared with TF. In contrast, RF-N significantly reduced soil mineral nitrogen, total nitrogen, and organic matter compared with TF, and inhibited plant growth, photosynthesis, and plant accumulation of nitrogen, phosphorus, and potassium. Wheat yields under RF and RF-K showed no significant differences from those under TF, whereas RF-N and RF-P significantly reduced yields by 42.68% and 22.69%, respectively, relative to RF, mainly due to decreases in spike length and grain number per spike. The increase in yield was associated with synergistic increases in grain number per spike, spike number per hectare, and spike length. Yield components were significantly positively correlated with soil organic matter, total phosphorus, and mineral nitrogen, with soil total phosphorus identified as the environmental factor most strongly associated with wheat yield. Grain protein content was significantly positively correlated with soil mineral nitrogen, while starch content was significantly negatively correlated, indicating that mineral nitrogen is a key factor regulating grain quality. In summary, nitrogen fertilizer is the primary limiting factor in this region. Applying nitrogen, phosphorus, and potassium together synergistically enhances wheat yield by increasing soil total phosphorus levels and improves grain quality by regulating soil mineral nitrogen. Thus, this combined fertilization strategy provides a foundation for precise nutrient management and the simultaneous improvement of both yield and quality.