Li, Yuan , Zhu, Yong-He , Lu, He-Peng , Wang, Neng , Cheng, Fu-Ying , Du, Yan-Lei , Li, Feng-Min
2026-06-01 FIELD CROPS RESEARCH 2026 343(卷), null(期), (null页)
Background: Global wheat production is experiencing a slowdown in genetic gains, posing a threat to long-term food security. In contrast, a dryland wheat system in China has shown an acceleration in yield growth over the past two decades, reaching a rate 2.44 times the global average. This surge coincided with a pivotal regional management shift, where mineral fertilizer use plateaued while organic amendments increased fourfold, driving a sustained rise in topsoil organic carbon. The unique co-evolution of cultivars and soil management in this system offers a compelling natural experiment for investigating genotype-environment-management interactions at a systems level. Objective: This study aimed to quantify the genetic gain rate in this system and elucidate the physiological mechanisms underpinned by dynamic cultivar-soil interactions in the dryland environments. Methods: We evaluated 18 historical wheat cultivars (2004-2022) across a soil-fertility gradient in a two-year experiment. A novel sequential retrospective-window analysis was used to detect changes in the genetic gain rate and its components. Results: It was revealed accelerating genetic gains for yield, a pattern that became increasingly pronounced as the retrospective analysis window was narrowed to more recent years. This acceleration stemmed from a dynamic cultivar-soil interaction: under regional baseline fertility, genetic gains were primarily driven by increases in grains per spike, whereas under improved fertility, gains decisively shifted to being driven by spike number per unit area. While soil enhancement consistently boosted biomass accumulation, it concurrently constrained the harvest index (HI) in modern cultivars. This key physiological trade-off, which intensified with higher fertility levels, fundamentally repositions HI as a critical leverage point for future yield improvement under high-fertility conditions. Conclusion: This work establishes that the trajectory of genetic improvement is critically co-determined by the soil environment. We demonstrate a fertility-dependent shift in the hierarchy of yield components and identify the suppression of HI as a constraint under high fertility. We advance cultivar-soil co-design as an essential paradigm that strategically utilizes these dynamic interactions to synchronize genetic potential with managed soil health, thereby breaking the cycle of diminishing returns from conventional breeding approaches. Significance: The findings suggest a promising paradigm: from the Green Revolution's model of predominantly maximizing harvest index toward cultivar-soil co-design. This reframes soil as an active partner in genetic improvement and provides a systemic strategy to overcome genetic gain stagnation in dryland agriculture.