Sequential natural rainfall erodes the efficacy of tillage practices on nitrogen loss: Insights from hydrological...

Agricultural nitrogen (N) non-point source pollution is one of the major threats to water environmental safety. Although previous studies have shown that appropriate tillage practices can effectively reduce runoff and associated N loss, it remains unclear how N loss from sloping farmland responds to changes in microtopography and hydrological connectivity under different tillage practices. In this study, natural rainfall observations were conducted from August 2024 to August 2025 under three tillage treatments-flat tillage (FT), artificial digging (AD), and contour ridge tillage (RT)-in the red soil region of southern China. By combining high-resolution UAV-SfM topographic reconstruction with the Index of Connectivity (IC), we examined the dynamic evolution of hydrological connectivity under different tillage practices and its regulatory role in N loss. The results showed that conservation tillage treatments (AD and RT) initially disrupted hydrological flow pathways and reduced N export by >50 %; however, this mitigation effect was short-lived. Continued rainfall and the progressive attenuation of microtopography led to different IC evolution trajectories, which in turn increased N loss. Under RT, abrupt ridge failure during extreme rainfall events triggered a sharp increase in IC, causing runoff and N losses to temporarily exceed those under conventional FT. In addition, heavy rain and rainstorm events were the main drivers of system losses, contributing 36.61 %∼54.75 % of total runoff and 26.40 %∼51.34 % of total nitrogen export, respectively. Partial least squares structural equation modeling (PLS-SEM) further confirmed that IC acted as the primary mediating variable regulating N export through changes in runoff and transport intensity. These findings suggest that maintaining ridge structure after extreme rainfall events is critical to preventing abrupt increases in nitrogen loss caused by sudden shifts in structural connectivity. This study provides a theoretical basis for water and nitrogen loss control and tillage optimization on sloping farmland.