To evaluate the process of nitrate accumulation and leaching in surface and ground water, we conducted simulated rainfall experiments. The experiments were performed in areas of 5.3 m² with bare slopes of 3° that were treated with two nitrogen fertilizer inputs, high (22.5 g/m² NH₄NO₃) and control (no fertilizer), and subjected to 2 hours of rainfall, with. From the 1st to the 7th experiments, the same content of fertilizer mixed with soil was uniformly applied to the soil surface at 10 minutes before rainfall, and no fertilizer was applied for the 8th through 12th experiments. Initially, the time-series nitrate concentration in the surface flow quickly increased, and then it rapidly decreased and gradually stabilized at a low level during the fertilizer experiments. The nitrogen loss in the surface flow primarily occurred during the first 18.6 minutes of rainfall. For the continuous fertilizer experiments, the mean nitrate concentrations in the groundwater flow remained at less than 10 mg/L before the 5th experiment, and after the 7th experiment, these nitrate concentrations were greater than 10 mg/L throughout the process. The time-series process of the changing concentration in the groundwater flow exhibited the same parabolic trend for each fertilizer experiment. However, the time at which the nitrate concentration began to change lagged behind the start time of groundwater flow by approximately 0.94 hours on average. The experiments were also performed with no fertilizer. In these experiments, the mean nitrate concentration of groundwater initially increased continuously, and then, the process exhibited the same parabolic trend as the results of the fertilization experiments. The nitrate concentration decreased in the subsequent experiments. Eight days after the 12 rainfall experiments, 50.53% of the total nitrate applied remained in the experimental soil. Nitrate residues mainly existed at the surface and in the bottom soil layers, which represents a potentially more dangerous pollution scenario for surface and ground water. The surface and subsurface flow would enter into and contaminate water bodies, thus threatening the water environment.

To increase the available land area, a large-scale land remediation campaign was carried out in the loess hilly and gully area. A large number of high and steep slopes have been produced in the construction of road engineering and water conservancy engineering, and these slopes will cause serious soil erosion under rainfall conditions. Because rainfall runoff is simultaneously affected by slope, bed surface and rainfall, the runoff movement characteristics are complex. It is difficult to consider all influencing factors in the existing models, especially for steep slopes. In this study, artificial rainfall experiments were conducted to study the rainfall-runoff hydraulic processes under different rainfall intensities and slope gradients, and a modified method was proposed to model the key hydraulic parameters (i.e., equilibrium time, water surface line, and runoff processes) on steep slopes. The results showed that (1) For steep slopes (a 70° slope compared to a 5° slope), the runoff generation time, confluence time and equilibrium time of the slope decreased significantly. At the same time, the single width runoff of the steep slope had a power function relationship with the rainfall intensity and gradient. (2) The runoff patterns of steep slopes were different from those on gentle slopes and runoff patterns were more likely to change. The Reynolds number and Froude number for slope flow changed slowly when the slope was less than the critical gradient and increased significantly when the slope exceeded the critical gradient. (3) Based on the analysis of the “double turbulent model theory of thin-layer flow on a high-steep slope”, combined with the dispersed motion wave model, a modified method for calculating the hydrodynamic factors of rainfall runoff was proposed. Then, this method was verified with indoor and outdoor experiments. The research results not only have theoretical significance, but also provide a more accurate calculation method for the design of high and steep slopes involved in land treatment engineering, road engineering and water conservancy engineering.