Soil water infiltration impacted by maize (Zea mays L.) growth on sloping agricultural land of the Loess Plateau

Increasing infiltration rates of sloping agricultural land from arid and semiarid regions not only affects water supply and precipitation transformations in soil directly, but also impacts erosion intensity. This is extremely important to the Loess Plateau regions of northwest China, where a majority of the rainfall occurs within three months of the year, and soils left bare during these months are considered the norm. Incorporating an actively growing crop during this period could protect the soil surface and increase water infiltration. Thus, the objective of this study was to evaluate the combined effects of landscape gradients and maize (Zea mays L.) growth stage on soil water infiltration from agricultural land using runoff plots with slopes ranging from 5.24% to 28.78% conducted under rainfall simulation based on the water balance method. A modified Green-Ampt model was also used under the artificial rainfall conditions to simulate infiltration at different maize growth stages. Results showed that maize plants can intercept precipitation and delay surface water runoff. The capacity of interception increased greatly with maize growth from seedling to the reproductive stage. Steady state infiltration rate improved significantly; cumulative infiltration before runoff and cumulative infiltration from the artificial precipitation were higher under an actively growing maize crop than bare ground. Infiltration rates decreased with increasing slope. Regardless of slope, this study showed that rainfall infiltration rates were enhanced in the presence of actively growing plants compared to bare soil and improved as the growth progressed over the season. The Green-Ampt model has been proposed as a tool to provide improved soil infiltration predictions, which are in theory supposed to be closely related to measured values. A good correlation was observed between the measured and simulated values. However, the predicted values were slightly overestimated using the Green-Ampt equations, which may be attributed to the model not taking into consideration soil compaction from raindrops hitting the soil surface.