Straw incorporation regulates rill erosion processes: Revealing multi-timescale hysteresis between runoff and sediment concentration in brown soil

Abstract

The rill erosion process in sloping farmland after straw incorporation is different from that under traditional tillage, including complex variations and hysteresis in runoff and sediment concentration. However, the hysteresis of original runoff and sediment concentration series is not enough to show the hidden multi-timescale information. Different straw incorporation strategies have varying degrees of change to the hysteresis on runoff and sediment content. The purpose of this study was to systematically identify the hysteresis characteristics and differences of runoff and sediment concentration in multi-timescales on rill erosion, and to reveal the effects of straw length, incorporation depth, and straw amount on hysteresis. In this study, scouring experiments were carried out in runoff plots for brown soil under nine maize straw incorporation treatments, i.e., two gradients of straw length (<3, 3−6 cm), incorporation depth (15, 20 cm), straw amount (4000, 8000 kg ha⁻¹), and traditional tillage without straw (CK). The signal decomposition method was used to extract runoff and sediment concentration components at multi-timescales for rill erosion, and hysteresis loops and cross-correlation were applied to analyze the multi-timescale hysteresis relationship of runoff and sediment concentration. The results showed that increases in straw length and straw amount could reduce runoff and sediment and increase runoff start time. The CK had the hysteresis loop with anticlockwise type, and the runoff and sediment concentration had complex hysteresis phenomenon under straw incorporation. The <3 cm straw length treatments exhibited a tendency for clockwise hysteresis loops. This was attributed to the limited supply of sediment during the later erosion stages. In addition, multiscale cross-correlation showed more detailed hysteresis information than the hysteresis loop of the original series, and better explained the variation of sediment concentration. In the >0.3 high-frequency component, the CK exhibited sediment concentration leading runoff (1−3 min), while the 8000 kg ha⁻¹ straw amount treatments intensified complex fluctuations in runoff and sediment, demonstrating variable leads or lags for the sediment concentration. At the sub-event scale, the 3−6 cm straw length treatments resulted in sediment concentration leading runoff, showing a hysteresis similar to that of the event scale. Clarifying the multi-timescale hysteresis of runoff and sediment helps improve understanding of runoff and sediment dynamics in rill erosion processes under straw incorporation for brown soil.