Abstract: To clarify the effects of drought and waterlogging stress on root of cotton at flowering and boll-setting stages, an experiment with drought, waterlogging and abrupt droughtflood treatments was carried out in the Irrigation and Drainage Experiment Station of Wuhan University in 2016. Minirhizotron systems were used to conduct in situ and long-time monitoring of root systems and digging method was used to validate the obtained data. The images obtained by minirhizotron were equivalent to a 3.3 mm depth of field (DOF) around the tube. From the squaring stage to boll opening stage, root growth resembled an S-shaped curve, with the fastest growth in the flowering stage. The cotton roots were mainly distributed in 0-30 cm soil layer, accounting for 70%-80% of the total amount of the roots. The total root length and total root surface area under different treatments were in the order of waterlogging > waterlogging after drought > control > drought. Waterlogging led to rapid growth of roots at the upper soil layer, especially for those with diameter of 0.2-0.4 mm. Drought decreased root length density, increased root diameter and the proportion of roots in deeper layers, which caused the barycenter of the roots to move downwards. The root growth rate was larger in the drought treatment followed by waterlogging treatment compared to the control group, implying that waterlogging has a compensatory effect on root growth following previous drought stress.

Key words: saline soil, silvopastoral system, edge effect, moving split-window technique, Yellow River Delta