Crop residue and soil water infiltration

<b>Study site and treatment</b>We conducted the experiment in two plastic film houses at the Japan International Research Center for Agricultural Sciences experimental field (24.38°N and 124.19°E) on Ishigaki Island. The climate is subtropical. The soil type was Ultisol (Soil Survey Staff, 2014) and the texture was sandy clay loam. The house was 5 m wide and 18 m long. We made three ridges (0.2 m high and 1 m wide) with a 0.5 m path on each side. We divided these ridges into three plots with 0.8 m paths between each plot. In this way, we created nine plots (1 m × 5.2 m) in each film house and randomly assigned them with nine treatments (3 × 3 factorial design). These treatments comprised three nitrogen levels (0, 10, and 40 kg N ha⁻¹; slow release type urea only, no other fertilizers were used) and three soil moisture levels (unmulched, weed barrier fabric, and black plastic film mulch). The effect of the soil moisture difference treatment was determined at the end of okra cropping by extracting soil core samples from 0 to 5 cm soil depth on the ridge. We replicated the treatments using two film houses (A and B). We cropped corn (<i>Zea mays</i>) without fertilizer before the experiment and collected the residue, then chopped the residue into approximately 3 cm pieces using a chopper and dried it for a month under a roof. We adjusted the soil moisture of the house at a suitable level for tillage by irrigating (25–40 mm) with mist irrigation tubes (Kiriko; Mitsubishi Chemical Agri Dream Co., Ltd., Tokyo) and then removed the tubes. We scattered 2 Mg ha⁻¹ of the corn residue, tilled by a rotary tiller, made the ridges, measured the soil water infiltration, set the irrigation tubes again, set the mulch films, transplanted rose grass (<i>Chloris gayana</i>) seedlings with fertilizer, and irrigated up to the field capacity. Additional irrigation was not provided. After harvesting rose grass, we repeated the above processes in the same way for okra (<i>Abelmoschus esculentus</i>). All the crop residues were collected in each house then evenly returned to the plots and thus the quantity of returning residue was the same in each house. The growing season of corn, rose grass, and okra were June 7 to Aug. 10 2016, Oct. 14 2016 to Jan. 11 2017, and Jan. 12 to Apr. 14 2017, respectively. An interval of 65 days was provided between the corn harvesting and the rose grass planting. There was no interval between rose grass harvesting and okra planting.<br><b>Infiltration rate measurement</b>We measured the soil water infiltration rate with a Mariotte bottle (20 cm high, 10 cm in diameter), with two holes in the bottom. We inserted a plastic ring of the same diameter into the ridge to a 10 cm depth and then watered from a 1 m height to the ring at a 60 mm min⁻¹ rate. We recorded the time needed to waterlog 50% of the soil surface area. We measured infiltration on the ridge at the initial stage (before the rose grass; with incorporated corn residue), after the rose grass (with incorporated rose grass residue), and after the okra (with incorporated okra residue). <br><b>Determination and analysis</b>Aboveground biomass was calculated by multiplying the plot’s whole fresh biomass weight to the average moisture content of the air-dried samples’ in each house. We performed Pearson’s product moment correlation analysis of the infiltration rate for the quantity of applied residue or for the aboveground biomass (dry weight). The correlation coefficients were calculated for the mean values of nitrogen levels and for that of mulch levels respectively. The mean values of nitrogen levels show the effects of aboveground biomass, which averaged out the effect of soil moisture. In contrast, the mean values of mulch levels show the effect of soil moisture.