Site description. In November 2020, the experiments were conducted in the Babaiqiao, Nanjing Agricultural University, Jiangsu Province, China (118°55′E,32°25′N). The tillage test was carried out in the field after the rice crop harvesting in autumn. The soils in the field were clay loam with yellow-brown colour, under rice-wheat rotations. Before the start of experiments, soil physical properties (cone index, moisture content, bulk density) and residue parameters (length, height, wet density) were measured and the results presented in Table 1.
Table 1. Soil properties and straw parameters of experimental site.
Parameter
|
|
Value
|
Soil
|
Texture
|
Clay loam ( 21.20, 39.67 and 38.96% sand, silt and clay, respectively )
|
Cone index
|
635, 1000, 987 kPa at 5, 10, and 15 cm depths, respectively
|
Moisture content
|
22.6, 23.4, 24.8% at depth of 0-5, 5-10 and 10-15 cm, respectively
|
Dry bulk density
|
1.29 g cm-3
|
Residue
|
Straw length
|
0-15 cm
|
Stubble height
|
0-20 cm
|
Wet density
|
8012 kg ha-1
|
Dry density
|
3943 kg ha-1
|
Description of the test bench and tillage tool. An electric and multi-functional field testing bench with precise parameter control was developed for this study. Its main features include a movable carriage, rotary tiller, traction motor, lifting motor, electric generator, power distribution box, and control system. The rectangular steel tubes of various sizes were welded to construct an 8000 mm long and 2000 mm wide test bench (Fig. 1). The movable carriage is transported on twin lead rails with an adjustable speed at 0.05-1 m s-1. A traction motor and four lifting motors drive the carriage to move forward and backward, and up and down, respectively. The test bench was powered by a 13.5kW electric generator, and there was a complex control system to complete power transmission and operation control.
A 225-mm-rotary radius C-type blade(IT225)was selected for its widespread application in the annual rice-wheat rotating fields managed in East China regions. The blades are fixed on the rotary tiller, and move with the movable carriage (Fig. 2). The moveable carriage in the test bench was equipped with a 6.3 kW drive motor for driving the rotary blades to rotate, and the rotary speed of the rotary tiller was adjustable from 0 to 600 rpm. The tillage depth, rotation rate, and forward speed are easy to adjust through a wireless control handle.
Residue preparation. Before the field test, straw length, stubble height, and the amount of residue after rice harvest were observed and measured. The straw lengths were found to range from 30-150mm, and the stubble heights ranged from 50 to 200 mm. The amount of total residue was 8012 kg ha-1. In this study, we want to determine an appropriate straw length and stubble height for residue incorporating. Therefore, the straw lengths of 30, 50, 75, 100, 125, and 150 mm, and stubble height of 50, 100, 150, and 200 mm were selected for the experiments. The stem of rice residues was collected from field and chopped into specific lengths with a chip cutter, and then laid evenly under the test plots after being dyed red (for better observation) with spray paint (Fig. 3a). There are two ways of residue preparation; one is to lay the straw on the soil face after stubble removal (Fig. 3b), and the other is to cut the stubble to the required height with a pair of scissors and then lay the straw on the soil face (Fig. 3c). The amount of residue laid in the two ways was identical, which was 8012 kg ha-1, to simulate the actual field state after harvesting.
Experimental design. In this study, three experiments were carried out to investigate the effect of residue parameters (straw chopping length and stubble height) and rotary speed on residue burying and residue spatial dispersion. The purpose of experiment 1 was to study the effect of straw chopping length on residue incorporation by rotary tillage. The data of residue burying and residue spatial dispersion was acquired using rotary tillage with the six straw lengths. A rotary tillage operation with the rotary speed of 280 rpm and forward speed of 0.5 m s-1 was implemented in the experiment.
The purpose of experiment 2 was to explore the effect of stubble height on residue burying and residue spatial dispersion using the four straw mixtures. Residue mixture 1 (M1) consisted of 50 mm high stubble and 50 mm long chopping straw. Residue mixture 2 (M2) consisted of 100 mm high stubble and 50 mm long chopping straw. Residue mixture 3 (M3) included 150 mm high stubble and 50 mm long chopping straw and residue mixture 4 (M4) comprised 200 mm high stubble and 50 mm long chopping straw. Each mixture had 8012 kg ha-1 of residue cover, and the operation parameters were the same as in experiment 1.
The purpose of experiment 3 was to investigate the effect of rotary speed on residue burying and residue spatial dispersion. The straw length of 50 mm, forward speed of 0.5 m s-1, and three rotary speeds of 240, 280, and 320 rpm were selected in this experiment. The tillage depth was 100 mm in all three tests, and each test was repeated three times. There were 39 field plots in three experiments, and each 2 m long and 0.5 m wide.
Measurements. Residue burying. The burying of residue is one of the important indexes to evaluate the quality of residue incorporation. The higher burying rate of residue implies a better quality of residue incorporation. The burying rate of residue was calculated using Eqs. (1) proposed Fang et al.19
(1)
Where mq (kg) is the total weight of residue before tillage and mh (kg) is the total weight of residue after tillage.
Residue dispersion. A) Sample collection and measurement. In the process, samples of soil-residue mixture were collected from the surface after rotary tillage, and the residue spatial coordinates were measured. Considering the average depth of the soil layer after tillage was about 150 mm, we made many steel sampling frames with dimensions of 300×300×150 mm. For collecting our sample, a sampling frame was placed in the middle of the tilled area and the knocked it utterly into the soil layer with a steel hammer. Finally, the sample of soil-residue mixture was taken out after a steel tray was embedded to the root of the sampling frame (Fig. 4a). After all samples were collected in this way, they were taken back to the laboratory for further measurement and analysis.
A residue spatial coordinate digitizer was developed to measure the residue spatial position and obtain the data of residue dispersion uniformity. It is mainly composed of four parts: an arc scale, a horizontal scale, a vertical scale and a pillar (Fig. 4b). The horizontal scale and vertical scale are fixed on the pillar, and above them is the arc scale. In the horizontal direction, the arc rotation arc and horizontal distance can be measured by rotating the horizontal scale along the pillar and moves along the slider. In the vertical direction, we also can obtain the vertical distance according to the slider position in the vertical scale. Because there was less residue bending after tillage since experimental materials were the main stalk of paddy residue with high moisture content and short length, so it was reasonable to use the coordinates of the residue head and tail instead of the position of the whole residue. Therefore, the digitizer is reliable to ascertain the location of the residue by measuring the coordinates of residue.
B) Analysis of residue dispersion. Firstly, the residue absolute coordinates were saved in the *. IBL format (a coordinate point file format) and inputted into the 3D software Pro/Engineer 5.0 (PTC, America) to create the 3D model of residue dispersion automatically (Fig. 5a). Secondly, a multi-scale segmentation of the 3D model of residue spatial dispersion was conducted to analyze the uniformity of residue dispersion. The model was not only divided with a 50 mm length scale in the depth direction (Fig. 5b), but also segmented with a 50× 50×50 mm cube scale in the overall direction (Fig. 5c). Finally, the total length of residue in each segment area was calculated by the 3D software.
The uniformity of residue dispersion could be accurately analyzed by the coefficient of variation of residue total length (CV) in each segment area, and the smaller the CV, the more uniform the residue dispersion, which implies the better quality of residue dispersion. The calculation of CV was shown in Eqs. (2):
(2)
Where SD is the standard deviation of residue total length and AV is the average value of residue total length.
The 3D model of residue spatial dispersion was segmented into three layers on average in the depth direction, namely the upper layer (UL), the middle layer (ML), and the lower layer (LL). Total length of all residues in each layer under different treatments was calculated, and then the proportion of residue in each layer was analyzed to evaluate the uniformity of residue spatial dispersion in the depth direction. A higher proportion of straw in ML and LL implies better residue dispersion in the depth direction. The 3D model of residue spatial dispersion was further divided into 108 small cubes in the overall direction, where the size of each cube was 50× 50×50 mm. The CV in each cube was analyzed to evaluate the uniformity of residue dispersion. It could easily be judged that the better quality of residue dispersion in the overall direction was that the smaller CV in each cube.
Data analysis. The data were subjected to statistical analysis by one way factorial analysis of variance (ANOVA) using IBM-SPSS Statistics 22 software (IBM Corp., Armonk, N.Y., USA). When the F-test indicated statistical significance at the P = 0.05 probability level, treatment means were separated by the least significant difference (LSD0.05) test.