, , ngang, , , ,

(1. College of Hydraulic and Civil Engineering, Xinjiang Agricultural University, Urumqi, Xinjiang 830052, China; 2. Xinjiang Academy of Water Resources and Hydropower Sciences, Urumqi, Xinjiang 830049, China)

Abstract：To study the effect of soil water and salt environment factors on the root growth of cotton under different moisture control, three different emergence water volumes (60, 105, and 150 m3/hm2), two different frequencies (high frequency and low frequency) and one double film cover winter irrigation control treatment (CK: 2 250 m3/hm2) were set up to analyze the spatial distribution patterns of soil water and salt environment and root density in dry sown and wet emerged cotton fields under diffe-rent moisture control conditions. The results show that the soil water content and water infiltration range gradually become larger with the increase of seedling water quantity, and the larger the seedling water quantity, the higher the soil water content. With the same seedling water quantity, the soil water content of the high-frequency(HF) treatment becomes obviously larger. The soil conductivity of each treatment tends to decrease gradually with the increase of seedling water and drip frequency, among which the distribution of soil conductivity of S6 treatment is closest to that of CK. With the increase in soil depth, the soil conductivity tends to increase first and then decrease. Compared with the low-frequency(LF) treatment, the high-frequency treatment shows a significantly deeper soil salt accumulation layer. The root length density (RLD) of cotton gradually increases with the amount of seedling water and the frequency of dripping. The soil layer of root distribution gradually deepens with the amount of seedling water in the vertical direction, and the RLD value in the horizontal direction is significantly greater in the mulched area than that in the bare area between films. This research can serve as a solid scientific foundation for the use of dry sowing and wet emergence techniques in cotton fields in southern Xinjiang.

Key words：cotton;double film mulching;dry sowing and wet germination;moisture regulation;water and salt distribution;root distribution

Water scarcity is a major factor limiting the sustainable development of agriculture in many arid and semi-arid regions of the world. Especially in the sou-thern Xinjiang region, where sunshine is strong, dryness and rainfall are scarce, and freshwater resources are severely lacking. With the promotion of the ″three red lines″ system of water resources management[1], the actual irrigated area in Xinjiang is controlled by the state to about 5.3×106hm2, but with the current limit of agricultural water saving on arable land, there will be a serious water shortage of 7.3 billion cubic meters in 2030, and conventional water for winter and spring irrigation cannot be met now. The key to sustainable agricultural development in the southern Xinjiang region is to develop new water-saving irrigation technologies. The ″dry sowing and wet emergence of double film cover″ technology does not require winter and spring irrigation, and the small amount of irrigation water greatly improves the utilization rate of water resources. The high seedling emergence rate ensures the yield of cotton fields and achieves the goal of saving costs and increasing income[2].

WANG et al.[3]found that high-frequency irriga-tion had an average impact on soil salt washing, the water consumption in the field was higher, the range of soil transportation was relatively smaller. In contrast, ZHANG et al.[4]concluded that high-frequency irriga-tion was effective in reducing soil salt content within the wetter as well as increasing cotton yield compared to low-frequency irrigation. A related study found that the growth of cotton roots under sub-membrane drip irrigation was concentrated directly below the drip line, and this localized growth was mainly due to the more abundant soil surface moisture near the drip zone and the significantly lower soil salinity near the drip zone by continuous leaching[5]. At present, research on dry sowing and wet germination double-membrane mulching technology in South Xinjiang is not mature, and the soil water and salt transport in drip-irrigated cotton fields under different water regulations are more complex, and there is little research related to the root system as an essential organ between soil and plant inte-raction. Therefore, the cotton variety ″Yuanmian No.11″ was used as the test material to analyze the different moisture treatments on soil water, heat, and salt environmental factors, as well as root density distribution patterns in dry sowing and wet germination cotton fields under double film mulching conditions, to improve the moisture management system suitable for cotton seedling emergence and growth, and to provide some guidance for local agricultural production.

1 Materials and methods

1.1 Overview of the test site

The experiment was studied in the field from April to October 2022 in a cotton field in the Aksu region of Xinjiang, China, in the town of Hailou, Shaya County, Northern Tarim Basin (82°47′E, 41°13′N). The study area is 986 m above sea level, with an ave-rage annual precipitation of only 47.3 mm, a maximum potential annual evaporation of about 2 000.7 mm, which is 42.3 times more evaporation than precipita-tion, and an average annual temperature of about 10.7 ℃. The soil in the cultivation layer is mainly sandy loam, with groundwater depth of about 3.8 m, soil pH value of about 7.8, N content of 174 mg/kg, and P content of 142.25 mg/kg.

1.2 Experimental design

The cotton variety ″Yuanmian No.11″ was selected as the test crop, and was sowed on April 10, 2022, using a planting pattern of 1 film, 3 tubes, and 6 rows with spacings of 66, 10 and 46 cm, respectively. Double film mulching of each treatment was conducted using machinery, for which the width of the film was 2.05 m, the plant spacing was 10 cm, and the planting density was 252 000 plants/hm2.

The experiment was designed using a two-factor, three-level combination scheme with three seedling water(SW) levels (60, 105 and 150 m3/hm2) and two drip frequencies (DF) (low frequency and high frequency) (Tab. 1), and irrigation was carried out at emergence period (EP) and strong seedling period (SSP), respectively. A total of 7 treatments were set up. The experiment was set up with three low-frequency treatments (S1: 60+225 m3/hm2, S2: 105+225 m3/hm2, and S3: 150+225 m3/hm2) with dif-ferent emergence water volumes, three high-frequency treatments(S4:60+45+225+150 m3/hm2, S5:105+45+225+150 m3/hm2, and S6:150+45+225+150 m3/hm2) with different emergence water volumes, and one local winter irrigation treatment (CK). The winter irrigation treatment was irrigated on 20 November 2021 with an irrigation quota (IQ) of 2 250 m3/hm2. Each treatment was replicated three times, resulting in a total of 21 plots, A protection line of 1 m was established between each plot field, and the area of each plot was 60 m2. The fertilizer level and agronomic measures were based on local experience, and the irrigation water was surface water with a drip rate of 2.1 L/h and a drip spacing of 30 cm.

Tab.1 2022 seedling irrigation trial program

1.3 Measurement items and methods

1.3.1 Soil moisture content

Soil moisture content was monitored using a PR2 moisture measurement device(Precision Technology Co., Ltd., U.K.). The instrument was deployed on April 12, 2022, and the PR2 tubing was placed in the wide and narrow rows of the soil crop and between the membranes at depths of 10, 20, 30, 40, 60, 80 and 100 cm, respectively, to monitor soil moisture daily.

1.3.2 Soil salinity and conductivity

Soil samples were taken using a 5 cm diameter soil auger in wide rows, narrow rows (below the drip head), and between the membranes in the same profile at depths of 0-10 cm, 10-20 cm, 20-30 cm and 30-40 cm, respectively. Soil salinity was characterized using a soil-to-water ratio of 1∶5 for the electrical conductivity of the soil leachate (EC1∶5). When the dried soil samples were ground into powder form, 18 g were added to a triangular flask to configure a soil-to-water mass ratio of 1∶5. The mixture was shaken for 10 min using a shaker and after standing for 15 min, the conductivity of the supernatant was measured using a conductivity meter typed F3.

1.3.3 Soil root system

The spatial layout of the root sampling was desig-ned according to the cotton plant and drip irrigation belt arrangement (Fig.1). Horizontally 60 cm was divided into three positions according to the field layout, 0-20 cm for wide row (W), 20-40 cm for narrow row (N), and 40-60 cm for inter-film (I). Root sampling was carried out in selected plots of 20 cm×60 cm soil area and 40 cm soil depth (h), with each plot measu-ring 10 cm×10 cm×10 cm and a total of 48 soil cubes collected. The soil cubes were collected and placed in individual self-sealing bags and sieved using a nylon film sieve (1 mm). The new roots were then separated from the soil particles, cleaned, air-dried, placed in root trays, and filled with an appropriate amount of water to disperse the roots. The root system was scanned using a flatbed image scanner, and the images were analyzed using WinRHIZO Pro root analysis software to determine the total root length.

Fig.1 Spatial layout of soil root sampling (unit: cm)

1.4 Data processing

The test data was counted and plotted using Microsoft Excel 2020 and Origin 2016 software. Signifi-cant ANOVA and correlation analysis were performed on the test data using SPSS 25.0, and the drip irrigation tape arrangement and soil root distribution were mapped using CAD 2016.

2 Results and analysis

2.1 Soil moisture spatial distribution characteristics

Fig.2 showed the distribution of soil water content (SWC) at 0-40 cm for each moisture treatment. In the figure,srepresents the distance to the midpoint of wide row. Since the wide rows are generally relatively wide, approximately 40 cm, in order to place the wide and narrow rows between the film placed in a plane, the midpoint of the wide row was used as a reference.It can be seen from Fig.2a that the drip frequency was the same, and the soil moisture content gradually increased with the increase of the amount of seedling water, showing S1

As can be seen from Fig.2b, in the vertical direction, soil moisture of low-frequency treatments was mainly concentrated in the 20-30 cm soil layer, while that of high-frequency treatments was mainly concentra-ted in the 30-40 cm soil layer. It can be seen that the soil moisture content of high-frequency treatments was significantly higher than that of low-frequency treat-ments. In the horizontal direction, the soil moisture content in each treatment film was significantly higher than that in the bare land between films. CK treatment compared with each dry sowing and wet extraction treatment, vertical and horizontal distribution of soil moisture content was significantly smaller. As a whole, the water distribution in the 0-40 cm of soil layer treated by S6 was the closest to that treated by CK, and the infiltration range of soil water was larger. During the strong seedling period, the moisture content of the soil under the dry sowing and wet extraction treatments was higher than that under the CK treatment, and the moisture content of 0-40 cm soil under the S6 treat-ment maintained a relatively high level.

2.2 Characteristics of spatial distribution of soil salinity by treatment

It can be seen from Tab.2 that the soilEC1∶5in the film-covered area (wide row and narrow row) was significantly lower than that in the non-film-covered area, showing I>W>N. Compared with CK treatment, the soilEC1∶5in the narrow row area of S6 treatment decreased by 5.9%. TheEC1∶5in 0-40 cm soil layer of each treatment decreased significantly due to the large drip volume during the seedling stage, and the difference gradually decreased. The soilEC1∶5in the film mulching area was lower than that in the non-film mulching area, and there was no significant difference in the soilEC1∶5value between the wide row and the narrow row, which showed W≈N>I.

Tab.2 Average soil conductivity at seedling stage for each treatment

Tab.3 Size of cotton root density at each treat-ment level

Tab.4 Size of cotton root density in vertical direction for each treatment

From the vertical distribution of salt in the 0-40 cm soil layer in Fig.3a and Fig.3b, it can be seen that with the increase of soil depth, the soilEC1∶5of each dry sowing and wet emergence treatment increased first and then decreased. S1, S2 and S3 treatments showed obvious salt accumulation in the 10-20 cm soil layer, while S4, S5 and S6 treatments mainly concentrated in the 30-40 cm soil layer. The drip frequency was the same, and the soilEC1∶5of each treatment gradually decreased with the increase of seedling emergence water, showing S1>S2>S3, S4>S5>S6. Compared with CK treatment, theEC1∶5of 0-20 cm soil layer in high frequency and large water S6 treatment was signifi-cantly lower. From the vertical distribution of salt in the 0-40 cm soil layer during the seedling stage in Fig.3c and Fig.3d, it can be seen that the 0-40 cm soil layer of each treatment remained at a low level due to the large drip volume during the seedling stage, and was significantly lower than CK treatment, and there was less variability between treatments.

Fig.3 Vertical distribution of soil conductivity at the seedling stage by treatment

2.3 Spatial distribution of root length density (RLD) in cotton at the seedling stage

Fig.4 shows the two-dimensional spatial distribu-tion of cotton root length density in each treatment. To simplify the analysis, the root sampling area in the ho-rizontal direction was divided into three dashed lines, which represented wide rows, narrow rows, and bare land between membranes. It can be seen that with the same drip frequency, the distribution range ofRLDgradually increased with the increase of seedling water volume. TheRLDof S1 and S4 was mainly distributed in the 0-20 cm soil layer (Fig.4a, 4d), and the larger the distance from the midpoint of the wide row, the smaller theRLDvalue. TheRLDof S2 and S5 treatments were mainly distributed in the 0-25 cm soil layer (Fig.4b, 4e), and the horizontal distribution range ofRLDincreased significantly. TheRLDof S3 and S6 was mainly distributed in the 0-35 cm soil layer (Fig.4c, 4f), and the horizontalRLDvalue and distribution range were significantly larger than those of other treatments. TheRLDvalue of high frequency treatment was significantly higher and the distribution range was significantly larger than that of low frequency treatment. Compared with CK treatment, theRLDdistribution of S6 treatment was the most similar, and theRLDvalues remained at a large level.

Fig.4 Two-dimensional distribution of RLD in cotton seedlings by treatment

Compared to the CK treatment, theRLDvalues for the low frequency treatment are significantly smaller and are mainly distributed between wide and narrow rows. Among the low frequency treatments, the S1 treatment had the smallestRLDvalue and the S3 treatment had the largestRLDvalue.

From the horizontal distribution ofRLDof cotton in Tab. 3, it can be seen that the difference ofRLDvalues between wide and narrow row positions of each treatment was small and significantly larger than that between films, showing W≈N>I. With the same drip frequency, theRLDof each treatment increased gra-dually with the increase of seedling water content, which showed S1

The high-frequency treatments had significantly largerRLDvalues and better root growth than the lowfrequency treatments, and a high drip frequency within the appropriate irrigation range could promote root growth.The low frequency treatments had smallerRLDvalues and slower root growth.

From the vertical distribution of cottonRLDin Tab. 4, it can be seen that theRLDof each dry sowing and wet emergence treatment increased first and then decreased with the increase of soil depth. Among them, theRLDvalue of S1 and S4 low emergence water treatment reached the maximum in the 10-20 cm soil layer, and S3 and S6 had large.

3 Discussions

Soil moisture content is one of the most important environmental conditions for cotton seedling emergence and growth[6]. In this study, it was found that in the horizontal direction, the soil moisture content in the mulched area (wide and narrow rows) was significantly greater than that in the bare ground between the membranes for each treatment, and the soil moisture content in the narrow rows was greater in the mulched area, showing N>W>I. This is consistent with the study by CHEN et al.[7], where the narrow rows were located closer to the drip tape than the wide rows and inter-membranes, and soil water infiltration into the wide rows and inter-membranes was significantly less than the narrow rows at lower irrigation levels. In the vertical direction, the soil moisture content of each treatment showed a low-high-low distribution with increasing depth of the soil layer, and the range of soil moisture infiltration became significantly larger with increasing seedling water and drip frequency, and the average soil moisture content in the 0-40 cm soil layer also gradually increased. LI et al.[8]showed that with the infiltration of irrigation water, the shape of the wetted body in cotton fields under drip irrigation was approximately semi-elliptical below the drip head, and the soil water content increased, and then decreased with the depth of the soil layer after irrigation. Soil salinity can affect seed germination and slow down the growth of cotton seedlings, and a suitable water and salt environment can promote the growth of cotton. WU et al.[9]found that the horizontal and vertical distribution of salinity showed desalination in both wide and narrow rows, and desalination was more obvious in narrow rows, while desalination changed to salt accumulation between films. The greater the amount of irrigation and frequency of dripping, the better the effect of soil salinity washing, which is consistent with the results of this study.

Many studies have shown that cotton root distribution has an important relationship with soil water and salt distribution, and that cotton roots grow better in soil conditions with high water content and low salinity. Studies have found that the distribution of soil moisture directly affects the root distribution and morphological characteristics of plants. When the soil moisture content of the shallow layer is low, the root system will extend to the lower layer of the larger soil moisture content to promote root development and improve the efficiency of deep soil water use, and horizontally to the mulched area because of the higher soil moisture in the mulched area. In the present study, the depth of soil water infiltration was different for the different moisture treatments, with a significantly larger range of soil water infiltration for the higher emergence volume and drip frequency treatments. The mulched area, however, has a significantly higherRLDvalue than the bare ground area between the membranes due to the higher water retention of the soil. One of the purposes of water irrigation is to wash away excess salts from the root zone, and too little irrigation may create a saline zone below the drip line that cannot be washed out of the root zone. KURTH et al.[11]found that salinity stress thinned the cotton root system and reduced root cell size, affecting root growth and increased salinity leading to lower averageRLDvalues in the root system. This also confirms the ″water and salt avoidance″ growth characteristics of the cotton root system.

4 Conclusion

1) The soil water content gradually increased with the increase of water volume of seedlings, and the soil water content of high frequency treatment was significantly larger than that of low frequency treatment. The soil water content of all treatments in the horizontal direction was as follows: mulching area> bare land between the films.

2) The soil conductivity of each treatment tended to decrease gradually as the amount of water emerged and the frequency of dripping increased, and the high frequency treatment showed a significantly deeper soil layer for salt accumulation compared to the low frequency treatment. In the horizontal direction, the conductivity of the soil in each treatment showed a trend as follows I>W>N. In the vertical direction, the conductivity of the soil increased and then decreased as the depth of the soil layer increased.

3) TheRLDvalue increased gradually with the amount of water and the frequency of dripping. TheRLDvalue in the horizontal direction was significantly greater in the mulched area than that in the bare area between the membranes, and theRLDvalue was signifi-cantly greater in the wide row area.