Nutrient Accumulation and Distribution in Cotton Promoted by Removal of Mulch Film

Zhang Zhan-qin , Zhang Li , Tian Hai-yan Niu Yuan, and Yang Xiang-kun * Xinjiang Academy of Agricultural and Reclamation Science, Shihezi 8000, Xinjiang, China

2 Key Lab of Xinjiang Production and Construction Corps for Cereal Quality Research and Genetic Improvement, Shihezi 832000, Xinjiang, China

3 College of Agriculture, Shihezi University, Shihezi 832000, Xinjiang, China

Abstract: Plastic film mulching affects changes in nutrient contents in soil and absorption and utilization of nutrients in plants were by changing hydrothermal condition of soil. The temporal and spatial variation of the total soil salt and nutrient contents with mulch film removed at three different times during the early cotton growth stage and its effects on nutrient absorption and accumulation in cotton plants were studied over 2015-2017. The film removal treatments reduced salt accumulation in normal rainfall year (2017). Film removal increased contents of soil organic matter, the total phosphorus and available potassium at the end of growth stage, increased contents of soil hydrolyzable nitrogen and the total nitrogen in the surface soil layer (0-10 cm), and increased the total nitrogen contents in the deep soil layer (40-50 cm). Film removal increased accumulation of nitrogen and phosphorus nutrients in cotton plants in 2017 and accumulation of nitrogen, phosphorus and potassium nutrients in cotton plants in heavy rainfall year (2016). These experimental results indicated that removal of mulch film at an appropriate and targeted time in the bud stage of cotton promoted nutrient absorption.

Key words: removal of mulch film, total soil salt, soil nutrient, nutrient accumulation, nutrient distribution

Introduction

Film mulching improves soil physicochemical properties, crop growth and yield by changing soil temperature, moisture, nutrients and microbial activity. Film mulching produces comprehensive effects on soil environment and crop growth (Jianget al., 2018; Filipovi?et al., 2016). The effects of mulch film on soil fertility and crop nutrient absorption vary from climate and farming practice. Film mulching can improve soil health by accelerating nutrient cycling and increasing carbon pool (Jinet al., 2018). In areas with heavy rainfall, plastic film mulching reduces surface runoff and the losses of the total nitrogen and phosphorus (Liuet al., 2012). However, the effects of mulching on soil nutrients in arid and semi-arid areas are different. Film mulching can promote the transformation of soil nutrients and increase the levels of soil organic matter (Cao, 2015). However, while film mulching can increase crop yields in arid and semiarid areas, it can also reduce soil fertility (Zhouet al., 2012). Wanget al. (1990) noted that film mulching can lead to nutrient deficiency during later growth stages and premature aging. In crop production, film mulching can increase the numbers and activities of microorganisms, accelerate the decomposition of organic matter and facilitate the mineralization of organic nitrogen and the release of phosphorus. However, long-term mulching and unsustainable farming methods will lead to depletion of soil organic matter (Jianget al., 2018).

Most studies (Geet al., 2016; Wanget al., 2004; Kumar andDey, 2011; Feng, 2013) concluded that mulching can promote crop accumulation of nutrients, promote the transport of nutrients from vegetative organs to reproductive organs and improve the efficiency of fertilizer utilization. However, some studies found that plastic film mulching has no significant effects on nitrogen, phosphorus and potassium contents in plants (Olaveet al., 2017).

Previous studies have concentrated on the effects of film mulching on soil fertility and crop nutrient absorption by crops (Zhang, 2012; Liet al., 2009; Liet al., 2007; Liet al., 2004; Zhou, 2009). Few studies have been conducted on the changes in soil fertility and nutrient absorption in crops after removing mulch film during the growth stages. Some studies have mentioned that the potential benefits of mulching in semiarid agricultural systems are great, but full realization of this potential depends on duration of film mulching during growing seasons (Liet al., 2004). Therefore, it is necessary to understand the changes in soil fertility and the pattern of nutrient absorption in crops, such as cotton after film mulching is removed at different time. This information will help to provide specific irrigation and fertilization management measures and enable high-yield and high-efficiency cotton cultivation after film mulching removal in arid and semiarid cotton producing areas.

Materials and Methods

Experimental design

The study was conducted in cotton-growing seasons of 2015-2017 (May-September) at No. 2 Experimental Site (44.3108°N, 85.986°E, 460 m) of Xinjiang Academy of Agricultural Reclamation Science, Shihezi, Xinjiang. This location is in an arid climate zone. During May-September of 2015-2017, the accumulated temperatures of ≥10℃ were 3 014℃, 3 165℃ and 3 341.2℃, respectively, with the average temperature of 22.9℃, 22.6℃ and 22.5℃, respectively, and the rainfall of 94, 120.2 and 96.5 mm, respectively. In 2016, the rainfall in April was 53.8 mm, 27.1 mm higher than that of the historical average (Weather data obtained from Shihezi Meteorological Service).

The planting mode was based on the report by Yanget al(2017). The sowing dates of 2015-2017 were April 28, May 6 and April 22, respectively. The harvest dates were September 10, September 26 and September 6, respectively, and the growth periods were 127, 134 and 131 days, respectively. During the growth periods, drip irrigation was used. In 2015-2017, there were eight, seven and eight irrigations, respectively, and the irrigation volumes were 382.5, 322.5 and 423 mm. The amounts of urea (46% nitrogen content) applied in each year were 844.5, 603 and 600 kg ? hm-2, respectively, and the amount of fertilizer (N : P2O5: K2O=6% : 30% : 30%) applied in each year were 555.75, 305.85 and 345 kg ? hm-2, respectively.

A variety ofGosspium hirsutumL. (Xinluzao 42) was used as the test material. A randomized block test design was used, with four treatments and three replications. The area of each block in this study was 42 m2(2.1 m wide and 20 m long). The treatments were: mulch film removed 10 days before first irrigation (T10); the mulch films were removed on the 19th day (May 25, 2015), on the 24th day (June 9, 2016) and the 33rd day (May 31, 2017) in 2015-2017 after seedling emergence, respectively. Mulch film removed 1 day before the first irrigation (T1), the mulch films were removed on the 29th day (June 4, 2015), on the 34th day (June 19, 2016) and the 42nd day (June 10, 2017) in 2015-2017 after seedling emergence, respectively. Mulch film removed 1 day before the second irrigation (E1), and the mulch films were removed on the 39th day (June 14, 2015), on the 44th day (June 29, 2016) and the 52nd day (June 20, 2017) in 2015-2017 after seedling emergence, respectively. Mulch film used during the whole growth stages was control treatment (CK) in 2015-2017. Seeding and film mulching were accomplished simultaneously by cotton planters commonly used in local area, but the mulch film was removed manually.

Sample collection and determination

In 2015-2017, soil samples were collected from cotton fields after planting and before harvesting. After planting, using a quincunx pattern sampling method, five points were selected in the field. The sampling depth was 50 cm and one sample layer was 10 cm. The soil samples from five sampling points at the same depth were mixed evenly to provide one soil sample. After sowing, a total of five soil samples was collected, and the soil samples were air-dried under shade. Before harvesting, one layer of sample was collected every 10 cm in vertical direction at each point in the middle of each wide row and narrow row and the sampling depth was 0-50 cm (Fig. 1). For each treatment, the soil samples from two sampling points at the same depth layer in the middle of wide row and narrow row were mixed evenly as one soil sample. There were five soil samples for every treatment. The soil samples were air-dried under shade. The total soil salt and nutrient contents were determined for five soil samples after planting and 12 soil samples from the three replications of four treatments before harvesting. The items to be measured were as the followings: the total salt, the organic matter, the total nitrogen, the total phosphorus, the total potassium, the hydrolyzable the nitrogen, the available phosphorus and the available potassium.

Fig. 1 Soil sample collection methods used in this study

In 2015-2017, starting from the 35th day (June 10, 2015), the 21st day (June 6, 2016) and the 33rd day (May 31, 2017) after seedling emergence, plant samples were collected every 14 days and dried to a constant weight using the method of Yanget al(2017). The plant samples were ground and mixed, and the contents of the total nitrogen, the total phosphorus and the total potassium of the whole plant were determined. At last sampling, for mature samples, contents of the total nitrogen, the total phosphorus and the total potassium of different organs (roots, stems, leaves (including petiole) and bolls) were determined for the three repeats of each treatment.

The measurement methods used of the soil samples were as the followings: the total salt (LY/T 1251-1999; here and below, the letter/number combinations referred the to the national standards and relevant industry standards of China), organic matter (LY/T 1237-1999), the total nitrogen (LY/T 1228-2015), the total phosphorus (LY/T 1232-2015), the total potassium (NY/T 87-1988), the hydrolyzable nitrogen (LY/T 1228-2015), the available phosphorus (LY/T 1232-2015) and the available potassium (NY/T 889-2004).

The measurement methods used of plant samples were as the followings: the total nitrogen (GB/T 6432-1994), the total phosphorus (GB/T 6437-2002) and the total potassium (GB/T 13885-2003). Soil and plant samples were measured by Food Quality Supervision, Inspection and Testing Center (Shihezi) of China's Ministry of Agriculture using the national standards and relevant industry standards of China.

Data processing

Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA, USA) was used for data entry and processing and calculation of means and standard deviations. Graphs were plotted using SigmaPlot 12.5 (Systat Software Inc, San Jose, CA, USA). Images were post-processed using Adobe Illustrator CS5 (Adobe Systems Incorporated, San Jose, CA, USA). Statistical analyses were performed using SPSS 23.0 (International Business Machines Corp, Armonk, NY, USA). For multivariate analysis of variance (MANOVA), a single-factor general linear model (GLM) was used to analyze the effects of different treatments on observed variables. The multiple post hoc comparisons were performed using the least significant difference (LSD) test. All the tests were considered statistically significant atP＜0.05 and highly significant atP＜0.01. The simulation of nutrient accumulation equation was performed using the DPS 16.05 software (Tang and Zhang, 2012) and Marquardt method.

Results

Changes in the total salt contents in different soil layers with mulch film removed

The total salt contents of 0-50 cm soil layers after planting in 2015-2017 are shown in Table 1. In the year with heavy rainfall (2016) and the drought year (2015), the film mulching treatment, during the period before harvest, reduced the total salt contents of each soil layer (Fig. 2). The total soil salt contents increased with removal of mulch film and T10 had the highest content. The earlier the mulch film was removed, the greater the salt accumulation. In the normal rainfall year (2017), T10 had the lowest total soil salt contents in each soil layer. In 2017, early removal of the mulch film inhibited salt accumulation and the most substantial effect was found in 0-30 cm soil layer.

Table 1 Soil total salt content (g ? kg-1) after planting in 2015-2017

MANOVA (Table 2) showed that treatment had no significant effects, but soil depth had significant effects on the changes in the total salt content. Except for significant differences between CK and T10 in 2015 and 2016, there were no significant differences among other treatments of 2015-2017.

Changes in soil organic matter contents in cotton fields with removal of mulch film

In Xinjiang cotton field, straw was returned to the field after harvest. After winter decomposition, organic matter was mainly distributed in 0-30 cm soil layer (Table 3) and there was only minor inter annual change. After a complete cotton growing season, the organic matter contents of most treatments in 30-50 cm soil layer increased before cotton harvest (Fig. 3). The treatments with the largest increase were present in different film removal treatments. These were T10 in 2015 (30-50 cm soil layer), T10 (30-40 cm soil layer) and T1 (40-50 cm soil layer) in 2016 and 2017 E1 (30-50 cm soil layer). In 0-30 cm ploughing layer, T10 in 2015 showed the largest increase in organic matter contents in 10-30 cm soil layer. In 2016, film removal treatment had a higher increase in organic matter contents than that of the CK.

In 2017, the largest increase in organic matter contents in 0-30 cm soil layer was in E1. MANOVA(Table 4) analysis showed that soil depth had a significant effect on the changes in organic matter contents during 2015-2017, and treatment had a significant effect on the changes in organic matter contents only in 2017. In 2017, only E1 was significantly different from the CK, T1 and E1. In 2015 and 2017, all the treatments were similar.

Fig. 2 Changes in the total salt contents of each soil layer before harvest relative to planting in different groups (10 day and 1 day before the first irrigation event (T10, T1) and 1 day before the second irrigation event (E1) after emergence) in 2015 and 2017Control group (CK) is film mulched throughout growth stage. Error bars represent standard error (n=3).

Table 2 F-value of MANOVA of changes in soil total salt in 2015-2017

Table 3 Soil organic matter content (g ? kg-1) after planting in 2015-2017

Changes in the total nitrogen, phosphorus and potassium contents in cotton fields with removal of mulch film

After planting in 2015-2017, the total nitrogen and the total phosphorus in soil were mainly distributed in 0-30 cm soil layer and seasonal changes were minor. However, the total potassium in soil was more evenly distributed among the soil layers and there was considerable inter-year variability (Table 5).

At the end of cotton growth (Fig. 4), in 0-10 cm and 40-50 cm soil layers, the highest increase in oil nitrogen contents was in film removal treatments. MANOVA (Table 6) showed that only soil depth had a significant effect on the changes in the total nitrogen contents in 2015-2017. There were no significant differences among treatments of 2015-2017.

Fig. 3 Changes in soil organic matter content of each soil layer before harvest relative to planting in different groups (10 day and 1 day before the first irrigation event (T10, T1) and 1 day before the second irrigation event (E1) after emergence) in 2015-2017Control group (CK) is film mulched throughout growth stage. Error bars represent standard error (n=3).

Table 4 F-value of MANOVA of changes in organic matter in 2015-2017

In most soil layers of film removal treatments, the increase in the total phosphorus contents was the largest before harvest. MANOVA (Table 6) showed that soil depth had a significant effect on the changes in the total nitrogen contents during 2015-2017, and treatment had a significant effect on the changes in the total nitrogen contents only in 2017. Significant difference between CK and T10, between CK and T1, between CK and E1 only occurred in 2017, and between T1 and CK only in 2015. There were no significant differences among any of other treatments.

Table 5 Soil total nitrogen, phosphorus and potassium contents (g ? kg-1) after planting in 2015-2017

The trends of inter-annual variations of the total soil potassium contents were considerably different. The total potassium contents pre- and postharvest were compared.

Fig. 4 Changes in soil total nitrogen, phosphorus and potassium contents sof each soil layer before harvest relative to sowing in different groups (10 day and 1 day before the first irrigation event (T10, T1) and 1 day before the second irrigation event (E1) after emergence) in 2015-2017The control group (CK) is film mulched throughout growth stage. Error bars represent standard error (n=3).

Table 6 F-value of MANOVA of changes in the total nitrogen, phosphorus and potassium contents in 2015-2017

The total potassium contents in 2015 had an increasing trend and in 2016 showed a decreasing trend at the end of cotton growth. In 2017, the potassium contents had a decreasing trend in most treatments at the end of cotton growth. At the end of cotton growth in 2015, the treatment with the largest increase in the total phosphorus contents in each soil layer was T1 (0-10 cm), T1 (10-20 cm), E1 (20-30 cm), E1 (30-40 cm) and E1 (40-50 cm). In 2016, the treatment with the smallest decrease in the total phosphorus contents in each soil layer was T1 (0-10 cm), CK (10-20 cm), T10 (20-30 cm), T10 (30-40 cm) and T10 (40-50 cm). In 2017, the smallest decrease in the total potassium contents in 0-10 cm soil layer was T10. Only CK had an increase in 10-20 cm soil layer. The largest increase in 20-50 cm soil layer was E1. MANOVA (Table 6) results showed that only soil depth had a significant effect on the changes in the total phosphorus contents in 2015 and 2017, and treatment had no significant effect on the changes in the total phosphorus contents in 2015-2017. The only significant difference in 2017 was between E1 and T10. There were no significant differences among other treatments.

Changes in hydrolyzable nitrogen, available phosphorus and rapidly available potassium in cotton fields after removal of mulch film

After planting in 2015-2017, soil hydrolyzable nitrogen, available phosphorus and rapidly available potassium were mainly distributed in 0-30 cm soil layer. The contents of hydrolyzable nitrogen and available potassium in each soil layer varied slightly among years, but the inter-annual changes in available phosphorus contents in 0-30 cm soil layer were greater (Table 7).

Table 7 Soil hydrolysis nitrogen, available phosphorus and rapidly available potassium contents (mg ? kg-1) after planting in 2015-2017

By the end of cotton growth (Fig. 5), most film removal treatments increased contents of hydrolyzable nitrogen. The exception was the decrease in the hydrolyzable nitrogen contents of T10 in 2017 in the 0-10 cm soil layer. In 2015, the available phosphorus contents of all the soil layers increased and most treatments of 2016-2017 also increased. In 2015-2016, the highest increase in soil available phosphorus within each soil layer occurred in film removal treatments. However, in 2017, the greatest increase in available phosphorus contents in all the soil layers was in CK. In 2015-2017, the greatest increase in soil available potassium of all the soil layers occurred in different film removal treatments. MANOVA (Table 8) showed that soil depth had significant effects on changes in levels of hydrolyzable nitrogen, available phosphorus and rapidly available potassium in 2015-2017 and film removal treatment had significant effects on changes in the levels of hydrolyzable nitrogen in 2015 and available potassium in 2015 and 2016.

In 2015, there was a significant difference in contents of hydrolyzable nitrogen between T10 and CK and between T10 and T1. There was a significant difference in the contents of available phosphorus between CK and T10, between CK and E1, and between T1 and E1 in 2015. The 2016-2017 treatments did not have significant differences in hydrolyzable nitrogen and rapidly available potassium contents. In 2015, there was a significant difference in contents of rapidly available potassium between T1 and CK (2015), between T10 and E1 (2015), between E1 and CK (2016), and between T10 and CK (2016).

Fig. 5 Changes in soil hydrolysis nitrogen, available phosphorus and rapidly available potassium contents of each soil layer before harvest relative to sowing in different groups (10 day and 1 day before the first irrigation event (T10, T1) and 1 day before the second irrigation event (E1) after emergence) in 2015-2017Control group (CK) is film mulched throughout growth stage. Error bars represent standard error (n=3).

Table 8 F-value of MANOVA of changes in available nitrogen, phosphorus and potassium in 2015-2017

Continued

Dynamic changes in nitrogen, phosphorus and potassium accumulations in cotton plants with mulch film removed

The highest amounts of accumulated nitrogen, phosphorus and potassium in cotton plants were usually found in the CK (Fig. 6). The changes in accumulation dynamics of nitrogen, phosphorus and potassium in each treatment appeared to follow an 'S'-shaped curve,and the nutrient accumulation increased with plant growth. The accumulation pattern was fitted by the Logistic equationY=K/(1+EXP(a+bt)), and the equation coefficientsa,bandKare shown in Table 10. The method of Ming (2006) was used to estimate the timeTmaxat which the nutrient accumulation rate reached the maximum value, the maximum accumulation rateRmaxatTmax, the amount of accumulated nutrientsWmatTmax, the start timet1 and the end timet2 of the linear accumulation, and the amount of accumulated nutrients betweent1 andt2 ΔWt2?t1.

Fig. 6 Nitrogen (N), phosphorus (P) and potassium (K) nutrient accumulation of cotton plants in various treatment groups (10 day and 1 day before the first irrigation (T10, T1) and 1 day before the second irrigation (E1) after emergence, with one control group of film mulching across growth stage (CK)) across cotton growth stage during 2015-2017 Error bars represent standard error (n=3).

Table 9 showed that the maximum accumulated amounts (Kvalue in the equation) andWmvalue of nitrogen, phosphorus and potassium in T1 treatments in the year with heavy rainfall (2016) were the highest.

Table 9 Parameters for logistic equation of cotton nitrogen (N), phosphorus (P) and potassium (K) nutrient accumulation in various treatment groups (10 days and 1 day before the first irrigation event (T10, T1) and 1 day before the second irrigation event (E1) after emergence)

TheKandWmvalues of nitrogen and phosphorus in E1 treatments were the highest in 2017. TheKandWmvalues of nitrogen, phosphorus and potassium in 2015 andKandWmvalues of potassium in 2017 in the CK treatments were the highest. TheTmaxandt1 of nitrogen, phosphorus and potassium accumulation in 2016 and nitrogen and phosphorus accumulation in 2017 in film-removal treatments were later than those of mulching treatment.

The linear accumulation time (t2?t1) of nitrogen, phosphorus and potassium accumulation in 2016 and the nitrogen and phosphorus accumulation in 2017 in filmremoval treatments lasted for a longer time. The opposite situations occurred for nitrogen, phosphorus and potassium in 2015 and potassium in 2017. The longest linear accumulation period of potassium (t2?t1) in 2017 was also present in film removal treatment. The CK exhibited the highestRmaxvalue for nitrogen, phosphorus and potassium in 2015 and 2016 and for potassium in 2017.

Discussion

Effects of mulch film removal on soil salt level

Film mulching combined with precision drip irrigation could provide salt control during cotton growth and salt release during non-growth stage (Yanget al., 2011; Tan,et al., 2017). Previous studies had examined the patterns of temporal and spatial migration of soil salt with or without film mulching, but it was unclear how the salt in soil changes after the film was removed during the growth stages. Zhaoet al. (2013) found that mulch film reduced the loss of soil water and lessened salt accumulation. Liet al. (2016) demonstrated that removal of mulch film could reduce the soil salt contents in 0-40 cm soil layer. Huanget al. (2001) showed that film mulching could inhibit the increase of soil salt contents in the late plant growth stage. Donget al. (2011) found that film mulching reduced the salt contents of surface soil of cotton fields compared with unmulched fields. Abd El-Mageedet al. (2016) showed that all the mulching materials were effective in reducing salt accumulation in root zone. However, Zhaoet al. (2016) discovered that film mulching increased the soil salt contents in late growth stage of sunflowers.

This study demonstrated that in a heavy rainfall year (2016) the film removal treatment increased the accumulation of soil salt, while the film mulching treatment inhibited salt increase. In the year with normal rainfall (2017), the film removal inhibited salt accumulation, and this trend was more pronounced in the late growth stage of cotton.

Qiet al. (2017) found that film mulching during the growth stages reduced the tendency of salt in soil water to move upward. Liu and Wu (2004) showed that desalination occurred in surface soil of a film-mulched dryland field, when the groundwater level was low. Bezborodovet al. (2010) found that the surface soil salt contents without mulching increased by 20% compared to a mulching treatment.

This study showed that in a heavy rainfall year (2016), the distribution of salt in each soil layer of mulching treatment was relatively uniform, but the salt contents were low. During the years (2016 and 2017) with mulch film removed, significant accumulation areas occurred in surface soil.

Zhanget al. (2014) showed that particle size distribution in soil influenced salt migration and distribution. The salt accumulated in relatively impervious soil layer. The salinity level was the lowest in the area under drip irrigation tube and the salt accumulated in the middle of the two films at the end of plant growth stage. Shao (2013) showed that in film-mulching conditions, inter-film soil had the highest salt contents. It was also found that in the initial flowering stage, flowering and boll-forming stage and boll-opening stage, salt accumulation areas were present in the middle of the two films in the CK. This trend was more pronounced in 2017.

Removal of mulch film effects soil nutrients in different growth periods

Film mulching changed soil nutrient contents and its spatial distribution in soil by changing soil physicochemical properties and microbial species, and thus affecting plant growth and nutrient cycling (Liet al., 2007). Songet al. (2002) showed that organic matter decreased by 21.2% during the whole film mulching process and decreased by 17.2% after 60 days of mulching. The decreases were relatively small after 30 days of mulching and no-mulching treatments (4.3% and 6.7%, relatively). Jianget al. (2018) found that mulching could cause depletion of soil organic matter. Mulching could lead to a decrease in soil organic matter contents (Liet al., 2007). Liet al. (2009) found that mulching treatment reduced organic matter contents of 0-5 cm surface soil layer. Zhou (2009) showed that mulching treatment reduced soil organic carbon (SOC) contents. Liet al. (2004) found that film mulching significantly increased microbial biomass carbon content within two years, but reduced the SOC contents. The present study compared the mulching treatment during the whole growth stage with mulch removal. Removal of mulch film increased the organic matter at the end of growth stage. This trend was most pronounced in surface (0-30 cm) soil layer during the year with heavy rainfall (2016).

Caiet al. (2006) found that in heading stage of rice, the total phosphorus, available phosphorus and rapidly available potassium in soil decreased due to the increased nutrient consumption by the crop. Zhang (2012) showed that under film-mulching conditions, the level of each nutrient decreased compared with the initial levels present before planting.

Liet al.(2007) found that the soil organic matter contents, the total soil nitrogen and available potassium contents of upland rice cultivation with mulching treatments decreased by 8.3%-24.5%, 5.2%-22.0%and 9.6%-50.4%, respectively compared with those of the traditional paddy rice cultivation. In four of the five test sites, the dry and mulching cultivation method also reduced the soil available nitrogen by 8.5%-26.5%. The total soil phosphorus contents of three test sites were decreased by 13.5%-27.8%, and the total phosphorus contents of other two test sites were increased by 6.6%-8.2%. However, in all the test sites, soil available phosphorus was increased by 20.9%-64.7%. Tianet al. (2013) indicated that mulching treatment reduced the nitrogen contents of 0-5 cm soil layer after wheat and rice seasons by 17% and 24%, respectively.

It was found that the film mulching treatment reduced the total phosphorus contents in 0-50 cm soil layer. Film removal treatment increased the total nitrogen contents in surface (0-10 cm) and deep (40-50 cm) soil layers. Film mulching treatment reduced soil hydrolyzable nitrogen contents in surface layer (0-10 cm) and rapidly available potassium content in every soil layer.

Liuet al. (2006) showed that the effects of different mulching durations on available phosphorus were relatively small. Zhaoet al. (2015) indicated that the effects of plastic film mulching on available phosphorus and rapidly available potassium in arable layer were not significant. It was also found no obvious pattern for the effects of mulch film removal on effective phosphorus content.

Qinet al. (2014) showed that the contents of available potassium in soil with film mulching treatment in the early and late growth stages of potato were lower than those in unmulched fields. This study demonstrated that the available nitrogen, phosphorus and potassium contents in soil at the end of growth stage with mulch film removed were higher than those of the CK.

Effects of removal of mulch film at different growth periods on nutrient accumulation and distribution in cotton

Chen (2003) showed that mulching promoted early nutrient absorption and accumulation in tobacco plants. Geet al. (2016) found that film mulching and nitrogen fertilizer application significantly increased nitrogen accumulation during each growth stage of maize. Wanget al. (2004) demonstrated that rainwater harvesting with mulches could coordinate the relationship between soil moisture and nutrients and promote wheat nutrient uptake by aboveground parts. Luet al. (2010) showed that the mulching combined with dry farming increased the total nitrogen, phosphorus and potassium contents in rice. Kumar and Dey (2011) indicated that both hay and black polyethylene coverings could effectively increase root growth, amount of absorbed nutrients and yield of strawberry. Qiu (2007) found that ridge covering mulches could increase nitrogen accumulation after wheat flowering and also promote nitrogen transport from vegetative organs to reproductive organs. Feng (2013) indicated that the film mulching treatment could significantly increase nutrient accumulation. The distributions of nitrogen and phosphorus in maturation period wheat with different treatments were as the followings: grain＞stem and leaf＞glume. The potassium distributions were as the following: stem and leaf＞grain＞glume. Liet al. (2014) demonstrated that timely removal of mulch film could effectively improve nutrient absorption and utilization in tobacco plants.

This study showed that in 2015, the film mulching treatment was favorable for accumulation of nutrients in cotton plants. In 2016, with heavy rainfall, removal of mulch film was favorable for the accumulation of nutrients.

Conclusions

In a year with heavy rainfall (2016) and a drought year (2015), removal of mulch film promoted the accumulation of salt in arable soil layer. In a year with normal rainfall (2017), removal of mulch film reduced salt accumulation. This trend was most pronounced in the late growth stages of cotton. Compared with postplanting levels, film removal increased soil organic matter, the total phosphorus and rapidly available potassium during harvest period and increased hydrolyzable nitrogen and the total nitrogen in surface soil layer (0-10 cm) and the total soil nitrogen in deep soil layer (40-50 cm). Film removal increased the accumulation of nitrogen and phosphorus nutrients in cotton plants in a normal rainfall year (2017) and the accumulation of nitrogen, phosphorus and potassium nutrients in cotton plants in a year with heavy rainfall (2016). In the drought year (2015), the opposite situations occurred.

Acknowledgements

We extend our gratitude to Mr. Wang Jin for providing Shihezi meteorological data.