WANG Dong , Xl Yue, SHl Xiao-yan, GUO Chao-li, ZHONG Yu-jie, SONG Chao, GUAN Yu, HUANG Lu, YANG Qi-feng, Ll Feng-min #

1 College of Ecology, Taiyuan University of Technology, Taiyuan 030024, P.R.China

2 College of Agriculture, Nanjing Agricultural University, Nanjing 210095, P.R.China

3 State Key Laboratory of Grassland Agro-Ecosystems, Institute of Arid Agroecology/College of Ecology, Lanzhou University,Lanzhou 730000, P.R.China

4 Teaching Affairs Office, Jilin Normal University, Siping 136099, P.R.China

5 Department of Resources and Environmental Science, College of Agriculture/The Key Laboratory of Oasis Eco-Agriculture of the Xinjiang Production and Construction Corps, Shihezi University, Shihezi 832000, P.R.China

6 Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, P.R.China

7 Gansu Provincial Department of Agriculture and Rural Affairs, Lanzhou 730000, P.R.China

Abstract Plastic film mulch in agricultural production becomes essential to maintaining crop yields in arid and semiarid areas.However, the presence of residual film in farmland soil has also drawn much attention.In this study, three experiments were conducted.The first two experimental designs included 0, 450, 1 350, and 2 700 kg ha-1 of residual film pieces of approximately 5 cm side length added to field soil (0-20 cm soil depth) for seven years and added to pots for four years.In the third experiment, 1 350 kg ha-1 of the residual film with different side lengths (2-5, 5-10, 10-15, and 15-20 cm) was added to field soil for six years to explore the effect of residual film fragment size on soil nutrients, soil microorganisms, crop growth and yields.The residual film had little effect on the soil moisture at a field depth of 0-2 (or 0-1.8) m.There were no significant effects on organic carbon, total nitrogen, inorganic nitrogen, total phosphorus or available phosphorus in the 0-20 cm soil layer.The presence of residual film decreased the richness and diversity of the bacterial community of the surface soil of the residual film, but it had no significant effect on the microbial community of the non-surface soil.The emergence rates of wheat and lentils occasionally decreased significantly with different amounts of residue fragments added to the field.At 450-2 700 kg ha-1, the residual film reduced the plant height and stem diameter of maize and significantly reduced the shoot biomass of harvested maize by 11-19%.The average yields of maize and potato over the seven years decreased, but there were almost no significant statistical differences among the treatments.These results provide important data for a comprehensive scientific understanding of the effects of residual film on soil and crops in dryland farming systems.

Keywords: residual film, soil nutrients, microbial community, crop yield

1.lntroduction

Plastic film is widely used in modern agriculture and has become essential in arid and semiarid areas (Zhanget al.2018).As the most prominent plastic film consumer globally, more than 20 million ha of farmland is mulched with plastic film in China every year (Ingmanet al.2017), and the mulched area will continue to increase.However, with the extensive use of mulched plastic film and its insufficient recovery, large amounts of plastic fragments accumulate in farmland yearly (Heet al.2018;Zhanget al.2020), and this “white pollution” has recently attracted increasing attention (Dinget al.2022).Because the traditional polyethylene-based plastic film is stable and difficult to degrade in the natural environment, it may produce physical barriers in the soil affecting the movement of soil water and nutrients (Liet al.2020).The residual film can also release toxic substances harmful to crops and humans (Yanet al.2014; Wanget al.2021b).Given this, some measures have been taken in recent years, such as increasing the thickness of the plastic film, improving the recovery rate of plastic film residues,and developing degradable film.However, many plastic residues remain in farmland due to the length of time that plastic film mulching has been practiced.Furthermore,these residual films also gradually break down into microplastics (Huanget al.2020), which are more challenging to be removed from the soil.

Although some studies have investigated the effects of residual film on soil and crops, the results have varied,and residual film fragments can migrate and change in size over time (Heet al.2018; Zhanget al.2020; Wanget al.2022).Further, the effects of residual film on soil may be a long-term and slow process.However, the research data on the effects of different amounts and fragment sizes of film residues on soil and crops over long periods are still lacking.Therefore, through several years of field and pot experiments, we studied: (1) the effects of different amounts and fragment sizes of plastic film residues on soil fertility and the microbial community,(2) the effects of different fragment sizes of residual films on seedling emergence rate and crop yield, and (3) the effects of different amounts of plastic film residues on crop yield and shoot biomass.The study results provide important data for a scientific understanding of the effects of the residual film in dryland farming soil.

2.Materials and methods

2.1.Effects of different amounts of residual film on crop yield and soil fertility

A field experiment was conducted in Xiaguanying Town,Yuzhong County, Gansu Province, China from 2012 to 2018 (altitude, 1 620 m; annual average precipitation,358.9 mm; annual average temperature, 7.5°C; soil pH, 8.2; soil bulk density of the 0-20 cm layer, 1.32 g cm-3).The local Agricultural Technology Extension Centre conducted a farmland survey in 2012 and reported that residual film in the 0-20 cm soil layer after one year of film mulching was about 45 kg ha-1.Accordingly, the amounts of residual film in the farmland after 10, 30, and 60 years of continuous film mulching were simulated,with treatment values of 450 kg ha-1(marked as F10),1 350 kg ha-1(F30), and 2 700 kg ha-1(F60), respectively.A treatment without residual film (F0) was the control.The specific operations were as follows: The plastic film used for one year was collected, washed, dried, and then cut into pieces with a side length of about 5 cm.The film fragments were plowed evenly into the 0-20 cm soil layer.During the experiment, no other residual film was added, and the mulched plastic film was recovered as far as possible after the crop was harvested yearly.Every treatment had three replicate plots (each 3 m×6 m), and two experimental fields with the same treatment but different crops were established in the exact location.Maize (ZeamaysL.)and potato (SolanumtuberosumL.) were rotated with film mulching every year, and specific information about the field management measures is given in Appendix A.The grain yield (2012-2018), shoot biomass of maize (dry weight,2015-2018), and tuber yield of potatoes (dry weight, 2012-2018) were measured at harvest.In 2015, the plant height,stem diameter of maize (every month), potato (flower stage),and the number of primary potato branches were measured.The soil nutrient levels for 2012, 2013, 2016, 2017, and 2018 were determined before sowing (April) and after harvest(October), while nutrient levels in 2014 and 2015 were determined in July.The soil water content for the 0-200 (or 0-180) cm soil depth was determined from 2014 to 2017.

A pot experiment was also conducted with the same experimental design as the field experiment.The amount of residual film added to each pot was calculated using the area of the filled soil (about 700 cm2) in the pots.The soil of 0-20 cm depth was collected from the farm used for the pot experiment, sieved through a 2 mm mesh sieve,then mixed evenly with vermiculite in a volume ratio of 3:1(soil/vermiculite powder).The same plastic film fragments as used in the field experiment were mixed with the soil(dry weight of 11 kg), and then the soil with the film added was placed into plastic pots (upper diameter of 31.7 cm,bottom diameter of 25.5 cm, and height of 27.0 cm).Each treatment had 24 pots, of which 12 were planted, and the other 12 were not planted and were used to determine the water consumption of the treated soil.The pots were planted with soybean (GlycinemaxL.Merr.) and wheat(TriticumaestivumL.) in rotation from 2012 to 2015.Other experimental information is given in Appendix B.The grain yield and straw biomass of crops were measured at maturity.

2.2.Effects of residual film fragment size on crop yield and soil fertility

Four sizes of residual film fragments with side lengths of 2-5 cm (marked as A1), 5-10 cm (A2), 10-15 cm (A3),and 15-20 cm (A4) were used in the experiment.A further treatment without residual film was set as the control(CK).The experiment was conducted from 2013 to 2018 in the same location as the field experiment described in Section 2.1.Each treatment had three replicates (each plot was 2 m×2 m).The method of adding residual film to the field was the same as that described in Section 2.1,and the amount of residual film added was 1 350 kg ha-1.The residual film was added in 2013, and no addition was made during the experimental period.This experiment design set two fields with the same treatments and replications but different crops.Lentil (Lensculinaris) and oat (AvenasativaL.) crops were rotated in 2012-2013,and lentils and wheat were rotated in the following years.Information about fertilization, sowing, and management is given in Appendix C.From 2015 to 2018, the seedling emergence rates were counted after 15 days.The soil nutrient levels were determined yearly after harvest (July).

2.3.Effects of residual film on the soil microbial community

In August 2016, residual film fragments from the experimental field treatments (Section 2.1) were collected and shaken to remove the soil attached to the residual film surface, washed with PBS buffer solution, and the washed soil was collected as the film surface soil (marked as S).At the same time, 0-20 cm layer soil samples (with residual film fragments removed) were taken from the F0 (marked as F0), F10, F30, and F60 (marked as nS)treatment plots.Multiple samples were taken from each plot and mixed.The microbial DNA of the soil samples was extracted with a Fast DNA Spin Kit for Soil (MOBIO Laboratories, Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions.The extracted microbial DNA samples were sequenced by Majorbio Bio-pharm Technology Co., Ltd.(Shanghai, China).Chao1 and Shannon indexes were used to evaluate the diversity and richness of the soil microbial community, and the calculation methods were:

whereSis the number of operational taxonomic units(OTUs);n1andn2are the numbers of OTUs with 1 read and 2 reads, respectively;Piis the ratio of theith OTUs to the total OTUs.

2.4.Determination of soil fertility and soil water content

The soil water content before April 2015 was determined using the soil drilling method, with samples taken every 20 cm to a depth of 200 cm.After April 2015, the soil water content was determined every 10 cm to a 170 or 180 cm depth using a TRIME-IPH soil water meter (IMKO Micromodultechnik GmbH, Germany).Soil organic carbon, total nitrogen, total phosphorus, and available phosphorus were determined according to methods described by Guet al.(2018).

2.5.Data analysis

The data are reported as the mean value of replicates.Multi-way analysis of variance was used to determine the effect of the external environment (interannual variation),crop species, and treatments on the observed data and whether there was an interaction among factors.One-way analysis of variance and the least significant difference test (P<0.05) were used to test for differences in the annual maize grain yield, annual potato tuber yield,annual soil nutrient levels, annual shoot biomass of maize,annual seedling emergence rate, microbial community diversity index, and water consumption in the pot.The ANOVA was conducted using GenStat (17th edition;VSN International Ltd., Rothamsted, UK) and R 4.1.2.SigmaPlot 12.0 software was used to create the graphics.

3.Results and discussion

3.1.Effects of residual film on soil nutrients and moisture

Residual film of different amounts or sizes had no significant effects on the soil organic carbon, total nitrogen, inorganic nitrogen, total phosphorus, or available phosphorus levels in the field soils (Appendices DG).This was consistent with the findings of Guanet al.(2018) that residual film had no significant effect on soil organic matter or water-stable aggregates (>0.25 mm).In contrast, other researchers found that residual film affected soil pH, soil conductivity, and soil nutrients (Qiet al.2020; Zhanget al.2020).It has been found that residual film could significantly increase soil organic matter and aggregates with the increase in particle sizes(Meulenet al.2006).The microplastics formed by the residual film could promote the formation of aggregates and increase the soil water-holding capacity (Machadoet al.2018).The effects of residual film on soils may be related to the soil properties such as soil texture, nutrients and water content, and the residual film amount, size, and states or shapes in soils.Machadoet al.(2018) found that these properties of plastic particles could alter their effects on soils.

The different amounts of the residual film had no significant effect on the water content of the 0-200 cm (or 0-180 cm) soil layer in the field experiment (Appendix H).In other studies, it has been found that residual film could reduce soil water holding capacity (Wanget al.2020), soil water storage (Huet al.2020), soil water permeability,and evaporation capacity (Zhanget al.2020).Meulenet al.(2006) speculated that residual film could conserve water by physically preventing the rise of capillary water while still allowing the downward gravity flow of water.We found that the residual film treatments had a waterconservation effect in the pot experiment (Appendix I).Liet al.(2020) found that when the amount of residual film was >360 kg ha-1, soil water, and nitrate movement were enhanced due to preferential flow.Thus, the water content was also high, especially when the residual film fragments were large.This might explain why the soil water content in the residual film treatments in our field experiment tended to be high after winter (Appendix H).However, the residual film was also found to accelerate soil water loss.This phenomenon was found to be much more evident with a decrease in the size and an increase in the amounts of film fragments (Wanet al.2019).It should be noted that this study only focused on the soil water in the layer with residual film, while it ignored the water movement between the soil layers.The residual film could affect water movement by increasing the proportion of macropores in the soil (Wanget al.2020)and reducing soil bulk density (Jianget al.2017).In contrast, Guanet al.(2018) found that residual film did not significantly change soil bulk density.The residual film’s density was much lower than that of the soil (generally,the density of polyethylene is about 0.910-0.925 g cm-3).Therefore, the addition of film would inevitably lead to a decrease in soil bulk density, but due to the low mass ratio of residual film in soil (e.g., 2 700 kg ha-1of residual film in the 0-20 cm soil layer in the present study amounted to a mass content<0.1%), the reduction of bulk density of the whole soil maybe limited.In future studies, it will be necessary to eliminate the influence of residual film fragments when determining the “real” bulk density of a soil containing residual film.

3.2.Effects of residual film on the soil microbial community

The Chao1 and Shannon indexes of the bacterial community in the soil collected from the residual film surface were significantly lower than those in the nonsurface soil samples or the control soil.At the same time, there were no significant differences in the fungal community between the treated soils (Table 1).Furthermore, the microbial composition in the soils collected from the film surface differed significantly from those of the control and the non-surface soil samples(Appendix J).These differences might influence the distribution of nutrients and water in the residual film surface soil (Liet al.2020).The residual film could also release toxic substances to affect the microorganisms(Wanget al.2016), thus affecting the soil microbial community attached to the residual film.The diversity of bacteria decreased with an increase in the amount of residual film added to the soil.However, the difference was insignificant (Table 1).At the same time, the different amounts of the residual film had no significant effect on the diversity of the fungal community (Table 1).These results indicated that the influence of residual film on soil microorganisms was limited to the bacterial community on the surface of the residual film, and there was no effect on the fungi and bacteria in the broader soil environment.The different responses of fungi and bacteria to residual film might be due to the residual film providing the substrate and space for microbial growth (Liuet al.2022), and fungi with dispersed hyphae might more easily colonize the surface of the film than bacteria (Sander 2019).Wanget al.(2016) found that the diversity of the microbial community was significantly decreased in a pot experiment with residual film amounts of 67.5 and 337.5 kg ha-1.Their study might have used new plastic film fragments as the residual film, which was very different from the residual film in the present study,which had been buried for four years before collection.The surface of the film and the concentration of toxic substances in the new and buried plastic films vary greatly(Wanget al.2016, 2021a; Hanet al.2020).It has also been found that when the amount of residual film reached a mass content of 0.2% and 1% in soil, the Shannon indexes of the bacterial community in rhizosphere soil were significantly affected (Qiet al.2020; Liuet al.2022).The highest treatment level of residual film in the present study (2 700 kg ha-1, mass content<0.1%) was far lower than the level of these two studies.However, the rhizosphere microbial community was not studied in the present study.

Table 1 Richness and diversity of the microorganism community in soils treated with different amounts of residual film

3.3.Effects of the size of residual film on crop emergence and yields

Residual film size significantly affected the emergence rates of wheat and lentils (Table 2).The emergence rates of crops displayed a decreasing trend with an increase in the size of residual film fragments.However,the reduction effects were inconsistent (Appendix K).It has been reported that residual film decreases crop emergence rates (Yanet al.2014), but varied results occur.Such variation might be related to the differences in cultivated crops and the timing of the emergence rate investigation.With 180-540 kg ha-1of residual film,the emergence rates of maize and flax were found to be reduced by 10.0-26.6 and 2.9-3.7% ten days after sowing compared with the control, respectively, while,when the seedling emergence rates were counted again 20 days after sowing, there were no significant differences between the treatments (Genget al.2019).The results indicated that residual soil film might delay crop germination and emergence but did not reduce the final emergence rate.The delayed effect might be related to the soil water, fertilizer, and other soil conditions.Our study counted the emergence rate after 15 days of seedling emergence (more than 20 days after sowing).The emergence rates differed significantly between years,possibly due to different soil conditions at sowing (Table 2;Appendix K).Therefore, when analyzing the effect of residual film on crop emergence, the emergence rate of the crop should be counted multiple times to ensure that the total emergence rate is determined.Further,interannual differences influence soil conditions, which should be considered when determining crop emergence rate.

Table 2 The results of ANOVA for grain yield, emergence rate, and shoot biomass of crops in fields treated with different amounts and sizes of residual film1)

There were no significant differences in grain yield of oat, wheat, and lentil crops treated with different sizes of residual film (Appendix L).This was consistent with theresults of Guanet al.(2018), who found that maize yield was not significantly affected by film fragments of <5 or 10-20 cm2applied at 1 680 kg ha-1.Yanet al.(2014)also found that adding residual film fragments less than 16 cm2in the area did not reduce crop yield.However, Qiet al.(2018) found that when the concentration of residual film added to the soil was 1% (mass content), both large(6.9 mm×6.1 mm or 6.9 mm×6.0 mm) and small (<1 mm)fragments affected crop yield and biomass.This might be because the amount of residual film was significant when compared with the amount applied in the present study(mass content less than 0.1%).

3.4.Effects of the amount of residual film on shoot biomass and yield

With increased residual film added to the soil, the harvested maize plant height and stem diameter decreased gradually (Appendix M).However, maize treated with the lower residual film amount had a faster growth rate in the first month (April to May), and the growth rate of maize in control was lower than that in the treatments with residual film in the later period (Appendix N).The residual film also reduced the plant height, stem diameter, and primary branch numbers of potatoes at the flowering stage, but the differences were insignificant(Appendix O).

The changes in maize plant height and stem diameter indicate that the soil’s residual film fragments act as a barrier to the root system just in the early stage of maize growth.After one month’s growth, the developed fibrous roots of maize would pierce or bypass the residual film and continue to grow.Henceforth, the impact of residual film on maize would be limited.In a rain-fed farming system, the slow growth in the early stage would mean that plants are already disadvantaged in the competition for limited resources.The residual film would also reduce the matching of soil water and the maize root system(Jianget al.2017), leading to a decreasing trend in the yield and aboveground biomass of the maize in residual film treatments.

In the present study, maize grain yields and potato tuber yields did not differ significantly between treatments with different amounts of the residual film (Table 2),except for maize grain yields in 2018 (Appendix P).For the amounts of residual film of 450, 1 350, and 2 700 kg ha-1, the average maize grain yield and average potato tuber yield over the seven years decreased in value by 15, 22, 26%, and 17, 20, 21% compared with the control,respectively (Appendix P).This might be the distribution of residual film fragments in the field being uneven by the tillage or winding of crop roots during the long-term experiment, which increased the difference in the value among the replications.Some researchers had observed that residual film significantly reduced crop yield when the amount of residual film was high (Donget al.2015;Zhanget al.2020).However, one study found that plants grew better in abandoned farmland containing residual film in Spain (Meulenet al.2006).This might be because the variable soil physicochemical conditions in different periods and locations can moderate the impact of the residual film, especially when the amounts of the residual film are high (600-2 700 kg ha-1) (Zhanget al.2020).For example, the residual film may affect the growth of maize seedlings under poor moisture conditions while having little effect under good moisture conditions (Geng 2018).Because the barrier effect of the residual film was limited in a better environment, the residual resources could still support crop germination and seedling growth.In the present study, the average precipitation in the experimental area was 408.5 mm from 2012 to 2018,significantly higher than that over the past 48 years(378.9 mm).Thus, the soil moisture condition during the experimental period was better than usual.In addition,the alkaline soil of the present study (pH=8.2) might also be a critical factor for the insignificant effect of residual film, because the impact of residual film on crop may be more significant in acidic soil than in alkaline soil (Liuet al.2022).

The residual film at levels of 450-2 700 kg ha-1significantly reduced the shoot biomass of maize(Table 2; Fig.1).By comparing the average values of each treatment with the control over all years, the shoot biomass of maize decreased by 11, 15, and 19% when the amount of residual film was 450, 1 350, and 2 700 kg ha-1, respectively.Huet al.(2020) believed that the residual film restricted the growth of roots, thus reducing the shoot biomass of maize.Zhanget al.(2020) found that with every 100 kg ha-1increase in residual film, the plant height decreased by 2.5%.In contrast, Machadoet al.(2018) found that residual film microplastics can increase aboveground crop biomass, which might be because microplastics can increase the soil water-holding capacity, and the barrier effect is limited.

Fig.1 The above-ground biomass (kg ha-1) of maize planted in the field treated with different amounts of residual film added to the soil.F0, F10, F30, and F60 indicate that the amounts of residual film added are 0, 450, 1 350, and 2 700 kg ha-1,respectively.The error bars are standard errors (n=3), and bars with different letters are significantly different (P<0.05).

The interannual changes in crop emergence rate,yield, and shoot biomass under the various treatments differed among crop species.The interannual changes were inconsistent among the treatment levels (Table 2;Appendices K, P and Q).This indicates that although the film fragments in the soil may change during the experimental period due to the inertness of polyethylene film in soil, the changes of the residual film in the 7-year experimental period of this study were not enough to change its effect on the soil and crops.Therefore, we speculate that interannual climate fluctuation and other external factors may cause the interannual variations in crop emergence rate, yield and shoot biomass.

The residual film effect on yield and straw biomass varied in the pot experiment.The residual film significantly increased the grain yield and straw biomass of soybean and wheat in 2012 and 2013, significantly decreased the wheat grain yield and straw biomass in 2015, and had no significant effect in 2014 (Table 3).The residual film increase in crop yield and straw biomass might be due to the regular watering of the pots.The loose soil in the pots provided better soil conditions than the field, thus moderating the effect of the residual film.Furthermore, the residual film in the pot also had a water conservation function (Appendix I).The pot experiment’s time extension allowed the residual film’s negative effect to appear.This negative effect might be associated with the harmful substances or pathogens that crop residues accumulate due to the long-term planting in the potting soil (Stamanet al.2001).It is also possible that some harmful substances may have been released by the plastic pots and residual film (Hahladakiset al.2018),resulting in a deterioration of the crop growth conditions.Moreover, it may also be one of the reasons that residual film had a positive effect on crop yield in the first two years, then later had no effect, and finally had a negative effect in the field experiment (Appendix P).

Table 3 The grain yields (g/pot) and straw biomass (g/pot) of soybean and wheat in the pot experiment treated with different amounts of residual film added to the soil

Plastic film mulching is essential in agricultural production in the dryland rain-fed area of the Loess Plateau.It was calculated that the amount of plastic film used in a typical agricultural field of the area is about 91 kg ha-1yr-1(for the calculation method see Appendix R).The recovery rate of used plastic film has reached 80%.Thus, the residual amount of film in the field is about 18 kg ha-1yr-1, much lower than the surveyed value in 2012 (45 kg ha-1yr-1).Therefore, if the recovery rate remains at 80%and there is no other loss of residual film, the residual film amounts of 450, 1 350, and 2 700 kg ha-1used in this study are equivalent to the amount of residual film that would accumulate after continuous film mulching for 25, 75, and 150 years, respectively.However, the experimental results showed that the impact of plastic film residue on soil and crops would still be limited even with such a large amount of residual film.Other risks, such as micro- or nanoplastics formed by the residual films, the effect of residual films on livestock, and the effect of residual films on the rural environment, need more attention.

4.Conclusion

Different amounts of plastic film residues and different sizes of residual film fragments had no significant effects on soil moisture and soil nutrient levels.Different amounts of the residual film only reduced the richness and diversity of the bacterial community in the soil attached to the surface of the residual film.Different sizes of the residual film fragments (each side length of 2-5 to 15-20 cm) had inconsistent reduction effects on the emergence rate of wheat and lentil crops.450-2 700 kg ha-1of the residual film added to the soil affected maize plant height andstem diameter in the seedling stage, thus significantly reducing the shoot biomass of harvested maize and causing a decreasing trend in maize and potato yield.In general, crop production under various treatments did not show significant linear changes within a year or increased residual film amount.However, residual film accumulation still impacted crop biomass or yield, which may be related to the random distribution of the residual film in the field and its accumulation in soil.Therefore, it is necessary to remove farmland residue film to keep the soil clean and maintain crop production.

Acknowledgements

This work was funded by the National Natural Science Foundation of China (31470496), the Fundamental Research Funds for the Central Universities, China(lzujbky-2021-sp42) and the ‘111’ Programme 2.0, China(BP0719040).We thank the International Science Editing service for editing this manuscript (http://www.internationalscienceediting.com).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on https://doi.org/10.1016/j.jia.2023.04.026