Yanhui ZHAOXiong XIONG and Chenxi WU∗

1State Key Laboratory of Freshwater Ecology and Biotechnology,Institute of Hydrobiology,Chinese Academy of Sciences,Wuhan 430072(China)

2University of Chinese Academy of Sciences,Beijing 100049(China)

(Received October 20,2019;revised September 14,2020)

ABSTRACT Periphytic biofilms are commonly presented at the water-soil interface in paddy fields.Different fertilization methods can affect the concentration and distribution of nutrients in paddy fields and thus affect the development of periphytic biofilms.In this study,the roles of periphytic biofilms in nitrogen(N)cycling in paddy systems and how they are affected by different fertilization methods were studied using microcosm experiments.Microcosms were prepared using soil samples from a paddy field and treated with surface and deep fertilization under light and dark conditions.Surface fertilization under light condition promoted the development of periphytic biofilms,while deep fertilization under dark condition inhibited their development.The development of periphytic biofilms increased the pH and dissolved oxygen levels in the overlying water.Surface fertilization resulted in high N concentrations in the overlying water and the topsoil layers,which enhanced NH3 volatilization and nitrification-denitrification but inhibited N fixation.The development of periphytic biofilms reduced NH3 volatilization loss but increased nitrification-denitrification loss and the overall N loss in the paddy system.The results from this work suggest that the presence of periphytic biofilms in paddy fields could increase N loss by 3.10%-7.11%.Deep fertilization is an effective method to retard the development of periphytic biofilms in the paddy system and can potentially increase the overall N use efficiency.

Key Words:deep fertilization,denitrification potential,N loss,N use efficiency,NH3 volatilization,paddy field

INTRODUCTION

Low efficiency of nitrogen(N)fertilizer is the main factor limiting rice culture yield in paddy systems(Fageriaet al.,2010;Lassalettaet al.,2014;Zhaoet al.,2015).The loss of N from the applied fertilizer is estimated to range from 10% to 65%(Cassmanet al.,1998;Zhaoet al.,2012),which can be mainly attributed to ammonia(NH3)volatilization,surface runoff,and nitrification-denitrification processes(Zhaoet al.,2012;Hudaet al.,2016;Suet al.,2017).In China,the broadcast application of fertilizers is still the most common practice for rice cultivation.The broadcast application of fertilizer results in high amounts of ammonium in floodwater,which increases NH3volatilization(Rochetteet al.,2013).Because N remains near the soil surface,the broadcast application of fertilizer causes considerable N loss and atmospheric pollution because of the emissions of nitrous oxide and nitric oxide through nitrification-denitrification(Singhet al.,1995;Wuet al.,2017).Moreover,the broadcast application of fertilizer increases the nutrient content in floodwater and subsequently increases the risks of surface runoffloss(Kapooret al.,2008),which leads to pollution of downstream water bodies.

Fertilizer deep placement in paddy fields has been widely recognized as an effective management practice for increasing N use efficiency(Bandaogoet al.,2015;Zhuet al.,2019).Deep placement of N fertilizer at 7 to 10 cm in a reduced soil layer decreases the N concentration in floodwater and reduces N loss by 30%-40%(Alamet al.,2013;Gaihreet al.,2018).Reduced N loss is related to the decrease in NH3volatilization,surface runoff(Rochetteet al.,2013),and nitrification-denitrification losses(Chienet al.,2009).In addition,deep placement of N fertilizer can also reduce nitrous oxide and nitric oxide emissions and increase grain yield by 15%-20%(Kapooret al.,2008).Therefore,fertilizer deep placement is considered a good management practice with multiple benefits.

In paddy fields,periphytic biofilms are ubiquitously developed at the soil-water interface,and these biofilms are aggregates that are comprised of algae,bacteria,protozoa,metazoans,epiphytes,and detritus(Suet al.,2017).Periphytic biofilms are known to play important roles in nutrient cycling in aquatic systems(Wuet al.,2016;Liuet al.,2017).A paddy field is an artificial wetland system subject to intensive human interference for rice cultivation.It provides a favorable habitat for the development of periphytic biofilms because of the high nutrient level and shallow water environment.Although these biofilms are commonly observed,the roles of periphytic biofilms in paddy fields remain poorly understood and are usually ignored in previous studies.

Different fertilization methods would change the distribution of nutrients in paddy fields and thus affect the growth of periphytic biofilms and their functions.Therefore,the aims of this study were to i)characterize the development of periphytic biofilms in paddy systems under different fertilization and illumination conditions;ii)assess the N lossviaNH3volatilization and denitrification;and iii)discuss the role of periphytic biofilms in N cycling as affected by the fertilization methods.The results from this work will improve the understanding of the role of periphytic biofilms and provide a scientific basis for establishing a better management practice to reduce N loss in paddy systems.

MATERIALS AND METHODS

Experimental setup

The soil(0-20 cm topsoil)used for the study is a silty clay soil collected using a shovel from a paddy field at the Hubei Academy of Agricultural Sciences,China.The soil sample was homogenized,sieved to＜2 mm to remove large debris,and then used for the experiment.An aliquot of 2 g fertilizer(Stanley Co.,Ltd.,China;N-P2O5-K2O≥580 g kg−1,B≥2.5 g kg−1,Zn≥1.5 g kg−1,Fe 0.5 g kg−1,Mn 0.5 g kg−1,Mo 0.1 g kg−1,and trace elements 5.0-30.0 g kg−1)was mixed with 0.17 kg soil(fresh weight)and used as the fertilized soil.The properties of the soils are provided in Table I.

Microcosms were set up using glass containers with a diameter of 12 cm and a height of 32 cm.Four treatments were designated as follows:surface fertilization under light condition(SL),surface fertilization under dark condition(SD),deep fertilization under light condition(DL),and deep fertilization under dark condition(DD).A total of 36 microcosms were prepared with 9 microcosms for each treatment.All microcosms were kept at 28±3°C.For SL and SD,each microcosm was filled with 2.17 kg soil,and 0.172 kg fertilized soil was added to the top.The soil surface was about 16 cm away from the bottom of the container.For DL and DD,each microcosm was first filled with 0.72 kg soil,and 0.172 kg fertilized soil was added,followed by 1.45 kg soil.The soil surface was about 16 cm away from the bottom of the container,and the fertilized layer was about 10 cm below the soil surface.The soil column was wrapped with aluminum from the outside of the container.An aliquot of 850 mL distilled water was added to each container,and the water surface was about 5 cm away from the soil surface.All containers were then covered with gas-permeable membranes on the top to prevent water evaporation while allowing gas exchange.The SD and DD microcosms were placed in the dark,and the SL and DL microcosms were illuminated with fluorescent lamps with a photosynthetic photon flux density of 40±5µmol photons m−2s−1at the surface of the microcosms.

The NH3volatilization was measured following a previous method(Wanget al.,2002).Sponges with a thickness of 2 cm and a diameter of 13 cm were evenly immersed in 15 mL phosphoglycerine solution(50 mL phosphoric acid and 40 mL glycerin diluted in 1 000 mL deionized water).Then,two sponges were placed in each microcosm.The first one was 4 cm away from the water surface,and the other one was placed on the top of the container.

Sampling and analysis

Sampling was carried out every 10 d for 30 d.The overlying water from each microcosm was siphoned into a conical flask.Dissolved oxygen(DO)and pH of water samples were measured with a Hach HQ40d meter(Loveland,USA).Total N(TN),and total P(TP)in water samples were analyzed following standard methods(MEPC,2002).

Samples collected using a core sampler(4.6 cm in diameter)were divided into periphytic biofilm layer(0-0.2 cm)and upper(0.2-5 cm),middle(5-10 cm),and deep(10-15 cm)soil layers.The collected samples were homogenized and freeze-dried.Total Kjeldahl N(TKN)was analyzed using an automatic Kjeldahl N analyzer(S5,Behr,Germany).TheN concentrations were measured spectrophotometrically following standard methods(MEPC,2002).Soil TN was calculated as the sum of the TKN,concentrations.Soil TP was measured as described previously(Rubanet al.,1999).The available P was measured after extraction with 0.5 molL−1sodium bicarbonate,and the available K was determined following extraction with 1 mol L−1ammonium acetate.

TABLE IPropertiesa)of the paddy soil used in the study

Sponges for NH3volatilization measurement were collected every day and analyzed using a continuous flow analyzer(SAN++;Skalar,The Netherlands)after extraction with 200 mL 1 mol L−1KCl(Wanget al.,2002).

The chlorophylla(Chl.a)and organic matter(OM)contents in the periphytic biofilm layer were also measured.The OM content was determined gravimetrically and the Chl.acontent was measured spectrometrically after extraction with organic solvent(Zhanget al.,2018).The denitrification potential(DNP)of the periphytic biofilm was determined using the acetylene inhibition method(Arnonet al.,2007).The nitrogenase activity(NA)of biofilms was measured using a modified acetylene reduction assay(Liet al.,2019)and expressed as the amount of acetylene reduced per unit dry weight(DW)of the periphytic biofilm per hour(nmol C2H2g−1DW h−1).The measurement was performed after the samples were incubated in the dark for 4 h.

The TN loss from the system was determined as the difference in TN at the beginning and the end of the experiment.The N loss from denitrification was calculated by subtracting the total NH3volatilization loss from the TN loss.

Statistical analysis

All results are presented as the means±standard deviations of three replicates.Differences between samples from different treatments were tested using one-way analysis of variance(ANOVA)with an LSDpost hoctest using SPSS software version 18.0(SPSS Inc.,USA).A confidence level of 95% was used.All figures were plotted using GraphPad Prism 6 software(GraphPad,USA).

RESULTS

Changes in DO and pH of overlying water

The changes in DO and pH of the overlying water during the experiment are presented in Fig.1.Over time,there was no significant difference in pH among the treatments,but DO increased significantly(P＜0.05)from day 10 to day 20 in DL.Among the different treatments,DO was significantly higher in SL and DL than in SD and DD,DO became higher in DL than in SL after day 20,and DO was always lower in SD than in DD.The pH showed no significant difference(P＞0.05)between SL and DL or between SD and DD throughout the experiment.However,at all sampling points,the pH of SL and DL was significantly higher(P＜0.05)than that of SD.

Development of periphytic biofilm

Chl.ais used as an indicator of algal biomass,and OM is an important indicator of soil fertility that is closely related to soil microbial biomass(Haoet al.,2008).The changes in Chl.aand OM in the periphytic biofilm developed over time in the different treatments are presented in Fig.2.The Chl.acontent was below the detection limit in SD and DD.Both OM and Chl.aincreased significantly(P＜0.05)over time in SL and DL,and OM in SD and DD did not show significant difference(P＞0.05)throughout the experiment.Among the different treatments,SD and DD showed the lowest OM values at all sampling times;SL always had the highest OM and Chl.avalues.

Nitrogen in overlying water

Fig.1 Changes in dissolved oxygen(DO)and pH of the overlying water on the paddy soil in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Vertical bars indicate standard deviations of the means(n=3).Bars followed by the same letter(s)for a given sampling time are not significantly different among treatments.

Fig.2 Changes in the chlorophyll a(Chl.a)and organic matter(OM)contents in the periphytic biofilm developed at the paddy soil-water interface in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Vertical bars indicate standard deviations of the means(n=3).Bars followed by the same letter for a given sampling time are not significantly different among treatments.

Fig.3 Changes in the total N(TN),-N concentrations in the overlying water on the paddy soil in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Vertical bars indicate standard deviations of the means(n=3).Bars followed by the same letter for a given sampling time are not significantly different among treatments.

Nitrogen in periphytic biofilm and soil profile

The changes in the different forms of N in the periphytic biofilm and soil profile over time are presented in Fig.4.In SL and SD,TKN was the highest in the periphytic biofilm layer and decreased with soil depth.In DL and DD,TKN in the soil profile was more evenly distributed.TKN in the periphytic biofilm decreased in SD but increased in the rest of the treatments over time.-N in the periphytic biofilm and upper soil layers was higher in SL and SD than in DL and DD except on day 30 in the upper soil layer.However,-N in the middle and deep soil layers was lower in SL and SD than in DL and DD.-N decreased over time in the periphytic biofilm and upper soil layers in both SL and SD,while the decrease in-N was greater in SL.-N in DD remained relatively stable but decreased slightly in DL in the periphytic biofilm layer and upper and middle soil layers during the experiment.-N in SL and SD was higher than that in DL and DD in the periphytic biofilm and upper soil layers but lower in the middle and deep soil layers except on day 30 in the periphytic biofilm layer.decreased over time in the periphytic biofilm and upper soil layers in SL and SD but increased in the upper soil layer in DL.-N was much lower thanand was close to the detection limit in the middle and deep soil layers in SL and SD.In the periphytic biofilm and upper soil layers,increased and was higher in SL than in SD.However,was the highest in the deep soil layer in DL and DD.N was generically higher in DL than in DD in the periphytic biofilm layer and upper and middle soil layers.Over time,-N decreased in all treatments except SD.

Fig.4 Changes in total Kjeldahl N(TKN),N concentrations in the periphytic biofilm layer(0-0.2 cm)and upper(0.2-5 cm),middle(5-10 cm),and deep(10-15 cm)paddy soil layers in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Horizontal bars indicate standard deviations of the means(n=3).

Nitrogenase activity in periphytic biofilm

Biological N fixation is a potential source of N.The changes in NA in the periphytic biofilm under different treatments during the experiment are presented in Fig.5.The NA in the periphytic biofilm was significantly higher(P＜0.05)in DL than in the rest of treatments,while NA was maintained at a relatively low level in SD.Over time,NA in the periphytic biofilm increased in SL and DL but was relatively stable in SD and DD.

Fig.5 Changes in nitrogenase activity in the periphytic biofilm developed at the paddy soil-water interface in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Vertical bars indicate standard deviations of the means(n=3).Bars followed by the same letter for a given sampling time are not significantly different among treatments.DW=dry weight.

NH3 volatilization loss in paddy systems

Fig.6 Changes in NH3 volatilization loss in each microcosm in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Vertical bars indicate standard deviations of the means(n=3).

The changes in NH3volatilization loss are presented in Fig.6.The values in SL were higher than those in SD,and they were higher than those in DL and DD at the beginning of the experiment.After that,the values in SL rapidly decreased until day 15,and SL had no significant difference(P＞0.05)with DL and DD after day 15.SD had the highest value after 5 d.The values in DL were significantly lower than those in DL during 6-20 d.The reason for this result is that the presence of periphytic biofilms under the light condition increased the floodwater pH(Fig.1),and periphytic biofilms assimilated and reduced-N in the floodwater in later(Figs.3 and 4).

Denitrification potential in periphytic biofilm

The DNP rate in the periphytic biofilm was the highest in SL,but it decreased significantly(P＜0.05)over time(Fig.7).The DNP rate in DL was lower than that in SD on day 10,while it was higher(P＜0.05)on day 30.DD had the lowest rate in all treatments over 30 d.

Fig.7 Changes in denitrification potential(DNP)rate in the periphytic biofilm developed at the paddy soil-water interface in the experiment with four treatments:surface fertilization under light condition(SL),deep fertilization under light condition(DL),surface fertilization under dark condition(SD),and deep fertilization under dark condition(DD).Vertical bars indicate standard deviations of the means(n=3).Bars followed by the same letter for a given sampling time are not significantly different among treatments.

N loss in paddy systems

The total N loss in different treatments varied from 16.24% to 44.16%(Table II).Surface fertilization resulted in higher NH3volatilization and denitrification losses than deep fertilization.Under the same fertilization condition,the light treatment resulted in a higher total N loss than the dark treatment.However,denitrification loss was the major N loss pathway under light treatment,while NH3volatilization loss was the major N loss pathway under dark treatment.

DISCUSSION

Different fertilization methods affect the nutrient levels and their distribution in paddy systems and thus affect the development of periphytic biofilms at the soil-water interface(Tripathiet al.,2008).Surface fertilization resulted in a high nutrient level in the soil surface and overlying water(Figs.3 and 4).Therefore,the development of periphytic biofilm waspromoted especially under the light condition because of the growth of phototrophic microorganisms.The rapid growth of algae at the water-soil interface was visually observed and evidenced by the increase in the Chl.acontent(Fig.2).Algae photosynthesis under light condition releases oxygen and increases pH because of the consumption of CO2(Caoet al.,2013).As a result,DO and pH were higher in the light treatment than in the dark treatment.Meanwhile,a high nutrient level may also stimulate microbial respiration(Fanget al.,2018).Therefore,DO was lower in SD than in DD and was also lower in SL and DL except on day 10.Deep fertilization retained N in the deep soil layer(Fig.4).Therefore,the development of periphytic biofilm at the water-soil interface was inhibited.

TABLE IIN loss in the different treatments,surface fertilization under light condition(SL),surface fertilization under dark condition(SD),deep fertilization under light condition(DL),and deep fertilization under dark condition(DD),during the experiment

Surface fertilization also resulted in a rapid conversion ofand produced moreespecially under the light condition because of an aerobic condition(Fig.3).NH3volatilization loss was much higher for surface fertilization than for deep fertilization especially in the first few days(Fig.6),owing to the increase in pH(Fig.1)and the decrease inconcentration in the overlying water.Total NH3volatilization loss was higher for the dark treatment than for the light treatment for both surface and deep fertilization(Fig.6).This trend may be related to the rapid conversion of,which rapidly decreased theconcentration in the overlying water.Additionally,the development of periphytic biofilms assimilatedinto their biomass,which further decreased theconcentration in the overlying water and reduced NH3volatilization(Figs.3 and 4)(Sheet al.,2018).

NA and DNP were measured to reflect the functions of the periphytic biofilms.NA indicates the biological N fixation potential,which can be a potential source of N for the paddy systems(Mårtenssonet al.,2009;Maet al.,2019).A high N level can suppress N fixation(Linet al.,2018).NA in the periphytic biofilms was inhibited in the surface fertilization treatment,while that in the deep fertilization treatment remained at a relatively high level(Fig.5).NA in SL increased over time,likely because of a decrease in N concentration over time.NA in the light treatment was significantly higher than that in the dark treatment,which might be because of the growth of N-fixing cyanobacteria under light condition(Choudhary and Bimal,2010;Maet al.,2019).Therefore,deep fertilization can not only reduce N loss but also maintain the N fixation function of the paddy system.DNP was significantly higher for surface fertilization than for deep fertilization under the same illumination condition.A higher DNP can be related to the higherlevel.A higherlevel at the soil-water interface stimulates denitrification in periphytic biofilms(Wanget al.,2013).With the same fertilization method,the light treatment resulted in a higher DNP,which can also be attributed to a higherlevel(Fig.3).In addition,the development of periphytic biofilms under light condition can release organic C sources that can be used by denitrifiers(Kalscheuret al.,2012).

Among all the treatments,total N loss was the highest in SL and the lowest in DD(Table II).Surface fertilization resulted in a much higher total N loss than deep fertilization in the paddy systems under the same illumination condition,while the dark treatment resulted in a lower total N loss than the light treatment under the same fertilization condition.Surface fertilization resulted in higher NH3volatilization and denitrification losses,as surface fertilization produced a highconcentration in the overlying water and an aerobic condition that favors nitrification.Theproduced by nitrification can be converted to(Fig.4)and N2by denitrification becausecan easily migrate into the anaerobic zone(Liuet al.,2015).Denitrification was the major N loss process in the light treatment,while NH3volatilization was the major N loss process in the dark treatment.This can be related to the development of periphytic biofilms under light condition.The development of periphytic biofilms might enhance denitrification loss by photosynthetic oxygen production,which increases nitrification(Xiaet al.,2018).Rapid nitrification and biological assimilation ofdecreasedconcentration in the overlying water and thus reduced NH3volatilization loss but increased the overall N loss.

CONCLUSIONS

In this work,we studied the effects of different fertilization methods on periphytic biofilm development and N cycle in paddy systems.Surface fertilization under light condition promoted the development of periphytic biofilms at the soilwater interface,while deep fertilization under dark condition was unfavorable for the development of periphytic biofilms.Surface fertilization resulted in high N concentrations in the overlying water and topsoil layers,which enhanced NH3volatilization and nitrification-denitrification while inhibited N fixation.The development of periphytic biofilms increased nitrification-denitrification loss but reduced NH3volatilization loss and increased the overall N loss by 3.10%-7.11% in the paddy systems.Therefore,controlling periphytic biofilms in paddy systems might be an effective way of reducing N loss.Deep fertilization can decrease N loss from the paddy system by 20.87%-24.82%,directly by reducing NH3volatilization and denitrification losses and indirectly by affecting periphytic biofilm development.Deep fertilization can be an effective way to increase N use efficiency in the paddy system.

ACKNOWLEDGEMENT

We would like to acknowledge the support from the State Key Development Program for Basic Research of China(No.2015CB158200)and the State Key Laboratory of Freshwater Ecology and Biotechnology of China(No.2019FBZ03).