Nitrogen slow-release behavior of oxamide granules in two different types of paddy soils

Xun LI,Jinlong DONG,Jingjing DUAN,Wenzhong SHEN and Zengqiang DUAN

1State Key Laboratory of Soil and Sustainable Agriculture,Institute of Soil Science,Chinese Academy of Sciences,Nanjing 210008(China)

2School of the Environment and Safety Engineering,Jiangsu University,Zhenjiang 212013(China)

3Agricultural Technology Promotion Center,Taicang Agriculture Committee,Suzhou 215400(China)

ABSTRACT Oxamide is a potential slow-release nitrogen(N)fertilizer,especially under waterlogged conditions,due to its low solubility in water and the slow-release of ammonium by soil amidases.To investigate the effects of oxamide granules(2.00–2.38 mm in diameter)as a single basal fertilizer(180 or 144 kg N ha−1)on rice growth,soil properties,and N use efficiency in terms of N recovery efficiency(NRE),we conducted field experiments on two different types of paddy soils over two rice-growing seasons.Results showed that the fertilization effects of oxamide granules varied between the two types of paddy soils.In the red clayey paddy soil,the grain yields for both rice-growing seasons were high with a significantly higher NRE in the oxamide treatment than in the urea treatment.However,in the alluvial sandy paddy soil,the grain yields in the oxamide treatment were slightly lower than those in the urea treatment.Furthermore,oxamide produced little improvement in NRE in the alluvial sandy paddy soil.Soil incubation experiments over 98 d were also carried out to evaluate the factors affecting the N release behavior of oxamide granules in the two types of paddy soils.We found that the amidase activity was higher and,therefore,the oxamide hydrolysis rate was faster in the alluvial sandy paddy soil,which had a higher soil pH value and organic matter content,compared to the red clayey paddy soil.The faster N release and the longer growth period resulted in a mismatch between N supply by oxamide and rice demand,which,in turn,led to little improvement in NRE and a decreased grain yield in the alluvial sandy paddy soil,especially in the reduced oxamide treatment.These results could help select the appropriate size of oxamide granules for use as a slow-release N fertilizer depending on the soil properties and growth period of rice.

Key Words:amidase activity,grain yield,nitrogen use efficiency,organic matter content,pH,slow-release fertilizer

INTRODUCTION

Rice(Oryza sativaL.)is one of the most important staple food crops in the world and is used to feed more than half of the world population(Muthayyaet al.,2014;Dinget al.,2018).In 2017,China accounted for 18.4%of the rice planting area worldwide and 27.6%of the global rice yield(Crop statistics,2017).However,the average apparent crop recovery efficiency of applied nitrogen(N)for rice production in China is only 28.3%(Miaoet al.,2011),which is much lower than the world average value(46%)(Ladhaet al.,2005).The main reason for the low N use efficiency(NUE)is the over use and improper application of N fertilizer,which is consequently lost through surface runoff,nitrate leaching,NH3volatilization,N2O emission,etc.(Xing and Zhu,2000;Lin D Xet al.,2007;Hayashiet al.,2008;Juet al.,2009).Excessive N inputs increase fertilizer and labor costs,and have caused a series of environmental problems.Therefore,many N fertilizer saving and enhanced-efficiency strategies have been developed(Akiyamaet al.,2010;Spiertz,2010;Denget al.,2012;Linquistet al.,2013).Among them,the use of slow-release N fertilizers,which can release N synchronized with crop demand and decrease environmental N losses,is an effective and promising approach(Shaviv and Mikkelsen,1993;Wanget al.,2007;Azeemet al.,2014).

Oxamide is a diamide of oxalic acid and contains 31.8%N.The amidogen connects a carbonyl group with an amido bond in oxamide,so oxamide can be hydrolyzed by amidase to liberate ammonium,which is similar to urea(Cantarella and Tabatabai,1983).The distinct difference between oxamide and urea is that the former is slightly soluble in water(4 g L−1at 25◦C)(DeMentet al.,1961).Therefore,the ammonium release process is much slower when oxamide,especially granular oxamide,is dispersed in soil compared to urea(Dilz and Steggerda,1962;Prasadet al.,1971).Most commercial slow-release N fertilizers are polymer-coated urea,but the polymer-coating can break down during longterm immersion in water,and the fertilizers fail to release N slowly(Azeemet al.,2014).However,because of the low oxamide solubility in water,the hydrolyzation of oxamide toammonium is determined mostly by amidase in both upland and wetland soils,which makes oxamide a promising slowrelease N fertilizer especially under waterlogged conditions.Since the late 1950s,the performance of oxamide used as a slow-release N fertilizer has been evaluated for some plants,such as rice,maize,wheat,sorghum,soybean,spinach,grass,woody plants,etc.(DeMentet al.,1961;Engelstadet al.,1964;Westermanet al.,1972;Rajale and Prasad,1973;Kuyper and Lambeth,1980;Sartain and Ingram,1984;Rubio and Hauck,1986;Mosdellet al.,1987;Schuler and Paulsen,1988;Miahet al.,1998).However,the high manufacturing cost of oxamide has limited its practical application as a fertilizer in agricultural production in the past.Accordingly,most previous studies on the slow-release N behavior of oxamide were conducted indoors or under pot culture conditions,whereas there have been few long-term field experiments.

Oxamide can be easily synthesized by reaction of aqueous ammonia with dimethyl oxalate(Dilz and Steggerda,1962;Hauck,1994).Nowadays,it is possible to efficiently and economically synthesize large quantities of the precursor,dimethyl oxalate,from coal in China(Zhaoet al.,2004,2005;Lin Qet al.,2007;Wanget al.,2017).Therefore,the manufacturing cost of oxamide has dramatically declined and is comparable to that of urea.This presents opportunities for the long-term field application of oxamide as a commercial slow-release N fertilizer.Furthermore,due to the great diversity of parent materials,there are various types of paddy soils in China.Consequently,two questions arise.To what extent is the NUE of rice improved by the application of oxamide?What are the major factors affecting the N release behavior of oxamide in different types of paddy soils?

The objectives of this study were:i)to investigate the effects of oxamide application on rice growth,soil properties,and NUE for rice growing on two different types of paddy soils over two seasons in the field experiments and ii)to evaluate the soil factors affecting the N release behavior of oxamide in a series of soil incubation experiments.

MATERIALS AND METHODS

Field experiment location and soil properties

Field experiments were conducted in two typical paddy soils in the middle and east of China.One experiment was located in Yujiang(28◦11′56′′N,116◦55′23′′E)in Jiangxi Province and the other was in Taicang(31◦31′22′′N,121◦05′37′′E)in Jiangsu Province.The soil in Yujiang was a paddy soil derived from Quaternary red clay(red clayey paddy soil)with a pH of 6.26measured at a soil:water ratio of 1:2.5(weight:weight),an organic matter content of 10.79 g kg−1,a total N content of 1.01 g kg−1,an alkali-hydrolyzable N content of 88.9 mg kg−1,an available phosphorus(P)content of 12.9 mg kg−1,and an available potassium(K)content of 113.7 mg kg−1in the topsoil(0–20 cm)at the beginning of the experiment.The soil in Taicang was a paddy soil derived from alluvial sand of the Yangtze River(alluvial sandy paddy soil)with a pH of 8.08(soil:water,1:2.5,weight:weight),an organic matter content of 15.07 g kg−1,a total N content of 1.43 g kg−1,an alkali-hydrolyzable N content of 122.1 mg kg−1,an available P content of 17.6mg kg−1,and an available K content of 118.3 mg kg−1in the topsoil(0–20 cm)at the beginning of the experiment.

Designs of field experiments

The crop succession was double-cropping rice(Oryza sativaL.)in Yujiang.The early-season rice was transplanted on May 12 and harvested on July 22,2017,and the lateseason rice was transplanted on July 29 and harvested on October 27,2017.The average soil temperature(at 10 cm depth in topsoil)was 24.8 and 24.4◦C in the early and late rice seasons,respectively.The crop succession was a rice-wheat(Triticum aestivumL.)rotation in Taicang.The rice was transplanted on June 16and harvested on October 21 in 2016,and transplanted on June 19 and harvested on October 30 in 2017.The average soil temperature(at 10 cm depth in topsoil)was 25.1 and 24.7◦C in 2016and 2017 rice seasons,respectively.

Both experiments in Yujiang and Taicang were designed as a randomized complete block(7.5 m×8.8 m for each plot)with three replicates and consisted of four treatments:i)control with no N fertilizer(CK),ii)urea(46%N)treatment at 180 kg N ha−1(Ur),iii)oxamide(31.8%N,granules with a diameter range of 2.00–2.38 mm)treatment at 180 kg N ha−1(Ox),and iv)reduced oxamide treatment at 144 kg N ha−1(OxR).Half of the urea was applied as a basal dressing and mixed with top soil one day before the rice was transplanted,while the rest of the urea was applied as two top dressings,once at the beginning of the tillering stage(30%)and the other at the beginning of the booting stage(20%).All the oxamide granules were basally applied and mixed with top soil one day before the rice was transplanted.All treatments also received 75 kg P2O5ha−1and 50 kg K2O ha−1.The field was alternately wetted and dried until two weeks before harvest,and the water level was allowed to fluctuate between 0 and 5 cm above the soil surface.Weeds were controlled by hand weeding,and insects and diseases were intensively controlled by chemicals.

Sampling and measurements in field experiments

At harvest,the crop was manually reaped from the entire area of each plot,and the grain was threshed and dried by machine to determine the yield.Five bundles of rice plants were randomly taken from each plot,mixed and divided into two parts(grain and straw),and then dried to determine thedry weight and the N content.The dry plant samples were ground to pass through a 0.5-mm sieve and digested with H2SO4-H2O2(Thomaset al.,1967).Total N contents in the grain and straw samples were measured using a discrete wet chemistry analyzer(SmartChem 200,Westco Scientific Instruments,Brookfield,USA).

Before fertilization and after harvest,five columns of topsoil(0–20 cm)were randomly taken from each plot,mixed and air-dried,and then ground through a 0.15-mm sieve.Soil pH was measured at a soil to CO2-free distilled water ratio of 1:2.5(weight:weight)using a pH meter(PHS-3C,Kangyi Instruments Co.,Ltd.,China).Soil organic matter was determined using the wet oxidation method with K2Cr2O7and concentrated H2SO4(Nelson and Sommer,1975).

Soil incubation experiments

Paddy soil samples were collected from the two sites of the field experiments before fertilization in Yujiang and Taicang.Briefly,100 g air-dried topsoil(0–20 cm for both red clayey and alluvial sandy paddy soils)was mixed with oxamide granules(2.00–2.38 mm in diameter)at a rate of 75 mg N kg−1soil.The amount of oxamide used was equivalent to the oxamide treatment(180 kg N ha−1)applied in the field experiments described above(assuming the bulk density of topsoil is 1.2 g cm−3).The mixtures were then transferred to a black 150-mL plastic cup covered with aluminum foil,and the soil moisture was adjusted to 100%water content(soil:water,1:1,weight:weight)every 3 d using distilled water.The incubator temperature was set at 25◦C based on the average top soil temperature of the field experiments.After 3,7,14,21,28,42,56,70,84,and 98 d,five cups of each type of soil were randomly removed.Then,all the visible oxamide granules remaining in the soil were picked out.After that,the soil was extracted with 2 mol L−1KCl at a soil:KCl solution ratio of 1:5(weight:weight),shaken for 1 h,and filtered to determine NH+4-N and NO−3-N contents using a continuous-flow injection analyzer(AA3,Bran and Luebbe,Germany).Net amounts of NH+4-N and NO−3-N released from the oxamide in each soil sample were calculated after subtracting the amounts in the control soil without oxamide.

After 7,28,and 98 d,the amidase activity(EC 3.5.1.4)in soil was also determined by a modified method described by Frankenberger and Tabatabai(1980a)and Dodor and Tabatabai(2003).The methods were based on the determination of NH+4released when soil was incubated with formamide without pH buffer at 25◦C for 2 h.The NH+4released was extracted with 2 mol L−1KCl containing Ag2SO4(to stop enzyme activity),followed by filtering and determination using a continuous-flow injection analyzer(AA3,Bran and Luebbe,Germany).Calculations and statistical analyses

The apparent NUE presents as N recovery efficiency(NRE)in rice aboveground biomass.Nitrogen accumulation in rice grain(GN)and straw(SN),total N accumulation(TNA),and NRE were calculated as follows:

whereCN-grainandCN-straware the N contents in rice grain and straw,respectively,and TNATand TNACKare the TNA of the treatments with and without N fertilization,respectively.

The data from two rice-growing seasons or years were statistically analyzed using the one-way analysis of variance(ANOVA)to compare differences between different treatments,and the means were separated using the least significant difference(LSD)test at a significance level ofP<0.05 by Duncan’s multiple comparison test.All statistical analyses were carried out using SPSS 22.0 software(IBM SPSS Statistics,USA).The figures were generated using Origin Pro 8.0 software(OriginLab Corp.,USA).

RESULTS

Rice yield,N content,and NREin field experiments

Rice yield was significantly affected by N fertilization,with different trends for the two types of paddy soils(Fig.1).Both urea and oxamide fertilization led to significantly higher rice yield compared to CK in the two rice-growing seasons for the two types of paddy soils.The grain yields of both the early-season and late-season rice grown on the red clayey paddy soil maintained a high level in the oxamide or even the reduced oxamide treatment as that in the urea treatment.However,the grain yield showed a declining trend in the oxamide treatment and was significantly lower(11.4%)in the reduced oxamide treatment compared to the urea treatment in the second rice season for the alluvial sandy paddy soil.

Fig.1 Rice grain yields under different fertilization treatments in the field experiments conducted on a red clayey paddy soil(a)and an alluvial sandy paddy soil(b).CK=control with no N fertilizer;Ur=urea treatment;Ox=oxamide treatment;OxR=reduced oxamide treatment.Error bars indicate standard errors of the means(n=3).Means with different letters are significantly different at P<0.05 for the same rice-growing season.

For both types of paddy soils,grain N content maintained or even increased after applying oxamide as a single basal fertilization compared to the split fertilization of urea(Fig.2).There was no difference in straw N content between the oxamide and urea treatments.

Fig.2 Nitrogen contents in rice grain(a and c)and straw(b and d)under different fertilization treatments in the field experiments conducted on a red clayey paddy soil(a and b)and an alluvial sandy paddy soil(c and d).CK=control with no N fertilizer;Ur=urea treatment;Ox=oxamide treatment;OxR=reduced oxamide treatment.Error bars indicate standard errors of the means(n=3).Means with different letters are significantly different at P<0.05 for the same rice-growing season.

For the red clayey paddy soil,the total amount of N accumulated in rice aboveground biomass was the highest in the oxamide treatment,although the difference was not significant for the late-season rice(Table I).However,the differences in the accumulated N between the oxamide andurea treatments were not significant for the alluvial sandy paddy soil in both 2016and 2017.Moreover,the accumulated N decreased by 5.8% and 8.8% in the reduced oxamide treatment when compared with the oxamide treatment in the alluvial sandy paddy soil in 2016and 2017,respectively.

In the red clayey paddy soil,NRE was significantly higher in the oxamide treatment than the urea treatment with the same N application rate,and the highest NRE always occurred in the oxamide or reduced oxamide treatment.The NRE increased by 28.5%and 22.1%in the early-season rice for oxamide and reduced oxamide treatments when compared with the urea treatment,respectively,and increased by 18.3%and 48.1% in the late-season rice,respectively.However,in the alluvial sandy paddy soil,oxamide application did not improve NRE,and there was little difference in NRE between oxamide and urea treatments over the two ricegrowing seasons.

Changes in soil properties in the field experiments

Over the two rice-growing seasons,soil pH decreased significantly in the fertilized treatments in the two types of paddy soils(Fig.3).The magnitude of this effect was less for the oxamide treatment than the urea treatment and the least for the reduced oxamide treatment.Specifically,in the oxamide and reduced oxamide treatments,the pH values decreased from 6.4 to 5.9–6.1 and from 8.1 to 7.5–7.7 afterthe two rice-growing seasons in the red clayey paddy soil and the alluvial sandy paddy soil,respectively.However,the pH values decreased to 5.8 and 7.3 in the urea treatment in the two paddy soils,respectively.Soil organic matter increased over the two rice-growing seasons,although fertilization effects were not significant.In both the urea and oxamide treatments,soil organic matter content increased at a similar rate after fertilization and cultivation with rice.In the red clayey paddy soil,soil organic matter content increased from 10 to 15 g kg−1,whereas it increased from 15 to 19 g kg−1in the alluvial sandy paddy soil.

Fig.3 Changes in soil pH(a and c)and organic matter(b and d)before and after each rice-growing season under different fertilization treatments in the field experiments conducted on a red clayey paddy soil(a and b)and an alluvial sandy paddy soil(c and d).CK=control with no N fertilizer;Ur=urea treatment;Ox=oxamide treatment;OxR=reduced oxamide treatment.Error bars indicate standard errors of the means(n=3).Means with different letters are significantly different at P<0.05.

TABLE ITotal N accumulated by rice aboveground biomass and N recovery efficiency of rice under different fertilization treatments in the field experiments conducted on a red clayey paddy soil(RC)and an alluvial sandy paddy soil(AS)

Nitrogen release and amidohydrolase activity in soil incubation experiments

Soil incubation experiments revealed more details about the N release behavior of oxamide granules.Net NH+4-N content initially increased with incubation time,reached a peak after around 28 d,but declined afterwards in the alluvial sandy paddy soil(Fig.4a).The appearance of this peak was delayed to 42 d of incubation in the red clayey paddy soil,and the peak value was lower than the alluvial sandy paddy soil.A similar,but more gradual pattern was found for net NO−3-N content(Fig.4b).The maximum NO−3-N content occurred at around 56d of incubation in the alluvial sandy paddy soil,and the peak occurred almost 28 d later in the red clayey paddy soil.Soluble N(NH+4-N and NO−3-N)content is shown in Fig.4c.The peak value occurred at around 28 d of incubation in alluvial sandy paddy soil,whereas it occurred at 56–70 d of incubation in the red clayey paddy soil.Moreover,the amidase activity was about 35.1%–77.6%higher in the alluvial sandy paddy soil than the red clayey paddy soil throughout the entire incubation period(Fig.4d).

Fig.4 Changes in net contents of soil NH+4-N(a),NO−3-N(b),and NH+4-N+NO−3-N(c)and amidase activity(d)with time in soil incubation experiments on a red clayey paddy soil and an alluvial sandy paddy soil mixed with oxamide granules.Error bars indicate standard errors(n=5)of the means.Asterisks***indicate significant differences at P<0.001 between the two types of paddy soils.

DISCUSSION

Oxamide has been recommended as a potential slowrelease N fertilizer for more than half a century due to its very low solubility in water and the slow-release of ammonium in soil(DeMentet al.,1961;Dilz and Steggerda,1962;Engelstadet al.,1964).However,high price limited its use in field experiments as a fertilizer on a large scale in early years;therefore,most previous experiments on oxamide were conducted under pot culture conditions(DeMentet al.,1961;Dilz and Steggerda,1962;Engelstadet al.,1964;Beatonet al.,1967;Rajale and Prasad,1973;Cantarella and Tabatabai,1983).Rubio and Hauck(1986)found that rice grain yield and NRE increased in the oxamide treatment compared to the urea treatment in pot experiments using15N-labelled fertilizer.A lysimeter experiment showed less N leaching,higher total residual fertilizer N,and improved NRE in the oxamide treatment than in the urea treatment after growing turf grass for a long period of time(>40 d)(De Nobiliet al.,1992).In rhizobox experiments,higher NO−3content,increased rooting density,and better wheat growth were obtained in the oxamide treatment compared to the(NH4)2SO4treatment(Miahet al.,1998,2000).In this study,delays in the NH+4-N and NO−3-N increases were observed in soil incubation experiments(Fig.4a–c),which confirmed that oxamide was an efficient slow-release N fertilizer.

This study also demonstrated that the application of urea and oxamide both resulted in a decrease in soil pH value,but the decline was smaller in the oxamide treatment(Fig.3).Due to the uptake of excess cations over anions by rice plants,the net removal of basic cations by rice grain and straw from soil,nitrification of NH+4,and loss of NO−3with a basic cation by leaching and runoff,the paddy soil will be acidified after fertilization with urea(Bolanet al.,1991;Shenet al.,2007;Guoet al.,2010;Kumaret al.,2018).Oxamide releases NH+4-N slowly,resulting in a low risk of supplying N in excess of rice demand.Therefore,slower nitrification of NH+4and reduced loss of NO−3through leaching and runoffwere observed in the oxamide treatment than the urea treatment,similar to the results on polymer-coated urea treatments,leading to a small decrease amplitude of soil pH(Miahet al.,1998;2000;Genget al.,2015;Sunet al.,2019).Soil organic matter content increase in the N fertilizer treatments was attributed to greater root exudate and stubble and root residues due to better rice growth(Mastoet al.,2006;Shenet al.,2007;Zhanget al.,2018).Since the stimulating effects of urea and oxamide on rice growth and yield were similar,there was little difference in the amplitude of the increase in soil organic matter between the urea and oxamide treatments.

There have also been a series of plot experiments to evaluate oxamide as a slow-release N fertilizer for turf grass.Higher NRE,better visual turf quality,and higher clipping yield were obtained in the treatment applied with oxamide granules after several years(Westerman and Kurtz,1972;Westermanet al.,1972;Allenet al.,1973;Landschoot and Waddington,1987;Mosdellet al.,1987;Wehner and Martin,1989).However,the plot experiments usually focused on high value crops,with the sizes of 1.0–5.5 m2for each plot.Results from larger-size plot experiments with oxamide,especially for food crops,have rarely been documented.The only field experiment with a plot size of 50 m2that investigated oxamide application effect on rice growth was conducted in our previous study(Tanget al.,2018).In that study,we found gaseous N loss(NH3volatilization)was reduced by 38.3%–62.7% in the oxamide treatment when compared with the urea treatment,accounting for only 4.1%–26.4%of the total N application.Therefore,the use of oxamide as a slow-release N fertilizer could reduce N lossviaNH3volatilization.

In this study,rice was grown in plots with a size of 66m2over two rice-growing seasons in the field experiments.In the red clayey paddy soil,the grain yield in the oxamide treatment was as high as that in the urea treatment,and the NRE of rice could be increased by more than 10 percentage points compared to the urea treatment(Table I).The results showed that a single basal fertilization with oxamide can increase NRE,maintain a steady rice grain yield,and save labor costs.Furthermore,the N release characteristics are synchronized with rice uptake,indicating that the improvement in the NRE of rice after oxamide application is similar to other commercial slow-release N fertilizers.In previous studies,slow-release N fertilizers,such as polyurethanecoated urea,thermoplastic resin-coated urea,cross-linked polyacrylamide-coated urea,urea formaldehyde,etc.,could increase the NRE of rice by 2%–18%compared to the urea treatment(Yanget al.,2012;Liet al.,2017;Huanget al.,2018;Miet al.,2018).The significant advantage of oxamide over other coated slow-release N fertilizers is that the decomposition products of oxamide in soil are ammonium and oxalic acid,without harmful or degradation-resistant materials left in soil(DeMentet al.,1961).Furthermore,oxamide costs less than other commercial slow-release N fertilizers recently,which makes it a promising slow-release N fertilizer for food crops.

However,as reported in our previous and present studies,the improvement in the NRE of rice caused by oxamide granule(2.00–2.38 mm in diameter)application was not significant in the alluvial sandy paddy soil(Tanget al.,2018).This indicates that the process of N release from oxamide is influenced by many factors.As the hydrolysis of oxamide is largely carried out by soil microbes,larger-size oxamide granules with a small specific surface area exhibit a slower N release behavior(DeMentet al.,1961;Engelstadet al.,1964).Engelstadet al.(1964)also found that oxamide granules with a higher degree of compactness and low porosity were less pervious to water and microbes,resulting in slower hydrolysis rates.However,less information is available on the effect of other factors,especially soil properties,on the N release behavior of oxamide.

Amidases(EC 3.5.1.4)are responsible for the hydrolysis of oxamide in soil.Therefore,factors influencing amidase activity also affect oxamide hydrolysis rate(Frankenberger and Tabatabai,1980a;Cantarella and Tabatabai.1983).Temperature is one of the most important factors that influence the enzyme-catalyzed reactions,and amidase activity has been found to increase with increasing temperature from 10 to 60◦C with an optimum temperature of 60◦C(Frankenberger and Tabatabai,1980a,1985).It has also been reported that the optimum temperature for amidase by mesophiles ranges from 25 to 30◦C,while the optimum range for amidase by thermophiles is 50–65◦C(Sharmaet al.,2009).Although the straight-line distance between the two field experimental sites in this study was more than 500 km,the average soil temperature(at 0–10 cm depth in topsoil)was similar for the rice-growing season,indicating that temperature was not the important factor involved in causing the significant difference in amidase activity between the two types of paddy soils in this study.The distinct differences between the two types of paddy soils attributed to pH value and organic matter content,both of which were significantly higher in the alluvial sandy paddy soil than in the red clayey paddy soil.The affinity constant of enzyme for a substrate is calculated as the reciprocal ofKm,which is the substrate content at which half the maximum catalytic reaction rate is reached.Therefore,a high affinity suggests that a high enzyme catalyzed reaction rate can be maintained,even if the substrate content is very low,and it also indicates a high catalytic efficiency for the substrate.Affinity constants of amidase for formamide,acetamide,and propionamide all reached their highest values at pH 8.5,where the highest amidase activity also occurred(Frankenberger and Tabatabai,1980a,b).It was reported that amidase activity was significantly positively correlated with soil pH(r=0.61,P<0.001)in the pH range of 4.6to 7.0(Acosta-Martinez and Tabatabai,2000).In this study,the pH value of the alluvial sandy paddy soil fluctuated from 7.5 to 8.1,whereas that of the red clayey paddy soil fluctuated from 5.9 to 6.4(Fig.3).Therefore,it is reasonable to conclude that the higher amidase activity in the alluvial sandy paddy soil is mainly due to the more favorable pH condition when compared with the red clayey paddy soil(Fig.4d).Furthermore,soil organic matter serves as both a substrate and an energy source.Higher level of organic matter stimulates microbial activity and,therefore,enzyme synthesis.It has been reported in many studies that amidase activity increases as the increase of organic matter content(Frankenberger and Tabatabai,1981;Dicket al.,1988;Dodor and Tabatabai,2003).In this study,soil organic matter content was much higher in the alluvial sandy paddysoil(15–19 g kg−1)than in the red clayey paddy soil(10–15 g kg−1)(Fig.3).Therefore,the higher organic matter content was also an important factor resulting in higher amidase activity in the alluvial sandy paddy soil.Higher amidase activity led to a faster hydrolysis of oxamide granules with the same diameter.

Nitrogen release behavior of oxamide in soil incubation experiments showed that the increase in NH+4-N content before the 28–42 incubation day period was caused by the hydrolyzation of oxamide,and the following decrease was caused by nitrification and the depletion of oxamide(Fig.4).As the peak of NH+4-N content appeared earlier in the alluvial sandy paddy soil than that in the red clayey paddy soil,it can be concluded that the hydrolyzation of oxamide was faster in the former soil.The increase in NO−3-N content was caused by the nitrification of sufficient NH+4during the earlier incubation stage,while the decrease was caused by denitrification loss and the depletion of NH+4.The faster hydrolyzation of oxamide in the alluvial sandy paddy soil could also be inferred from the earlier appearance of the NO−3-N peak.

It has been reported that the N uptake rate of transplanted rice increased from midtillering,reached a maximum during panicle initiation,and then decreased after heading(Yinget al.,1998;Yuet al.,2013;Zhanget al.,2013).Therefore,in a double-cropping rice system with a rice growth period of 70–90 d in the red clayey paddy soil,oxamide granules could supply sufficient N to match the keen N demand for rice plants from midtillering(20–24 d after transplanting(DAT))to the heading stage(50–70 DAT).In a rice-wheat rotation system with a rice growth period of 120 to 130 d in the alluvial sandy paddy soil,the plant available N released from the oxamide granules was synchronized to rice uptake from midtillering(25–32 DAT)to the panicle initiation stage(55–65 DAT).After that,the plant available N content in the alluvial sandy paddy soil was less than the red clayey paddy soil.Therefore,N impoverishment was more likely to happen at and after the heading stage(70–85 DAT).The faster N release and longer growth period resulted in the mismatch of N supply by oxamide and rice N demand,which in turn caused little improvement in NRE and a decreased grain yield in the alluvial sandy paddy soil especially in the reduced oxamide treatment(Fig.1,Table I).

CONCLUSIONS

In this study,oxamide granules showed excellent N slowrelease behavior and could be used during rice cultivation as a slow-release N fertilizer.Intriguingly,fertilization effects of oxamide granules varied in the field experiments conducted on different types of paddy soils.For the red clayey paddy soil,the grain yields of both early-season and late-season rice were higher with a significantly higher NRE in the oxamide treatment than in the urea treatment.However,for the alluvial sandy paddy soil,the rice grain yield in the oxamide treatment was slightly less than that in the urea treatment,and little improvement in NRE was obtained by oxamide fertilization.In soil incubation experiments,soil pH value and organic matter content were found to be the two critical factors influencing the amidase activity and concomitant hydrolysis rate of the oxamide granules.Those results also implied that the N release patterns of oxamide granules with the same size might be quite different due to the distinct amidase activity.Therefore,before applying oxamide granules as slow-release N fertilizer,soil properties and amidase activity and N release patterns of different sized oxamide granules should be evaluated.The length of rice growth period should also be carefully considered.The optimum size of oxamide granules needs to be determined to ensure the N release pattern well synchronized with rice N uptake.For example,due to the high amidase activity,larger-size oxamide granules with a longer release period are more suitable for application in the paddy soil with higher soil pH,organic matter content,or temperature.The well matching N supply by oxamide to rice demand could decrease environmental loss of N and help maintain a high grain yield,which,in turn,improves the NRE of rice.

ACKNOWLEDGEMENTS

The authors thank the National Key Research and Development Program of China(No.2017YFD0800103)and the Strategic Priority Research Program of Chinese Academy of Sciences(No.XDA23020401),for funding support.The authors also thank the Jiangsu Danhua Group Co.,Ltd.(Danyang,China)for providing the oxamide fertilizer.