Impacts of Fertilization Optimization on Ammonia Volatilization, Soil Nitrication, Denitration Intensity From Wheat Fields and Nitrogen Utilization Under Water-saving Irrigation

Scholars have proposed the practice of split N fertilizer application (SNFA), which has proven to be an effective approach for enhancing N use eciency. However, the effect of SNFA on NH 3 volatilization, nitrication and denitration in soil, remain largely unknown. As such, the current study assessed soil NH 3 volatilization, nitrication and denitrication intensities, abundance of nitrogen cycle-related funetional genes, and invertase activity for different treatments. We applied a rate of 240 kg·ha -1 of N, and the following fertilizer ratios of the percent base to that of topdressing under water-saving irrigation: N1 (basal/dressing, 100%/0%), N2 (basal/dressing, 70%/30%), N3 (basal/dressing, 50%/50%), N4 (basal/dressing, 30%/70%), and N5 (basal/dressing, 0%/100%). N3 treatment resulted in a signicant decrease in rate of NH 3 volatilization. This treatment also signicantly reduced nitrication and denitrication intensities, primarily owing to the reduced functional genes abundance involved in the nitrogen cycle (Amoa-AOB, nirK and nirS) and reduced invertase activity (urease, nitrate reductase, nitrite reductase) in wheat-land soil. 15 N tracer studies further demonstrated that N3 treatments signicantly increased the grain nitrogen accumulation by 9.50-28.27% compared with that under other treatments. This increase was primarily due to an increase in the amount of N absorbed by wheat from soil and fertilizers, which was caused by an enhancement in total N uptake (7.2-21.81%). Collectively, these results suggest that the N3 treatment (basal/dressing, 50%/50%) improves N uptake by wheat, reduces the soil NH 3 volatilization rate, and has the potential to reduce the amount of N 2 O generated by nitrication and denitrication. in volatilization in soil amended with nitrogen fertilizer (2) dene the impact of these amendments on nitrication and denitration intensity, as well as the relative abundance of genes for the N cycle in wheat-land soil; characterize the primary properties of soil chemicals for that inuence soil nitrication and denitrication intensity, and the relative abundance of genes involved in the N cycle ; (4) elucidate the manner by which changes in the N cycle and fertility of the soil affect the availability of mineral nitrogen to be taken up by wheat plants. treatment maintains the contents of NH 4+ -N and NO 3− -N in the soil at a low level, and thus, denitrifying bacteria only have a low level of activity. Along with the increased soil denitrication rates following split nitrogen fertilizer application, these results suggest that N3 treatment decreases soil N 2 O emission by reducing nitrication and denitrication rates. Based on these data, we conclude that the optimal split nitrogen fertilizer application ratio decreases soil N losses by decreasing the concentrations of labile N, the activities of N-cycling enzymes, and the abundance of N-cycling key genes, as well as the rates of denitrication in wheat-land.


Introduction
NH 3 volatilization is the major pathway by which nitrogen is lost from wheat cultivation systems (Abdo, 2021). Moreover, the proportion of total nitrogen lost as NH 3 from nitrogen fertilizers varies from 9 to 40% (Fu et al., 2020). One of the major anthropogenic sources of NH 3 release into the atmosphere is agricultural elds, primarilay owing to nitrogen based fertilization and the associated management of this practice. Meanwhile, the primary air pollutants now include atmospheric NH 3 , which is rapidly deposited on the earth's surface within 4 to 5 km of its sources (Behera, 2013). It is, therefore, crucial to manage nitrogen fertilizers in a manner that minimizes its effects on the environment. The simultaneous measurement of NH 3 emissions has the potential to provide valuable information on the processes responsible for their formation, as well as their contribution to both environmental and air pollution. . However, soil nitri cation and denitri cation are highly vulnerable to the environment in the external soil (Neal et al., 2017;Wang et al., 2018). In addition, the effects of nitrogen fertilization on these processes can be modulated by the physiochemical properties of soil, including inorganic nitrogen (Fraser et al., 2017) and N cycle gene abundance and the critical enzyme activity (Neal et al., 2017). Thus, knowledge of the effects of soil physicochemical factors on soil nitri cation and denitri cation is vital to promote the e cient utilization of N in agriculture through the application of various strategies, including the amendment of soils and rational fertilization.
Soil moisture is an important element that affects NH 3 emissions and soil nitri cation and denitri cation. Our previous research proposed a watersaving irrigation technology (WCT), which is based on measuring soil moisture at the key stage of wheat growth (man et al. In this experiment, 15 N isotope tracer technology was used to compare nitrogen uptake and utilization in different split nitrogen fertilizer treatment schemes. To explore the potential effects of split nitrogen fertilizer on NH 3 emission ux and the rates of denitri cation and nitri cation, as well as the utilization to plant nitrogen, a eld experiment was performed in a crop of winter wheat. The goals of this study were as follows (1) determine the primary changes in ammonia volatilization in soil amended with split nitrogen fertilizer (2) de ne the impact of these amendments on nitri cation and denitration intensity, as well as the relative abundance of genes for the N cycle in wheat-land soil; (3) characterize the primary properties of soil chemicals for that in uence soil nitri cation and denitri cation intensity, and the relative abundance of genes involved in the N cycle ; (4) elucidate the manner by which changes in the N cycle and fertility of the soil affect the availability of mineral nitrogen to be taken up by wheat plants.  The eld experiment consisted of ve different split nitrogen fertilizer applications at an application rate of 240 kg·ha −1 (basal/dressing, 100%/0%, 70% /30%, 50%/50%, 30%/70%, 100%/0%; hereafter referred to as N1, N2, N3, N4 and N5, respectively) with a randomized plot design (Table 1). Each application was performed in triplicate, resulting in a total of 15 plots (plot area 20 m 2 ). The rate of fertilizer application that was selected was 240 kg·ha −1 , as it is commonly used by the local farmers. Single superphosphate (P 2 O 5 12%) and potassium chloride (K 2 O 60%) were applied to provide P (P 2 O 5 150 kg·ha −1 ) and K (112.5 K 2 O kg·ha −1 ), respectively. The basal fertilizer was comprised of P and K, while the nitrogen was applied in two split applications. All potash and phosphate fertilizers, as well as the basal nitrogen fertilizer were spread over the soil surface before the wheat was sown.
A rotary cultivar was used to immediately mix the soil to a depth of 20 cm. During the jointing stage, nitrogen fertilizer was applied to create furrows that were immediately covered.
The soil moisture was measured to manage this parameter based on a WCT. The relative water content in the 0~40 cm soil layer was supplemented to 70% at the jointing and anthesis. The amounts of irrigation was calculated using the method of Man (Man et al., 2014). All irrigation processes involved the use of a hose, and the amount of water used to irrigated each event was determined manually and recorded with a water meter. The detailed apply nitrogen fertilizer and irrigation regimes are shown in Table 1. The elds were managed according to the local practices of farming with standard applications of herbicides and pesticides. The solution of NH 4 + -N that had been extracted was analyzed with an AA3 continuous ow analyzer (Bran Luebbe company, Germany). The ux of NH 3 was calculated using the formula described by Yang et al (2020). The cumulative volatilization of NH 3 was calculated as the integral sum of the gas emissions from the stages that were sampled. The NH 3 volatilization factor and yield-scaled volatilization rate were calculated using the following formulas (Yang et al., 2020): Yield-scale NH 3 volatilizaiton (kg N t −1 grain)=N fertilizer /G Where N fertilizer and N control are the total cumulative NH 3 that had volatied during the entire wheat growing season under the treatment with nitrogen fertilizer and the control, respectively. F n is the total nitrogen that had been applied (240 kg·ha −1 ). G is the grain yield.

15 N measurement
We

Soil sample collection and preparation
The soil from each treatment was sampled at the stages of wheat heading, anthesis and maturity . Each soil sample was a mixture of 5 randomly selected locations in a given plot. The soil samples were then passed through a 1 mm sieve for division into three fractions, mixed thoroughly, and stored at 4 ℃ for subsequent microbial biomass carbon or nitrogen (MBC or MBN), and inorganic N content (NH 4 + and NO 3 − ). An aliquot was air dried and urease activity and 15 N abundance were determined through a 1-mm sieve. The last sample was freeze-dried and stored at -80 ℃ for subsequent DNA extraction and real-time PCR analysis.
2.6 Total soil DNA extraction, q-PCR, and cloning of bacterial genes Soil DNA was extracted from each samples using a FastDNA Spin Kit for Soil (MP Biomedicals, LLC., Solon, OH, USA) according to the manufacturer's instructions. The DNA was then stored at -80℃ and analyzed within 3 days. A Nanodrop®ND-2000 UV-vis spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) was used to quantify the DNA and examine its purity. To quantify the abundance of Amoa-AOA, Amoa-AOB, nirS and nirK genes, quantitative poly-merase-chain-reaction (qPCR) assays were performed in triplicate using real-time PCR with a LightCycler 480 (Roche Applied Science). The conditions and primers are given in Table 2. The standard curves for real-time PCR were prepared as previously described (Li et al., 2017). A plasmid that contained 102-109copiesµL − 1 was obtained by serial 10× dilutions. A q-PCR assay was then performed in triplicate to provide an external standard curve for determine the numbers of unknown gene copies. The e ciency for ampli cation of target genes in the assays ranged from 92.3 to 105.2% and the R values were from 0.996 to 0.999. 2.7 Potential nitri cation rates (PNRs) and potential denitri cation rates (PDRs) We soil d −1 . The activities of nitrate reduxtase and nitrite reductase activities were assayed as described by Dominchin (2021). The MBC and MBN were measured by chloroform fumigation-extraction. The MBC was calculated as the difference in DOC levels between the samples that had been fumigated and a control thta had not with an e ciency factor of 0.45. The MBN is calculated as the difference of total extractable nitrogen contents between fumigated soil and unfumigated soil, and the e ciency coe cient is 0.54.

Statistical analyses
The effects of different split nitrogen fertilizer treatments on gene expression and biochemical parameters of key enzymes in nitrogen cycle were analyzed by SPSS v. 18.0 (IBM Corp., Armonk, NY, USA) 3 Results

Ammonia volatilization
The peak of daily NH 3 was detected 2 days after each application of nitrogen fertilizer and then decreased to a relatively low levels 6-7 days after each application of nitrogen fertilizer (Figure 4). During this period, the NH 3 uxes ranged from 0.2 to 3.0 kg N ha −1 d −1 and increased with increasing basal/topdressing nitrogen fertilizer ratio. Furthermore, the NH 3 ux increased with increasing basal nitrogen fertilizer proportion between the sowing stage and jointing stage, as well as with increasing topdressing nitrogen fertilizer proportion from the jointing stage to the end of the experiment. This indicates that different treatments had signi cantly effect on NH 3 volatilization losses from soil.

Soil nitri cation intensity and denitration intensity
Fertilization strategy signi cantly affected soil nitrogen conversion ( Figure 5). At the jointing stage, the nitri cation and denitri cation intensities increased with increasing topdressing nitrogen fertilizer proportion. At anthesis stage, the nitri cation and denitri cation intensities under the N3 treatment were signi cantly lower than those under the N4 or N5 treatments, while the N1, N2, and N3 treatments did not differ signi cantly. At the maturity stage, the split nitrogen treatments did not result in any signi cant differences in nitri cation intensity. Additionally, the denitri cation intensity under the N3 treatment was signi cantly lower than that under N4 or N5 treatment, while that under N1, N2, and N3 treatments did not differ signi cantly.

Abundance of nitrogen cycle functional genes
Q-PCR based on the 16S rRNA gene was used to estimate the abundances of nitrogen cycle functional genes in soil under different split nitrogen fertilizer treatments (Figure 6). During the entire sampling process, changes in the Amoa-AOA, Amoa-AOB, nirK, and nirS counts in each group were similar, showing gradual decreases. The fertilization strategies did not signi cantly affect the Amoa-AOA counts. In all treatments, the Amoa-AOB, nirK, and nirS counts at each stage increased with increasing proportion of topdressing nitrogen fertilizer at the jointing stage. During the anthesis and maturity stages, the Amoa-AOB, nirK, and nirS counts were signi cantly lower under the N3 treatment than those under the N4 or N5 treatments, and the N1, N2, and N3 treatments did not differ signi cantly. Hence, the one-time addition of excess nitrogen fertilization can increase the abundance of key genes for nitri cation and denitri cation. Moreover, a reasonable ratio of basal to topdressing nitrogen fertilizer (N3) can ensure that the numbers of these genes remain at a low level.

Soil nitrogen invertase activity
Differences were observed in soil nitrogen invertase activity among the split nitrogen fertilizer treatments (Figure 7). At the jointing stage, the urease content, nitrate reductase activity, nitrite reductase activity, and protease activity increased with increasing topdressing nitrogen fertilizer proportion. At the anthesis stage, the urease and protease activities under the N3 treatment were signi cantly higher than those under the N1 treatment, whereas the N2, N3, N4, and N5 treatments did not differ signi cantly. Moreover, the activities of nitrate reductase and nitrite reductase under the N3 treatment were signi cantly lower than those under the N4 or N5 treatments, however, no signi cant differences were observed among the N1, N2, and N3 treatments. The changes in soil nitrogen invertase activity during the maturity stage and the soil nitrogen invertase activity in the anthesis stage showed similar trends. These results demonstrate that N3 treatment leads to decreased activity of nitrate reductase and nitrite reductase in the soil, causing low intensity denitri cation.

Soil nitrate accumulation, ammonium nitrogen accumulation, soil microbial biomass N and C
At the jointing stage, the soil MBN, MBC, nitrate accumulation, and ammonium nitrogen accumulation increased with increasing topdressing nitrogen fertilizer proportion (Figure 8). Meanwhile, during the anthesis stage, the soil MBN under the N3 treatment was signi cantly higher than those under the other treatments. The soil MBC under N3 treatment was signi cantly higher than that under N1 or N2 treatment. However, no signi cant difference was observed among the N3, N4, and N5 treatments. Furthermore, the split nitrogen fertilizer treatments did not signi cantly affect nitrate or ammonium nitrogen accumulation. At the maturity stage, the soil MBN and MBC under the N3 treatment were higher than those of other treatments.
Changes in the accumulation of nitrate and ammonium nitrogen showed similar trends.
In addition to the nitrogen obtained from other crops that x nitrogen, the nitrogen in winter wheat primarily originates from the soil and nitrogen fertilizer. The plants treated with N3 absorbed the greatest amount of nitrogen from fertilizers and soil, followed by plants treated with N2 and N4, whereas the plants treated with N1 or N5 had the lowest levels. Moreover, the amount of nitrogen absorbed from basal/topdressing nitrogen fertilizer increased with increasing basal/topdressing nitrogen ratio. These results demonstrate that N3 treatment was more conducive to nitrogen nutrient uptake by wheat plants, resulting in high-yield and high-e ciency.

15 N fertilizer in wheat plant-soil system
We compared and analyzed the ratios of nitrogen fertilizer residue, plant recovery, and potential loss to nitrogen fertilizer application under the ve treatments with split nitrogen (Figure 11). No signi cant difference was observed in the different ratios of nitrogen fertilizer residue to nitrogen fertilizer application. The ratio of plant recovery to nitrogen fertilizer application in N3 increased by 5.16-28.59% compared with those under N1, N2, N4, or N5 treatment. Similarly, under N3 treatment, the ratio of potential loss to nitrogen fertilizer application decreased by 12.81-24.73% compared with those under N1, N4, or N5 treatment. Increases in potential loss indicate that the signi cant loss of nitrogen to the atmosphere occurs through other channels, resulting in environmental issues. nitrogen inputs over the two years. In contrast, it was evident that N3 treatments has the ability to reduce NH 3 losses by 5.22~13.42% of the cumulative NH 3 volatilization, with signi cant reductions of 9.20~21.36% and 9.91~22.78% in both the volatilization of NH 3 and that of yield-scaled NH 3 , respectively, under the N3 treatment over the two years. Hence, an appropriate split nitrogen fertilization scheme may decreases the rate of urea hydrolysis, which could contribute to decreases in NH 3 volatilization.
It is evident that nitrogen loss via NH 3 volatilization increases in conjunction with increasing proportions of basal or topdressing nitrogen . Indeed, the current study was purposefully conducted in this region to demonstrate that NH 3 volatilization from basal nitrogen fertilizer is higher than that from topdressing nitrogen fertilizer (Zhong et al., 2021). This may be due to immediate irrigation following topdressing nitrogen application causing in ltration of fertilizer into the deep soil, with a subsequent reduction in the amount of ammoniacal nitrogen in the top soil layer.
This would consequently lower the losses from the volatilization of NH 3 , which supports the ndings from previous studies (Holcomb et al., 2011). In our study, the peak of daily NH 3 uxes was identi ed 2~3 days after the application of N fertilizer, and they subsequently decreased to relatively low levels 6~7 days after application. These results suggest that the loss of NH 3 primarily occurs during the early period following nitrogen application. In general, the duration of our gas sampling measurements following nitrogen fertilizer application could effectively capture most of the NH 3 volatilization induced by fertilizer application.

The effect of split nitrogen fertilizer on nitri cation intensity
The nitri cation intensity is a metric with the goal of determining the maximum capacity of nitri ers in the transformation of ammonium . Nitrogen fertilizer addition signi cantly alters the nitri cation intensity at the stages of wheat jointing, anthesis and maturity, suggesting that nitrogen fertilizer affects the current season's wheat. The nitri cation intensity at heading is signi cantly and negatively correlated with N uptake by wheat, showed that nitri cation is an critical factor in the growth period of wheat (Yang et al., 2017).
AOA and AOB are two critical groups that participate in nitri cation. The ndings of our study indicate that nitri cation activity was stimulated by large topdressing nitrogen fertilizer proportions and was accompanied by a signi cant increase in the abundance of AOB. AOB can frequently outcompete AOA for the inorganic nitrogen fertilizer (Hink et al., 2017). This competition can include the inhibition of AOA functions and growth, which prefer to use native soil N in contrast to an exogenous N source as a substrate (Fisk et al., 2015). The increase in soil nitri cation could result in loss of nitrogen from agricultural systems and subsequent pollution of groundwater owing to nitrate leaching and denitri cation. Our results also show that larger topdressing nitrogen fertilizer proportions produce a higher average accumulation of nitrate and ammonium in soil than other split nitrogen fertilizer ratios.
We demonstrated the inhibitory effect of the optimum split nitrogen fertilizer ratio on nitri cation intensity from another perspective. The consequences of split nitrogen fertilizer on the microbial community structure in soil nitri cation merit further study. Generally, the primary action of excess topdressing nitrogen fertilizer is to improve urea hydrolysis (Cantarella et al., 2018), and the results of this study suggest that the activity of urease also increases with the increasing topdressing nitrogen proportion.

The effect of split nitrogen fertilizer on denitration intensity
A substantial amount of previous research has shown that total N 2 O emissions positively correlated with soil denitri cation intensity (Wang et al., 1991), which is closely related to the size of the pool labile N forms, such as NH Zhang et al. (2010b) also reported that nirK and nirS are the predominant genes in soil denitri cation, the abundance of which was signi cantly reduced in our study following application of an optimal split nitrogen fertilizer ratio. A possible explanation for this observation could be that N3 treatment maintains the contents of NH 4 + -N and NO 3 − -N in the soil at a low level, and thus, denitrifying bacteria only have a low level of activity. Along with the increased soil denitri cation rates following split nitrogen fertilizer application, these results suggest that N3 treatment decreases soil N 2 O emission by reducing nitri cation and denitri cation rates. Based on these data, we conclude that the optimal split nitrogen fertilizer application ratio decreases soil N losses by decreasing the concentrations of labile N, the activities of N-cycling enzymes, and the abundance of N-cycling key genes, as well as the rates of denitri cation in wheat-land.

The effect of split nitrogen fertilizer on wheat nitrogen utilization by 15 N tracer technique
In the 15 N tracer experiment, separate applications of basal fertilizer 15 N and topdressing fertilizer 15 N were used to overcome an issue with the utilization of traditional fertilizers, which only examine the utilization of the total nitrogen fertilizer during the process of wheat growing. Based on measurements in soil samples and wheat plants, we estimated the applications of basal and topdressing nitrogen fertilizers in the wheat-soil system and the accumulation of nitrogen from basal/topdressing fertilizer in the system of wheat and soil. The results indicated that the N3 treatment resulted in a higher TNAA of wheat to fertilizer 15 N than that under other split nitrogen fertilizer treatments. The TNAA in wheat was increased by 7.20-21.81% relative to that from fertilizer 15 N, and NAAG (9.50-28.27%) was the highest under N3 treatment. In addition, under N3 treatment, the nitrogen accumulation from soil increased by 7.20-27.45%, compared with that under other split nitrogen fertilizer treatments. These results indicate that the N3 treatment contributes to a high accumulation of nitrogen by improving the absorption and utilization of soil and fertilizer nitrogen by wheat (Shi et al., 2012). In fact, a single application of excess fertilization resulted in a soil nitrogen surplus due to a difference in the supply of N supply and demand of the crop (Fageria et al., 2005).
Previous studies have shown that NO 3 -N leaching will lose excess N and pollute the environment (Oborn et al., 2003;Sieling et al., 2006). Our results indicate that the plants accumulate higher amounts of nitrogen when it is applied as topdressig rather than basal fertilizer. Thus, applications of basal N and high-level topdressing with N lead to a surplus of soil N and possibly loss via the leaching of NO 3 -N, the loss of basal N results in a loss of N throughout the entire growing season due to the poor synchrony between the supply of N and the demand of crops. Further evidence suggests that altering the type of N fertilizer and applying it at the optimal rates for fertilization can meet the dual goals of sustaining the accumulation of nitrogen in crops and mitigating the volatilization of NH 3 and greenhouse gases in winter wheat systems.

Conclusion
Compared with current conventional strategies, a cleaner nitrogen fertilization strategy for winter wheat production should decrease wheat-land soil NH 3 volatilization while increasing grain yields and nitrogen accumulation to achieve sustainable agricultural development. In this study, N3 treatment signi cantly decreased soil NH 3 volatilization as well as nitri cation and denitri cation intensities, while increasing nitrogen accumulation in grans in the winter wheat cropping system, resulting in a lower overall environmental burden. Thus, appropriately splitting nitrogen fertilizer applications under water-saving irrigation conditions is an effective fertilization strategy with bene ts for both agronomy and the environment.  Map showing the study site.

Figure 2
Effective precipitation and temperature during wheat growth period. Map showing the micro-plots.

Figure 4
Effect of split nitrogen fertilizer on NH3 ux.
The effects of split nitrogen fertilizer on the soil nitrogen invertase activity.

Figure 8
The effects of split nitrogen fertilizer on the nitrate accumulation, ammonium nitrogen accumulation, soil microbial biomass N and C.

Figure 10
Effect of split nitrogen fertilizer on the plant nitrogen accumulation form fertilizer or soil. NAAG, nitrogen accumulation amount in grains; TNAA, Total plant nitrogen accumulation amount; SN, soil inorganic nitrogen content.

Figure 11
Effect of split nitrogen fertilizer on the fate of 15N fertilizer in plant-soil system.