20 Years nitrogen dynamics study by using APSIM nitrogen model simulation for sustainable management in Jilin China

The tremendous increase in industrial development and urbanization has become a severe threat to the Chinese climate and food security. The Agricultural Production System Simulator model was used to simulate soil nitrogen in black soil in Yangling Jilin Province for 20 years. The observed values are consistent with the simulated values. The predicted values of total soil NO3−–N and NH4+–N nitrogen are 10 kg ha−1 and 5 kg ha−1 higher than the observed values. The total soil NO3−–N loss has the same trend as the rainfall, and it increases with the number of rainfall days over the years. The average 20 years losses of NO3−–N and NH4+–N observed were 1375.91 kg ha−1, and 9.24 kg ha−1, while in the simulation increase was 1387.01 kg ha−1 and 9.28 kg ha−1, respectively. The difference between the observed and simulated values of NO3−–N and NH4+–N of mean loss was 11.15 kg ha−1 and 0.04 kg ha−1 respectively. Moreover, our findings highlight the opportunity further to improve management policies (especially for nitrogen) to maintain crop yield.

crop yield 8,9,[11][12][13] . Soil type and topography are relatively constant within time, while land use and vegetation are influenced by anthropogenic activities and climate 14 . Therefore, China had big challenge with food security 15 . Agricultural production plays a significant role in global climate change 16 . China's food production is highly dependent on irrigation, but water scarcity has challenged irrigated agriculture 17 . Agricultural production has a significant impact on global climate change 16 . Approximately 65% of global nitrous oxide (N 2 O) emissions are from agricultural soils. this primarily comes from nitrogen (N) fertilizer, which is misused at the national level and the optimization of N management, therefore, a target for sustainable agriculture 18 . Furthermore, large inputs of mineral N fertilizer above crop demands can result in low N use efficiency and lead to several negative impacts on the environment, such as surface water eutrophication, groundwater nitrate pollution, soil acidification, and emitting greenhouse gases. Such environmental problems are getting worse because mineral N fertilizer will increase predictably in the future 19 .
APSIM (Agricultural Production Systems Simulator) is a software system that provides a versatile framework for simulating climate and soil management effects on crop growth in agricultural systems and soil resource changes. APSIM's predictive accomplishment for soil water and nitrate nitrogen simulation in contrasting soils (clay loamy) and environments 20 . APSIM allows modular configuration of crop models, pastures, soil water, nutrients, and erosion to simulate the different production systems 21 . Many variables, including weather fluctuations, will restrict producers' revenues and reduce yield by moderate inputs 22 . A key feature of APSIM, which differentiates it from other models, is the soil's central position rather than the crops. In response to management and weather, alteration in the status of soil state variables are simulated crops constantly come and go, find the soil in a given state and leave it in a changed state. Another aspect is its integrated structure: high order mechanisms (i.e., soil water balance, crop production and soil N dynamics) as distinct modules. Several processbased models are in use or are being developed, for example, APSIM 23 , DAISY 24 , DNDC 24 , and WNMM 25 . These models present the ability to examine the role of particular process towards system outputs. They help to understand better that how environmental conditions shared with management strategies interrelate to control N cycling and losses. However, the conceptualization of N transformations in APSIM and DNDC models are different. DNDC is a microbial growth model. APSIM defines denitrification and nitrification processes using equations of empirical reaction conveyed using Michaelis Menten type equation 26 . APSIM model correlated well with measurements (r = 0.97), while NZ-DNDC performed well on the Otago soils (r = 0.83 and 0.92 for Wingatui and Otokia, respectively 24 . Much of the code that comprises the APSIM SOILWAT and soil nitrogen has developed from prior experiences with, firstly, models of the CERES family, notably CERES-Maize 27 and, secondly, PERFECT 28 was primarily established to create the effects of erosion on productivity of vertisols in the Australian subtropics but did not address N. PERFECT involved schedules for simulating impacts of surface residues on soil runoff and evaporation. Surface residues decay was modelled as a simple time function. The CERES models treat and residues from previous crops as integrated into the soil and therefore cannot involve any effect on soil water balance from surface residues. CERES deals with nitrogen, and the simulation of fresh residue decomposition takes into account the residue ratio C: N and the moisture and temperature environmental factors. These models' code had been re-engineered into discrete modules. APSIM was a broadly used model of the agro-ecosystems. It stimulates plant growth on a daily basis, based on incoming solar radiation and depends on both water content and Nitrogen supply. In APSIM the standard model for soil, water uses a tipping bucket approach 29 . It simulates nitrate movement linked to the vertical water flow, and eventually drainage and leaching. Evapotranspiration potential is calculated using Priestley Taylor 27 and the runoff is calculated using an approach to the USDA soil curve number.
This study aimed to study the dynamics of nitrogen fertilization and its effect on nitrification and denitrification. The vertical movement of nitrate in soil was investigated. To explore the novelty of the simulation, to observe and predict future nitrogen loss under the same conditions of nitrate content change, to reduce the 20-year environmental consequences and its impact on nitrogen loss, the nitrogen model was calibrated and validated. New to this study is examining nitrogen dynamics and its response to the environment and the use of the APSIM nitrogen model.

Materials and methods
Site description and experimental data. Data for the study was measured from the long-term soil monitoring research station in Gongzhuling-Jilin, China. The province is located between (43° 30′ 23′ N, 124° 48′ 33′ E) latitude and longitude while the altitude of (220 m) the study area in China. Jilin's mean annual temperature and precipitation were found as 4-5 °C & 500 mm with cool temperate, highly continental, and 13.5 °C & 1000 mm respectively. The crop was maize and which holds a relatively large area and main crop. The soil type is clay loam, and pH was 7.6. The organic carbon C (g kg −1 ), Total N (g kg −1 ), Total P (mg kg −1 ), Available N (kg ha −1 ), Available P (mg kg −1 ), Available K (mg kg −1 ) were 20, 1.34, 0.546, 15.3, 10, 119. Soil pH was measured in a 1:5 soil: ratio, the organic carbon g kg −1 was measured by the walkley wet combustion method 30 , total N measured by the Kjeldahl method 31 , available P with 0.5 M NaHCO 3 32 , and exchangeable K with N ammonium acetate 33 . The long-term experimental sites in Gongzhuling, Jilin, China was measured from 1990 to 2010. To check the -long-term status on the soil nitrogen dynamic and s improve this soil better for estimating the pollution level and crop productivity. A composite soil was selected for the soil sampling, and the soil was sampled in October every year, up to the depth of 20 cm, and for APSIM modeling, simulation depth was kept at up to 180 cm. The APSIM soil parameters include soil property and hydrodynamic values; these values were calculated by the formula. These values were used by APSIM nitrogen model to simulates soil properties and nutrient cycling by each soil layer. APSIM model overview. APSIM includes a series of modules that simulate biological and physical processes in farming systems, including crop growth, soil water, and carbon and nitrogen dynamics due to climate variation and interaction 20 . This analysis performed in APSIM version 7.5. Maize crop production is measured as a function of photoperiod-modified thermal time ( Table 1). The Agricultural Production Systems Simulator (APSIM) is a modular modeling framework that has been developed by the Agricultural Production Systems research unit in Australia 36 . Four modules, a soil-water module (SOILWAT2), the soil nitrogen module (SOILN2) and the fertilizer module (FERTILIZ), and a specific crop module (APSIM-maize), were linked within APSIM to simulate the cases described in this paper (Table 2). APSWIM is based on the numerical solution of the equation of Richards coupled with the formula of convection dispersion to the movement of the model solute. The APSIM model's application is based on the 'standalone' SWIMv2.1 (soil water infiltration and movement) 37 . Soil water properties parameterization for APSWIM includes specification of the relationships of moisture characteristics and hydraulic conductivity in each soil layer. Runoff is handled by considering the roughness of the soil.
The ability to hold surface water might be change over time, e.g., increasing due to agricultural practice or decreasing due to raindrop effects. The surface roughness of soil also effects rainfall effect. This also varies in response to tillage performance. The soil-water module (SOILWAT2) is a model of water balance in cascades 23,29 . Water movement is defined as using separate saturated or unsaturated flow algorithms 23 . In this module, the redistribution of solutes, such as nitrate-and urea-N, is done. Evaporation is a two-stage process based on potential evaporation (Priestly-Taylor) (energy-limited and water-limited). The APSIM model is distinctive to soil organic matter. The CERE model divided the soil organic matter into two subgroups: fresh organic matter (FOM) and humus (HUM), which boosts the microbial pool. HUM, is not susceptible to decomposition; this is specified as Finert (Figs. 1a, 2b).
Although soil microbial biomass only represents a small part of organic matter, it has a significant effect on the nutrient cycle and soil nitrogen 38,39 . The APSIM soil pH module provided a representation of soil acidification Table 1. List of APSIM Modules used in the study and simulate in the nitrogen model. Intellectual property remains that of the original developer. APSWIM (APSIM-Soil water infiltration and movement) it measures soil water behavior in APSIM. a Four modules, a soil-water module (SOILWAT2), the soil nitrogen module (SOILN2) and the fertilizer module (FERTILIZ), and a specific crop module (APSIM-maize).  www.nature.com/scientificreports/ and how pH changes are disseminated through the profile due to cation and anion absorption imbalances, nitrate leaching, and changes in soil organic matter content 40 and soil pH refers to the proton balance 41 . APSIM's evolution was primarily as a modeling framework for climate response and management simulation of cropping systems. Soil nitrogen is the module that simulates nitrogen mineralization and thus the N supply available to a soil crop and residues/roots from previous crops 29 . The model observed value for NO 3 − -N was 15 kg ha −1 , and NH + -N was 0 kg ha −1 , while in predicted NO 3 − -N was 25 kg ha −1 and NH 4 + -N was 5 kg ha −1 . Fertilizer amount N, P, and K at a soil monitoring station in Gongzhuling, Jilin from 1990 to 2010 for NPK treatment N kg ha −2 was 165 kg ha −2 , P 2 O 5 82.5 kg ha −2 , and K 2 O 82.5 kg ha −2 . The Control has no fertilizer. APSIM parameterization. The soil module parameterization is critical to correctly simulate the nitrogen and water balance of the soil-crop system. The soil textures, drained upper limits (DUL) and lower limits (LL15) obtained from laboratory measurements were used to set these parameters. The initial water content was selected to the soil water storage capacity. The crop was maize, and the soil organic carbon and nitrogen value is written in the site description in the material method. The initial surface residue was zero, and the organic matter pool name was maize.
Vertical movement in the soil profile of nitrate. To check the distribution of nitrate through the soil profile, throughout the depth after 31 days of fertilization, and again after five months. To visualize this, we created a depth plot. Two values needed for depth plot, the "dlayer" variable. It holds the depth of each variable. (a) Volumetric water content and whereas (SAT) is the saturated water content, (DUL) is the field water holding capacity and lower water absorption limit (LL15) (b) Soil SOC (soil organic carbon), fraction of biomass_C, inert_C,and volumetric water content in soil profile in Jilin. The data was from initial soil sample collected from Jilin. Soil parameters in APSIM. There are two soil parameters such as hydrodynamic parameters and soil property. The different models in APSIM simulate nutrient cycling and soil properties by soil layers. These various soils can also give of different results and all the changes that happened in the layers. This obliged the study of the vertical soil distribution of nitrogen and water dynamic changes in the complete soil profile and root zone. The average saturated water content of the 0-180 cm soil depth is 59.1%, the field water holding capacity is 47%, and the soil bulk density is 1.07 g cm −3 . The soil organic carbon, soil microbial biomass, and passivated organic matter was shown in Fig. 1a, and the soil hydrodynamic parameters are shown in Fig. 1b. The soil properties here SAT is the saturated water content, DUL is the field water holding capacity, BD is the soil bulk density, and 2.65 is the soil-specific gravity. The soil pH was 7.6; bulk density was 1.07 and change to 1.35 up to the depth of 180 cm. The total depth was divided into seven layers. SAT and DUL reduced with the depth from (59 to 49) % and (47 to 39) %. Depth layer is the soil layers divided into seven layers from top to bottom as shown in Fig. 1a.
Crop parameters in APSIM. The crop in the study site was maize. The maize module also involves two crops, soil parameters, namely the crop's lower water absorption limit (LL) and the crop water absorption coefficient (KL) at various root depths. LL characterizes the residual value of crops after different soil water depths after the water supply was discontinued during the vigorous crop growth period. KL characterizes the day-today absorption of effective moisture by crop layer. Since there was no measured value of LL, LL15 was used instead of LL, and only KL is adjusted. Since there was no moisture calculation in this experiment, KL's estimate was based on the APSIM estimation process and value in Australia (Table 3).

Statistical analysis. Model calibration and validation.
To test the accuracy of modeled, soil nitrogen data was compared with observed and simulated data and correlate with various measures such as 45 . These included the (R 2 ) and (RMSE) coefficient of determination, root mean square error. To determine model performance, we calculated the following four indicators: coefficient of determination (R 2 ), Root Mean Square Error (RMSE), and index of agreement (D-index). Willmott concordance index (d) describes the deviation between observed and predicted values. If d = 1, it means that the model performance is perfect while d = 0, it indicates a model for which the mean square error is just matching the variability in the observed.
In our study, the crop was maize, the average sowing date and maturity date was observed from the field of long-term of Gongzhuling (43° 30′ N, 124° 48′ E), it was validated with 46 dates. Calibration is a significant step for the adjustment of model parameters under location agro-climatic conditions. The model calibration of the phonological parameters of maize, sowing dates, and harvesting dates were observed values of the site. The model was run when soil and weather data of the site was input and the year of the simulation was from 1990 to www.nature.com/scientificreports/ 2010. The model was parameterized through adjustment of soil and crop and weather file factors for maximum matching of observed and simulated data. The field result was used for the model calibration from the 1990-2010 field experiment with model outputs. In this step, for the parameterization of the models, we used the derived parameter. The process of calibration followed phases: soil data, climate data, fertilizer data, and crop data. We then used the supplementary data set to independently validate the models. For calibration and validation, we analysed the goodness-of-fit among models observed and simulated the values of nitrogen and total nitrate and total ammonia as well as denitrification. In addition, the conventional R 2 regression is for comparison calculation of the mathematical (least-squares decision coefficient), which is fundamental, which is vital when testing simulation model output.

Regression.
We used SPSS to check the regression model for estimating the climatic change in precipitation, temperature. We also used trend analysis to check the change in 20 years.

Plant material collection and use permission. No permission is required for plant material as it was
purchased from certified dealer of local area.
Ethics approval and consent to participate. We all declare that manuscripts reporting studies do not involve any human participants, human data, or human tissue. So, it is not applicable.

Complies with international, national and/or institutional guidelines. Experimental research and
field studies on plants (either cultivated or wild), comply with relevant institutional, national, and international guidelines and legislation.  -N from soil were not the same for observed and simulated results by APSIM. It has directly related to the application of nitrogen rates and rainfall events. In general, the APSIM nitrogen model significantly predicted the changes in soil NO 3 − -N and NH 4 + -N over different application rates. The difference in precipitation significantly changes over-fertilization years, while essential soil water content also wildly fluctuates with rainfall events (R 2 = 0.171). The denitrification in (Fig. 4) showed that the changes were significant with a year of fertilization in both observed and predicted. To study the impact of individual year, we divide the years into five scenarios and predicted the NO 3 − -N and NH 4 + -N losses with depth, denitrification, and nitrification with precipitation for observed and simulated values in the APSIM model. The average 20 years of losses of denitrification and urea losses in observed and predicted were not changed significantly. The essential soil water content remained the same and did not change over 20 years, as shown in Table 3. Because these loss mechanisms are biological, soil N 2 O losses were caused by temperature and soil water conditions. Natural N 2 O emissions will occur as a result of whether fertilizer is used because soil organic matter decomposition often contributes to the same microbial soil processes that produce N 2 O. Nevertheless, the appli- www.nature.com/scientificreports/ cation of fertilizer would increase the amount of direct N 2 O emissions and indirect nitrogen losses significantly due to greater availability of N. The graph in Fig. 5c showed the distribution of nitrate through the soil profile on 31 January and 16 June after fertilization. The distribution of nitrate in the soil profile on days 1, 15 and 31 days after the addition of urea fertilizer and at 5 months. The result revealed that with depth on two different days of the year, simulation of leaching and losses of NO 3 − -N and NH 4 + -N occur. The losses increased with depth, the denitrification on 16th June is higher than the 31th January.

Nitrogen result for the year 1995. Nitrogen directly or indirectly affected the metabolism and growth
of plants in many aspects and was the main element affecting crop yield. In crop production, nitrogen and phosphorus fertilizers can improve and regulate soil nitrogen and phosphorus supply capacity and promote crop growth, which has become essential elements and means to increase crop yield. The graph in Fig. 6a  The soil nitrogen directly influences the crop growth and metabolism in the crop and one of the main elements in crop yield. The graph in Fig. 6b between day versus total NO 3 − -N, rain, essential soil water (esw) and total NH 4 + -N showed that rainfall in 1995 is relatively low and frequent rainfall occurs in the middle of the year. The total NO 3 − -N (kg ha −1 ) in the seven layers was 59.68 followed by 59.68, 23.31, 13.58, 6.63, 2.39, 2.45, and 2.52, the same decreasing trend was observed in total NH 4 + -N (kg ha −1 ) 8.11, 0.43, 0.30 and 0.14 respectively and zero up to 7th layer. The rainfall is the main factor in losses of nitrogen. When the soil is saturated with water, a process called denitrification and cause nitrogen was loss from the soil zone. Denitrification was the transformation of nitrate into one of the gaseous types of nitrogen that will be lost to the environment. This process usually occurs when there were conditions for the absence of oxygen (anaerobic), such as in saturated soils. A method called denitrification will cause a loss of N when the soil was saturated with water. Figure 6b showed denitrification losses between 150 and 200 days due to high rainfall between this time. When the long-term improper use of nitrogen fertilizer increases considerably, the accumulation of nitrate-nitrogen in deep soil, even more than the single application of nitrogen fertilizer.
The chance of leaching has increased considerably. To investigate the impact of nitrogen fertilizer on the . The nitrogen was lost through denitrification in large amounts of nitrate available in the soil when it's in saturated conditions. Figure 6c showed that the losses of the total (NO 3 − -N and NH 4 + -N) with different depth showed that the losses of NO 3 − -N losses depend on the rainfall events; at 16-6-1995, the losses were higher as compared with the January 31. The NH 4 + -N also followed the same pattern because it was also affected by rainfall and temperature. Compared with the soil layers, the change of the total nitrogen content with the soil layer depth gradually increased. This indicates that with the increase of nitrogen application rate, the accumulation of soil nitrogen below the root layer (generally, the corn root layer is 120 cm) increases, which increases the leaching loss of nitrogen. Fig. 7a showed that in the year 2000, the losses of nutrients were the same and followed the 1995 year with minimum variation. The rainfall event is also high from 150 to 220 days of the year, so the essential soil water also fluctuates, which results in the changes in total  Figure 8b showed that the day versus total NO 3 − -N, rainfall, essential soil water (esw), and denitrification (dnit) showed that the rainfall was high in the mid-year range from 150 to 215 days. The total soil NO 3 − -N losses were highly dependent on the rainfall event. The essential soil water also changes with the rainfall and constant when rainfall did not occur in the start and end of the year. The total soil NO 3 − -N losses were high, and Fig. 8c showed that the Nitrate distribution through the soil profile at 31 days after fertilization and again at 5 months. To help visualize this, we build a depth plot. Layered variables were kept in ranges at all times. For this reason, we included NO 3 − -N and NH 4 + -N as layered variables, and the nitrogen leach downward up to the soil depth. Figure 8c showed the distribution of nitrate in the soil profile after 21 days of fertilizer addition and at 5 months. www.nature.com/scientificreports/ showed that the losses were higher than previous years and the rain fall in this year was higher, which affected the soil saturation level and led to nitrogen losses. The denitrification (dnit) losses were higher among 200-250 days, and rainfall was also high. Soil essential water also fluctuated with the rainfall throughout the year. Figure 9c showed that the nitrogen losses with depth occur in the soil, and the graph showed that the NO 3 − -N losses were higher in the 5th month of the year compared to the start of the year. The NH 4 + -N losses also suggest that the losses were higher on the 5 th of the month than the start of the month. This showed that characteristics of nitrate leaching in this region occurred, and it needs to address during the summer maize season.

Discussion
The globe's average temperature increased over the last decades and continues to increase and rise in prediction, with the great chance to experience hot days-this increase in temperature solar radiation, and precipitation will have a high impact on agriculture 47 . The world's annual fertilizer nitrogen consumption has reached up to 70 million tons. China has an annual consumption of over 15 million tons and is the world's largest consumer of fertilizer nitrogen.
The result of simulation indicates that for enhancing crop production and APSIM mineralization, the chemical composition of biochemical needs to address and accounted for, moreover by enhancing the soil N pool from crop residues which come from C and N partition (CARB, CELL etc.), the crop biochemical composition of residual crop or by adding a conceptual pool or imitated SMM. Various studies also support the result by separating the chemical digestion of organic material into fractions 48,49 . However, nitrogen fertilizer efficiency was low, and there were large losses to the environment. It is estimated that agricultural nitrogen losses can be as high as 40-60% of nitrogen in our country 50 . Overuse of chemical N fertilizers, high net mineralization and nitrification, and predominance of rainfall during the summer season with a light soil texture were the main control factors responsible for the heavy nitrate leaching 51 . In this study, we assessed the dynamic of nitrogen fertilization and climate change with respect to solar radiation, temperature, precipitation, and its impact on nitrification, denitrification, and nutrient losses with depth. We used the nitrogen model in APSIM and used irrigation level constant to check the soil runoff, which plays a key role in nitrate leaching, depending primarily on precipitation and irrigation levels. For example, when drainage declined from 570 to 79 mm, the irrigation rate decreased from 500 mm to no irrigation 52 . Water was an important feature in nitrogen losses, and nitrate was transported by the flow of soil water and can lead to loss of leaching if there were abundant water movement out of the root zone. Leaching of nitrates also occurs during the drainage season when precipitation and irrigation surpass evaporation 53   www.nature.com/scientificreports/ nitrous oxide (N 2 O) and dinitrogen (N 2 ). In this study, the nitrogen model uses rainfall parameters to check the denitrification, the higher the rainfall higher will be the denitrification. Our result was similar to the findings of 55 , which stated that pasture growth in the deep soils was not affected by irrigation frequency, while denitrification increased with higher frequency irrigation, particularly in the poorly drained soil, resulting in increased N 2 O emissions. All soils and climates showed significantly higher denitrification and N 2 O emissions under high-frequency/low-intensity irrigation (Irr1) compared to low-frequency/high-intensity (Irr6), because soils exceeded a critical moisture content that favors denitrification. The rainfall cause runoff, and it result in losses of nitrogen; our result was also similar to the finding of 56,57 where he stated that nitrogen losses by leaching and surface runoff were the highest among all climatic scenarios for each treatment under the RCP 8.5 scenarios. The explanation may be due to the expected more regular heavy precipitation events and high soil nitrate concentration due to the introduction and mineralization of mineral Nitrogen. In our study, the higher average annual denitrification rates under some years under APSIM scenarios, this result was similar to the finding of 56,58 , he stated that the climate change would increase N 2 O emissions globally. Based on the simulating findings in Fig. 11, inter annual variability steadily enhanced application of nitrogen rates greater than 25 kg N ha −1 similar result was also reported by [59][60][61] . The Nitrogen budget or balance was often calculated by comparing different N inputs and outputs in plant-crop systems, taking into account shifts in soil mineral Nitrogen 57 . In most cases, in spite of the great uncertainty associated with its calculation, denitrification was also seen as an important process of nitrogen loss. A long-standing issue in soil N research was the direct quantification of denitrification nitrogen loss from nitrogen fertilized soils 57 . This result was similar to the findings of our nitrogen model simulation of nitrogen losses; for example, willigen P compared 14 nitrogen cycle models and found that these 14 models could not be simulated in late spring and early summer. Loss of soil inorganic nitrogen after fertilization 62 . However, the study provided the adverse effects of climate change and give a clear picture of nitrogen dynamics under long-term weather conditions. In most cases, despite the great uncertainty associated with its measurement, denitrification was also seen as an important process of nitrogen loss. Clear denitrification quantification, the loss of nitrogen from N-fertilized soils were a long-standing issue in soil nitrogen research 57 . Because the soils were above a critical moisture content that favors denitrification 55 . Yet expected N losses have increased by leaching, denitrification, and N 2 O emissions. Pasture growth in the deep soils was not affected by irrigation frequency, while denitrification increased with higher frequency irrigation, especially in the poorly drained soil, resulting in increased N 2 O emissions. Based on these modeling results, it was possible to reduced nitrogen losses with little effect on 54 . The change in the soil microsites under the incubation conditions may also illustrate the response of N 2 O of emissions from denitrification response to the availability of temperature and soil NO 3 − -N. As observed, the increase in the temperature of soil respiration levels is likely to result in O 2 depletion, affecting NO 3 − -N as the terminal electron acceptor during denitrification. The nitrous oxide emission from denitrification and partitioning of gaseous losses as affected by nitrate 63 .
The study indicated that the use of synthetic nitrogen (N) fertilizer has played a critical role in boosting food production to an increasingly growing population of the world. Furthermore, high inputs of mineral N fertilizer overcrop demands can lead to decreased N use efficiency and affect several negative impacts on the environment, such as surface water eutrophication, groundwater nitrate pollution releasing greenhouse gases, and soil acidification. These environmental issues were getting worse because the use of mineral nitrogen fertilizer would grow predictably in the future 19 . Complex interactions between soil properties, weather patterns, crop growth, and nitrogen loss networks make it more challengingto sync up fertilizer management with crop nitrogen demand, leading to under-or above-N utilization 64 . As such, the overall objective of this study was to provide a clearer understanding of increasing dynamics in the supply and demand balance of water, its effects on food production 17 . Appropriate strategies for fertilization must be adopted for further optimization to fit in with future climate change. Future more, climate change could significantly impact the soil nitrogen, especially the combined effects of elevated temperature, increased losses, and increased precipitation event increase denitrification. The APSIM nitrogen simulation model was used to simulate the influence of climate change on soil nitrogen dynamics based on future climate change across maize cropping regions in China.

Conclusion
Quantitative information on nitrogen and its impact on soil under long-term fertilization, the best strategies, is essential for the assessment of nitrogen loss and availability. The results of the model showed that the nitrogen model between observed and simulated values for nitrogen losses, denitrification, and nitrogen losses through different depths. The nitrogen simulation through APSIM showed that after urea fertilization, it depletes in the soil after 50 days. This result in the uptake of nitrogen by the plant was high at the start of the year. The total NO 3 − -N showed that losses started from 80 days of the year while NH 4 + -N losses started after 100 days. The model predicted that with an increase in rainfall events in the year, losses of nitrogen increase. This study illustrates the potential for using crop management and nitrogen simulation models as an information technology tool for maintaining the suitable management strategies for maize production in Jilin province, China. This can provide alternative management strategies to overcome the nitrogen losses in maize crops. The weather record and long-term soil data offer the best management scenario for the analysis and prediction of nitrogen. The future recommendation about the study area was to 20-year, long-term simulation with APSIM validated model exhibited that N application (15 kg N ha −1 ) improves both the long-term average nitrogen losses. The nitrogen content 25 (kg ha −1 ) showed increased N losses with the same climatic condition. This implies a lower average yield under an increased amount of nitrogen application; hence 15 kg N ha −1 appear more appropriate for farmers, therefore and a higher yield with minimum losses of N will be observed, thus vulnerability will be reduced. www.nature.com/scientificreports/