Effect of nitrogen application on nitrogen leaching pattern and leaching loss from drip irrigated cotton fields in Southern Xinjiang

Farmers provided excessive nitrogen fertilizer to obtain high cotton yield in Xinjiang, Northwest China. Although drip irrigation could save water resources and improve nitrogen use efficiency, it is not known if leaching is occurring or not and whether leaching will harm the water environment under different nitrogen application. The purpose of our study was to estimate the effect of different nitrogen (N) application on the N leaching loss in drip irrigated cotton fields in South Xinjiang. A field experiment was conducted with N application rates of 317(N317), 395(N395) and 476(N476) kg hm−2 from 2021 to 2022. The characteristics of N pattern and leaching amount were analyzed. The concentration of NO3- and total nitrogen in the leaching water significantly decreased with the decrease of N application. The proportion of NO3–to total nitrogen reached the highest values of 51.52%–58.16%, 49.53%–56.91%, and 57.52%–62.87% at bud, flower, and boll stages of cotton, respectively. Moreover, the proportion of NO3– to total nitrogen remained at a stable level, which indicated that NO3–was the main pattern of N leaching. The N loss in the treatment N395 and N317 was 79.67% and 67.48% of that in the N476 treatment (1.23 kg hm−2), while the yield was 120.56% and 112.46% of that in the N476 treatment. Compared with conventional fertilizer application (476 kg hm−2), the reasonable N fertilizer application would not only reduce the amount of N leaching, but also increase the cotton yield.


Introduction
Currently, non-point source pollution (NPSP) has become a major threat to the water quality of global water resources [1]. NPSP is formed by lots of factors, such as soil type, land use pattern, climate, hydrology, and management [2], especially when chemical fertilizers and farm manures are used in production and daily life in agroecosystems [3].Nitrogen (N) is one of the essential plant nutrients and it is taken up by crops throughout the growing season, thereby affecting the plant growth and yield [4,5].
Cotton (Gossypium hirsutum L.) is the main cash crop in Xinjiang, North-western China. Cotton income is the main source of income for farmers in South Xinjiang [6]. Farmers apply large amounts of nitrogen chemical fertilizers in order to obtain high yield. According to the reasearch, the average amount of N fertilizer applied to cotton in South Xinjiang is as high as 420-430 kg hm −2 , which is much higher than the actual amount of nitrogen fertilizer required by cotton [7]. Excessive application of N fertilizer will cause a large accumulation of N in soil, surface runoff and leaching, which significantly increase the risk of N leaching losses [8]. Studies have shown that a 20% reduction in conventional fertilizer N application can effectively reduce N loss in surface water and the total N discharged from the environment [9]. Namratha (2014) found if N fertilizer application was reduced by 168 kg hm −2 on farmland, the inorganic N content was reduced by 69% through run off [10]. In addition, there is a stronger relationship between the concentration of NO 3 − in runoff and the amount of fertilizer applied compared with the other N pattern and the concentration of NO 3 − increased with the increase of N fertilizer [11]. Ha et al (2015) showed that application of chemical N fertilizer significantly increased the content and output of N in agricultural runoff water [12]. Babar et al (2014) found there is a significant positive correlation between N loss and 27fertilizer application [13]. Yin et al (2007) N fertilizer application in most cotton areas in Xinjiang exceeded the actual level required for cotton growth, and the excess N fertilizer application easily led to a 'surplus' of soil N content, resulting in increased N leaching losses [14]. Although various studies have investigated N leaching, most focused on the effects of soil nitrogen losses, whereas few considered the losses of nitrogen in leachate water,especially under different N fertilization of drip irrigation in arid regions. Therefore, the objectives of this study were: (1) to verify the characteristics of N leaching patterns under different treatments; (2) to define the specific amount of N leaching losses during the different cotton growth stages; (3) to assess the main N loss pattern in drip-irrigated cotton field in order to provide a scientific basis for reducing N leaching loss and ensuring cotton yield security by reducing N fertilizer application.

Site description
The experiment was conducted at the No. 2 test site (85°52′N, 41°41′E) of the Water Conservancy Experiment Station of the Bazhou Water Pipeline Division, Korla City, Xinjiang.The average annual rain fall is 58.3 mm and the evaporation is 2572 mm. The accumulated temperature (10°C) is 4315.3°C. The frost-free period is 208 d. The groundwater level is 2.13m.The previous crop was cotton. Field soil belongs to brown desert soil with a sandy loam texture,with medium fertility level. The properties of the surface soil (0-20 cm) at the trial site were: soil organic matter = 7.38 g kg −1 , total nitrogen = 0.48 g kg −1 available nitrogen(N) = 34.12 mg kg −1 , available phosphorus (P) = 16.21 mg kg −1 , exchangeable potassium(K) = 118.47 mg kg −1 , soil pH = 8.08.

Experimental design
The experiment was conducted in 2021-2022. Drip-irrigated cotton fields were selected for N leaching experiments. Three nitrogen application rates of 317 (N317), 395 (N395), and 476 (N476) kg hm -2 were tested, respectively. The results established the nitrogen application rate as 476 kg hm −2 . Urea with a nitrogen content of 46.4% was selected as a nitrogen fertilizer for the experiment. Base fertilizers of 68 kg hm −2 potassium sulfate (K 2 O) as well as 186 kg hm −2 calcium phosphate (P 2 O 5 ) were used to treat the soil. Before sowing the seeds, the mixture of base fertilizers was sprinkled into the soil. Furthermore, the nitrogen fertilizer was utilized as base fertilizer, with 40% incorporated during the initial application, and the remaining 60% employed as top dressing [15]. The three test fields of area 4.8 × 6.9 m each were randomly selected.
Cotton seeds (Xinluzhong No. 21) were sown in early April; the crop was harvested in late October [15][16][17]. The three links of sowing, laying drip irrigation pipes, and plastic mulch were done sequentially. The seeds were sown in double rows, maintaining a distance of 30 cm between the rows and 66 cm between the seeds in each row. The sowing was carried out using a custom-made tractor-pulled seeder. Polyethylene film was used for covering the ground given its benefits: it is air-tight, but its transparency allows light to reach the ground [15][16][17].
The growing season for cotton is from mid-June, the budding period, to late August, the boll-opening period. The crop required 10 times the water and 8 times the nitrogen fertilizer during this period. The irrigation quota for the season was set at 4,800 m 3 hm −2 . Nitrogen fertilizer was not applied for the first watering and the last watering of the crop. Besides, techniques such as topping, pest control, use of chemical regulators etc.were consistent with those that are adopted in previous studies [15,17].

Sample and analysis
The field infiltration tank method was used for the study of subsurface leaching, and the field infiltration tank installation in each plot was operated according to the 4specification [10], and the field infiltration tank installation was schematically shown in figure 1. The total volume of leaching was measured in each time with a cylinder. After shaked the leachate water, two mixed water samples were taken, one of which was used for analysis, and the other was used to store. The volume of each sample was 500 ml. If the volume of leachate water was less than 500 ml, all the leachate water would be collected as a sample for analysis.The concentration of nitrate nitrogen (NO 3 − ) and ammonium nitrogen (NH 4 + ) in the leachate water was determined using a continuous flow analytical system (TrAAcs 2000). The concentration of total nitrogen (TN) in the leachate water was determined by potassium persulfate oxidation ultraviolet spectrophotometry [16].

Dissolved organic nitrogen
Cotton in each field trial plot was harvested during cotton harvesting period, and the economic yield of cotton in each plot was recorded.

Calculation method of N leaching and leaching rate
The amount of N lost by surface runoff or subsurface leaching is equal to the sum of N in each runoff (or leaching) in the whole monitoring cycle (a complete anniversary), where the N in each runoff with unit kg hm −2 is obtained by the concentration of N in each runoff (or leaching) water multiplied by the volume of runoff water (or leaching) [18].
The calculation equation is as follows: P Ci Vi Where P is the amount of N loss; Ci is the concentration of N in the leaching; Vi is the volume of the leaching.
here N n represents nitrogen fertilizer leaching coefficient; Cn represents the amount of nitrogen loss under conventional fertilizer treatment; Rn represents the amount of nitrogen loss under reduced fertilizer treatment; Nr represents the amount of N fertilizer under reduced fertilizer treatment.

Statistical analysis
Differences among different treatments for each year were analyzed by one-way parametric analysis of variance using SPSS version 19.0 (SPSS, Chicago, IL, USA). LSD-test were carried out to determine if there were significant differences between individual treatments at P < 0.05.

Effect of N application on N leaching
According to the N loss formula, the N loss values of N476, N395 and N317 treatments in 2021 and 2022 were calculated and analyzed at the bud, flower and boll stage (tables 4-6). The order of N losses in N476, N395, and N317 treatment at bud, flower, and boll stages of cotton growth was N476 > N395 > N317. In 2021, the N loss in N476 treatment reached a maximum of 1.23 kg hm -2 during the whole cotton growing period, while the N loss in N395 and N317 treatments accounted for 79.67% and 67.48% of the N loss in N476 treatment. The loss of N decreased with the reduction of N application. The order of N loss was N476 > N395 > N317. The N loss in N395 treatment was 0.98 kg hm -2 and 1.02 kg hm -2 in 2021 and 2022, respectively. The N317 treatment had N    2). The order of cotton yield was N395 > N317 > N476. The results in 2022 was the similar with that in 2021. The same as above. The same as above. The same as above.

Discussion
Nitrogen which is not absorbed by crops because of the excessive irrigation or nitrogen is likely to be leached. Nitrogen leaching depends on the irrigation amount,fertilizer type, and fertilizer rate, and there is a strong interaction between the irrigation treatment and fertilizer rate [16].The results showed that the concentration of inorganic N(NH 4 + and NO 3 -) in leachate water decreased significantly with the reduction of N fertilization [18][19][20][21][22]. This result is consistent with our study. Li P et al (2012) found that NO 3 is the main N form lost from rice-wheat crop rotation system [22]. The proportion of NO 3to TN was 69.1% and 78.4% in CK and FP treatment, respectively, and the proportion of NH 4 + to TN was 8.2% and 2.4% in 5 years. In our study, it showed that the proportion of NO3 to TN was 51.52%-58.16%, 49.53%-56.91%, and 57.52%-62.87% at bud, flower, and boll stage, respectively. This is verified that NO 3 is the main form of N elements in leaching water.
It was reported that a 20% reduction of the conventional N application could effectively reduce soil NO 3 loss, which indicates that N application of 240 kg hm -2 was more effective. Jie X et al (2016) found that the NO 3 loss of different fertilizer treatments ranged from 1.76 to 5.25 kg hm -2 , and the leaching coefficient ranged from 2.57% to 4.11% [23]. A related study on corn (Zea mays L.) in Northeast China showed that the N leaching loss under different fertilizer treatment ranged from 1.76 to 5.71 kg hm -2 and the N leaching rate was about 5% [24]. Lin et al (2011) concluded that the N leaching loss from N fertilizer was about 2% and the leaching coefficient was 2% [25]. In our study, the N leaching rates of N395 and N317 treatments were 1.02 kg hm -2 and 0.86 kg hm -2 , respectively, and the leaching rates were 0.36% and 0.28%. Compared with the previous studies, the N loss and N fertilizer leaching rate were lower, and reducing N application could reduce N loss and N fertilizer leaching rate. The effect of N application on the accumulation and leaching of NO 3 from soil in cotton fields in Xinjiang was less reported. Therefore, the study of the effect of N application on N leaching and N fertilizer leaching rate can provide a scientific basis for N fertilizer reduction in cotton field in Xinjiang.
The investigation of groundwater NO 3 content in Shandong Province proved that the areas with high groundwater NO 3 concentration own high fertilizer application, and reducing the amount of N application can reduce the NO 3 concentration in groundwater [25]. Studies have shown that the relationship between NO 3 pollution and agricultural production is inextricably linked. Reducing the amount of N fertilizer application can reduce NO 3 leaching by more than 20% compared to conventional N application treatments [26]. In this study, N application could directly affect the NO 3 concentration in the subsurface leachate water. Compared with the conventional N application (N476), N395 and N317 treatments accounted for 79.67% and 67.48% of the N loss from the conventional treatment. The NO 3 loss was lower by reducing N fertilizer application compared with conventional N management, and the crop yield was not reduced [27,28]. This result is consistent with our result. The annual N losses of N476, N395 and N317 treatment varied from 0.83 to 1.31 kg hm -2 , and the N loss rate varied from 0.25% to 0.36% under drip irrigation. The advantage of drip irrigation is small amount and many times, which is adopted to effectively reduce the N loss and obtain high cotton yield. Therefore, reasonable water and fertilizer management and reduction of N fertilizer application under drip irrigation can reduce soil N loss and N fertilizer loss rate.

Conclusions
The concentration of NO 3 in the leachate water was a sensitive indicator of the variation of N application. The proportion of NO 3 to TN in the leachate water was higher than other N pattern during the cotton growth period, and NO 3 was the main form of N loss. The N leaching loss under N476 treatment was 1.23 kg hm -2 , the N leaching loss under N395 and N317 was 79.67% and 67.48%, and the yield was 120.56% and 112.46% of that under N476 treatment. The nitrogen leaching rate decreased from 0.31% to 0.25% with the decrease of the N application from 395 to 317 kg hm -2 . Reduce N fertilizer not only could reduce N loss, but also increase cotton yield and N fertilizer utilization rate.