Nitrogen Fertilizer and Nitrapyrin for Greenhouse Gas Reduction in Wolfberry Orchards on the Qinghai–Tibetan Plateau

: Wolfberry production has become a major agro-industry on the Qinghai–Tibetan Plateau, causing increased nitrogen (N) pollution and greenhouse gas (GHG) emissions. Appropriate N fertilizer rate and nitrification inhibitors may mitigate GHG emissions and improve N use efficiency. A 2-year field experiment was conducted to measure the effects of N application rate and nitrapyrin on GHG emissions, to reduce GHG emissions and N pollution. We used eight treatments: Control (CK), 667 kg·ha −1 N (Con), 400 kg·ha −1 N (N 400 ), 267 kg·ha −1 N (N 267 ), 133 kg·ha −1 N (N 133 ), 400 kg·ha −1 N plus 2.00 kg·ha −1 nitrapyrin (N 400 I 2.00 ), 267 kg·ha −1 N plus 1.33 kg·ha −1 nitrapyrin (N 267 I 1.33 ) and 133 kg·ha −1 N plus 0.67 kg·ha −1 nitrapyrin (N 133 I 0.67 ). Compared with Con treatment, N 400 maintained fruit yield and increased net income, but saved 40% of N fertilizer and decreased the cumulative N 2 O emission by 14–16%. Compared to N 400 , N 267 and N 133 treatments, the cumulative N 2 O emission of N 400 I 2.00 , N 267 I 1.33 and N 133 I 0.67 treatments was reduced by 28.5–45.1%, 26.6–29.9% and 33.8–45.9%, respectively. Furthermore, N 400 I 2.00 resulted in the highest wolfberry yield and net income. The emissions of CH 4 and CO 2 were not significantly different among treatments. Moreover, the global warming potential (GWP) and the greenhouse gas emission intensity (GHGI) of N 400 I 2.00 declined by 45.6% and 48.6% compared to Con treatment. Therefore, 400 kg·ha −1 N combined with 2.00 kg·ha −1 nitrapyrin was shown to be a promising management technique for maintaining wolfberry yield while minimizing GWP and GHGI.


Introduction
Global warming due to excessive GHG emissions is now a serious global issue drawing unprecedented scientific and political attention. China is one of the largest emitters of GHG. The country aims to decrease GHG by 60-65% per unit of gross domestic product by 2030 [1]. Agriculture, the largest anthropogenic source, directly releases considerable amounts of GHG to the atmosphere [2,3]; this accounts for 17% of total release in the world, but close to 30% in China [4]. On a 100-year scale, the global warming potential of CH4 and N2O is 28 and 265 times than that of CO2, respectively [5]. It is predicted that annual GHG emissions in arable land will be about 564 Tg CO2-equivalent in 2030 [6].
To meet the growing demand for food, N fertilizer input is increasing in agriculture, especially in China. Wolfberry (Lycium barbarum L.), a salt-tolerant and drought-resistant shrub, is mainly grown in northwest China [7,8]. Because of the unique climatic condition and geographical environment, wolfberry production has become a major agro-industry in Qaidam, on the Qinghai-Tibetan Plateau. Plantings reached 49,900 ha in 2020 [9]. However, in the search for high yields, excessive N fertilizer input in wolfberry production has reduced N use efficiency (NUE) and increased costs, N pollution and GHG emissions [10][11][12]. Moreover, previous studies have shown that high N inputs negatively affect soluble solids and other quality indicators of wolfberry fruit [13][14][15]. Therefore, rational N fertilization is of great significance to provide a theoretical basis for reducing GHG emission and improving wolfberry yield and quality.
Optimizing N fertilizer applications and using nitrification inhibitors may be effective strategies to improve N use rate and reduce the environmental costs of agricultural production [16,17]. It was revealed that N2O emission significantly decreased due to reduced N fertilizer rates [18][19][20][21]. The cumulative N2O emission decreased by 68.8% during the experiment when N fertilizer was decreased by 53.3% [22]. Wang et al. [23] found that the N2O loss was lowered by 72.2 g·ha −1 when fertilizer N input was 33.3% lower. In apple orchards, when N fertilizer was decreased from 800-400 kg·ha −1 N, N2O cumulative emission decreased by 43.3% [24]. On the basis of standard current farmer practice, N application rate decreased by 22.9% and 21.9%, resulting in the reduction of GHGI [25]. A reduction in N application rate also helped to decrease CH4 emissions [26].
Nitrapyrin can delay the conversion of ammonium (NH4 + ) to nitrate (NO3 − ) by inhibiting the activities of nitrifying bacteria in soil. This improves N retention in the soil and enhances N utilization rates [27,28]. Zhou et al. [29] found that nitrapyrin could decrease N2O emissions and GHGI by 61.4% and 28.5%, respectively, in paddy fields. In the Midwest of the United States, applying nitrapyrin significantly lowered GHG emissions, while the yield and soil N retention increased by 7.0% and 28.0%, respectively [30]. Moreover, a nitrification inhibitor effectively undermined the denitrification in an apple orchard by 15.8-53.1% [31]. It is worth noting that there is a low residual amount of the nitrification inhibitor itself in the soil, and there is no ecological risk [32,33]. Consequently, nitrapyrin and optimum N fertilizer rate have broad potential in mitigating GHG emissions and preventing environmental pollution.
At present, excessive N fertilizer significantly increases GHG emissions, which has negative effects on wolfberry quality and the environment. In the Qaidam area, the application of N fertilizer in production of wolfberry lacks scientific theoretical basis, and the problems of low N utilization rate, waste of agricultural resources and high production cost caused by blindly investing a large amount of N fertilizer are common. Furthermore, few reports have focused on the effects of optimum N fertilizer rate and nitrapyrin on GHG emissions in wolfberry on the Tibetan Plateau. Therefore, we hypothesized that reasonable N fertilizer rate and nitrapyrin, in combination, could mitigate greenhouse gas emissions. A 2-year field study was carried out in a wolfberry orchard on the Qinghai-Tibetan Plateau to: (i) investigate greenhouse gas fluxes across optimum N fertilizer rate combined with nitrapyrin; (ii) elucidate the effects of optimum N fertilizer rate combined with nitrapyrin on GHG emissions and wolfberry yield; (iii) quantify GWP and GHGI of CH4 and N2O emission under optimum N fertilizer rate combined with nitrapyrin.

Description of Study Site
A field experiment was conducted at Nomuhong Farm (96°20′ E, 36°25′ N) in Golmud City, Qinghai Province, Qinghai-Tibet Plateau from 2019 to 2020, with an altitude of 2760 masl and a typical plateau continental climate. The average annual sunshine duration is 3600 h and the evaporation is 2800-3000 mm per annum. About 66.8% of the annual precipitation occurs from June to October. The soil type is sandy loam classified as grey-brown desert soil according to Chinese Soil Taxonomy [11]. The physicochemical properties of 0-20 cm soil were determined according to Bao [34], and the main parameters were: 1.51 g·cm −3 bulk density, 8.49 pH, 1.43 g·kg −1 total N, 3.05 g·kg −1 total P, 23.1 g·kg −1 total K, 19.5 g·kg −1 organic matter, 82.6 mg·kg −1 Olsen-P, 69.76 mg·kg −1 alkaline hydrolyzed nitrogen and 210.8 mg·kg −1 available K. Wolfberry is the predominant economic plant in Nuomuhong farm with an increasing area planted per annum.
Commercial organic fertilizer (Qinghai Enze Agricultural Technology Co., Ltd., Xining, China; the fertilizer is the product of sheep manure fermentation after maturity: organic matter, 51.7%; N, 3.7%; P2O5, 1.1%; K2O, 2.0%) was applied at 1667 kg·ha −1 and calcium triple superphosphate (Yuntianhua Group Co., Ltd., Kunming, China, P2O5, 46%) at 724 kg·ha −1 (equivalent to 333 kg·ha −1 pure P2O5) were applied in each treatment, which is the typical rate used by farmers. Half of the total N (Urea, Yuntianhua Group Co., Ltd., Kunming, China, N, 46%) and 50% nitrapyrin (Zhejiang Aofutol Chemical Co., Ltd., Shaoxing, China, 70%) input with commercial organic fertilizer and calcium triple superphosphate were applied as basal fertilizers, which were evenly distributed into fertilization holes (0.30 m away from root, 0.20 m deep × 0.50 m long × 0.25 m wide) between rows on 19 May 2019 and 15 May 2020. The remainder (50% total N and nitrapyrin) was topdressed in the same location on 30 June 2019 and 5 July 2020. Urea and nitrapyrin of each treatment were mixed thoroughly before fertilization. The wolfberry orchard was flood irrigated 7 times a year with a total irrigation quota of 6000 m 3 ·ha −1 . Other field managements, such as irrigation, were the same as that used by local farmers.

Gas Sampling and Measurements
Greenhouse gas samples were collected in each plot from May to October in 2019 and 2020, using a closed static chamber-gas chromatography method [35]. Each gas collecting equipment included a chamber and a base. The chambers (0.5 m deep × 0.5 m long × 0.5 m wide) were composed of a stainless-steel frame, and covered with cystosepiment to prevent dramatic temperature changes inside the chamber. The chamber base was inserted 20 cm into the soil before the experiment. A three-way valve was connected at a distance of 35 cm from the bottom of the chamber, and a temperature detection port was arranged beside the valve. A small fan was installed at the top of the box's diagonal corner to mix gas. The chamber was fitted into the frame base with a groove (2 cm width) during sampling, and the groove was full of water to seal. Sampling was conducted from 9:00-12:00 am. At 0, 15, 30 and 45 min after the chamber was closed, the gas was extracted with a 50 mL syringe and injected into the sealed air bags. Gas samples were determined by gas chromatography (Agilent 7890A, Agilent Technologies, Palo Alto, USA) equipped with a 63 Ni-electron capture detector for N2O detection at 350 °C and a flame ionization detector for CH4 and CO2 detection at 200 °C. Frequency of gas collections was daily for 7 days after fertilization, 3 days after irrigation, 1 additional measurement at the time of daily precipitation more than 20 mm; the remaining experimental time samples were collected once a week [35].

Environmental Factors Measurements
Air temperature and precipitation data were obtained from a meteorological station adjacent to the experimental site. A thermometer was used to measure the 10 cm soil temperature before the first gas sample was collected and after the last gas sample was taken. A digital thermometer within the chamber provided the average air temperature inside the chamber, which was used to calculate the gas flux. Wolfberry fruit from each plot was collected to determine the weight of fresh fruit on 29 July, 20 August and 14 September in 2019 and 4 August, 25 August and 19 September in 2020. The fresh fruits were dried in a greenhouse and then weighed.

Calculation
Soil gravimetric water content was calculated to 10 cm depth according to Equation (1). Gravimetric water content and soil bulk density were separately determined with the oven-drying method and cutting ring method [36].
Fluxes of N2O, CH4 and CO2 were estimated using Equation (2)  for CO2) is the initial rate of change in N2O, CH4 or CO2 concentration in the chamber enclosure [38]. Gas fluxes and cumulative emissions were calculated using a linear or non-linear model [39,40] and the direct interpolation method [41,42], respectively. Direct N2O emission factor was calculated according to the following formula: where R and RCK separately represent the cumulative N2O emission (kg·ha −1 N2O-N) from applied N fertilizer per plot and CK plot, and F is the N fertilizer rate for one year (kg·ha −1 N).
GWP is calculated using Equation (4) where the unit of GWP is CO2-eq kg·ha −1 ; RCH 4 and RN 2 O are the cumulative CH4 and N2O emissions (kg·ha −1 ). On the time scale of 100 years, the GWP of unit mass CH4 and N2O is 265 times and 28 times of CO2. 265 GHGI is an indicator for comprehensive evaluation of greenhouse effect. As shown in Equation (5), its unit is CO2-eq kg·Mg −1 ; the yield is in wolfberry dried fruits (Mg·ha −1 ).

Statistical Analysis
Using IBM SPSS Statistics 23.0, and all raw data were verified with respect to exhibiting normality (Kolmogorov-Smirnov test) and homogeneity of variance (Levene's test). For multiple comparisons, we used Fisher's protected least significant differences (LSD) at the 5% level of probability. Pearson's correlation was used to analyze the correlation between greenhouse gas emission and soil water-filled pore space (WFPS) or 10 cm soil temperature (Tsoil). We performed linear, quadratic and exponential curve fittings to simulate the response of greenhouse gas emission to N rate, then used the determination coefficient (R 2 ) to select the optimum curve. Origin 2018 software (OriginLab, Northampton, USA) was used for drawing figures.

Environmental Factors
The total rainfall during the wolfberry growing season was 48. WFPS increased following seven flood irrigation events. Due to heavy rain on 3 July 2019 and 21 June 2020, WFPS peaked after the second irrigation in two years. There was no difference in average WFPS between 2019 (58.3%) and 2020 (58.2%), nor was there a significant difference in WFPS among treatments in the same year. The 10 cm soil temperature decreased and then increased with the occurrence of irrigation, reaching the highest value in August in both years ( Figure 2). The peak values of the 10 cm soil temperature were 19.5 °C and 21.1 °C in 2019 and 2020, while the minimum was 8.1 °C and 5.6 °C, respectively. The mean value of the 10 cm soil temperature was 13.7 °C in 2019, with an increase of 6.2% over 2020.

N2O Emission
The N2O emissions after N fertilizer application or irrigation significantly increased to various degrees ( Figure 3a). Peak N2O emission was observed for 3-5 days after N fertilization following irrigation. The N2O peaks of Con treatment reached 1765.40 μg·m −2 ·h −1 on 19 May 2019 (fertilization combined with irrigation on 16 May 2019) and 1783.30 μg·m −2 ·h −1 on 8 July 2020 (top-dressing N fertilizer following irrigation on 5 July 2020). N400I2.00 resulted in lower N2O peaks than Con over the two years. Moreover, emission peaks of N400I2.00, N267I1.33 and N133I0.67 treatments were lower than that of N400, N267 and N133 treatments. Pearson's correlation showed N2O emission was significantly positively correlated with WFPS and 10 cm soil temperature (p < 0.05) (Figure 4).    Nitrogen fertilization significantly increased the cumulative N2O emissions compared with the CK treatment and the extent also varied with the addition of nitrapyrin and year (p < 0.05) ( Table 1). When no nitrapyrin was added, the cumulative N2O emissions in 2019 and 2020 were significantly positively correlated with N application rate; the determination coefficients were 0.952 and 0.902, respectively ( Figure 5). In contrast to Con treatment, the cumulative N2O emission of N400 and CK decreased by 13.87-16.37%, 86.70-90.13%, respectively. Excluding the reference treatment, the cumulative N2O emissions were the highest when the N application rate was 400 kg·ha −1 in 2019 and 2020, which were 5.72 kg·ha −1 and 8.38 kg·ha −1 , respectively. Nitrapyrin addition significantly reduced the cumulative N2O emissions and EFN 2 O under the same N application rate. Compared with N400, N267 and N133 treatments, N400I2.00, N267I1.33 and N133I0.67 treatments with nitrapyrin decreased cumulative N2O emissions by 45.1%, 26.6% and 45.9% in 2019 and 28.5%, 29.9% and 33.8% in 2020. In two years, in contrast with N400, N267 and N133 treatments, the EFN 2 O of N400I2.00-N133I0.67 treatments were significantly lower than N400, N267 and N133 treatments, which was significantly decreased by 53.3-69.6% in 2019 and 32.3-42.6% in 2020. Table 1. Cumulative emissions of GHG, wolfberry yields, GWP and GHGI from wolfberry orchard with different treatments.

Year
Treatment

CH4 Uptake
CH4 emissions were influenced by year (p < 0.05), but nitrapyrin and N fertilizers were not. Likewise, the combined effects of N application rate, nitrification inhibitors, and years did not affect CH4 emissions ( Table 1). The CH4 emissions were negative during the growing season (Figure 3b). The highest values of CH4 emission were 75 μg·m −2 ·h −1 and 170 μg·m −2 ·h −1 , while absorption peaks were 205 μg·m −2 ·h −1 and 211 μg·m −2 ·h −1 in 2019 and 2020. The CH4 cumulative emissions were not obviously different among treatments. Moreover, CH4 emissions showed no significant correlation between soil moisture and 10 cm depth soil temperature (Figure 4).

CO2 Flux
Similar to CH4 emissions, the CO2 flux was affected by year (p < 0.05) and the law of CO2 flux was in line with the atmospheric variation with peaks in July and August ( Figure  3c). CO2 flux increased to 314 mg·m −2 ·h −1 from May to July in 2019, then slowly decreased to −35 mg·m −2 ·h −1 in early October. The peak value of CO2 flux was 253 mg·m −2 ·h −1 in July and the minimum was 10 mg·m −2 ·h −1 at the beginning of October in 2020. In both years, cumulative CO2 flux was not significantly different among treatments. During the growing season, CO2 flux was positively correlated with soil temperature (Figure 4).

Wolfberry Yield, GWP and GHGI
Although the yield of wolfberry varied greatly among different years, the trend in yield variation among different treatments was consistent within both years. Besides, the yield was influenced by nitrapyrin and N application rates (p < 0.05) ( Table 1). The yield ranged from 7.03-7.87 Mg·ha −1 and 7.58-8.46 Mg·ha −1 in 2019 and 2020, respectively. When no nitrapyrin was added, wolfberry yield increased first and then decreased with N application rates increasing from 0 to 667 kg·ha −1 . When the N application rate was 400 kg·ha −1 , the yield of wolfberry reached the highest, which were 7.44 Mg·ha −1 and 8.01 Mg·ha −1 in 2019 and 2020, respectively. Compared with Con and CK treatment, the wolfberry yield of N400 treatment increased by 1.01-3.33%, 16.80-19.37%, respectively. Meanwhile, the change in net income was similar with yield. At the same N application rate, the yield and net income were significantly higher in the treatment with nitrapyrin than that of without nitrapyrin, and N400I2.00 treatment produced the highest yield and net income in 2019 (7.87 Mg·ha −1 , 40,841 USD·ha −1 ) and 2020 (8.46 Mg·ha −1 , 43,995 USD·ha −1 ). Relative to N400 treatment, N400I2.00 treatment had 5.7% and 5.6% higher yield in both years. When the combined N application rates were 423 kg·ha −1 and 524 kg·ha −1 in 2019 and 2020, the fruit net income was both the highest, 38,880 USD·ha −1 and 41,483 USD·ha −1 , respectively ( Figure 6).  Similar to the effect of yield, N application rates, nitrification inhibitors and different years can also affect GWP and GHGI independently (p < 0.05) while the two or three together did not (Table 1). GWP and GHGI had significant differences among treatments. N fertilizer input significantly improved GWP and GHGI, while the addition of nitrapyrin decreased GWP and GHGI under the same N application rates. The Con treatment produced the highest GWP of 2822 kg·ha −1 and 3998 kg·ha −1 in 2019 and 2020, respectively. In comparison to Con treatment, GWP of N400, N267 treatments declined by 16.6%, 46.2% and 14.0%, 36.5% in two years. Meanwhile, compared with N400, N267 and N133 treatments, GWP of N400I2.00, N267I1.33 and N133I0.67 treatments decreased by 45.6%, 26.8% and 47.0%, while the GHGI decreased by 48.6%, 30.2% and 47.1%, respectively. GHGI ranged from 52.35-504.11 CO2-eq Mg·kg −1 in the study period. The GHGI peaked in Con treatment, with the lowest at N133I0.67 treatment. N400I2.00, N267I1.33 and N133I0.67 treatments lowered GHGI by 30.2-48.6% vs. N400, N267 and N133 treatments without nitrapyrin.

Effects of N Fertilizer Rate on Yield, GHG Emissions, GWP and GHGI
Scientific N management may reduce N fertilizer rate without compromising fruit yield [44,45]. Compared with no N fertilization, the input of N-fertilization can increase the yield of wolfberry; however, our study shows that excessive N application (e.g., Con) will reduce the yield of wolfberry. This may be due to the synergistic and antagonistic effects between nitrogen and other elements. High N input may lead to excess nutrient growth of wolfberry, causing unbalanced nutrient absorption [46]. Moreover, a previous study showed that long-term high N level led to reduced soil pH and root vigor [47]. In our study, compared to 667 kg·ha −1 N treatment, a reduction of approximately 40% in N rate did not reduce wolfberry yield, which indicated that 667 kg·ha −1 N was excessive for a one-year growth cycle of wolfberry. Therefore, on the basis of maintaining or increasing the yield, appropriate N amount and timing can improve economic benefits for the sustainable development of wolfberry industry. In combination with economic benefits, when no nitrification inhibitor is applied, an N application rate of 423 kg·ha −1 in 2018 and 524 kg·ha −1 in 2019, which we recommend in wolfberry orchards with similar fertility, maximized net income, increasing net income by 1-5% but savign N fertilizer by 21-37%, respectively, compared to Con treatment. Therefore, excessive N input would reduce net income, while reducing N input reasonably could maximize net income and improve wolfberry production.
Zhang et al. [48] and Liu et al. [49] stated that N application in agriculture led to substantial GHG emissions in China, contributing nearly 50% of the total GHG emissions from agriculture in China. In our study, the main contribution of GWP was N2O. The N2O emission was significantly positively correlated with nitrogen application rate (p < 0.05). N2O emission peaks occurred within one week after N application following irrigation. The occurrence of emission peaks typically coincided with high soil moisture (such as 19 May 2019 and 15 May 2020). One likely reason was that N input disrupted the C/N balance of microorganisms and stimulated microbial activity to some extent, which promoted N2O emission [50]. Moreover, N application provided sufficient substrates for microorganism, so that the intermediate products, NH2OH and NO2 − were released and improved in the process of soil ammonia oxidation, which was positively correlated with N2O emissions [51][52][53]. Another likely reason is high temperature improved the activity of soil microorganism and flood irrigation resulted in an anaerobic environment conducive to denitrification [54,55]. Denitrifying microorganisms under high temperature and humidity conditions increase N2O emissions. In this study, average daily N2O emissions from wolfberry receiving 0-667 kg·ha −1 N ranged from 6.55-67.57 g N2O-N ha −1 ·d −1 . Similarly, in eastern China, average daily N2O emissions from an apple orchard were 71.78 g N2O-N ha −1 ·d −1 [20]. At a similar nitrogen rate, we report lower N2O emissions. This may be due to the lower temperature and humidity on the Qinghai-Tibet Plateau, which reduces the N2O generation by inhibiting microorganisms to a certain extent. The soil N surplus due to high N application might increase N2O emissions; therefore, the cumulative emissions of N2O in 2020 was higher than that in 2019 in this study. Therefore, it is reasonable to expect that a reduction in the N fertilizer rate to an optimum level could substantially reduce N2O emissions. Our study showed that N fertilizer application had no significant effect on CH4 uptake. A meta-analysis showed that stimulation or inhibition of CH4 oxidation depended on N additions, with a threshold of 100 kg·ha −1 N [56]. However, the N application rate in our study was 133-667 kg·ha −1 , so the excessive N input might conceal the effect of inorganic N on methanotrophy. Moreover, CH4 uptake is an extremely complex process, which is affected by environmental conditions and agricultural management [57]. Soil environment can affect soil respiration [46,58,59]. In our study, a significant positive correlation was detected between CO2 flux and soil temperature. This might be due to the fact that CO2 is the product of plant root and soil microbial respiration. High temperature stimulated this process and promoted CO2 emissions. The temperature ranged from 5-25 °C in our study, and the increase in temperature might enhance the activity of soil microorganisms and accelerate the decomposition of organic matter, leading to an increase in CO2 flux [60].
Increasing the N application rate greatly contributed to GHG emissions. Previous research indicated that N application increased annual N2O and CH4 emissions, although it benefited atmospheric CO2 sequestration into soil. The net effect is an increase in GWP and GHGI [61]. In our study, both GWP and GHGI of N400 treatment were 40% lower compared with Con treatment, indicating that 667 kg·ha −1 N was excessive for wolfberry production, causing high GHG emissions, low NUE and waste of agricultural resources. Excessive N fertilizer input is the main source of N2O emission and had a noticeable impact (increase) on GWP and GHGI. It is possible to reduce greenhouse gas emissions through optimizing N application rate, thereby further reducing the GWP of wolfberry orchard [62,63]. Results from this study also prove reasonable regulation on soil temperature and moisture was beneficial for mitigating GHG emissions.

Effects of Nitrapyrin on Yield, GHG Emissions, GWP and GHGI
Nitrification inhibitors have been used to regulate N conversion to nitrate-N and subsequent nitrous oxides. These inhibitors are also considered an effective measure to increase NUE. In our study, with the same N application rate, combining treatments with nitrapyrin increased yield. Because it inhibits the activity of ammonia-oxidizing bacteria, consequently delaying the conversion of NH4 + into NO3 − , leading to the reduction in substrate that produces N2O emission [64]. Furthermore, the uptake of NH4 + required lower energy for plant N assimilation than NO3 − [65].
Our experiment showed the addition of nitrapyrin under the same N fertilizer application rate reduced the cumulative N2O emissions. The main reason was that nitrapyrin reduced the activity of soil nitrifying bacteria, and then inhibited the oxidation process of NH4 + to NO2 − in nitrification, thereby reducing the accumulation of NO3 − -N and the emission of N2O [66]. Nitrification inhibitors have been widely reported as an effective management measure to reduce N loss and improve NUE, but the effect of nitrification inhibitors on CO2 and CH4 was rarely studied [67][68][69]. In our experiment, no significant difference was detected in net CO2 and CH4 between treatments. The effects of nitrification inhibitors on emissions may be jointly influenced by many factors, such as soil temperature, pH and soil oxidation-reduction potential. These factors were combined and lead to no significant effect of N fertilization on CO2 and CH4 emissions. The specific influencing process and interaction mechanism of these factors need further study.
GHGI reflected the total GHG emissions per unit of yield. This study showed that optimum N application combined with nitrapyrin significantly reduced GWP and GHGI, compared with N fertilizer alone. When CH4 and N2O emissions were expressed as CO2 equivalents, N2O was the main contributor to GWP. The contribution of N2O emission to GWP was as high as 99%. Nitrapyrin inhibited nitrification and reduced N2O emission, thereby reducing GWP and GHGI. These results showed that N application combined with nitrapyrin was a recommended cultivation method for high productivity and sustainable agriculture, considering environmental and economic efficiency.

Effects of N Fertilizer Rate Combined with Nitrapyrin on Yield, GHG Emissions, GWP and GHGI
Fertilization is an important means to maintain soil fertility and ensure stable food production. However, the phenomenon of excessive chemical fertilizer (especially N fertilizer) and low fertilizer utilization rate is widespread in crop production. Nitrification inhibitors are a widely used synergist for N fertilizer, effectively reducing N loss. In our study, 400 kg·ha −1 N combined with 2.00 kg·ha −1 nitrapyrin maximized net income in both years, which was the recommended N management. Furthermore, compared to 400 kg·ha −1 N combined with 2.00 kg·ha −1 nitrapyrin, 267 kg·ha −1 N combined with 1.33 kg·ha −1 nitrapyrin did not result in a great decrease in net income (by 2-4%), but that reduced GHG emission by 13-27%, and saved N fertilizer by 33%. Therefore, on the base of output and environmental impact assessment, 267 kg·ha −1 N combined with 1.33 kg·ha −1 nitrapyrin was the recommended N management in the study site. The combination helps to mit-igate GHG emission and reduce N loss, thus improving wolfberry N uptake, and increasing NUE. An appropriate N fertilizer rate combined with nitrapyrin not only reduced GHG emissions, but also saved agricultural resources, which was beneficial for increasing economic benefits and achieving agricultural safe production. It was also the key to realizing the coordinated development of soil production and environmental function. If the optimized method was adopted across the Qaidam, the wolfberry yield would effectively increase, and N fertilizer input would decline by roughly 60%. Consequently, it was considered as a powerful strategy for the development of an agronomic system, benefiting both agriculture and environment.

Conclusions
Compared with Con treatment (i.e., standard current farmer practice), appropriately reducing N application rate could maintain wolfberry fruit yield and improve net income, but greatly reduce GHG emissions and GWP. The recommended N application rate was 423 kg·ha −1 in 2018 and 524 kg·ha −1 in 2019, that maximized net income in wolfberry orchards with similar fertility. The nitrapyrin combined with N application could increase wolfberry fruit yield but reduce GHG emissions and N loss. Both management practices had positive effects, decreasing N2O emission, while having no significant impact on CH4 uptake and CO2 emission. This strategy should be considered when the main concern is mitigating N2O emission. The recommended N management was 400 kg·ha −1 N combined with 2.00 kg·ha −1 nitrapyrin maximized net income in the production of wolfberry in Qaidam. However, if simultaneously considering on economic profit and ecological benefit, 267 kg·ha −1 N combined with 1.33 kg·ha −1 was the optimal N application management that could maintain net income, but save N input and reduce N pollution.