Effects of Different Land Use Types on Soil N2O Fluxes on the Loess Plateau

DU Jun¹, LI Guang¹,²,³, DU Mengyin¹, YAO Yao¹, MA Weiwei¹, YUAN Jianyu¹,²,³

¹College of Forestry, Gansu Agricultural University, Lanzhou 730070, Gansu, China
²State Key Laboratory of Arid Land Crops Science, Gansu Agricultural University, Lanzhou 730070, Gansu, China
³College of Prataculture, Gansu Agricultural University, Lanzhou 730070, Gansu, China

Abstract

Nitrous oxide (N₂O) is a significant greenhouse gas in the atmosphere that exerts a pronounced influence on global climate warming. Changes in land use patterns constitute a critical factor affecting N₂O emissions, particularly in ecologically fragile semiarid regions where the underlying mechanisms are more complex. However, systematic research remains lacking on how the diverse land use types in China's semiarid regions influence soil N₂O emissions and what key driving factors control these emissions. To address this knowledge gap, this study examined four typical land use types in the semiarid Loess Plateau of central Gansu Province: Picea asperata forest, Medicago sativa grassland, abandoned land, and wheat field. Soil N₂O fluxes were monitored using the static chamber–gas chromatography method, combined with soil physicochemical property data, to reveal the key drivers regulating soil N₂O emissions under different land use patterns.

The results demonstrated that: (1) Compared with abandoned land, the Picea asperata forest and Medicago sativa grassland significantly increased soil water content, while wheat fields elevated ammonium nitrogen (NH₄⁺-N) and nitrate nitrogen (NO₃⁻-N) concentrations. (2) The Medicago sativa grassland and wheat fields markedly enhanced nitrate reductase (NR) and nitrite reductase (NiR) activities, with enzyme activities in all treatments decreasing with soil depth. (3) Soil N₂O fluxes under different land use types exhibited an initial increase followed by a decline during vegetation growth stages. Total soil N₂O emissions decreased by 34.2% and 23.3% in the Picea asperata forest and Medicago sativa grassland, respectively, but increased significantly by 32.47% in wheat fields compared to abandoned land. (4) Random forest analysis identified soil temperature as the most influential factor affecting soil N₂O flux. Overall, artificial forest and grassland systems demonstrated superior emission reduction effects compared to abandoned land and wheat fields. Future vegetation restoration and ecological rehabilitation efforts should prioritize optimizing the proportional allocation of "forest-grass-cropland" land use patterns and appropriately increasing the coverage of artificial forests and grasslands to achieve dual objectives of ecological benefits and emission mitigation.

Keywords: Loess Plateau; global climate warming; land use types; N₂O emission fluxes; soil temperature

Nitrous oxide (N₂O) ranks among the three most important greenhouse gases in the atmosphere, with a global warming potential approximately 298 times that of carbon dioxide on a century timescale. It persists in the atmosphere for extended periods, participates in various photochemical reactions, and contributes to ozone depletion [1]. The N₂O emission rate has been increasing at 0.2–0.3% annually, reaching approximately 17.0 Tg·a⁻¹ in recent years [2]. Anthropogenic activities, including land use changes and extensive fossil fuel consumption, represent significant causes of increased greenhouse gas emissions and global warming. Reducing greenhouse gas emissions from production and daily life has become an essential development goal for all nations [3]. Consequently, the impact of N₂O emissions on global climate change and its role in biogeochemical cycles are receiving growing international attention.

Nitrification and denitrification serve as the key processes for N₂O production in soils. Nitrification refers to the microbial conversion of ammonium to oxidized nitrogen forms (NO₂⁻, NO₃⁻) under aerobic conditions, while denitrification is a microbial ecological process in which denitrifying bacteria reduce nitrate and nitrite to gaseous nitrogen oxides (NO, N₂O, N₂) under anaerobic conditions [4]. Land use changes directly or indirectly affect N₂O production, consumption, and diffusion by altering plant community composition and soil properties, thereby modifying emission patterns [5]. Regulating plant and microbial activities can significantly influence N₂O emissions [6]. In grassland ecosystems, different land use types can alter plant community and soil characteristics—such as soil temperature, moisture, bulk density, and nitrogen availability—leading to increased carbon and nitrogen availability in grassland soils and consequently enhanced N₂O emissions [7]. In cropland soils, N₂O emissions are primarily influenced by soil texture, mineral nitrogen, and organic carbon [8]. Soil enzymes involved in organic matter decomposition and transformation also affect N₂O emission rates [9]. Thus, different land use types regulate N₂O emission dynamics by modifying soil nitrogen cycling and enzyme activities. Elucidating the interactions between soil carbon and nitrogen metabolism under human disturbance is crucial for optimizing land management models, maintaining regional carbon-nitrogen balance, mitigating greenhouse gas emissions, and curbing climate warming [10].

1 Materials and Methods

1.1 Study Area Description

The experimental site was located in the Soil and Water Conservation Monitoring Station (34°26′–35°35′N, 103°52′–105°13′E) in Anding District, Dingxi City, Gansu Province [FIGURE:6]. This region features a temperate semiarid climate with abundant sunlight and large temperature variations, belonging to the rain-fed agricultural zone of the Loess Plateau in central Gansu, with an elevation of 2000 m and mean annual precipitation of approximately 391 mm concentrated between June and September [11]. The area experiences chronic drought and water scarcity with sparse vegetation. Since 1999, the Grain for Green Project has been implemented, converting original cropland to forest and grassland. The main tree species include Picea asperata and Platycladus orientalis, herbaceous vegetation primarily comprises Medicago sativa and Onobrychis viciifolia, and major crops include spring wheat (Triticum aestivum), maize (Zea mays), and potato (Solanum tuberosum).

1.2 Experimental Design

Based on field surveys of ecological and vegetation characteristics and literature review, sample plots were established in July 2021 in vegetation areas with similar soil types and disturbance histories. Four land use types were selected as treatments: Picea asperata forest (PA), Medicago sativa grassland (MS), abandoned land (AL), and wheat field (WF) (Table 1). The Picea asperata forest was restored through the Grain for Green Project in 1999 without further management. The Medicago sativa grassland was planted and then enclosed after establishment. The abandoned land was left fallow since 2015 and naturally recovered with sparse weeds, receiving no management. The wheat field was converted from wasteland in 2015 and planted with spring wheat "Ganchun 41" using conventional tillage, with base fertilizer (150.0 kg·hm⁻² urea and 62.5 kg·hm⁻² calcium superphosphate) applied at sowing.

For each treatment, three 20 m × 20 m fixed sampling areas were randomly selected. Within each plot, a 0.5 m × 0.5 m fixed gas sampling area was established. During the 2022 vegetation growing season, soil (0–20 cm) and gas samples were collected biweekly (at the beginning and middle of each month) and transported to the laboratory for analysis.

1.3 Soil Sample Collection

Soil samples were collected three times during the 2022 plant growing season, corresponding to the initial growth stage (mid-April), mid-growth stage (mid-July), and late growth stage (mid-October) [12]. Using a soil auger, samples were taken from 0–10 cm and 10–20 cm depths following the five-point sampling method. After removing impurities, samples were placed in ziplock bags, stored in ice-filled coolers, and transported to the laboratory. Samples were divided into two portions: one portion of fresh soil was sieved through a 2 mm mesh, stored in a 4°C refrigerator for determination of soil water content, nitrate nitrogen, and ammonium nitrogen; the other portion was air-dried and sieved for analysis of nitrate reductase and nitrite reductase activities. Soil water content was determined by the aluminum box drying method, soil temperature was automatically monitored every 30 minutes using a data logger, and soil nitrogen forms were analyzed using the Devarda's alloy distillation method for simultaneous determination of nitrate and ammonium nitrogen [13]. For enzyme activity analysis, 0.5–1 g of air-dried soil sieved through 0.5–1.5 mm mesh was used. Nitrate reductase activity was measured by the sulfanilic acid–naphthylamine colorimetric method, and nitrite reductase activity was determined by the naphthylamine colorimetric method and sulfanilic acid method.

1.4 Gas Collection

Soil N₂O flux was measured using a 50 cm × 50 cm × 50 cm static dark chamber made of thin stainless steel plates, with two air-mixing fans installed on the top to ensure uniform air mixing during flux measurement. The chamber walls were wrapped with insulation material to minimize temperature changes. During gas sampling, the chamber and base were sealed with water. After chamber closure, 100 mL air samples were collected at 8-minute intervals (0, 8, 16, 24, and 32 minutes) using a 60 mL polypropylene syringe equipped with a three-way stopcock. Samples were immediately transported to the laboratory and analyzed using gas chromatography to determine N₂O flux variations under different treatments.

1.5 Data Analysis

To compare differences in soil physicochemical properties among treatments, soil layers, and growth stages, one-way ANOVA and independent samples t-tests were performed using SPSS 26.0, with figures generated using Origin 2021. The Hmisc package in R 4.3.3 was used to calculate correlation coefficients between soil environmental factors and N₂O emissions, while the rfPermute package was employed to analyze key physicochemical factors influencing N₂O emissions. All statistical analyses were completed in R 4.3.3, with visualizations based on ggplot2.

2 Results

2.1 Effects of Land Use Types on Soil Physicochemical Properties

2.1.1 Soil Temperature

Land use types significantly affected soil temperature. All treatments showed similar temperature trends, increasing initially then decreasing with vegetation growth stages. Throughout the growing season, soil temperature in the wheat field (WF) was higher than other treatments, following the order WF > AL > MS > PA. At 10 cm depth, WF soil temperature was significantly higher than other treatments (P < 0.05).

2.1.2 Soil Water Content

Land use types significantly influenced soil water content. Compared to AL, PA and MS significantly increased soil water content (P < 0.05), while WF significantly decreased it (P < 0.05), following the order PA > MS > AL > WF. Across the soil profile (0–20 cm), water content gradually increased with depth. During the growing season, soil water content showed a decreasing then increasing trend.

2.1.3 Effects on Soil Nitrate and Ammonium Nitrogen

Compared to AL, WF significantly increased soil NO₃⁻-N content by 25.20% and NH₄⁺-N content by 21.28% (P < 0.05), while PA decreased NO₃⁻-N by 2.39% and NH₄⁺-N by 31.56% (P < 0.05). During the growing season, NO₃⁻-N content in PA, MS, and AL showed a decreasing then increasing trend, while WF exhibited an increasing then decreasing pattern. NH₄⁺-N content in PA, MS, and AL also showed a decreasing then increasing trend, while WF displayed a gradual decline. In the 0–10 cm layer, NO₃⁻-N and NH₄⁺-N contents were significantly higher than in the 10–20 cm layer across all land use types (P < 0.05).

2.2 Effects of Land Use Types on Soil Enzyme Activities

2.2.1 Nitrate Reductase Activity

Land use types significantly affected soil nitrate reductase (NR) activity during the growing season [FIGURE:6]. In the initial and mid-growth stages, WF showed the highest NR activity, while in the late growth stage, MS exhibited significantly higher NR activity than other treatments (P < 0.05). In the vertical soil profile, NR activity in the 0–10 cm layer was significantly higher than in the 10–20 cm layer (P < 0.05). Compared to AL, WF and MS increased soil NR activity by 15.56% and 30.73%, respectively.

2.2.2 Nitrite Reductase Activity

Land use types significantly influenced soil nitrite reductase (NiR) activity during the growing season [FIGURE:7]. In the initial and mid-growth stages, WF showed the highest NiR activity, while in the late growth stage, MS had significantly higher NiR activity than other treatments (P < 0.05). Compared to AL, WF and MS significantly increased soil NiR activity by 30.73% and 15.56%, respectively (P < 0.05). Across different land use types, soil NiR activity showed an initial increase then decrease trend with growth stages, being significantly higher in the mid-growth stage than in initial and late stages (P < 0.05). In the vertical profile, 0–10 cm NiR activity was significantly higher than in the 10–20 cm layer (P < 0.05).

2.3 Dynamic Changes in N₂O Emission Fluxes Under Different Land Use Types

During the vegetation growing season, soil N₂O emission fluxes under different land use types showed distinct seasonal variation patterns [FIGURE:8]. In the initial growth stage, N₂O flux increased gradually with rising temperatures. In the mid-growth stage, all treatments reached maximum emission fluxes. In the late growth stage, N₂O flux decreased slowly and stabilized with falling temperatures. Overall, WF N₂O flux and cumulative emissions were significantly higher than other treatments throughout the growing season, with a pulse emission peak occurring in the initial growth stage.

2.4 Effects of Land Use Types on Cumulative N₂O Emissions

Cumulative N₂O emissions differed significantly among land use types [FIGURE:9]. Compared to AL, PA and MS significantly reduced cumulative N₂O emissions by 34.2% and 23.3%, respectively (P < 0.05), while WF significantly increased emissions by 32.47% (P < 0.05).

2.5 Relationships Among Soil Environmental Factors, Enzyme Activities, and N₂O Emission Fluxes

Correlation analysis revealed that soil temperature was significantly positively correlated with N₂O emission flux (P < 0.05), while soil water content was significantly negatively correlated with N₂O flux (P < 0.05) [FIGURE:10]. Random forest results indicated that soil temperature was the most important factor influencing N₂O emissions, representing the key driving factor.

3 Discussion

3.1 Effects of Land Use Types on N₂O Fluxes

During the vegetation growing season, soil N₂O emissions under different land use types showed an initial increase followed by a decrease, with peaks occurring in the mid-growth stage. This pattern likely resulted from frequent low-intensity precipitation and higher temperatures during the mid-growth stage, creating moist soil conditions that facilitated microbial survival and activity, accelerating nitrification and denitrification processes and thereby promoting N₂O emissions [14]. Additionally, frequent wet-dry cycles created numerous anaerobic microsites in the soil surface. When soil water content was high, denitrification dominated N₂O production; when water content decreased, nitrification became the primary process. However, denitrification should not be overlooked during the mid-growth stage.

Notably, WF exhibited a pulse emission peak in the initial growth stage. Different land use types altered fundamental soil properties, leading to significant differences in cumulative N₂O emissions. In this study, WF had the highest cumulative N₂O emissions, consistent with Wu et al. [15]. This high emission resulted from two synergistic effects: (1) Base fertilizer application significantly increased nitrogen concentration and availability, providing substrates for nitrifying and denitrifying bacteria and ultimately causing substantial increases in soil N₂O cumulative emissions [16]; (2) Long-term tillage promoted organic matter mineralization, releasing large amounts of ammonium and nitrate that further stimulated microbial metabolism and increased N₂O emissions [17]. In contrast, PA and MS had relatively low cumulative emissions, likely due to minimal anthropogenic disturbance [18].

3.2 Effects of Soil Physicochemical Properties on N₂O Fluxes

Soil moisture regulates nitrification and denitrification by altering soil oxygen conditions and is a critical factor affecting nitrogen substrate diffusion to microbial communities. Tang et al. [19] demonstrated that lower moisture content favored nitrification compared to high moisture conditions across different land use types. In this study, soil water content was significantly negatively correlated with N₂O emissions, consistent with previous research [20]. The PA treatment had the highest soil water content, while WF had the lowest. This difference may relate to vegetation cover and water movement—PA's high vegetation coverage effectively shaded the soil surface, reducing evaporation, while its developed root system increased water infiltration, enhancing nitrate leaching and reducing denitrification substrates, thereby decreasing N₂O emissions [21]. Additionally, forest litter with high lignin concentrations reduced nitrogen availability, further decreasing N₂O emissions [22].

Besides soil moisture, soil temperature is another key driver of N₂O emissions. Although some researchers consider soil temperature less important, our random forest analysis identified it as the most influential factor, with correlation analysis showing a significant positive relationship with N₂O flux. This result stems from temperature's regulatory effect on soil microorganisms—nitrifying bacteria activity increases 1.5–3 times for every 10°C temperature rise, accelerating nitrification and N₂O release [23]. Zhang et al. [24] also found that nitrogen fertilizer addition activated nitrification enzymes. In this study, WF had higher soil temperatures than PA throughout the growing season because WF's lower vegetation cover allowed more solar radiation to reach the soil surface. The longer phenological period of PA vegetation also contributed to lower temperatures. Growth stage changes further affected N₂O emissions, with mid-growth stage fluxes higher than initial stages due to optimal hydrothermal conditions promoting microbial growth and enzyme synthesis [25].

3.3 Effects of Soil Enzyme Activities on N₂O Fluxes

Nitrate reductase and nitrite reductase are two key enzymes in the denitrification process, and their activity levels serve as important indicators of soil denitrification capacity. This study found a significant positive correlation between soil N₂O emission flux and both NR and NiR activities, consistent with most research findings [26]. The underlying mechanism may involve high NR activity generating substantial NO₂⁻, providing ample substrate for hydroxylamine formation and subsequent N₂O production. However, when NiR activity increases beyond a certain threshold, the catalytic reaction becomes more complete, reducing accumulation of the intermediate product N₂O [27].

Significant differences in NR and NiR activities were observed among land use types. WF significantly increased both enzyme activities due to nitrogen fertilizer application, which promoted accumulation of soil inorganic nitrogen and provided sufficient substrates for nitrification. The relatively short phenological period of WF crops meant that nitrogen demand decreased gradually, while surface litter decomposition returned some nitrogen to the soil, further increasing N₂O emissions. In contrast, PA's large vegetation biomass absorbed substantial inorganic nitrogen, reducing soil N₂O emissions. Additionally, denitrification enzymes are influenced by soil temperature and moisture, which affect denitrifying microorganisms and regulate denitrification rates and enzyme production [28]. During the mid-growth stage, favorable hydrothermal conditions promoted microbial growth and denitrification enzyme synthesis, increasing NiR activity. In the late growth stage, MS showed significantly higher NiR activity than other treatments, likely because abundant alfalfa litter input substantial effective resources to the mineral soil layer through leaching, thereby enhancing NiR activity.

4 Conclusions

This study systematically investigated N₂O emission characteristics and driving factors under typical land use types in the Loess Plateau of central Gansu. The main conclusions are:

1. Land use types significantly affected soil physicochemical properties. Compared with other treatments, the Picea asperata forest significantly reduced soil temperature and increased soil water content, while wheat fields significantly increased soil ammonium and nitrate nitrogen contents.

2. Compared with other treatments, wheat fields significantly increased soil nitrate reductase and nitrite reductase activities due to nitrogen fertilizer application, with enzyme activities in the 0–10 cm layer significantly higher than in the 10–20 cm layer.

3. Soil N₂O fluxes under different land use types showed an initial increase then decrease trend with vegetation growth stages. Compared with other treatments, the Picea asperata forest significantly reduced total N₂O emissions, while wheat fields showed a significant increasing trend. Random forest analysis identified soil temperature as the most important factor influencing N₂O emissions, representing the primary driving factor.

References

[1] Ravishankara R A, Daniel S J, Portmann W R. Nitrous Oxide (N₂O): The dominant ozone depleting substance emitted in the 21st century[J]. Science, 2009, 326(5949): 123-125.

[2] IPCC. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change[R]. Cambridge: Cambridge University Press, 2013.

[3] Snyder C, Bruulsema T, Jensen T, et al. Review of greenhouse gas emissions from crop production systems and fertilizer management effects[J]. Agriculture, Ecosystems and Environment, 2009, 133(3-4): 247-266.

[4] Zhang Ruoyu. A Study on Radiative Forcing and Global Warming Potential of Long-life Greenhouse Gases CH₄ and N₂O[D]. Nanjing: Nanjing University of Information Science, 2011.

[5] Xu Yuxiu, Guo Liping, Xie Liyong, et al. Characteristics of background emissions and emission factors of N₂O from major upland fields in China[J]. Scientia Agricultura Sinica, 2016, 49(9): 1729-1743.

[6] Wu Jianli, Liu Mengyun, Zhao Guoqing, et al. Effects of land use types on soil organic carbon mineralization and greenhouse gas emissions in Loess tableland[J]. Journal of Agro-Environment Science, 2016, 35(5): 1006-1015.

[7] Du Ziyin, Cai Yanjiang, Zhang Bin, et al. Research progress on livestock excreta returning on soil nitrogen transformation and nitrous oxide emission in grasslands[J]. Acta Ecologica Sinica, 2022, 42(1): 45-57.

[8] Milin, Ye Guiping, Xu Chao. Effects of forest types and nitrogen addition on soil nitrous oxide emissions in a subtropical forest[J]. Journal of Soil and Water Conservation, 2023, 37(6): 262-267.

[9] Yang Chuanjie, Li Guang, Yan Lijuan, et al. Effects of different vegetation types on ecosystem respiration in semiarid Loess Hilly Region, Central Gansu Province, China[J]. Ecological Indicators, 2022, 145: 109683.

[10] Yuan J, Yao Y, Guan Y, et al. Effects of land use patterns on soil properties and nitrous oxide flux on a semiarid environmental condition of Loess Plateau China[J]. Global Ecology and Conservation, 2024, 51: e02899.

[11] Qi Xiaoping, Li Guang, Yuan Jianyu, et al. Effects of conservation tillage on the water storage, moisture conservation, and yield of land wheat fields of central Gansu Province[J]. Arid Zone Research, 2022, 39(1): 312-321.

[12] Song Wenhao, Yang Shiwei, Gao Xiaodong, et al. Characteristics of CO₂ and N₂O emissions under different land use types in Loess Hilly Region of China[J]. Research of Soil and Water Conservation, 2017, 24(1): 68-74.

[13] Cao Xinde, Chen Ying, Wang Xiaorong. The effects of environmental conditions on the release and species transformation of rare earth elements (REEs) in soil[J]. China Environmental Science, 2000, 20(6): 486-490.

[14] Sun Zhiqiang, Hao Qingju, Jiang Changsheng, et al. Advances in the study of nitrous oxide production mechanism and its influencing factors in agricultural soils[J]. Chinese Journal of Soil Science, 2010, 41(6): 1524-1530.

[15] Chen Wei, He Mingzhu, Li Chengyi, et al. A review of research progress on nitrous oxide emissions from desert soil and its driving factors[J]. Journal of Desert Research, 2023, 43(2): 184-194.

[16] Meng Yi, Liao Ping, Wei Haiyan, et al. Effects of gypsum application on grain yield and methane emissions in rice paddies: A global meta-analysis[J]. Chinese Journal of Eco-Agriculture, 2023, 31(2): 280-289.

[17] Xiao Wei, Wu Qingfeng, Li Fusheng. Effect of drip irrigation and nitrogen management on inorganic nitrogen content and nitrous oxide emission in maize planting soil[J]. Journal of South China Agricultural University, 2023, 44(3): 410-419.

[18] Žurovec Ognjen, Sitaula Bishal Kumar, Čustović Hamid, et al. Effects of tillage practice on soil structure, N₂O emissions and economics in cereal production under current socio-economic conditions in central Bosnia and Herzegovina[J]. PloS One, 2017, 12(11): e0187681.

[19] Tang Ruijie, Hu Yujie, Zhao Caiyue, et al. Effects of land use conversion on soil nitrification and NO & N₂O emissions in tropical China under different moisture conditions[J]. Environmental Science, 2022, 43(11): 5159-5168.

[20] Shi Hongai, Li Lujun, You Mengyang, et al. Impact of soil temperature and moisture on soil N₂O emission from mollisols under different land use types[J]. Journal of Agro-Environment Science, 2013, 32(11): 2286-2292.

[21] Zhao Hairong, Shuai Wei, Li Jing, et al. Interception of NH₄⁺-N and NO₃⁻-N during rainfall processes in several typical plantations in the rainy area of western China[J]. Acta Scientiae Circumstantiae, 2015, 35(11): 3710-3718.

[22] Li Yuan, Wang Tao, Zhou Ying, et al. Study on the seasonal dynamics of N₂O release flux in three cultivated grasslands[J]. Acta Agrestia Sinica, 2015, 23(4): 703-709.

[23] Cao Wenchao, Song He, Wang Yajing, et al. Key production processes and influencing factors of nitrous oxide emissions from agricultural soils[J]. Journal of Plant Nutrition and Fertilizers, 2019, 25(10): 1781-1798.

[24] Zhang Yulan, Chen Zhenhua, Ma Xingzhu, et al. Response of soil enzymes to different amounts of N and P fertilizers in rice ecosystems of meadow brown soil[J]. Chinese Journal of Soil Science, 2008, 39(3): 518-523.

[25] Wang Suping, Hui Jianchun, Lin Lijin, et al. Effects of fertilizer system on soil denitrification enzyme activity at different growth stages of maize in hilly regions of middle Sichuan[J]. Research of Soil and Water Conservation, 2012, 19(3): 274-277, 283.

[26] Li Jiachen, Hou Lei, Wang Yanxia, et al. Effects of nitrogen input on N₂O production and enzyme activity in Luoshijiang Wetland sediments, China[J]. Chinese Journal of Applied Ecology, 2023, 34(2): 405-414.

[27] Fang Huajun, Cheng Shulan, Yu Guirui, et al. Study on the responses of nitrous oxide emission to increased nitrogen deposition in forest soils: A review[J]. Acta Pedologica Sinica, 2015, 52(2): 262-271.

[28] Yang Yongkai, Ma Weiwei, Chen Hao, et al. Response of soil enzyme activities in marsh meadow in Gahai wet area to short-term warming and nitrogen addition[J]. Acta Agrestia Sinica, 2022, 30(12): 3263-3271.