Publications
Sort:
Issue
Inventory and Mitigation Potential of Nitric Oxide Emissions from Agricultural Soils in China Based on Localized Emission Factors
Scientia Agricultura Sinica 2026, 59(19): 4289-4300
Published: 01 October 2026
Abstract PDF (571 KB) Collect
Downloads:2
Objective

Nitric oxide (NO) is not only a crucial precursor for tropospheric ozone formation but also participates in acid rain formation and nitrogen deposition processes, posing a severe threat to ecological security. Agriculture is a key anthropogenic source of NO. Given China’s substantial nitrogen fertilizer consumption, NO emissions from its agricultural soils warrant attention. Addressing the high uncertainty associated with current estimations that rely on global default emission factors (EFs) or a single national EF, this study aims to construct a localized EF dataset reflecting the characteristics of China's geographical regions, crops, and fertilizer types. Furthermore, it seeks to calculate the 2021 NO emission inventory for Chinese cropland soils and assess mitigation potentials, thereby providing a scientific basis for green agricultural development.

Method

This study systematically retrieved 54 field in-situ observation papers from the CNKI and Web of Science databases, extracting 440 pairs of full-growth period data. Three EF calculation methods-Maximum Likelihood Estimation (M1), Linear Regression With Intercept (M2), and Linear Regression Without Intercept (M3)-were compared. Adopting a “70% calculation+30% independent validation” mode, the optimal algorithm was screened to calculate localized EFs for China. By combining statistical data on crop planting areas and nitrogen application rates across provinces in 2021, the NO emission inventory for Chinese croplands was calculated, and uncertainty was quantified using the error propagation method. Finally, optimized management scenarios, such as fertilizer reduction and organic substitution, were established to predict NO mitigation potential.

Result

Independent validation results indicated that the M1 method yielded the highest prediction accuracy (R2=0.56). The calculated localized EFs demonstrated significant differences across crops and fertilization practices. Under sole synthetic fertilizer application, tea plantations exhibited the highest EF (1.91%), while rice paddies had the lowest (0.27%). Combined application of organic and inorganic fertilizers generally resulted in lower EFs compared to sole synthetic fertilizer application. Spatially, EFs presented a pattern of being higher in the southeast and lower in the northwest. It was estimated that the total NO emissions induced by nitrogen fertilizer application in Chinese croplands in 2021 were 150 kt NO-N, with an uncertainty of 9.1%. Regarding crop contribution, grain crops were the primary emission source, accounting for 49.9% of the total emissions. Among them, maize contributed the most (31.8 kt NO-N) due to its large planting area and high nitrogen application rates. Despite their smaller planting areas, tea plantations and sugarcane exhibited high emission intensities per unit area due to intensive nitrogen inputs, reaching 7.02 kg NO-N·hm-2 and 2.55 kg NO-N·hm-2, respectively. Regionally, East China contributed 40.9% of the total NO emissions from cereal crops nationwide (30.6 kt NO-N). Scenario analysis indicated significant mitigation potential for NO emissions from Chinese agricultural soils. Through the implementation of scientific nitrogen reduction and the promotion of organic fertilizer substitution technologies, the national NO mitigation potential could reach 24%, equivalent to an annual reduction of 36 kt NO-N. East, North, and Northeast China together accounted for 78.9% of the national mitigation potential from cereal crops.

Conclusion

By constructing a localized emission factor (EF) dataset based on field measurements, this study controlled the uncertainty of NO emission estimates from Chinese cropland soils to within 9.1%. However, due to the scarcity of field measurement data in regions such as Northeast and Northwest China, there remains some uncertainty in regional estimates. This study clarified the significant spatial heterogeneity and crop specificity of NO emissions from Chinese agricultural soils, identifying East China and high-nitrogen-input cash crops as priority areas for future NO mitigation and control. The findings confirm that optimizing chemical fertilizer application rates and implementing organic fertilizer substitution are effective pathways for reducing agricultural soil NO emissions. These results provide scientific support for formulating differentiated, region- and crop-specific agricultural nitrogen management policies and air pollution control measures.

Issue
A Review of Soil Carbon Sequestration and Greenhouse Gas Emissions Quantification in Cropland
Scientia Agricultura Sinica 2025, 58(11): 2190-2205
Published: 01 June 2025
Abstract PDF (565.8 KB) Collect
Downloads:53

As significant sources and sinks of greenhouse gases in terrestrial ecosystems, cropland can effectively mitigate climate change through scientific management practices. Developing methods for quantifying soil carbon sequestration and greenhouse gas emissions is crucial for accurately measuring the contribution of agriculture to carbon neutrality strategies. These methods are also essential for greenhouse gas inventory preparation, evaluating carbon sequestration and emission reduction effects, voluntary emission reduction trading, and carbon-related certification. This paper systematically reviewed the methods for quantifying soil carbon sequestration and greenhouse gas emissions in cropland, comparing the differences among various methods in terms of data requirements, acquisition costs, estimation accuracy, and the challenges faced. It focused on the applicability of these methods in different scenarios and their optimization pathways. The methods discussed for soil carbon sequestration and greenhouse gas emissions included the Two-Period Difference Method (for soil carbon sequestration only), Data Integration Method (for soil carbon sequestration only), Parameter Method/Emission Factor Method, Empirical Models, and Process Models. The Two-Period Difference Method and Data Integration Method estimate changed in soil organic carbon using empirical data, but their accuracy was directly affected by the matching of datasets and the amount of data. The Parameter Method/Emission Factor Method estimated soil carbon sequestration and greenhouse gas emissions with minimal data requirements, but its accuracy depended on the localization and refinement of parameters/emission factors. Due to these limitations, the Parameter Method/Emission Factor Method was more suitable for low-resolution estimates at regional scales and could not accurately capture high-resolution spatial variations. Empirical Models and Process Models considered the effects of meteorological, soil, and management factors on soil carbon sequestration and greenhouse gas emissions, making them suitable for high-resolution spatial simulations and field-scale measurements. The performance of empirical models depended on the quantity, quality, and accuracy of variable selection in the modeling database. In contrast, Process-based Models could capture the spatiotemporal dynamics of soil organic carbon and greenhouse gases under complex conditions, but required local parameter optimization for accurate performance. Given the challenges of different methods, this paper highlighted the importance of balancing data requirements, costs, and estimation accuracy in specific-scenarios. It noteed that model simulation methods would become the main trend in future agricultural carbon quantification. Addressing how to balance data requirements, acquisition costs, and estimation accuracy for different scenarios was a significant challenge. Future research should focus on building comprehensive field monitoring networks, promoting standardized database construction and sharing, deepening the localization and refinement of emission factors, and exploring innovative applications of machine learning in multi-algorithm integration and process model parameter adjustment.

Total 2