Interannual variation in grassland net ecosystem productivity and its coupling relation to climatic factors in China

Grazing decreases net ecosystem carbon exchange by decreasing shrub and semi-shrub biomass in a desert steppe

Livestock grazing can strongly determine how grasslands function and their role in the carbon cycle. However, how ecosystem carbon exchange responds to grazing and the underlying mechanisms remain unclear. We measured ecosystem carbon fluxes to explore the changes in carbon exchange and their driving mechanisms under different grazing intensities (CK, control; HG, heavy grazing; LG, light grazing; MG, moderate grazing) based on a 16-year long-term grazing experimental platform in a desert steppe. We found that grazing intensity influenced aboveground biomass during the peak growing season, primarily by decreasing shrubs and semi-shrubs and perennial forbs. Furthermore, grazing decreased net ecosystem carbon exchange by decreasing aboveground biomass, especially the functional group of shrubs and semi-shrubs. At the same time, we found that belowground biomass and soil ammonium nitrogen were the driving factors of soil respiration in grazed systems. Our study indicates that shrubs and semi-shrubs are important factors in regulating ecosystem carbon exchange under grazing disturbance in the desert steppe, whereas belowground biomass and soil available nitrogen are important factors regulating soil respiration under grazing disturbance in the desert steppe; this results provide deeper insights for understanding how grazing moderates the relationships between soil nutrients, plant biomass, and ecosystem CO2 exchange, which provide a theoretical basis for further grazing management.

Spatio-temporal pattern of urban vegetation carbon sink and driving mechanisms of human activities in Huaibei, China

Carbon neutrality is a strategic choice for the sustainable development of global cities. Quantitatively assessing the spatio-temporal patterns of urban vegetation carbon sink and the impact of human activities has become an essential basis for adjusting urban carbon balance. We used Huaibei, a typical city with vigorous human coal resource mining activities, as the case study area. We regarded the net ecosystem productivity (NEP) as an indicator parameter of vegetation carbon sink and calculated it based on the improved Carnegie-Ames-Stanford approach (CASA). We then revealed the spatial-temporal evolution of vegetation carbon sink through trend analysis, coefficient of variation, and standard direction. Finally, we used geographic detectors to evaluate the impact of human activities on NEP. We found that net primary productivity (NPP) accuracy was good, and the R² value was 0.755 compared with MODIS NPP products. NEP was characterized by the first decrease and then increase, showing a slow increase overall, with an average trend coefficient of 0.15 gC·m^-2·a^-1. The average value in 2010 was the lowest at 18.30 gC·m^-2·a^-1. In terms of spatial characteristics, NEP showed a gradual decrease from north to south. High and severe fluctuations were distributed along the southeast, mainly concentrated in Duji District, Xiangshan District, and Lieshan District. The driving factors with reliable explanatory power for NEP were population density, GDP, and road density, while land use type, soil erosion intensity, and mining and collapse area had weak explanatory power. Meanwhile, factors of cooperative interaction enhanced the explanatory power of the results.

Quantifying the Ecosystem Services of Soda Saline-Alkali Grasslands in Western Jilin Province, NE China

This study aimed to quantitatively describe the ecosystem services of soda saline-alkali grasslands based on literature research, the InVEST model, a transition matrix, and Spearman's correlation analysis. The chosen methodology could provide insight into the relationships between different services to provide empirical evidence for decision-making concerning the protection and restoration of saline-alkali grasslands. The research provided several insights into the ecological situation in western Jilin Province. First, the area of saline-alkali grassland in western Jilin Province had noticeably decreased from 1990 to 2018. Moreover, the threat of grassland degradation in western Jilin Province has increased year by year, and has become the main problem facing the ecological environment of this region. Second, the results demonstrated how the amount of grassland area, and coverage, are intricately linked to the provided ecosystem services, and maintaining the stability of ecosystem services is the basis for future efforts to increase grassland area and coverage. A trade-off relationship exists between water supply services and other ecosystem services, which indirectly confirms a climatic cause for grassland salinization in western Jilin Province. The analyses identified various types of grassland ecosystem service hotspots, but the share of hotspots representing all four assessed ecosystem services was small; this indicates that the grassland ecosystem of western Jilin Province is of generally poor quality. In conclusion, increasing grassland salinization has reduced vegetation coverage, which leads to the degradation of the grassland ecosystem and, in turn, affects the relationships between various ecosystem services.

Simulation of Soil Organic Carbon Content Based on Laboratory Spectrum in the Three-Rivers Source Region of China

Soil organic carbon (SOC) changes affect the land carbon cycle and are also closely related to climate change. Visible-near infrared spectroscopy (Vis-NIRS) has proven to be an effective tool in predicting soil properties. Spectral transformations are necessary to reduce noise and ensemble learning methods can improve the estimation accuracy of SOC. Yet, it is still unclear which is the optimal ensemble learning method exploiting the results of spectral transformations to accurately simulate SOC content changes in the Three-Rivers Source Region of China. In this study, 272 soil samples were collected and used to build the Vis-NIRS simulation models for SOC content. The ensemble learning was conducted by the building of stack models. Sixteen combinations were produced by eight spectral transformations (S-G, LR, MSC, CR, FD, LRFD, MSCFD and CRFD) and two machine learning models of RF and XGBoost. Then, the prediction results of these 16 combinations were used to build the first-step stack models (Stack1, Stack2, Stack3). The next-step stack models (Stack4, Stack5, Stack6) were then made after the input variables were optimized based on the threshold of the feature importance of the first-step stack models (importance > 0.05). The results in this study showed that the stack models method obtained higher accuracy than the single model and transformations method. Among the six stack models, Stack 6 (5 selected combinations + XGBoost) showed the best simulation performance (RMSE = 7.3511, R2 = 0.8963, and RPD = 3.0139, RPIQ = 3.339), and obtained higher accuracy than Stack3 (16 combinations + XGBoost). Overall, our results suggested that the ensemble learning of spectral transformations and simulation models can improve the estimation accuracy of the SOC content. This study can provide useful suggestions for the high-precision estimation of SOC in the alpine ecosystem.