The net ecosystem CO₂ exchange (NEE) is a critical parameter for quantifying terrestrial ecosystems and their contributions to the ongoing climate change. The accumulation of ecological data is calling for more advanced quantitative approaches for assisting NEE prediction. In this study, we applied two widely used machine learning algorithms, Random Forest (RF) and Extreme Gradient Boosting (XGBoost), to build models for simulating NEE in major biomes based on the FLUXNET dataset. Both models accurately predicted NEE in all biomes, while XGBoost had higher computational efficiency (6~62 times faster than RF). Among environmental variables, net solar radiation, soil water content, and soil temperature are the most important variables, while precipitation and wind speed are less important variables in simulating temporal variations of site-level NEE as shown by both models. Both models perform consistently well for extreme climate conditions. Extreme heat and dryness led to much worse model performance in grassland (extreme heat: R² = 0.66~0.71, normal: R² = 0.78~0.81; extreme dryness: R² = 0.14~0.30, normal: R² = 0.54~0.55), but the impact on forest is less (extreme heat: R² = 0.50~0.78, normal: R² = 0.59~0.87; extreme dryness: R² = 0.86~0.90, normal: R² = 0.81~0.85). Extreme wet condition did not change model performance in forest ecosystems (with R² changing −0.03~0.03 compared with normal) but led to substantial reduction in model performance in cropland (with R² decreasing 0.20~0.27 compared with normal). Extreme cold condition did not lead to much changes in model performance in forest and woody savannas (with R² decreasing 0.01~0.08 and 0.09 compared with normal, respectively). Our study showed that both models need training samples at daily timesteps of >2.5 years to reach a good model performance and >5.4 years of daily samples to reach an optimal model performance. In summary, both RF and XGBoost are applicable machine learning algorithms for predicting ecosystem NEE, and XGBoost algorithm is more feasible than RF in terms of accuracy and efficiency.

Nitrogen is the limiting nutrient for plant growth in peatland ecosystems. Nitrogen addition significantly affects the plant biomass, diversity and community structure in peatlands. However, the response of belowground microbe to nitrogen addition in peatland ecosystems remains largely unknown. In this study, we performed long-term nitrogen addition experiments in a permafrost peatland in the northwest slope of the Great Xing’an Mountains. The four nitrogen addition treatments applied in this study were 0 g N·m⁻²·year⁻¹ (CK), 6 g N·m⁻²·year⁻¹ (N1), 12 g N·m⁻²·year⁻¹ (N2), and 24 g N·m⁻²·year⁻¹ (N3). Effects of nitrogen addition over a period of nine growing seasons on the soil microbial abundance and community diversity in permafrost peatland were analyzed. The results showed that the abundances of soil bacteria, fungi, archaea, nitrogen-cycling genes (nifH and b-amoA), and mcrA increased in N1, N2, and N3 treatments compared to CK. This indicated that nitrogen addition promoted microbial decomposition of soil organic matter, nitrogen fixation, ammonia oxidation, nitrification, and methane production. Moreover, nitrogen addition altered the microbial community composition. At the phylum level, the relative abundance of Proteobacteria increased significantly in the N2 treatment. However, the relative abundances of Actinobacteria and Verrucifera in the N2 treatment and Patescibacteria in the N1 treatment decreased significantly. The heatmap showed that the dominant order composition of soil bacteria in N1, N2, and N3 treatments and the CK treatment were different, and the dominant order composition of soil fungi in CK and N3 treatments were different. The N1 treatment showed a significant increase in the Ace and Chao indices of bacteria and Simpson index of fungi. The outcomes of this study suggest that nitrogen addition altered the soil microbial abundance, community structure, and diversity, affecting the soil microbial carbon and nitrogen cycling in permafrost peatland. The results are helpful to understand the microbial mediation on ecological processes in response to N addition.

Wetland conversion to cropland substantially suppresses methane (CH₄) emissions due to the strong suppression of methanogenesis, which consists of various pathways. In this study, we evaluated the cultivation impacts on four predominant CH₄ production pathways, including acetate, carbon dioxide (CO₂), methylamines, and methanol, in a wetland and cultivated cropland in northeastern China. The results showed significant suppression of CH₄ production potential and the abundance of genes for all four methanogenic pathways in cropland. The consistency between CH₄ production and methanogenesis genes indicates the robustness of genomic genes in analyzing methanogenesis. The suppression effects varied across seasons and along soil profiles, most evident in spring and 0 to 30 cm layers. The acetate pathway accounted for 55% in wetland vs. 70% in the cropland of all functional genes for CH₄ production; while the other three pathways were stronger in response to cultivation, which presented as stronger suppressions in both abundance of functional genes (declines are 52% of CO₂ pathway, 68% of methanol pathway, and 62% of methylamines pathway, vs. 19% of acetate pathway) and their percentages in four pathways (from 20 to 15% for CO₂, 15 to 9% for methylamines, and 10 to 6% for methanol pathway vs. 55 to 70% for acetate pathway). The structural equation models showed that substrate availability was most correlated with CH₄ production potential in the wetland, while the positive correlations of acetate, CO₂, and methylamine pathways with CH₄ production potential were significant in the cropland. The quantitative responses of four CH₄ production pathways to land conversion reported in this study provide benchmark information for validating the CH₄ model in simulating CH₄ cycling under land use and land cover change.