Managing forest understory has a significant impact on soil greenhouse gas (GHG) emissions and the ecosystem’s capacity for carbon sequestration. However, its specific impacts and mechanisms within hickory (Carya cathayensis Sarg) forests remain unclear. The objective of this study was to examine the effects of different understory vegetation treatments on hickory stands with similar growth history, site conditions, and slopes: Cinnamomum chekiangense (Cinnamomum chekiangense Nakai) and strip-sown ryegrass (Lolium perenne L.) (CR1), Cinnamomum chekiangense and scattered ryegrass (CR2), Torreya grandis (Torreya grandis’ ‘Merrillii’ Hu) and strip-sown wild rapeseed (Brassica napus L.) (TW1), Torreya grandis and scattered wild rapeseed (TW2), and removal of understory vegetation (CK). Twenty experimental plots were established at the Lin’an Forestry Carbon Sink Pioneer Base, and after 12 months of monitoring, the responses of GHG emissions, vegetation, and soil organic carbon (SOC) sequestration were analyzed, revealing the differences in ecosystem carbon sequestration capacity. Compared to CK, CR1, CR2, TW1, and TW2 increased the global warming potential (GWP) by approximately 26%, 55%, 26%, and 16%, respectively. The SOC increased by approximately 76%, 102%, 51%, and 32%, respectively, while the vegetation carbon sink increased by approximately 30%, 27%, 53%, and 62%, respectively. In summary, ecosystem carbon sequestration increased by approximately 109%, 98%, 95%, and 92%, respectively. The findings indicate that managing understory vegetation in pecan forests significantly enhances ecosystem carbon sequestration but also increases soil GHG emissions. To enhance future research priorities, it is essential to focus on increasing carbon sequestration in hickory forests and managing soil GHG emissions through effective and rational understory vegetation management.

Agroforestry management has immense potential in enhancing forest carbon sequestration and mitigating climate change. Yet the impact and response mechanism of compound fertilization rates on carbon sinks in agroforestry systems remain ambiguous. This study aims to elucidate the impact of different compound fertilizer rates on soil greenhouse gas (GHG) emissions, vegetation and soil organic carbon (SOC) sinks, and to illustrate the differences in agroforestry systems’ carbon sinks through a one-year positioning test across 12 plots, applying different compound fertilizer application rates (0 (CK), 400 (A1), 800 (A2), and 1600 (A3) kg ha⁻¹). The study demonstrated that, after fertilization, the total GHG emissions of A1 decreased by 4.41%, whereas A2 and A3 increased their total GHG emissions by 17.13% and 72.23%, respectively. The vegetation carbon sequestration of A1, A2, and A3 increased by 18.04%, 26.75%, and 28.65%, respectively, and the soil organic carbon sequestration rose by 32.57%, 42.27% and 43.29%, respectively. To sum up, in contrast with CK, the ecosystem carbon sequestration climbed by 54.41%, 51.67%, and 0.90%, respectively. Our study suggests that rational fertilization can improve the carbon sink of the ecosystem and effectively ameliorate climate change.

Forests play a crucial role as the primary sink for greenhouse gases, and forest mortality significantly impacts the carbon sequestration capacity of forest ecosystems. A single type of forest mortality model has been developed, and its model variables are incomplete, leading to significant bias in mortality prediction. To address this limitation, this study harnessed data collected from 773 permanent plots situated in Zhejiang Province, China, spanning a period from 2009 to 2019. The primary objectives were to pinpoint the key variables influencing forest mortality and to construct forest mortality prediction models utilizing both traditional regression methods and machine learning techniques, ultimately aiming to provide a theoretical basis for forest management practices and future predictions. Four basic linear regression models were used in this study: Linear Regression (LR), Akaike Information Criterion (AIC) Stepwise Regression, Ridge Regression, and Lasso Regression. Four machine learning models, Gradient Boosting Regression (GBR), Random Forest (RF), Support Vector Regression (SVR), and Multilayer Perceptron (MLP), were used to model stand mortality. Mortality was used as the dependent variable, and environmental factors such as topographic factors, soil composition, stand characteristics, and climatic variables were used as independent variables. The findings unveiled that soil and stand-related factors exerted significant effects on the mortality rate, whereas terrain-related and climate factors did not exhibit statistical significance. The Random Forest model established by using stand age, tree height, ADBH, crown cover, humus layer thickness, and the biodiversity index has the highest fitting statistics such as R² and Mean Squared Error, indicating that it has a good fitting and prediction effect, which effectively predicts mortality at the stand level, and is a valuable tool for predicting changes in forest ecosystems, with practical value in estimating tree mortality to enhance forest management and planning.