Non-point source (NPS) pollution poses a significant threat to the water environment, yet controlling it at the watershed scale remains a formidable challenge. Understanding the characteristics and drivers of nitrogen (N) and phosphorus (P) outputs at the watershed scale, along with identifying cost-effective best management practices (BMPs), is crucial for effective pollution control. In this study, we utilized the Wangjiaqiao watershed within the Three Gorges Reservoir Area (TGRA) as a case study to explore the characteristics of N and P load outputs and their dominant drivers by combining the SWAT model and a geographic detector. Based on our analysis of N and P loads within the watershed, we employed the entropy weight method to evaluate the reduction efficiency and cost-effectiveness of 64 BMP scenarios, encompassing seven measures (vegetative filter strips, parallel terraces, 10% fertilizer reduction, 30% fertilizer reduction, residue cover tillage, grass mulching, and returning farmland to forest) and their combinations. Our findings revealed the following: (1) spatial heterogeneity in NPS loads within the watershed, primarily influenced by land use, fertilizer application, and surface runoff, with interactive enhancement effects among driving factors; (2) the differential effectiveness of BMPs at the watershed level, with structural measures, particularly terracing, exhibiting higher efficacy and achieving reduction rates of 28.12% for total nitrogen (TN) and 37.69% for total phosphorus (TP); the combined BMPs showed improved reduction efficiency, but not merely additive; and (3) in terms of cost-effectiveness, 30% fertilizer reduction emerged as the most beneficial among the individual measures. Moreover, a combination of vegetative filter strips, parallel terraces, and 30% fertilizer reduction demonstrated significant improvements in TN and TP reductions (48.05% and 61.95%, respectively), suggesting their widespread applicability. Overall, our study provides insights into developing a cost-effective BMP strategy for the Wangjiaqiao watershed and offers valuable guidance for NPS pollution management in similar small watersheds within the TGRA.

As a critical agroforestry crop in Southern China, Moso bamboo, maintains regional timber security and bamboo shoot production, with its net ecosystem productivity (NEP) directly determining dry matter accumulation and economic yield. This study integrates 2024 continuous flux observations with XGBoost and SHAP explanations to characterize the subtropical bamboo forest carbon budget and its nonlinear driving mechanisms. The results show a weak carbon sink in 2024 with an annual cumulative NEP of 120 g C m⁻², as high respiration of 860 g C m⁻² limited organic matter conversion by consuming nearly 88% of the 980 g C m⁻² total primary production. The peak production period during May and June was offset by growth stagnation in August, caused by extreme heat and drought. Net radiation served as the primary driver, with a positive contribution threshold of 75.28 W m⁻², whereas precipitation exceeding 1.85 mm or air temperatures over 17.85 °C hindered carbon accumulation through radiation attenuation and metabolic heat loss. Strong radiation–precipitation interactions confirm that water’s impacts on yield are deeply contingent upon radiation backgrounds. These nonlinear regulatory pathways provide a scientific foundation for stabilizing bamboo forest productivity through synergistic water-radiation management and structural optimization during extreme climate events.

Under the global climate governance framework, the Paris Agreement and the China–U.S. Glasgow Joint Declaration established a non-negotiable target of limiting 21st-century temperature rise to 1.5 °C. To date, over 130 nations have pledged carbon neutrality by mid-century, with agricultural activities contributing 25% of global greenhouse gas (GHG) emissions. The spatiotemporal dynamics of these emissions critically determine the operational efficacy of carbon peaking and neutrality strategies. While China’s Nationally Determined Contributions (NDCs) commit to achieving carbon peaking by 2030, a policy gap persists regarding differentiated implementation pathways at the county level. Addressing this challenge, this study selects the Three Gorges Reservoir (TGRA)—a region characterized by monocultural cropping systems and intensive fertilizer dependency—as a representative case. Guided by IPCC emission accounting protocols, we systematically evaluate spatiotemporal distribution patterns of agricultural CH₄ and N₂O emissions across 18 county-level units from 2006 to 2020. The investigation advances through two sequential phases: Mechanistic drivers analysis: employing the STIRPAT model, we quantify bidirectional effects (positive/negative) of critical determinants—including agricultural mechanization intensity and grain productivity—on CH₄/N₂O emission fluxes. Pathway scenario prediction: We construct three developmental scenarios (low-carbon transition, business-as-usual, and high-resource dependency) integrated with regional planning parameters. This framework enables the identification of optimal peaking chronologies for each county and proposes gradient peaking strategies through spatial zoning, thereby resolving fragmented carbon governance in agrarian counties. Methodologically, we establish a multi-scenario simulation architecture incorporating socioeconomic growth thresholds and agroecological constraints. The derived decision-support system provides empirically grounded solutions for aligning subnational climate actions with global mitigation targets.