The Chinese Loess Plateau has approached a critical threshold in vegetation restoration, requiring optimized land-use strategies for sustainable ecosystem management. This study examines soil organic carbon (SOC) content and associated soil properties across 0 to 500 cm soil profiles under 4 land-use types (grassland, shrubland, forestland, and cropland) in the Gutun Basin. SOC dynamics were assessed using acid-soluble carbon (F1) and carbon components extracted at combustion temperatures of 25 to 400 °C (F2), 400 to 600 °C (F3), and 600 to 900 °C (F4). Based on ¹⁴C ages, F1 to F3 were classified as labile components, whereas F4 was considered nonlabile. Although cropland exhibited the highest SOC stock (0.88 ± 0.38 kg m⁻²), it also had the highest proportion of nonlabile carbon (75%). The similar ¹⁴C ages of deep-soil carbon components in shrubland and forestland, along with higher carbon content in deep layers, suggested that carbon sequestration could occur at depths of at least 300 to 500 cm. However, soil water content (SWC) in shrubland and forestland was significantly lower (93.71 ± 44.94 and 71.15 ± 17.62 g kg⁻¹, respectively), and forestland exhibited lower SOC content (1.77 ± 0.56 g kg⁻¹). Among all land-use types, grassland had a significantly higher SOC content (3.96 ± 2.56 g kg⁻¹), a greater proportion of labile carbon (77%), and a higher SWC (132.06 ± 21.71 g kg⁻¹), suggesting that it was the most suitable land-use type in this context.

The interplay of extreme precipitation and high temperature markedly diminishes the carbon sink capacity of China's terrestrial ecosystems. Although the spatiotemporal evolution characteristics of these extreme events have been initially identified, their impacts on gross primary production (GPP) remain inadequately quantified. With the multisource data (1981 to 2100) of 4 shared socioeconomic pathways (SSPs), the Mann-Kendall test, empirical orthogonal function, and center of gravity shift models were used to identify the spatiotemporal evolution of GPP, and the structural equation model was used to elucidate the pathways and contributions of extreme events to GPP changes. GPP increased nationwide (0.02 to 0.07 kg C m⁻² a⁻¹ decade⁻¹, P < 0.05), with growth varying by scenario and there is regional heterogeneity; the first principal mode of GPP was positive in most areas (59.55% to 100%), with the same increasing and decreasing characteristics; the spatial centroid of GPP exhibited a shift from the east and south toward the west and north, accompanied by marked ecosystem greening; extremely high temperature exerted the most -substantial negative influence on GPP under SSP5-8.5 (standardized path coefficients: −0.580), whereas extreme precipitation duration in SSPs 1-2.6 and 3-7.0, and intensity in SSP2-4.5. The total effect of extreme events on GPP change shifts from a positive to a negative effect (0.966 → −0.655, −167.81%) with increasing carbon emissions. Negative effects are concentrated in Northeast China, East China shows alternating positive and negative effects, and the other regions shift from positive to negative effects. This study emphasizes the importance of differentiated climate change adaptation capacity building.

Highly urbanized cities have undergone substantial land use changes driven by intensive human activities, leading to increasingly prominent ecological security challenges. This study employed landscape pattern indices to evaluate the landscape ecological risk (LER) across 275 prefecture-level cities in China from 2011 to 2019. We hypothesized that the mechanisms influencing LER vary according to the level of urbanization. An integrated panel threshold regression model was applied to examine this hierarchical influence and identify potential thresholds in the urbanization rate (UR). In addition, the spatial spillover effects of LER were investigated using a spatial panel econometric model. Our findings revealed an initial increase in LER from 2011 to 2013, followed by a notable stabilization between 2015 and 2019, with evident spatial heterogeneity. Using UR as the threshold variable, the normalized difference vegetation index, population density, and per-capita gross domestic product (PGDP) were found to exert nonlinear impacts on LER, which were consistent across multiple model specifications. A critical threshold was identified at a UR of 62.1%, beyond which the influence of PGDP on LER altered substantially. Spatial spillover effects of LER were also stronger in cities below this UR threshold, implying that highly urbanized cities may possess self-regulatory ecosystem mechanisms that mitigate LER. Based on the results, the development of secondary and tertiary industries, alongside the construction of infrastructure, could mitigate LER for low-urbanization cities. In highly urbanized cities, blue-green space planning should be strategically aligned with industrial transformation and upgrading, thereby bolstering urban ecological protection and contributing to long-term environmental sustainability.

Improving the zonal management of the ecosystem service (ES) supply-demand balance requires a thorough understanding of its spatiotemporal variability and the nonlinear characteristics of the driving mechanisms. Coordinated social-ecological development at the regional level depends on this understanding. Consequently, we delineated bundles of the ES supply-demand relationship within the Harbin-Changchun urban agglomeration, examined interactions between ES supply and demand across these bundles, and investigated the mechanisms by which social-ecological drivers shape the spatial pattern of this relationship over an extended time series. The findings show that deficit areas for the 5 ESs are concentrated in construction zones and have expanded annually; synergies outweigh trade-offs among the 5 bundles, and the trade-off relationship intensifies markedly over time. Social drivers dominate bundles 1 (key synergetic bundle), 2 (carbon fixation-water yield-crop production synergetic bundle), and 5 (carbon fixation-crop production-habitat quality synergetic bundle), whereas ecological drivers prevail in bundles 3 (water yield bundle) and 4 (ecological transition bundle). The ES supply-demand relationship is increasingly strained by social disturbances over time. A normalized value near 0.5 emerges as a recurrent threshold at which the effects of several metrics such as digital elevation model, average precipitation, and normalized difference vegetation index shift over a 20-year period. Implementing precise optimization and rational planning for bundle-level ecological management benefits sustainable development and human well-being.