The electric power industry is one of the major industries in terms of carbon dioxide (CO₂) emissions, and it is necessary to explore low-carbon green power generation models. In recent years, more research has focused on the difference in carbon emissions in fossil energy versus renewable energy but ignored the impact of energy on human well-being. The life cycle assessment (LCA) method is a better method for assessing the impact of the low-carbon model on human well-being. In this paper, the carbon footprints of coal power plants and photovoltaic power (PV) plants generating 1 Kilowatt hour (kWh) of electricity are compared to analyze the degree of carbon emissions at different stages of the two models, and the environmental impact potential of the two models is analyzed using the LCA method. The differences between the two models in terms of human well-being were analyzed through questionnaires and quantified using the hierarchical analysis method. The impact of the different models on human well-being was compared using LCA method. The results of the study were as follows: the total CO₂ emissions from coal-fired power generation at the 1 kWh standard were 973.38 g, while the total CO₂ emissions from PV power generation were 91.95 g, and the carbon emission intensity of coal-fired power plants was higher than that of PV power plants. The global warming potential and eutrophication potential of coal-fired power plants were higher than those of PV power plants, and the rest of the indicators were lower than those of PV power plants. The composite human well-being index of PV power plants was 0.613 higher than that of coal-fired power plants at 0.561. The per capita income–global warming potential of PV power plants was higher than that of coal-fired power plants, indicating that PV power plants were a low carbon-emission and high well-being model. In conclusion, the PV power plant model is a low-carbon and high human well-being industrial model that is worthy of application in the Qilian Mountains region. The low-carbon industrial model proposed in this study can have a positive effect on regional ecological environmental protection and human well-being enhancement.

In the context of high-quality development, scientifically and objectively assessing regional ecosystem health (EH) is important for ecological civilization. However, the commonly used EH assessment framework typically neglects intrinsic connections, mutual adaptability, and coordination among interrelated indicators. The coupling coordination model was utilized to improve the classic pressure–state–response assessment (PSR) model. The carbon footprint, water footprint, landscape pattern, and response status of the Manas River Basin were used to construct a medium-scale regional EH assessment framework linking natural ecosystems with human socioeconomic elements. A quantitative assessment was conducted on the EH conditions of the Manas River Basin from 2000 to 2020. Over the past 21 years, the EH conditions of the Manas River Basin have fluctuated upward. The ecosystem health index (EHI) increased from 0.18 to 0.37. Compared with the conventional PSR model, the coupling coordination pressure–state–response model (CC–PSR) better reflected the fluctuations in EH conditions caused by “pressure”, “state”, and “response” level changes. In the early stage (2000–2006), increasing human activity strongly pressured the regional ecosystem, limiting EH improvements. The increase in “pressure” was reflected in the increasing trends of the water footprint, carbon footprint, and ecological footprint. During the middle to late period (2009–2020), as the “response” level improved, the regional EH condition continued to increase, and the EHI stabilized between 0.29 and 0.38. Ecosystem resilience improvements and human afforestation projects enhanced the “response” level, but their impacts were noticeably delayed. Over the past 21 years, regional landscape diversity, landscape connectance, and landscape contagion have remained high. The well-maintained landscape pattern has laid the foundation for consolidating and improving the regional EH. The EHI is increasing; its fluctuations stem from periodic fluctuations in the regional water yield and carbon sequestration capacity, which are constrained by the basin climate and vegetation coverage. This study provides a scientific model for basin EH assessment.

Ecosystem services (ESs) are a multiple whole composed of multiple services and their multiple relations, which can be expressed as tensors (multiple functions of multiple vectors). This study attempts to introduce the concept and method of tensors into ES research to solve problems caused by the multiplicity of ESs, such as multiple descriptions and perceptions of ESs, repeated calculation of ES values, and cascading relationships with the social economy. Taking the Manas River Basin composite ecosystem as an example, we constructed five different types of ES tensors based on different understandings and applications: (1) As multiple vectors, three eigenvectors were extracted from the ES state tensor (ESST), including farmland service (FS), vegetation service (VS) and water service (WS). According to the ES response tensor (ESRT), an increase in FS may lead to a decline in overall services. (2) As multiple functions, the ES value (ESV) of the basin was measured by the ESV metric tensor (ESVMT), with a gross value of 14.8 billion USD and a net value of 10.17 billion USD. From different stakeholders perceptions constructed by the ecosystem services to human well-beings (ES-HW) tensor, the human well-being values (HWV) were ranked as citizens > farmers ≈ herdsmen > public. (3) The HWV output efficiency of different LULC per unit of water use was calculated by a fourth-order mixed tensor constructed by water–LULC–ES–HW multiple cascading relations. Among them, the HWV efficiency of water areas and wetlands was the highest, but the area was the smallest. Cultivated land and unused land had the lowest HWV efficiency and largest area. In general, the ES tensor is the extension and integration of the ES scalars/indicators to the ES vectors/bundles, which can provide tools for the integral expression, objective measurement and multiple perceptions of ESs.