To have a more comprehensive understanding of the water ecological services of Xiangxi, the index system of water ecological service is constructed and the intensity of its influencing factors founded on the geographic detector model (GDM) is obtained in this paper. Then, the water ecological service index of eight cities in Xiangxi to evaluate the function level is determined. It is proved that: (a) The overall service value of water cultural function as a key factor in the construction of water ecological services, soil conservation value, and water supply value are relatively poor. (b) The value of the water ecological culture function type is the largest, with an overall value of 353.32 billion yuan. The value of water resources supply is the smallest, with a value of only 4.37 billion yuan, which indicates that human activities have a strong interference with the service function of the water ecosystem. (c) Based on GDM, four factors, including precipitation, resident population, sewage discharge, and water quality compliance rate, have a significant impact on water ecological services. Among them, per capita GDP(X7) > vegetation coverage (X6) > precipitation (X2) > resident population (X4), which further clearly explains the impact intensity of mountain climate basic conditions and human socio-economic activities on water ecological services. The results can offer a practical basis for the spatial development and protection of the water ecosystem to coordinate humans and the environment.

Maintaining regional landscape ecological stability and enhancing rural landscape ecosystem services are critical research priorities. This study selected Xiangxi Autonomous Prefecture (XXAP), a representative mountainous region in Southwestern (SW) China, as the case study area. This study aims to construct a rural landscape ecological risk (RLER) evaluation index system based on five-period remote sensing data (2000–2020), analyze the spatio-temporal characteristics of RLER, and provide a scientific basis for landscape ecological management and rural spatial governance. The results show that the RLERI exhibited a balanced multi-ring development trend, decreasing slightly from 0.295 in 2000 to 0.282 in 2020, suggesting a slight alleviation of overall risk. Medium-risk areas of the RLERI consistently accounted for the largest proportion (over 37%). Notably, the share of high-risk areas remained relatively stable, fluctuating narrowly between 7.98% and 8.73%. Meanwhile, high-risk areas of the landscape disturbance degree (LDD) expanded markedly from 1.87% to 10.28%. Correspondingly, high-risk areas of the landscape fragility degree (LFD) also increased significantly, rising from 0.93% to 2.8%. Spatially, RLER displayed significantly positive spatial autocorrelation, with high–high (H-H) clusters concentrated in the central-southern part and low–low (L-L) clusters distributed in the northern and southern margins, indicating pronounced spatial differentiation. In conclusion, this study provides a transferable framework for ecological risk assessment in mountainous regions. Furthermore, it underscores the importance of optimizing landscape patterns in ecologically fragile areas, strengthening ecological risk management, and mitigating ecological risks in rural settings.