Traditional villages have received widespread attention from all walks of life based on important carriers of Chinese rural culture. The mutual superposition of natural and cultural factors may exacerbate the evolution of traditional village geographical patterns. To understand such relationships and effects, factors and degrees influencing traditional villages need to be determined. Here, we analysed the data of 724 traditional villages in Guizhou recognised by relevant national ministries and commissions in China using average nearest neighbour analysis, Tyson polygon analysis, nuclear density analysis and Geodector. The geographic pattern feature revealed that traditional villages, in general, are highly clustered regionally and have significant edge effects on administrative units. Different substrate environments result in significant spatial heterogeneity in village spatial density, clustering, surface undulation, sun exposure, and waterfront. The geographic pattern of traditional villages is mostly affected by the closest distance to river valleys, the types and number of intangible cultural heritage resources in the county, river gorge density, edge effect index, degree of county ethnic language use, and proportion of paddy fields to the regional area; and their combined effects influence and control the community structure. The results highlight the impact of nature and culture on the distribution of traditional villages, which helps traditional village conservation and scientific exploration of human-land relationship issues in the mountainous areas of Southwest China.

The protection and utilization of traditional villages have received widespread attention from all walks of life and have become an essential grip of the current rural revitalization strategy. This paper takes 724 traditional villages in Guizhou recognized by relevant national ministries and commissions as the research objects, uses spatial analysis and geographic probes to characterize the spatial distribution of traditional villages and reveal the distribution patterns in different natural and humanistic environments, and quantitatively detects the dominant factors of spatial differentiation of traditional villages and their driving forces from a microscopic perspective. The results show that: (1) traditional villages, in general, are highly clustered locally, forming hotspots with high first place in some districts and counties of Qiandongnan, Anshun and Qiannan, and more than 90% of the villages have significant edge effects and spatial distribution characteristics of crossing the boundaries of 2-4 county administrative units. (2) 98.07% of the villages are distributed in the range of 250-1500m above sea level. There is a significant negative spatial correlation between the elevation and longitude of the villages; their surface undulation is generally higher than the average level of natural villages in the province, 76.24% of the villages are concentrated in the slope of 8°-24°, the characteristics of sunwardness and waterfront are obvious, the number and aggregation degree of clastic rock areas are significantly higher than those of carbonate rock areas, the buffer zone of 1500m along the fault line controls more than half of the number of villages, which are mostly extended in the northeast-southwest direction. 99.17% of the villages are distributed in areas with high physical comfort, and the overall level of vegetation index is high. (3) Traditional villages are more distributed in the gathering places of Miao, Dong, Buyi, Shui, Tujia and Gelao ethnic minorities, and have significant spatial correspondence with the degree of minority language use, the type and number of Intangible Cultural Heritage resources, the distribution of rice fields, the lower GDP level of regions such as Qiandongnan, and the years of female education. (4) Different substrate environments result in significant spatial heterogeneity in village spatial density, clustering, surface undulation, sunwardness and waterfront. (5) The geographic pattern of traditional villages is mainly affected by the closest distance to river valleys (x7), the types and number of provincial-level Intangible Cultural Heritage resource in the county (x26, x27, x28), river gorge density (x6), edge effect index (x18, x19), degree of county ethnic language use (x25), and proportion of paddy fields to the regional area (x14) and their combined effects influence and control. Through the detection of the mechanism of the distribution process of traditional villages in Guizhou, it helps to further clarify the reciprocal feedbacks mechanism between traditional villages and natural, human elements in the mountain ecosystem, and provides theoretical reference and reference for the conservation and sustainable use of villages as well as the scientific exploration of human-land relationship issues in the mountainous areas of Southwest China.

World heritage is the common wealth of all mankind. How the public perceives the “outstanding universal value(OUV)” under the framework of the expert knowledge system and whether they can consciously recognize, which is already related to the interpretation of heritage value and protection management. Taking the South China Karst World Natural Heritage property as case sites, this paper analyzes the tourists’ perception of OUV and further explore the underlying reasons behind the perception based on online text data. The results showed that: (1) Tourists do not have a high degree of perception of the OUV in the South China Karst World Natural Heritage property, and there are obvious differences; (2) The landscape features of the heritage site at different stages of karst landform evolution have become a prerequisite for the difference in the tourists' perceptions of the karst landscape and aesthetic features of the heritage site. The insufficient OUV display during the tourism development of the heritage site has resulted in an overall low level of tourists’ perception, in addition to the service and experience are gradually valued by tourists, thus the OUV landscape of the heritage site is no longer a hot spot. The image of the tourist destination constructed by the heritage site runs through the pre-tourism motivation, willingness and the post-tourism evaluation of tourists, and then deepens the perception of heritage OUV. By understanding the tourist group's perception of heritage OUV, the results of the study can provide a certain theoretical and practical reference for the development, operation and management of heritage tourism.

Accurate and non-destructive identification of the weathering degree of speleothems in tourist caves holds critical implications for establishing scientific conservation criteria. To address the limitations inherent in conventional rock weathering classification systems, this study employs hyperspectral data acquired from speleothems exhibiting varying weathering intensities. A random forest algorithm was utilized to identify diagnostically significant spectral features, which were subsequently leveraged to construct a decision tree model. The model's performance was rigorously evaluated using metrics including the F1-score and the Area Under the Curve (AUC). The results demonstrate that: (1) lichen colonization does not markedly alter the overall spectral profile of speleothems but leads to a concentration of the wavelength at the absorption trough minimum; (2) among eight spectral features analyzed, the combination of \(\:{Q}_{SAI}\),\(\:{\lambda\:}_{1}\), and \(\:{\lambda\:}_{m}\) exhibits the highest discriminative capacity; (3) the decision tree model developed based on this feature combination effectively classifies speleothems according to their weathering degree. These findings provide a theoretical foundation for advancing the weathering classification of speleothems.

Accurately and non‑destructively identifying the weathering degree of speleothems in show caves is critical for their effective management and conservation. To overcome the limitations of traditional rock weathering classification methods, this study extracts basic waveform features from hyperspectral data of speleothems with different weathering degrees and builds machine learning classification models. The results show that: (1) In terms of spectral morphology, hyperspectral reflectance generally decreases as weathering degree intensifies; (2) Among the evaluated models, the neural network (NN) achieves the highest overall accuracy, F1 score, and generalization capability on this dataset. For weathering categories, the model classifies slightly weathered and moderately weathered samples best, followed by surface‑weathered and strongly weathered samples; (3) Wavelength parameters contribute relatively little to the machine learning classification, whereas reflectance and waveform morphology dominate the NN’s classification decisions. These findings provide a theoretical basis for non‑contact, high‑precision, and quantitative weathering classification of speleothems.