The Chinese Eastern Railway (CER) has been identified as a heritage site that has witnessed industrial and social developments and changes over the past 100 years. Many old infrastructures have transitioned from civil use to historical conservation, but the spatial distributions and driving forces of ecosystem service functions provided by local ecological infrastructures are still unclear. The green view index (GVI) is a flexible parameter that measures the ratio of pixels accounted for by greeneries to those in an intact street view image; hence, it is widely recognized as a reliable variable for assessing the magnitude of ecosystem services provided by ecological infrastructures at a regional scale. In this study, a total of 42 urban parks were selected from regions covered by CER routes and adjacent cities without its involvement. No difference in GVI was found among cities regardless of distance from CER routes, and the distance of a park to the nearest CER line was found to have only indirect and negative effects on GVI. Correlation analysis indicated that the logarithms of both green and blue space areas had positive relationships with GVI. Regression analysis indicated that the logarithm of the blue space area, together with elevation and longitude, had positive effects on GVI, while latitude resulted in a negative effect. Overall, low GVI values (~6.0%) in the parks of Tsitsihar resulted from the effect of high closeness with the CER route, and high GVI values (over ~40%) in the parks of Changchun were indicated by the joint characteristics of local landscape metrics.

The intricate relationship between land use and ecological changes is a critical area of eco-environmental research. However, there is a notable lack of quantitative, analytical studies that explore this relationship in depth. This study aims to bridge this gap by applying a contrastive learning approach to a practical land use application, utilizing big data. We analyzed 2,197 publications from the Web of Science Core Collection spanning from 1998 to 2022. Our methodology correlates textual and visual data, enhancing the categorization and thematic analysis of existing literature. The contrastive learning technique uncovers nuanced relationships between textual descriptions and visual data, providing a comprehensive view of research trends. Key findings include an increasing volume of publications, indicating growing interest and potential in this field; predominant focus areas such as the relationship between climate change and land use, dynamic land use modelling, urbanization impacts, land use management, ecosystem services, and biodiversity. Emerging research hotspots identified through contrastive learning include the interactions between urban thermal environments and land use changes, land use effects on biological communities and soil organic matter, and the relationship between ecological conditions and land use in vulnerable regions. This study not only maps the current research landscape but also employs a contrastive learning approach to predict future research directions, thus enriching the methodological framework for analyzing land use applications with big data.