This project explores the integration of artificial intelligence (AI) and geographic information systems (GIS) in visualizing climate change in nineteenth-century China. It examines how GIS spatial analysis, combined with generative AI for text-to-image conversion, can generate new historical layers to trace the intensifying El Niño–Southern Oscillation (ENSO) impacts on the East China Sea from the 1860s onward. With a post-Taiping Rebellion context, the study investigates the contingent consequences of rising sea levels, stronger tidal surges, and increasingly frequent summer cyclones, all of which strained social infrastructures and challenged local governance across the Yangzi River estuary. In particular, it underscores the profound transformation of the northern Yangzi Delta, where once-arable farmland turned into marshy enclaves afflicted by floods, disease, and salinization—disasters that in turn displaced rural populations and created large waves of underclass refugees. By remapping shoreline dynamics through the integration of AI and GIS, this project offers new ways to visualize the contested waterscapes of land use, labor regimes, and local politics.

Methodologically, this project advances the use of text-image data processed through AI tools to enhance GIS mapping across multiple historical scales. In contrast to conventional remote-sensing datasets, which typically privilege large-scale urban and peri-urban regions, this study focuses on smaller spatial units such as counties, towns, and villages in rural settings. This micro-scale perspective provides an analytical interface through which scholars can examine climate change patterns at a fine-grained level, enabling closer investigation of how local communities perceived, responded to, and adapted to recent environmental challenges over time. Building on this approach to historical scaling, the project also demonstrates how generative AI can assist in processing diverse types of evidence and converting them into layered spatial data suitable for GIS georeferencing. In doing so, it not only highlights the utility of AI in historical research but also points toward broader applications of AI-assisted GIS mapping in the digital humanities, with significant implications for research, pedagogy, and distance learning.

The source base of this research extends beyond Chinese-language materials to include multilingual records in English, French, and Japanese. These comprise a wide range of archival and textual materials, including those local records left by French Jesuit missionaries who arrived in Shanghai in the 1850s. The coincidence of their co-presence with intensified ENSO events positioned these Catholic priests as direct witnesses to the environmental transformations along the shoreline. The religious landscape of their newly established churches thus provides valuable evidence for mapping the shifting waterfront in local contexts. In addition, meteorological data collected by the Jesuit Observatory in Shanghai constitutes one of the earliest modern scientific records of climate in East Asia. When processed through AI technologies, these data sets—covering wind direction, rainfall, and barometric pressure—enable detailed analysis of climatic variability across time and space, including the tracing of heightened summer cyclone activity and its impact on coastal communities. Taking together, these materials demonstrate the potential of religious and scientific sources to serve as underutilized datasets for advancing scholarship in both environmental humanities and GIS-based historical mapping.