Abstract: Accurate and complete spatial distribution data of rural settlements provide a critical support for studying the evolution of rural settlements. Since rural settlements are characterized by small-scale and scattered distribution, it is hard to acquire a long-term and high-resolution rural settlement dataset. Focusing on the development of rural settlement data preparation, this article reviews the sources, acquisition methods, spatiotemporal resolution, and accuracy of different rural settlement datasets for the long-term, medium-term, and short-term temporal spans, respectively. It also summarizes the applicability, advantages, and disadvantages of different data acquisition methods. The results show that for studies on a medium and a long term, historical documents and digitized topographic maps are the main sources for acquiring spatial distribution of rural settlements, and missing regional data is the main difficulty in creating such datasets. For short-term studies, spatial data acquired through modern technologies, such as satellite remote sensing and aerial remote sensing, can provide high-resolution data to delineate rural settlements. However, acquiring high-resolution imageries often becomes the primary difficulty in the process of data preparation owing to its high price and poor accessibility. In response to the above problems, we should pay attention to the integration of history, archaeology, and other disciplines, and expand the sources of historical information about rural settlements. We should also make full use of free high-resolution imageries, such as Google Earth and Gaofen-2, coupled with image interpretation, oral history, field surveys, and household visits in order to construct a complete picture of spatial distribution of rural settlements and fully achieve data sharing.

Key words: rural settlement, evolution, data preparation, historical data, topographic map, remote sensing data, review