In modern agricultural production, the efficient utilization of nutrients and the protection of arable land resources require massive soil information, while traditional soil testing methods are inefficient and costly, failing to meet the demands of modern agricultural production. There is an urgent need for rapid and efficient acquisition of soil information technology and methods, and modern spectroscopy provides a new option to meet this demand. Spectroscopic sensing techniques can rapidly obtain soil information, adaptive intelligent prediction models were developed applying chemometric methods and multivariate data fusion algorithms, which enabled rapid assessment of soil properties. Typical farmland soil samples were collected in this study, and a soil spectral information system of Nanjing has been constructed through utilizing soil mid-infrared and laser induced breakdown spectral data and self-adaptive functional models, which demonstrated broad application potential in digital soil management, providing new technological support for the development of modern sustainable agriculture.