Geospatial knowledge bases (GKB) make a new geoinformation product. At the same time, it has already proved its effectiveness as a decision support tool in various industries, from geology to medicine. GKB are designed to form solutions of various problems based on the formalized geospatial knowledge and ensuing procedural rules. The content of geospatial knowledge bases can be visually represented on maps using GIS tools. The purpose of the research is to study the essence, the experience of development and implementation, as well as the prospects for using them as a decision support tool. The authors examine the essence of the said bases, define their features and suggest a definition of the concept of a "geospatial knowledge base". Examples of development and implementation of these bases for various fields of activities created at the Siberian State University of Geosystems and Technologies (SSUGT) are given. They confirm the relevance and applicability of geospatial knowledge bases. Technologies such as probabilistic block-chains and geospatial knowledge graphs were identified and studied. Using them will help reduce the complexity of the process of creating geospatial knowledge bases; make them better and more accessible. The necessity of abandoning the relational model at creating GKB in favor of open ones such as the EAV is noted