Thematic maps are fundamental for decision-making in natural resources and ecosystem management and planning. Thematic information, land cover in particular, is dynamic and changeable with the passage of time. Hence, providing up-to-date and precise data in a short period of time is considered as a key factor for successful utilization of these maps. Field inspection and visual interpretation of satellite images and topographic maps are necessary for this purpose. These processes are labor intensive, time consuming, and costly. An expert system, as a computer system with the capability of an expert’s performance simulation, could be an assistant for experts involved with the thematic map updating process. In this paper, a knowledge-based expert (KBE) system was designed, implemented, and tested to automate, facilitate, and expedite the conventional thematic map updating procedure. The developed system in this research is based on using existing thematic map and satellite images together with the new acquired satellite image of the desired region. It entails the following five major steps: (1) relative radiometric normalization, (2) change detection, (3) spectral database construction, (4) identifying the content of the changes, and (5) cartographic operations. In order to assess the effectiveness of the proposed KBE system for automatic updating of thematic maps, two different case studies related to Urmia Lake and surrounding areas were considered. Near-anniversary bi-temporal Landsat TM 4,5 imagery, acquired on 30 June 1989 and 24 June 2007 over the eastern coast of the lake alongside the existing land cover map of 1989, were used in the first case study. Utilized data sets in the second case study were near-anniversary Landsat 8 satellite images (OLI sensor) taken on 20/7/2014 and 23/7/2015 over the southern coast of lake alongside the outdated land cover map of interest region in 2014. Test results of different parts of the system were confirmed by the experts, and finally the system could satisfy the assessment criteria from the experts’ point of view in this particular application. Overall accuracy of the updated thematic map in the first and second case studies was 97.18 and 87.81 %, respectively.