<p>In the Yangtze River Basin, the birthplace of civilization in China, it is necessary to further discover and investigate ancient remains. The archaeological site prediction model is important for discovering and investigating archaeological sites. In this paper, we focused on the ancient city sites of the Neolithic and Bronze Age in the Jianghan region located along the middle reaches of the Yangtze River, annotated the specific locations and extents of 33 ancient city sites using the Google Earth Engine (GEE) cloud platform, and proposed a machine learning-based ancient city site prediction model that couples geographic element features and temporal spectral features. The results indicated that the ancient city sites were recognizable in different geographic elements and were separable in Sentinel-2 multispectral bands and spectral indices. The coupled time series spectral features could improve the ability of the model to recognize the regions with ancient city sites. Ultimately, the percentage of pixels with a high probability of prediction (greater than 0.57) within the extent of the ancient city sites reached 80.0% and enabled to obtain high probability regions for the distribution of the ancient city site. The proposed model can be used to predict the potential geographic locations of ancient city sites and identify key areas for future archaeological field survey work.</p>