The terrains of China, showing a trend of being high in the west and low in the east, can be categorized largely into three types of stepped terrains that consist of different geomorphic units — namely, glaciers and mountains (the first [highest] step), basins and plateaus (the second [less high] step), and plains (the third [lowest] step). The source-to-sink systems in the first and second steps are closed systems from land to lake, whereas those of the third step are mainly open systems from land to ocean. We have based our analytical methods mainly on the parameters of the source-to-sink systems for the three types of stepped terrains using the Google Earth database. We acquired data for the catchment area in the source unit, the longest channel and coefficient of river-network development in the conduit unit, the fan area and fan length: width ratio in the sink unit, and the gradient of the entire system. On the basis of our comparison of these parameters in the three types of stepped terrains, results indicate differential source-to-sink system parameters (catchment area, longest channel, coefficient of river-network development, fan area, fan length:width ratio, and gradient) and a transition among these parameters from the first step to the third. The gradient varies in different geomorphic units, being relatively small on the plains (third step) and relatively large in the basins and on plateaus (the first or second step). The other five parameters become higher from the first step to the third. By analyzing the correlations of these parameters, we found out that (1) the coefficient of river-network development declines with the gradient increasing, (2) the catchment area and the coefficient of river-network development show a positive correlation, (3) the fan area increases with the catchment area increasing, and (4) the fan area and gradient show a negative correlation. From the perspective of controlling factors, we focused on the formation of the geomorphic units, tectonic conditions, and the climate resulting from the obvious differences between the three types of stepped terrains. A new approach to compare the parameters of each step and analyze the correlations between parameters has been used in this study with the goal of predicting the correlations between parameters of different units.