Abstract: It is inevitable to bring about uncertainty during the process of data acquisition,derivation,and visualization.Different from data error and data conflict,uncertainty resides in and propagates along each stage of the visualization pipeline.Uncertainty is an important ingredient of data.Visualizing uncertainty can help the analysts gain insight into the data to make better decisions.In this paper,first we introduce the main sources of uncertainty;then we summarize the state-of-the-art uncertainty visualization approaches into four categories: glyph,visual variable encoding,geometry,and animation;at last we outline some directions requiring further studies.