Capturing the variation of lake-water area using remotely sensed imagery is an essential topic in many related fields. There are a variety of remote sensing data that can serve this purpose. Generally speaking, higher spatial resolution data are able to derive better results. However, most high spatial resolution data are sometimes defective because of their low temporal resolution and limited scene coverage. Visible Infrared Imaging Radiometer Suite onboard Suomi National Polar-orbiting Partnership (Suomi NPP-VIIRS) provides a newly-available and appropriate manner for monitoring large lakes because of its frequent revisit and wide breadth. But its spatial resolution is relatively low, from 375m to 750m. This study introduces a two-step method that integrates spectral unmixing and sub-pixel mapping to map lake-water area at sub-pixel scale from NPP-VIIRS imagery. Accuracy was assessed by employing corresponding Landsat images as the reference. Five plateau lakes in Yunnan province, China, were selected as the case study areas. Results suggest that the proposed method is able to derive finer resolution lake maps that show more details of the shoreline. The accuracy was significantly improved comparing to traditional classification method. Analysis also reveals that errors and uncertainties also exist in this method. Most of them come from the spectral unmixing procedure that retrieve water fraction from NPP-VIIRS data. Therefore, in order to achieve better lake mapping result, future work should concentrate more on improving this part to produce a better water fraction map first.