To explore the way to use remote sensing data to monitor the thermal plume from a nuclear power plant and verify the application ability of the thermal infrared data from Landsat-8 satellite in thermal plume monitoring of a nuclear power plant, a sea area of 10×10 km$^2$ near Hongyanhe Nuclear Power Plant is selected as the research area. The cross calibration method consisting of the thermal infrared data from Landsat-8 satellite and the temperature product from the Moderate Resolution Imaging Spectroradiometer (MODIS) is used to establish a temperature information extraction model. By referring to the 'Sea Water Quality Standard', different temperature rising intensity is classified into five ranks. With the supplemented in-situ hydrological and meteorological data, the temperature field distribution and temperature rising intensity variation of the thermal plume from Hongyanhe Nuclear Power Plant are analyzed. The temperature rising curves of a step of 1℃ in different tidal states are mapped out. The areas affected by different temperature rising are obtained. The data are compared with the comprehensive result of the mathematic simulation and physical model test. The results show that with the Landsat-8 satellite data, precise monitoring of thermal plume from a nuclear power plant can be implemented.