A multichannel thermal infrared radiometer, called CE312, was used to study the infrared characteristics of the Dunhuang radiometric correction field. Site surface radiance and the downwelling atmospheric radiance were obtained by measuring the target site and the infrared standard plate. The multichannel temperature and emissivity separation algorithm was used to calculate the site channel emissivity and temperature. Finally, the optimal offset method was employed to obtain the site emissivity spectrum. The same target area was measured using a 102F Fourier transform infrared spectrometer. The results separated by the iterative spectral smooth temperature and emissivity algorithm were then compared with those separated by the multichannel temperature and emissivity separation algorithm. The comparison results show that the maximum deviation of the channel emissivity obtained by the two methods is within 0.011, and the site temperature deviation is within 0.104 K, indicating that the usage of the multichannel thermal infrared radiometer can separate the site temperature and emissivity in addition to obtaining high-precision thermal infrared site parameters. This test provides a reference for the automated observational absolute radiometric calibration of the satellite remote sensing thermal infrared band based on the Dunhuang radiometric correction field.