To investigate the influence of sample temperature on the characteristic parameters of laser-induced Cu plasma, brass was used as the study object. A Nd: YAG nanosecond pulsed laser with 532 nm wavelength was adopted under optimized experimental conditions to excite and to induce the breakdown of massive brass under different sample temperatures, and the characteristic spectral line intensity and signal-to-noise ratio of the Cu plasma were measured. The Boltzmann diagonal line and Stark broadening methods were used to analyze and calculate the electron temperature and electron density of the plasma under different sample temperatures. When the laser power was 60 mw, the characteristic spectral line intensity and signal-to-noise ratio of Cu gradually increased as the sample temperature increased and tended to be saturated after reaching maximum values under 130 ℃. The relative intensities of eight spectral lines—Cu Ⅰ 329.05, Cu Ⅰ 427.51, Cu Ⅰ 458.71, Cu Ⅰ 510.55, Cu Ⅰ 515.32, Cu Ⅰ 521.82, Cu Ⅰ 529.25, and Cu Ⅰ 578.21 nm—of Cu in the brass sample increased 11.55, 4.53, 4.72, 3.31, 4.47, 4.60, 4.25, and 4.55 times under 130 ℃ compared with those under room temperature (18 ℃), and the spectral signal-to-noise ratios increased 1.35, 2.29, 1.76, 2.50, 2.45, 2.28, 2.50 and 2.53 times respectively. Elevating the sample temperature would increase the ablation mass of the sample and plasma particle concentration compared with those under relatively low temperature and consequently would enhance the plasma emission spectral intensity. Therefore, appropriately elevating the sample temperature could increase the spectral line intensity and signal-to-noise ratio to enhance the measurement accuracy of LIBS technology in detecting and analyzing weak spectral signals and to improve its detection sensitivity for trace elements. The variation tendency of the electron temperature and electron density with the change of sample temperature was investigated. In the calculation, the electron temperature of the plasma was basically unchanged when it was increased from 4 723 to 7 121 K when the sample temperature increased from room temperature to 130 ℃. This change law was consistent with the variation tendency of the emission line intensity and signal-to-noise ratio. This condition was mainly due to the increase in the laser ablation quantity and internal energy of the plasma at the initial phase of the sample temperature rise, which increased the electron temperature of the plasma. No additional change in the sample quantity under laser ablation was observed after reaching a certain value, and the laser energy was absorbed, scattered, and reflected by excited and sputtered sample evaporants and dust particles; such process reduced the laser energy density. As a result, the electron temperature tended to be saturated, and several dynamic balances were reached. The Stark broadening coefficient of the 324.75 nm Cu atomic spectral line was selected in this study to calculate the electron density of the plasma. The variation tendency of the plasma electron density with the change of sample temperature was evaluated. The plasma electron density that corresponded to Cu Ⅰ 324.75 nm when the sample temperature was 130 ℃ increased by 1.74×1017 cm-3 compared with that under room temperature (18 ℃). This variation tendency was consistent with that of electron temperature. Appropriately elevating the sample temperature increased the electron density and the probability for electron and atom collision, which excited many atoms. This process was one of the reasons for the enhancement of spectral line density. Thus, elevating the sample temperature is a convenient and effective means to improve LIBS detection sensitivity.