Abstract: To evaluate the feasibility of using seawater as an alternative coolant in pressure suppression pools, this study conducts a systematic comparative experimental investigation on the thermal stratification characteristics of steam-seawater and steam-freshwater submerged jets. The experiments were performed under low steam mass flux conditions (G=10, 15, 20 kg/(m²·s)), and the pool was initially filled with water at 40 ℃. By measuring the temperature variation at multiple vertical positions within the pool, the temporal evolution of the temperature field was examined, along with its spatial structure and the thermocline erosion velocity. The results show that in both seawater and freshwater, the pool undergoes a three-stage temperature evolution: thermal mixing, thermal stratification, and re-mixing. During the initial stage, the degree of subcooling is high, and intense mixing induced by the chugging regime results in a uniform pool temperature. With further heating of the pool, the condensation mode transitions to bubbling condensation, the jet loses much of its stirring capacity, and a stably stratified layer forms. Thereafter, the thermocline gradually moves downward, the cold layer is eroded layer by layer, and the entire pool eventually returns to a fully mixed state. Comparing the thermal stratification behavior between the seawater and freshwater conditions, the differences are mainly reflected in the stability and dissipation rate of the stratification. At G=10 kg/(m²·s), the entire process from thermal mixing to stratification and back to re-mixing is observed. The onset temperature of stratification under seawater conditions is not significantly different from that under freshwater conditions, but the stratification lasted approximately 1 100 s longer in the seawater pool than in the freshwater pool, meaning that a stratified layer, once formed, is harder to break down in seawater. At G=20 kg/(m²·s), when the temperature near the nozzle approaches 90 ℃, the freshwater pool has essentially completed re-mixing, whereas a distinct cold water zone persists in the lower region of the seawater pool. Quantitative analysis of the thermocline erosion velocity reveals that for both the steam-seawater jet and the steam-freshwater jet, the downward movement velocity of the thermocline increases with increasing steam mass flux and decreases with increasing distance between the thermocline and the nozzle outlet. However, the thermocline erosion velocity of the steam-seawater jet is generally lower than that of the steam-freshwater jet. The lower erosion velocity causes more heat to accumulate in the upper region of the pool, leading to two directly observable consequences: First, the upper hot layer in the seawater pool heats up faster, with the time required for the nozzle region to reach 90 ℃ being approximately 1 000 s shorter than that under freshwater conditions (at G=20 kg/(m²·s)); Second, the stratified state under seawater conditions is more persistent, requiring either a longer time or a higher steam flow rate for the cold water zone to be completely eliminated. Two factors account for the above differences. First, the density of seawater is higher, resulting in a larger density difference between the hot and cold layers under the same temperature difference, which means that downward erosion of the thermocline must overcome a stronger buoyancy resistance. Second, salt ions in seawater inhibit bubble coalescence, and the presence of a large number of dispersed microbubbles increases the dissipation of jet kinetic energy along its path, weakening the effective disturbance reaching the vicinity of the thermocline. Additionally, a dimensionless correlation for predicting the thermocline erosion velocity was established. This correlation demonstrates satisfactory predictive performance (–15% to +25%) and can provide a reference for the design of pressure suppression systems in nuclear power plants. The results of this study can provide a reference for the design of pressure suppression systems using seawater as the coolant, and are helpful for assessing and mitigating the risk of condensation efficiency degradation caused by thermal stratification.