Sea Surface Temperature (SST) is a basic parameter in characterizing the ocean-atmosphere system and serves an important function in climate change. Many types of cloud-free, high-spatial, and temporal coverage merged SST products have been generated by the Croup for High Resolution Sea Surface Temperature. These products provide important data sources that can be used in a wide variety of operational and scientific applications. However, differences are existed among these products, due t。 their specific research requirements, different blending algorithms, different satellite SST sources for blending, and quality control methods. Therefore, monitoring the quality of these products is necessary, particularly at shelf and coastal seas around China, which are characterized by complex atmospheric conditions and hydrodynamics. This study compares four types of merged SST products in the South China Sea and adjacent waters in the years 2008 and 2009. Four multi-satellite merged SST products-the Operational SST and Sea Ice Analysis (OSTIA), microwave/infrared optimally interpolated SST, microwave optimally interpolated SST, and new generation SST (NGSST)-are validated with the Argo SST in the shelf sea and Argos SST in the shallow coast. The match-up data are collected on the same day and location. The Root Mean Square (RMS), bias, and correlation coefficients are calculated and used to quantify the errors. These products are projected int。 the same grid of NGSST using the nearest-neighbor sampling method for comparison. OSTIA is selected as the basis, and the rela- tive differences between OSTIA and the other three products are computed and visualized using maps, box-plot, and time series plots. The statistical results show that the RMS between the merged SSTs and Argo temperature ranged between 0.3℃ and l. 0 ℃, whereas the bias ranged between -O. 1 ℃ and 0.6 ℃ in the shelf sea (water depth > 80 m). The other three merged SSTs were consistent with the in situ data in the coastal area, except for NGSST, which had a significantly warm bias (- 1 ℃)and the largest RMS (- 1.5 ℃). The bias and RMS of OSTIA were the smallest. An inter-comparison indicates no significant differences among the four merged SST products in the shelf sea. Their biases were within 0.3 ℃. However, the deviation increases in shallow water. The largest bias was found in winter because of the poor weather conditions, whereas the smallest bias was found in summer. In summary, the four merged SST products were consistent with in situ data in the study region, except for the NGSST in the shallow coastal sea and the OSTIA product exhibited the best performance. This study has provided a reliable basis for the effective application of these merged SSTs with high spatial and temporal coverage in the South China Sea and its adjacent waters.