The present study is focused on remote quantification of total suspended matter (TSM) for turbid inland waters. In situ remote sensing reflectance (Rrs) and TSM at 863 stations over 10 inland water bodies from China, Australia, and USA were collected and examined. Four empirical regression models based on sensitive reflectance bands (SB), derivatives (SD), the band ratio proposed by Doxaran et al. (2002; Rrs850/Rrs550: DM), and optimal band ratios (OBR) were examined to estimate TSM. The performance varies due to TSM concentration and the Chl-a: TSM ratio. The four models perform well when the water bodies are dominated with non-algal particles at high TSM concentration and yielded higher accuracy (R-2 ranged from 0.83 to 0.91) with both DM and OBR models, while the OBR model outperformed other models when waters are dominated by phytoplankton. Our findings also indicate that phytoplankton in the water column affects the band ratio algorithm for TSM estimates. When data from all water bodies are considered collectively, the OBR model (R-2 = 0.92) marginally outperforms the other three models (0.89, 0.87, and 0.88 for DM, SB, and SD, respectively). Future studies should be undertaken to analyze the influence of phytoplankton abundance on water-leaving signals for TSM estimates. The results of the present study also need further analyses to gain a more in-depth understanding of inherent optical properties for optically active constituents (OACs), such as absorption and backscattering to interpret the observed variations.