Study region Situated in the arid region of northwestern China, Ebinur Lake exemplifies a highly turbid inland water system. Changes in the distribution and intensity of suspended particulate matter (SPM) over space and time provide critical insights into the area's ecological dynamics and hydrological functions. Study focus To address the limitations of empirical models in optically complex waters, this study develops a multi-parameter spectral lookup table (LUT) based on the Hydrolight radiative transfer model. The LUT incorporates key optical constituents, including chlorophyll-a (Chla), SPM, colored dissolved organic matter (CDOM), and the backscattering probability of suspended particulate matter (zeta(SPM)). An SPM retrieval algorithm based on spectral similarity matching is proposed. Sensitivity analysis indicates that SPM and zeta(SPM) are the primary drivers of remote sensing reflectance (R-rs), while CDOM strongly absorbs in the shortwave region, and Chla exhibits weak, easily masked signals. Validation against Sentinel-3 OLCI reflectance shows strong agreement with simulated spectra (RMSE < 0.027 sr(-)(1) across 15 bands; R-2 > 0.91). New hydrological insights for the region The model was applied to Sentinel-3 OLCI imagery during the ice-free period of 2024 to estimate SPM concentrations in Ebinur Lake. Results consistently showed elevated SPM levels in nearshore areas and a slight decline in model performance during autumn. Retrieved SPM distributions align well with in-situ observations. The proposed model demonstrates high accuracy, physical consistency, and robustness for remote sensing inversion of SPM in turbid inland lakes.