The near infrared reflectance spectra (NIRS) of interstitial water samples of lake sediments in Chaohu lake were determined by near-infrared reflectance spectrometry. The respective near NIRS calibration models for predicting total nitrogen (TN), total organic carbon (TOC), phosphorus (PO3-3), ammonia nitrogen(NH3-N) and silicate(SiO-3) were built using partial least squares (PLS) algorithm with two spectral pretreatment tools including, wavelet compression combining orthogonal signal correction (OSC) and orthogonal signal correction (OSC) combining wavelet compression. The correlation coefficients between measured values and predicted values in calibration set for TN, NH3-N, PO3-3, TOC and SiO-3 were 0.975, 0.989, 0.937, 0.862 and 0.888, respectively. RMSEC(root mean square error of the calibration）for TN, NH3-N, PO3-3, TOC and SiO-3 were 0.353, 0.238, 0.031 3, 2.005 and 2.674 mg·L-1, respectively. The correlation coefficients between measured values and predicted values in validation set for TN, NH3-N, PO3-3, TOC and SiO-3 were 0.912, 0.918, 0.773, 0.337 and 0.856, respectively. RMSEP(root mean square error of the prediction）for TN, NH3-N, PO3-3, TOC and SiO-3 were 1.424, 0.945, 0.081, 7.866 and 4.273 mg·L-1, respectively.