This study was conducted to investigate the effects of soil moisture content on the absorbance spectra of sandy soils with different phosphorus (P) concentrations using ultraviolet (UV), visible (VIS), and near-infrared (NIR) absorbance spectroscopy. Sieve sizes were 125, 250, and 600 mu m for fine, medium, and coarse, respectively. The medium size of the samples was usedfor the study. Investigations were conducted at 0, 12.5, 62.5, 175, 375, 750, and 1000 mg kg(-1) P application rates. Three soil moisture contents (4%, 8%, and 12%) were investigated. P concentrations of the soil samples were analyzed and reflectance of the samples was measured between 225 and 2550 nm with a 1 nm interval. Dried soil samples reflected more light than wet soil in the 225-2550 mn range. As moisture content of the soils increased, reflectance from the soil sample decreased, which indicates that water is a strong light absorber in sandy soils. Dry soil spectra were reconstructed from the wet soil spectra by removing the moisture content effect and compared with the dry spectra of the same soil sample. Absorbance and reconstructed absorbance data were prepared as calibration and validation data sets in order to measure the Performance of the spectral signal processing used for removing the moisture content effect on absorbance spectra. A partial least squares (PLS) analysis was applied to the data to predict P concentration before and after processing the spectra. The results showed that removing the moisture effect by spectral signal processing considerably improved prediction of P in soils.