Reduction of model complexity by orthogonalization with respect to non-relevant spectral changes

A method is presented to remove changes in the calibration spectra that are known to be not related to the property of interest. This can lead to mult ivariate calibration models that require fewer latent variables and are eas ier to interpret. This method requires the spectra of a sample to be measur ed under the different conditions that modify the spectra (for example, at different temperatures). These variations-are not related to the concentrat ion of the analyte and can therefore be removed before modeling with an ort hogonalization step. The method has been used to remove the effect of tempe rature in the determination of NaOH in aqueous solutions by using near-infr ared (NIR) spectra and partial least-squares (PLS) regression. This approac h reduced the number of latent variables of the final model and made the in terpretation of the PLS scores simpler.