Chemometrics and vibrational spectroscopy as green tools for mine phytoremediation strategies

► We used NIR to obtain chemical profiles of Combretum erythrophyllum growing on metal contaminated soils at different seasons. ► We applied chemometric models to reveal patterns and trends in chemical profiles. ► Principal component analysis (PCA) models revealed clustering of data points, according to the month of harvest. ► Partial least squares (PLS) models linked the variations in the chemical composition and levels of soil contamination. ► Harsh conditions on mine sites induced the production of phenolic compounds.