Abstarct: Sarcheshmeh is of one of the largest opencast porphyry copper mines in the world. The amount of mine waste produced by extraction and processing is very high which is mainly due to low copper grade in the host rock. Weathering and oxidation of sulfide minerals in wastes cause variability of the natural concentrations of elements and increasing their transfer rate in these environments. In order to evaluate source, type and the extent of contamination, sampling and laboratory analyzes should be performed. Existing standard methods for estimating the total concentrations of toxic elements in soil require mixture of several acids to perform acidification steps. These methods are accurate and well documented by the researchers. However, they are expensive, time consuming and laborious since requiring substantial sample preparation and digestion with abrasive chemicals. Given the disadvantages of classical measurement methods, a quick, low-cost, and easy way to continuous monitoring and assessment of elemental changes in the region is required. As an alternative to traditional sampling and chemical analysis methods, the capability of laboratory (visible, near and short wave infrared) and imaging (remote sensing) spectroscopy was investigated for prediction of toxic elements concentrations. 120 waste samples were gathered from dump number 31 in Sarcheshmeh copper mine and various experiments were carried out to determine the mineralogy and the toxic elements contents of the samples. Concentrations of all elements followed lognormal distribution. The mineral composition of samples which determined by X-ray diffraction was classified in primary, clay and secondary minerals groups. Environmental indices of toxic elements showed high contamination of some mextals, especially cadmium, molybdenum, lead, zinc and copper in the studied dump. The reflectance spectra of the samples were also measured in the laboratory using a portable spectroradiometer with mesurment range of 350 to 2500 nm. After performing the first and second derivative preprocesses on the raw spectra of the samples, concentrations of toxic elements were predicted along with investigating the prediction mechanism and elements' speciation. baxsed on the results of mechanism and speciation, nine toxic elements were divided into two groups related to Fe oxy / hydroxides (including arsenic, cadmium, cobalt and nickel) and clay minerals (copper, chromium, molybdenum, lead and zinc). It was also found that, some important phases for controlling toxic elements in the dump environment can be accurately determined by spectroscopy. Regarding the results of evaluating regression models, nonlinear methods were the most frequent among the best methods estimating the concentrations of elements. Among all preprocessing methods, first derivative presented the best results for 7 elements of arsenic, copper, nickel, cobalt, cadmium, lead, and zinc. In remote sensing part of the thesis, the best linear regression model obtained for each element by spectroscopy was applied to the images. They were then classified using a binary fitness function. The classified Sentinel 2, OLI, and fusion of OLI and ALI images were more similar to the distribution maps of elements with high correlation with oxy/ hydroxides. Furthermore, the compatibility of the classified images of Aster, Hyperion and Hyperion-ALI was higher with distribution maps of elements closely related to clay minerals. The similarity between distribution maps and classified Hyperion and ALI fused image was acceptable for nearly all elements mainly due to the improved spatial resolution of the image and it's higher spectral resolution. This similarity was greater for elements related to the SWIR range, so that the best prediction results were obtained for chromium, copper and lead by applying regression models on fused hyperspectral image. The findings of this study showed that spectroscopy (laboratory and imaging) can be considered as a novel method for assessing the concentration of toxic elements in highly heterogeneous environments.