The Pierre Auger Observatory started collecting data in 2004. The obtained dataset is of unprecedented quality and quantity enabling information about ultra high energy cosmic ray flux and origin. However, due to very limited duty cycle of Fluorescence Detector, the reliable information about UHECR mass compostion in the region of the highest primary energies is still missing. Therefore, the accent has been put on finding a method to infer mass composition from measurements of the Surface Detector which duty cycle is unlimited. Two approaches were investigated for obtaining mass composition information from the SD measurements. Correlation between muon content sensitive observables and structural and temporal properties of signal recorded in SD stations was studied in details. Muon fraction estimator was researched and determined. Its performance was studied on wide range of air shower observables and potential applications. It was applied to the PAO SD data and estimate of muon signal was obtained. Comparison to simulations of different hadronic interactions models was perfomed. The deficit of predicted muon signal was confirmed and quantified with respect to obtained estimated value. Detailed study of correlation between the depth of the shower maximum and SD observables related to shower front was performed. The Xmax estimator based on SD observables was determined and its performance was studied. In addition, the empirical start time variance model was obtained and applied in reconstruction of shower geometry related SD observables.