$Z \to \tau_h \tau_l$ cross section measurement and $\tau \tau$ mass reconstruction with the ATLAS detector at the LHC

In this dissertation, the $Z \to \tau \tau$ decay where one $\tau$ decays leptonically and the other one hadronically is presented. The process is studied with data collected from the ATLAS experiment in $pp$ collisions at $\sqrt{s}$ = 7 TeV at the Large Hadron Collider (LHC). A first observation of the process and an initial measurement of its cross section is done with 36 pb$^{−1}$ of data recorded during the 2010 run. The measured cross section value in the mass region 66-116 GeV is $857.6 \pm 84.3 \text{(stat)} \pm 124.7 \text{(syst)} \pm 29.2 \text{(lumi)} \pm 2.8 \text{(theo)}$ pb for the muon channel and $1142 \pm 138.6 \text{(stat)} \pm 197.7\text{(syst)} \pm 38.9 \text{(lumi)} \pm 2.6 \text{(theo)}$ pb for the electron channel. The measurement is updated with 1.55-1.34 fb$^{−1}$ of data taken in 2011 and is $912.4 \pm 15.0 \text{(stat)} \pm 94.7 \text{(syst)} \pm 33.7 \text{(lumi)}$ pb for the muon channel and $998.1 \pm 23.7 \text{(stat)} \pm 131.9 \text{(syst)} \pm 36.9 \text{(lumi)}$ pb for the electron channel. Finally, the clean $Z \to \tau \tau$ sample is used to study ditau ($\tau \tau$) mass reconstruction methods. The considered methods are effective mass, collinear approximation, missing mass calculator, 'true' transverse mass and bound mass. All methods are found to have advantages and disadvantages and their applicability depends on the needs of the analysis, such as whether it is for discovery or mass measurement purposes.