Software Compensation for Hadronic Showers in the CALICE AHCAL and Tail Catcher with Cluster-based Methods

The hadronic energy resolution of the CALICE setup, consisting of a silicon tungsten electromagnetic calorimeter, an analog hadron calorimeter and an analog tail catcher has been studied using data taken in 2007 at CERN and simulations. To improve the energy resolution of hadronic showers in the hadron calorimeter and the tail catcher, a weighting procedure based on the energy density of the hadron shower is studied. The shower itself and its energy density was reconstructed using a simple clustering algorithm. Furthermore, the use of a neural network has been studied for the same purpose. Both methods use simulated data to determine weights which are then applied to test beam data. These first preliminary studies yield an relative improvement of the energy resolution by roughly 15 % for the shower weighting technique and 23 % for the neural network approach compared to the energy resolution of hadronic showers without software compensation applied.