The tail is empirically modelled by a complex function (defined only above 1.255 keV) :

T_{i}(E_{j}) = D_{4}+ D_{5}exp( (PI-D_{6})/D_{7}) if PI<D_{6}T_{i}(E_{j}) = D_{5}exp(-0.5 [3(PI-D_{6})/(D_{2}-D_{6})]^{2}) otherwise

where the symbols are as follows :

- PI indicates the i-th output channel
*D*is the peak of the gaussian component of given energy, given by the gain relation_{2}*D*is computed indirectly as a function of energy as a quadratic polynomial of the Xenon absorption efficiency (3 coefficients_{4}*L*,_{1,1}*L*,_{1,2}*L*necessary)_{1,3}- D
_{5}depends on other terms and is given by formula*D*_{5}=(C/D_{7})/(1-exp[(L_{4}-D_{6})/D_{7})]) - where
*C*is a quadratic polynomial of energy up to 8.04 keV, then remains constant (3 coefficients*L*,_{2,1}*L*,_{2,2}*L*necessary)_{2,3} *D*is given by a linear function of energy in two separate stretches (of same slope_{6}*L*, but with two different intercepts_{5,3}*L*and_{5,1}*L*below and above the Xenon L-edge)_{5,2}*D*is computed via the gain relation from a single coefficient_{7}*L*_{3}*L*is an independent coefficient giving the start channel above which the tail is defined_{4}

The notation

Fig. 4.7.2-II : Energy dependency of tail parameters
*D _{4}* and

Additional representations of simulated spectra can be obtained via the following form. Data is always generated for the three MECS units.