Here we apply, for the first time, the Cappellari & Copin (2003; hereafter CC03) Voronoi tesselation algorithm (hereafter VT) to build the XMM EPIC temperature map of A3266. We first extracted source and background images in the 0.5-7.5 keV energy band. The chosen energy range optimises the cluster signal over the particle background in the temperature range of the cluster. These two images were then used to estimate the Signal-to-Noise (S/N) of each pixel.

We were unable to use the CC03 algorithm in a single step, since X-ray events are distributed following Poisson statistics and many of the pixels have a low S/N. Our implementation of the algorithm thus involves two steps. We first selected only those pixels with a sufficiently high S/N (i.e. (S–N)/N ≥ 1.05) and used the CC03 algorithm to bin these pixels into meta-pixels with a S/N ~130 (our final goal for the temperature map). Since these meta-pixels were obtained from a high S/N subset, they are not generated from a continuous set of pixels. The second step consists of assigning all of the so-far unbinned pixels to their closest meta-pixel. Obviously, the addition of these lower S/N pixels adds scatter to the S/N of the final distribution of meta-pixels. However, the resulting distribution of convex meta-pixel cells covers the whole image without significantly degrading the S/N.

Application of this technique to the mosaic event list of A3266 results in 138 cells. We fitted the spectrum of each cell with an absorbed mekal model using XSPEC v11.2. The absorption was fixed to the Galactic value (N_H = 1.6×10²⁰ cm^-2; Dickey & Lockman 1990), and the abundances were fixed to 0.2 Z/Z_sol. This abundance value was obtained by fitting a global spectrum, extracted in a circle of 10 arcmin and excluding point sources, with a two-temperature mekal model with abundances tied together. Our best fitting global abundance is in reasonable agreement with the value found by Henriksen & Tittley (2002) in the central region mapped by Chandra (see also De Grandi & Molendi, 1999). We note that a single temperature model is an acceptable fit to all 138 cells.