We present a detailed spatial and dynamical analysis of the central \sim 2.2 { { h } _ { 75 } } ^ { -1 } Mpc region of the galaxy cluster Abell 521 ( z=0.247 ) , based on 238 spectra ( of which 191 new measurements ) obtained at the 3.6 m Telescope of the European Southern Observatory and at the Canada-France-Hawaii Telescope .
From the analysis of the 125 galaxies that are confirmed members of the cluster , we derive a location ( “ mean ” velocity ) of { C } _ { BI } = 74019 ^ { +112 } _ { -125 } km/s and detect a complex velocity distribution with high velocity scale ( “ dispersion ” , { S } _ { BI } = 1325 ^ { +145 } _ { -100 } km/s ) , but clear departure from a single Gaussian component .
When excluding a possible background group of four galaxies , the velocity dispersion remains still large ( \sim 1200 km/s ) .
The general structure of the cluster follows a North-West/South-East direction , crossed by a perpendicular high density “ ridge ” of galaxies in the core region .
The Northern region of the cluster is characterized by a lower velocity dispersion as compared to the whole cluster value ; it hosts the BCG and a dynamically bound complex of galaxies , and it is associated with a group detected in X-ray ( Arnaud et al . 2000 ) .
This region could be in a stage of pre-merger onto the main cluster .
The small offset ( { \sim } +250 km/s ) in the mean velocity of the northern region as compared to the whole cluster suggests that the merging occurs partly in the plane of the sky .
These results , taken together with the fact that most of the clumps detected on the isodensity maps , as well as the early-type galaxies and the brightest ones ( L > { L } ^ { * } ) are aligned , suggest that this North-West/South-East direction is the preferred one for the formation of this cluster .
The central high dense region ( “ ridge ” ) shows a lower velocity location ( { C } _ { BI } ~ { } = ~ { } 73625 ^ { +344 } _ { -350 } km/s ) and significantly higher scale ( 1780 ^ { +234 } _ { -142 } km/s ) as compared to the whole cluster values .
This is due to the presence of a low-velocity group of galaxies with a high fraction of emission line objects .
This can be explained in a scenario in which a merging of subclusters has recently occurred along the direction of the “ ridge ” with a significant component along the line of sight .
The low-velocity group would then be a high-speed remnant of the collision which would have also triggered an episode of intense star formation responsible for the large fraction of late-type objects in this region .