We present first results from a large-area ( \sim 80 ^ { \circ } \times 20 ^ { \circ } ) , sparsely sampled two-filter ( B , R ) imaging survey towards the Canis Major stellar over-density , claimed to be a disrupting Milky Way satellite galaxy .
Utilizing stellar colour-magnitude diagrams reaching to B \sim 22 mag , we provide a first delineation of its surface density distribution using main sequence stars .
It is located below the Galactic mid-plane , and can be discerned to at least b=-15 ^ { \circ } .
Its projected shape is highly elongated , nearly parallel to the Galactic plane , with an axis ratio of \gtrsim 5:1 , substantially more so than what Martin et al .
originally found .
We also provide a first map of a prominent over-density of blue , presumably younger main sequence stars , which appears to have a maximum near [ l , b \sim 240 ^ { \circ } , -7 ^ { \circ } ] and extends in latitude to b \sim -10 ^ { \circ } .
The young population is markedly more localized .
We estimate an upper limit on the line-of-sight depth , \sigma _ { los } , of the old population based on the main sequence width , by comparing with a simple stellar population ( Pal 12 ) , obtaining \sigma _ { los } < 1.8 \pm 0.3 kpc , at an adopted D _ { \odot } = 7.5 \pm 1 kpc .
For the young stellar population , we find \sigma _ { los } < 1.5 kpc .
The overall picture presented is one of a young stellar population that is less extended , both in terms its line-of-sight depth and angular size , than the older population .
In the literature three explanations for the CMa over-density have been invoked , namely ( a ) a partially disrupting dwarf galaxy on a low-latitude orbit , ( b ) a projection of a warped outer Galactic disk , and ( c ) a projection of an out-of-plane spiral arm .
While the data provide no firm arguments against the less well-defined third scenario , they have clear implications for each of the others : ( a ) We infer from the strong elongation of the over-density in longitude , and simulations in the literature , that the CMa over-density is unlikely to be a gravitationally bound system at the present epoch , but may well be just a recently disrupted satellite remnant .
A complexity for the satellite origin may arise from the ‘ flattening ’ of the young MS population , which is possibly more pronounced than the older one .
( b ) Based on modeling , we find that the line-of-sight depth of the MS over-density in old stars is clearly inconsistent with published locally axi-symmetric descriptions of the warped Galactic disk , such as those considered in Momany et al .
( 2006 ) .
Without detailed modeling , the data-set itself does not provide sufficient leverage to distinguish between an interpretation as sub-structure in the warped outer Galactic disk or a disrupted satellite .