To study the accretion phase for local massive galaxies , we search accreting satellites around a massive compact galaxy ( M _ { * } \sim 3.9 \times 10 ^ { 10 } M _ { \odot } ) , spectroscopically confirmed ( z _ { spec } = 1.9213 ) in the eXtreme Deep Field , which has been originally reported in Szomoru et al .
We detect 1369 satellite candidates within the projected virial radius ( r _ { vir } \sim 300 kpc ) of the compact galaxy in the all-combined ACS image with 5 \sigma -limiting magnitude of m _ { ACS } \sim 30.6 ABmag , which corresponds to \sim 1.6 \times 10 ^ { 7 } M _ { \odot } at the redshift .
The photometric redshift measured with 12 multi-band images confirms 34 satellites out of the candidates .
Most of the satellites are found to have the rest-frame colors consistent with star forming galaxies .
We investigate the relation between stellar mass and star formation rate ( the star formation main sequence ) , and find the steeper slope at the low-mass end ( < 10 ^ { 8 } M _ { \odot } ) , while more massive satellites are consistently on the sequence reported in previous studies .
Within the uncertainties of star formation and photometric redshift , we conjecture possible scenarios for the compact galaxy which evolves to a local massive galaxy by way of significant size and mass growth .
While merging of the existing total stellar mass of the satellites is not enough to explain the mass growth predicted by observations and simulations , the contribution by in-situ star formation in the satellites would compensate the deficit .
Provided that most satellites keep the observed in-situ star formation and then quench before they accrete by , e.g. , environmental quenching , the compact galaxy would become a massive early-type galaxy consistent with the local size-mass relation .