One of the main challenges for current cosmology is the conciliation, in a single theory, of the formation of large-scale structures in the Universe and the properties of nearby galaxies, including the Milky Way. The most widely accepted scenario is based on cold dark matter, in which a significant fraction of the galaxies observed at present were formed by assimilation and accumulation of smaller stellar systems, similar to the Local Group dwarf galaxies. In the Milky Way, and more specifically in its halo, there are a multitude of substructures that are considered to be  direct evidence of the hierarchical galaxy formation. The ancient remnants of those extinct dwarf galaxies are found as tidal streams, coherent structures of stars associated both spatially and kinematically. The comparison of observations of these streams with their corresponding numerical simulations allows us to derive important physical parameters, probe the shape of the Galactic potential and estimate the total mass of the Galaxy. It is for this reason that new detections of these substructures provide clues about the origin and evolution of the outer regions of the Milky Way.

The globular clusters, systems composed of millions of stars, have played a key role in the study of the Milky Way because their properties contain valuable information about when and where they were born. The Galactic globular cluster system appears to be formed by two differentiated populations, old and young, where the latter may be composed of those elements formed in protogalactic systems, subsequently accreted becoming  part of the Galactic halo. In this scenario, it is expected that some of the Galactic globular clusters were surrounded by the remnants of their parent galaxies as tidal streams or in a more dispersed distribution. Identifying the clusters in the Milky Way that have been accreted and those that were formed with the Galaxy has been an important line of research in the field of stellar clusters in the last few years.

The main objective of this thesis is to locate new sections of the known tidal streams or new underlying stellar systems around globular clusters in the Milky Way. For this purpose, we have designed an observational study of a sample of 24 of these systems between 10 and 40kpc from the center of the Galaxy, representing the 2/3 of those found in this distance range. The observations were made with the Wide Field Camera and Wide Field Imager mounted on the Isaac Newton Telescope (Roque de Los Muchachos, La Palma) and ESO2.2m (La Silla, Chile) respectively. These instruments allow us to study wide areas around each target in the search for tidal remnants with deep photometry, as well as the obtention of an unprecedented database of wide-field observations of globular clusters. We detect the presence of underlying stellar populations around a significant fraction of the clusters, which manifest as overdensities in the observed color-magnitude diagrams compared with those obtained from a Milky Way stellar populations model. Because of its spatial coincidence, we associate the most of these detections with known halo substructures with available theoretical models. Based on our results in this work, we suggest that at least 1/4 of the entire Galactic halo globular cluster system was formed in smaller building blocks later accreted into the halo of the Galaxy. Interestingly, our accreted clusters candidates do not include only those classified as young halo clusters, contrary to the classic scenario of old/young halo globulars, in which the latter ones would be the systems with an external origin.

\newpage

We found 3 globular clusters compatible with the position of the tidal stream generated by the disruption of the Sagittarius dwarf spheroidal galaxy, which is surrounding the Milky Way in an almost polar orbit. In addition, we unveiled tidal debris associated to that stream in the background of 5 globular clusters but at a different radial distance, supporting the scenario in which the orbits of accreted clusters and that of the progenitor galaxy differ over time as expected in a time-dependent potential. Therefore, the Sagittarius tidal stream has contributed with 8--12 clusters the Galactic globular cluster system, which means that this system may be the member of the Local Group forming globular clusters more efficently taking into account its specific frequency (24 < Sn < 36), more than the large spirals (Sn < 2) and only comparable to the Fornax dwarf galaxy (S= 29). Other clusters in our sample are surrounded by a subjacent stellar population associated to the low Galactic latitude Monoceros ring and at least 3 of these systems are compatible in location with that substructure.  This result would be the first empirical association of globular clusters with Monoceros and would indicate that this ring was generated by the disruption of a smaller galaxy and not by the distorsion of the Galactic disk. In particular, NGC1851 may play a relevant role in the resolution of the Canis Major overdensity controversy.

In order to have a criterion to separate the stellar content of the globular cluster from the rest of the observed field, we derived complete radial profiles for these systems, including their outer regions that are usually poorly explored in comparison with their central areas. Our results suggest that dynamical King models, extensively used in the study of these systems, are not able to reproduce the outer parts of the globular clusters observed, in some of the cases, for the first time in this study. The classical tidal radius is found to be a 40% larger than the values derived in previous works (based on shallower surface-brightness profiles) and a power-law template appear to be a better representation for 2/3 of the observed profiles. The slope of this function "gamma" is used to classify the globular clusters into two well differentiated categories: those with a flatter profile and high density ("gamma "< 4; tidally unaffected) and those with lower density and a more pronounced radial profile ("gamma" > 4; tidally affected). We propose that "gamma" could be used as an indicator of the evolutionary state of a globular cluster and could possibly provide clues about its origin.