A bewildering variety of eruptive and ejective phenomena continually take place in
the solar atmosphere on a wide range of space and time scales. Particular attention
was devoted in the past decades to those associated with the reconnection of magnetic field
lines of separate plasma systems that come into contact in the atmosphere, especially
when this is the result of the emergence of magnetic flux from the solar interior. Such
events can cause a large disruption of the solar atmosphere, lead to the release of
magnetic energy which is turned into kinetic and internal energy of the plasma
and radiation, launch impulsive mass ejections, and bring about the reconfiguration
of the magnetic domain structure in the chromosphere and corona. Even though
observationally known for many decades now, among those dynamic phenomena
there is one whose understanding has progressed very slowly: the surges.

Surges are cool, dense and non-collimated ejections typically observed in
chromospheric lines, like Hα 6563 Å, with velocities of a few
to several tens of km s-1 and lengths of 10-50 Mm. They frequently arise in the
solar atmosphere related to other transient events like UV bursts or EUV/X-Ray coronal jets.
Through idealized numerical experiments, the surges have been explained as a by-product
of magnetic reconnection taking place between emerging and preexisting magnetic
systems in the atmosphere that eventually causes chromospheric plasma to be
dragged into higher layers. In spite of their interest, those experiments miss a number
of fundamental physical mechanisms at work when the surges are ejected, so
they constitute only a preliminary step in this field.

This thesis addresses the surge phenomenon under a fourfold perspective: (1)
its basic approach is theoretical, and is carried out by modeling the formation and
evolution of the surges using a radiation-magnetohydrodynamics (R-MHD) code that includes a
realistic treatment of the material plasma properties and radiation transfer;
(2) from an observational point of view, we analyze coordinated high-resolution observations
of the chromosphere and transition region (TR), exploring the response of TR
lines to a surge ejection and the close relationship of surges to other events like UV bursts;
(3) from a forward modeling perspective, synthetic observations are created that
permit us to understand some of the peculiar features seen in the actual observations
and provide a theoretical basis for them; and
(4) from the point of view of code development, we have created a Fortran module
that improves the computational efficiency of the ambipolar diffusion term and opens
up the possibility of including partial ionization effects on the electrodynamics of the
system via the generalized Ohm's law.

The first objective is achieved thanks to the possibilities afforded by the Bifrost
code. Via 2.5D R-MHD experiments of magnetic flux emergence from the solar layers
beneath the surface up to the corona, we have found that a surge is formed in
spite of the prior interaction of the the emerging field with the granular cells at and
below the surface. The surge detaches from the emerged
region as a consequence of strong shocks that develop following the impact of plasmoids
ejected along the current sheet in the reconnection site. The surge plasma
elements experience accelerations well in excess of the solar gravity value in the onset
phase, while it undergoes free-fall in the central and decay phases. Using detailed
Lagrange tracing, we have discerned different populations with distinct evolutionary
patterns within the surge, some of them directly related to the heating/cooling processes
included as part of the Bifrost code. In fact, we have found that a non-small
fraction of the surge could not be obtained in previous and more idealized experiments
because of the lack of a proper treatment of the thermodynamics and entropy
sources.

The second goal of the thesis has been accomplished through the observation
of an episode of simultaneous occurrence of an Hα surge and of a UV burst obtained
with the Swedish 1-m Solar Telescope (SST) and the Interface Region Imaging
Spectrograph (IRIS), respectively. Although surges are traditionally related to
chromospheric lines, we have found that they can exhibit enhanced UV emission in
TR lines of Si IV with brighter and broader spectral profiles than the
average TR. Furthermore, we have provided additional observational evidence to
support the common origin and relationship between surges and UV bursts.

The third objective is achieved by means of forward modeling of numerical experiments
that include nonequilibrium ionization of key elements in the TR
like silicon and oxygen. The results show that the TR enveloping the surge is strongly
affected by nonequilibrium ionization effects, noticeably increasing the number of emitters
of the main lines of Si IV and O IV.  The departure from statistical equilibrium is due to
the short characteristic times of the optically-thin losses and heat conduction during
the surge evolution. In addition, we have concluded that line-of-sight effects are important to
understand prominent and spatially intermittent Si IV and O IV brightenings within the surges.

Finally, the last goal of this thesis has been accomplished through the creation of
a new module in the Bifrost code that implements the Super Time-Stepping (STS)
method to speed up the calculation of the ambipolar diffusion term. After carrying
out an in-depth analysis of the method, we have found the optimum combination
of parameters that ensures stability and efficiency. As a first instance of work to be
completed in the postdoctoral phase, in the final chapter we have started to explore
the effects of ambipolar diffusion on the magnetic flux emergence
process leading to the surge phenomenon and in the surge itself.