[eng] The onset of COVID-19 in late 2019 had a wide and profound effects on our lives. Some of the
outstanding impacts of the pandemic are on the way people interact and travel. In response
to the sanitary crisis, many countries have implemented containment policies that have proven
effectiveness in controlling and mitigating the spread of the disease. However, the increase in
the frequency of epidemics observed in recent years, underlines the importance of knowledge on
the effects that restrictive policies on human mobility have on epidemic spreading. This is useful
not only for understanding and predicting the dynamics of COVID-19 infection, but it is rather
essential to better cope with similar scenarios in the future. The purpose of this Master’s Thesis is
to assess how the human mobility network, on a country-wide basis, has evolved in response to
restrictive measures and how these changes affect the ability of the network to support diffusion.
For that purpose, we use movement data, of mobile phone users, that account for the number
of trips between each pair of locations in Germany and Spain. In Germany, we focus on the first
outbreak, while in Spain, we extend the coverage period to more than a year later, allowing us
to identify long-lasting changes. The study of mobility patterns in both countries showed that
traffic was effectively reduced. In Germany, pre-pandemic values were re-established in early
June, 2020, whereas in Spain they are no longer reached. Furthermore, this reduction did not
occur homogeneously in the network, with long-distance flows showing a greater reduction than
short-distance flows. Furthermore, in Spain, we find that long-distance travel has been one of the
primary drivers of the epidemic across the country during the early stages of the pandemic. To
better understand these results, we go one step further and, for Spain, we study and characterise
weekly mobility networks, using the tools provided by Network Science. Our analyses reveal that,
since the implementation of the containment policies, local connections have a higher probability of
being retained and hence paths are generally longer, since more local steps have to be included.
Moreover, we find evidence of profound structural changes in the networks that remain present in
the long term. Such changes cannot be explained by a uniform reduction of mobility alone. The
resulting mobility networks are less dense, more clustered and local, and hence, more homogeneous.
Finally, we study how these changes affect the spread of epidemics. To this end, we implement an
epidemiological metapopulation model that takes into account different containment scenarios. We
find that lockdown policies have a significant impact on the spread of the outbreak. In addition,
a reduction in long-distance travel, reduces the geographical spread of the disease. This effect is
primarily due to structural changes in mobility networks. All of this suggests that specific mobility
restrictions, that target long-distance connections, should be used to limit the spread of a disease.
We hope these findings will be of great assistance in mitigating the current COVID-19 pandemic
and lead to a better preparedness for similar future scenarios.