This dissertations pursues three research objectives: (1) map how transnational the online public discourse on the global phenomenon of climate change is, (2) understand the role of the (trans)nationalized online public discourses on climate change in today’s hybrid media system, and (3) find, implement, and validate computational methods to study public discourses across different political and language spaces. Devoted to these objectives, the three articles included in this thesis produced the following results: (1) The public discourse on climate change is transnationalized to a considerable degree. First, the same topics define the issue in the countries studied. However, some of the topics are of different importance to the actors in these countries. Second, the discourses in the countries are shaped by both domestic and foreign actors. However, the scope of transnationalization is restricted to countries of the Global North, with a clear bias towards the United States. The Global South is thus a blind spot. (2) For the studied case of Germany, there is no evidence for continuous resonance among climate change skeptics’ online communication and legacy media. However, there are occasions of selective resonance when climate change skeptics manage to exploit specific events to push their perspectives and positions onto the mass media’s agenda. The influence of the transnational skeptical counter-movement on German mainstream discourse is therefore limited. (3) The combination of machine translation and topic models is a great option when it comes to the automated analysis of large multilingual corpora. Regardless of whether full texts or only the vocabulary of a corpus is translated, the approach produces reliable and robust results. Moreover, the analysis of transnational discourse convergence has shown that machine translation and topic models can also be used for comparative research.