This doctoral thesis is concerned with how upper secondary students in Sweden develop system geographical knowing through Geography teaching, where complex issues such as climate change are dealt with. In many countries, these issues are particularly relevant for the school subject of Geography since the subject is responsible for issues that involve system changes. These issues involve a complex web of aspects and dimensions so it is challenging for students to develop indepth holistic understandings. Commonly, students are left alone to synthesise the various parts into a whole. Systems thinking is suggested as an approach to dealing with these challenges, but there is little research on what such thinking implies in Geography teaching, or what students need to experience to develop this knowing. This educational design study departs from practice theory to explore and broaden systems thinking in Geography teaching by introducing a model task with connection web models that combine system dynamic causality, soft systems thinking, and spatial dimensions in a new way. The model task was developed and tested empirically in two upper secondary schools in Sweden. Research data consist of interviews and recordings of the students’ work with connection web models as a way of exploring complex issues. Findings from the study are presented in four peer-reviewed articles that together answer the two research questions: i) what does it mean to become system geographically knowledgeable when managing complex geographical issues? and ii) how can Geography teaching be designed to develop students’ system geographical knowing? 

Article I proposes a subject didactic model where the systems concept and other geographical concepts were used as organising tools to design Geography teaching. The model was developed in collaboration with Geography teachers and was used for designing the teaching interventions in the present study. Article II contributes findings from a series of lessons where the connection web model was introduced and presents four qualitatively different aspects of knowing that are critical for students to be able to use the model. For instance, being able to discern different causal characters of connections, to use place and scale as analytical tools, and to interpret the holistic pattern of interwoven connections. Article III analyses three different teaching designs and presents four design principles that teachers can use to support the development of the students’ system geographical knowing and overcome challenges that have been identified. Article IV presents findings about how the students experience the phenomenon of using the connection web models as a way of managing complex issues. Three different ways of experiencing the phenomenon are presented as well as critical aspects that students need to experience to qualify their knowing. 

This doctoral thesis broadens the view on what systems thinking can mean in the integrated and spatial school subject of Geography and contributes didactic models for teachers and researchers to use to develop the students’ knowing. The study argues that system geographical knowing is relevant for students to better understand complex issues concerning system changes. Finally, it is argued that connection web models have the potential to function as learning models if the model work is considered as a long-term process.