This report focuses on challenges for adaptive, reflective and legitimate regional environmental governance related to environmental risks in the regional context of the Baltic Sea. The point of departure is the assumed challenges for establishing mutually fruitful regional collaboration among a diverse group of neighbouring countries. The Baltic Sea countries differ considerably in terms of socioeconomic development, institutional structures and procedures, power relations, policy styles, political cultures and history. How can regional governance arrangements cope with such differences and establish robust and sustainable modes of risk management? The report places its focus on regulatory frameworks for identified environmental risks as well as decision-making forms and processes. It builds on a discursive comparative case-study design where five key risks for the long-term ecological integrity of the Baltic Sea previously have been studied in detail: oil discharges from marine transportations, chemical pollution, over-fishing, eutrophication and invasive alien species. The analysis is based on case studies undertaken for the international research project ‘Risk governance of the Baltic Sea’ (RISKGOV). The analytical framework is primarily based on mappings of problem structures (i.e. bio-geophysical features affecting collaborative patterns), existing international conventions, regulatory institutions and to what extent civil society actors take part in governance arrangements. Although our primary focus is on the regional scale, the analysis takes into consideration the interplay (in terms of synergistic or conflicting effects) of such regional arrangements with national, EU, and international risk management. Our findings suggest that whereas comprehensive regulatory frameworks in most cases are in place, enforcement and implementation often lags behind. Moreover, regional institutional mechanisms for systematic reflection among relevant stakeholders on long term improvement of environmental safety within the individual issue-areas as well as between different sectors are largely lacking. This tends to lead to – via mecchanisms such as path dependency, sectoral management, too narrow coneptions of uncertainty, static rather than dynamic approaches, neglect of self-moitoring activities and inadequate appreciation of governance plurality – reactive rather than forward-looking policy responses, legitimacy deficits and sub-optimal social and institutional learning