The COVID-19 pandemic caused unprecedented global disruptions, with the hotel and aviation industries—two critical pillars of tourism—among the hardest hit. This study analyzed 451 hotel-related and 336 aviation-related records from the Web of Science database, applying semantic network analysis to uncover eight clusters of crisis management knowledge: basic functions, crisis response, operational strategies, epidemic prevention and control, crisis perception, innovative services, scope of influence, and internal and external environments. Latent Dirichlet Allocation (LDA) topic modeling identified distinct thematic strategies for each sector. In hotels, these included Digital Innovation Transformation, Monitoring Management Procedures, Emotional Awareness Incentives, and Resilience Mechanism Establishment. In aviation, strategies focused on Green Economic Transformation, Co-creation Value Realization, Passenger Incentive Mechanisms, and Balancing Health Risks. By visualizing co-occurrence relationships and mapping thematic intersections and divergences, this study provides actionable insights into the recovery strategies of these industries. The findings offer robust support for developing targeted management approaches and decision-making frameworks to ensure the sustainable growth of the tourism sector.

Sustainable tourism research has generated substantial insights into environmental, social, economic, governance, and cultural dimensions of sustainability. However, existing approaches provide limited explanation of how different capacities interact through stewardship processes to shape divergent sustainability trajectories under conditions of uncertainty and change. This paper develops the Adaptive Socio-Ecological Stewardship Framework (ASES) to conceptualize sustainable tourism as an adaptive process emerging from interactions among interconnected stewardship capacities within complex socio-ecological systems. Drawing on interdisciplinary scholarship from sustainable tourism, socio-ecological systems, adaptive governance, resilience theory, sustainability transitions, and community development, the paper synthesizes existing knowledge into an integrated conceptual framework. ASES identifies five stewardship capacities: community agency, institutional capacity, technological capacity, ecological capacity, and cultural capacity. Rather than treating these capacities as separate sustainability dimensions, ASES explains how their reinforcing, constraining, and conditional interactions generate alternative sustainability pathways. The framework proposes four adaptive stewardship configurations: regenerative sustainability, technocratic sustainability, fragile sustainability, and unsustainable lock-in. By shifting analytical attention from sustainability pillars toward the mechanisms through which stewardship capacities interact, ASES contributes a tourism-specific framework for understanding adaptation, transformation, and sustainability transitions. The framework provides a foundation for future empirical testing, adaptive governance, and regenerative tourism planning.

Effective multi-agency collaboration (MAC) is essential for disaster risk reduction; however, its implementation remains profoundly challenging in contexts where modern state structures intersect with traditional governance systems. Grounded in polycentric governance theory, this study investigates the barriers to MAC in Samoa, a Small Island Developing State characterized by the coexistence of formal agencies and the indigenous fa’a-Samoa governance system. Using a qualitative case study design, semi-structured interviews were conducted with 15 senior officials across Samoa’s disaster management landscape. Findings reveal that challenges stem not from a lack of policies, but from a fundamental institutional misalignment between bureaucratic systems and socio-cultural structures, resulting in informational silos, fragmented communication, and a disconnect between national and village-level response. The study identifies the critical need for bridging institutions that can translate between these different logics of operation. The paper argues that for MAC to be effective, governance models must move beyond technical solutions and engage in institutional bricolage, creatively blending modern and traditional systems. This involves formally integrating matai (chiefs) and pulenu’u (village mayors) into the command structure and fostering configurational boundary work through joint simulation exercises. This offers a new perspective on disaster risk reduction in the Pacific and other post-colonial contexts, demonstrating that resilience is built by designing for polycentricity, rather than against it.