• Our review examines how local knowledge is used to produce usable climate services. • Studies on local knowledge are growing, yet real integration cases remain limited. • Local knowledge is often narrowly defined and excludes diverse knowledge holders. • Exploration of local knowledge is often explored for its own sake, not connected to decisions. • We propose a framework that broadens and embeds local knowledge across the climate service value chain. Climate services (CS) often face usability challenges, with many climate information products lacking uptake and use. Integrating local knowledge (LK) can enhance the credibility, salience and legitimacy of CS. Although international frameworks increasingly call for incorporating LK, how it is studied and applied in practice remains unclear. We conducted a systematic review of CS and climate change adaptation literature (119 articles) and operational CS projects (22 projects) to examine where LK is explored, whose knowledge is considered, and how it is used within CS. Our review shows a growing body of scholarship focusing on LK related to risk knowledge, monitoring and prediction, risk management, and collective action. These dimensions add value by improving the usability of CS and providing a resource for communities or individuals to build resilience. Despite these advances, the actual integration of LK into operational CS remains limited. This gap stems from insufficient conceptual clarity regarding what constitutes LK, approaches that separate LK from the context of its use, and the tendency to recognise LK only at the end-user level, neglecting more tacit forms of LK mobilised by other actors throughout CS development. We argue that addressing this challenge requires reframing LK within the context of CS production and the decisions CS aim to support. Exploration of LK should start from decision-making and the ways in which LK is mobilised. A decision-focused perspective applied throughout the CS value chain can enable a more systematic and sustained integration of LK across all actors engaged in CS provision. Climate services (CS) are typically produced through collaboration among stakeholders on the CS value chain. The actors on the value chain perform crucial functions, converting raw climate data to downscaled sector-specific information that is not only locally relevant but is communicated through channels and in ways that align with knowledge generation and sharing mechanisms of the end user. The end user, in this case, may be a local community, business or an institutional agency and organisation (e.g., water management agencies, line ministries, civil society organisations, etc.). Usability of CS, therefore, hinges on these being embedded in the local decision-making processes of stakeholders. Knowing what works on the ground, what information is needed to inform decisions, when and how, as well as building trust with users, is critical for the uptake of CS. This is precisely why integrating local knowledge (LK) is crucial. We find that even though CS are produced through collaboration among stakeholders on the CS value chain, the exploration and integration of local knowledge remain disconnected from this reality. Efforts to incorporate LK largely confine it to the end-user level, often reducing LK to a set of indicators of interest. Moreover, documented examples of LK integration in operational CS remain scarce. The narrow understanding of what LK entails and who holds it has constrained its practical use. In doing so, the central aim of producing decision-oriented information remains only partially realised. To address this gap, we argue the need to reframe LK through the lens of the CS value chain, recognising that CS are co-produced through collaboration and negotiation among diverse knowledge systems. Adopting a broader conception of LK means acknowledging the different ways it is generated: through personal experience, tradition, culture and norms, recent human-climate interactions, and structured or formalised processes. This broader framing of LK allows us to identify relevant knowledge not only at the end user level but throughout the climate services value chain. Doing so can help embed climate services more effectively in the local context and improve their credibility, salience and legitimacy. Practical ways in which our reconceptualisation can help in identifying and articulating LK across different levels of the CS chain include (but are not limited to): • Data providers, integrators, modellers: Engaging with the LK of global and local data providers can help sharpen decisions on data and model selection, improve model calibration, help bridge gaps between global datasets and local realities, and produce a more nuanced understanding of what climate services can or cannot do. • Service providers: Stakeholders like government agencies or private sector companies are often responsible for providing climate sectoral information tailored to a specific sector. Given the role of service providers in translating and tailoring the climate information, engaging with their tacit knowledge is critical in understanding the broader landscape within which climate services are embedded, how information can be made more salient and more closely linked to action, and managing knowledge flows across institutions, sectors and scales. • Service purveyors: Their proximity to the local context provides them with the know-how of both the end user and their physical environment. This positioning enables them not only to translate information into usable formats, identify trusted communication channels, and build legitimacy, but also to validate climate data and model outputs against local insights. • End users: Valuing their diverse experiential and intergenerational knowledge ensures services are aligned with real needs, enabling better-informed and more resilient decisions. Furthermore, our recommendation on how to operationalise this reconceptualisation is not by viewing integration of LK as a goal in itself, but rather to understand how knowledge is mobilised for decision-making. Therefore, the entry point to integrating LK should not be the knowledge itself, but rather the relevant decisions or processes that CS aims to support. Such integration is key to overcoming the usability gap that continues to limit the effectiveness of many CS.
Rastogi et al. (Wed,) studied this question.