Reimagining the scale in climate services 

The problem of scale and how to link phenomena within and across scales is an important scientific question in many fields, and is particularly relevant for climate change governance (e.g. Levin, 1992, Kolbert, 2006). However, climate science still commonly applies one-dimensional and static approaches to scale, which have been criticised for being apolitical and detached from the social construct and relevance of the phenomena being studied.   We investigated the concept of scale used by the climate change community and sought to identify what can be learned from other disciplines and practitioners that frame and use scale differently. Following Howitt (1998), we investigated three framings of scale: i) scale as size, (ii) scale as level and (iii) scale as relation. Scale as size focuses on measurement units, such as space and time and is the common framing in physical sciences. Scale as level, commonly used in physical geography, refers to the conceptual hierarchy of spaces and their interplay that reflects actual levels of organisation in the real world. Related, but less studied and applied, scale as relation emerged from ecology and now is also common in social sciences.We found that scale as size is commonly used in climate science and services, especially related to space and time. For spatial scales, downscaling techniques and, more recently, kilometre-scale global climate models, focus on ever-finer scales that seek to align the scale of observation or model with the operational or impact scale of a (climate) phenomenon. For temporal scales, besides the commonly used long term climate projections, there has been an increased interest in (sub)seasonal and multi-annual predictions. Even more relevant are aspirations to provide seamless climate predictions and projections that directly link these different modelling approaches. We found less evidence of the climate science community using other interpretations of scale, but some climate services practitioners appear to engage with these other framings. Interpreting scale as level could help link processes and phenomena that operate at different levels. Equally, the relational framing of scale is perhaps the most important for sustainable, resilient and just aspects of climate change. For example, it is key to explore and communicate the relationship between global GHG emissions and large scale phenomena, on one side, and the local impacts and climate-related decision-making contexts, on the other. These types of relational framings alongside providing climate information across timescales will increasingly be required to render climate services more relevant and applicable.With these findings in mind, we present results from a first attempt to jointly rethink the concept of scale with various climate practitioners and scientists. We conclude by arguing that shifting to a multi-level, multi-dimensional interpretation of scale would allow climate services to better support decision-making for multi-faceted climate change adaptation action.