Search for a command to run...
Summary Litter decomposition is a biogeochemical process fundamental to element cycling within ecosystems, influencing plant productivity, species composition and carbon storage. Climate has long been considered the primary broad‐scale control on litter decomposition rates, yet recent work suggests that plant litter traits may predominate. Both decomposition paradigms, however, rely on inferences from cross‐biome litter decomposition studies that analyse site‐level means. We re‐analyse data from a classical cross‐biome study to demonstrate that previous research may falsely inflate the regulatory role of climate on decomposition and mask the influence of unmeasured local‐scale factors. Using the re‐analysis as a platform, we advocate experimental designs of litter decomposition studies that involve high within‐site replication, measurements of regulatory factors and processes at the same local spatial grain, analysis of individual observations and biome‐scale gradients. Synthesis . We question the assumption that climate is the predominant regulator of decomposition rates at broad spatial scales. We propose a framework for a new generation of studies focused on factoring local‐scale variation into the measurement and analysis of soil processes across broad scales. Such efforts may suggest a revised decomposition paradigm and ultimately improve confidence in the structure, parameter estimates and hence projections of biogeochemical models.