Vegetation carbon turnover processes in forest ecosystems and their dominant drivers are far from being understood at a broader scale. Many of these turnover processes act on long timescales and include a lateral dimension and thus can hardly be investigated by plot-level studies alone. Making use of remote sensing-based products of net primary production (NPP) and biomass, here we show that spatial gradients of carbon turnover rate (k) in Northern Hemisphere boreal and temperate forests are explained by different climate-related processes depending on the ecosystem. k is related to frost damage effects and the trade-off between growth and frost adaptation in boreal forests, while drought stress and climate effects on insects and pathogens can explain an elevated k in temperate forests. By identifying relevant processes underlying broadscale patterns in k, we provide the basis for a detailed exploration of these mechanisms in field studies, and ultimately the improvement of their representations in global vegetation models (GVMs).