Biodiversity–ecosystem functioning (BEF) experiments have shown that local species richness promotes ecosystem functioning and stability. Whether this also applies under real-world conditions is still debated. Here, we focus on larger scales of space, time and ecological organization. We develop a quasi-experimental design in which we relate land-cover type richness as measure of landscape richness to 17-year time series of satellite-sensed functioning in 4974 landscape plots 6.25 or 25 ha in size. We choose plots so that landscape richness is orthogonal to land cover-type composition and environmental conditions across climatic gradients. Landscape-scale productivity and temporal stability increase with landscape richness, irrespective of landscape plot size. Peak season near-infrared surface albedo, which is relevant for surface energy budgets, is higher in mixed than in single land-cover type landscapes. Effect sizes are as large as those reported from BEF-experiments, suggesting that landscape richness promotes landscape functioning at spatial scales relevant for management.