Ruben Remelgado , Christopher Conrad, Carsten Meyer

Satellite remote sensing is vital to monitoring, research, and policy addressing sustainability challenges from climate and ecosystem changes to food and water security. Here, Landsat satellite data play a crucial role, given their unique global and long-term historical coverage at high resolution. Yet, severe but mostly disregarded biases in the Landsat data archive threaten the validity of their applications. Here, we demonstrate massive spatiotemporal gaps and quality limitations in Landsat data, with low, seasonally incomplete, or interrupted data coverage affecting large fractions of the world. These limitations affect various prominent data products, leading to biased perceptions of environmental-change patterns that impair reliable assessments of sustainability challenges. Moreover, the magnitude and prevalence of both data limitations and their biasing effects disproportionately affect lower-income countries. The global data-quality information we provide here supports their explicit consideration in future mapping efforts. Our results call for better data-bias reporting and control in satellite-based monitoring.