Structural biases in global land degradation research: Climatic, indicator, and methodological selectivity

Land degradation (LD) research has expanded rapidly over the past three decades, yet inconsistencies across indicators, methods, and regions suggest structural biases within the global evidence base. Using a quantitative synthesis of 1701 peer-reviewed studies published between 1992 and 2025, this study examined the spatial distribution, indicator frequencies, co-occurrence networks, climatic context, and methodological structure of the LD literature. The novelty of this study lies in its systematic quantification of observability-driven biases across indicators, methods, and climatic contexts within global land degradation research. The results reveal four consistent structural patterns. First, studies are spatially concentrated in regions undergoing active degradation or recovery, while stable systems remain substantially underrepresented by approximately 14%. Second, erosion dominates the indicator structure of the literature, accounting for nearly one-third of all reported indicators, with co-occurrence analysis identifying it as the central organizing variable linking hydrological and productivity-related processes. Third, the LD literature exhibits a strong climatic skew toward arid and semi-arid environments, where key indicators reach up to approximately 30% representation. Fourth, methodological approaches are increasingly dominated by remote sensing, which accounts for more than half of studies in recent years. These findings demonstrate that land degradation knowledge is shaped not only by biophysical processes, but also by observability constraints and methodological choices that influence what can be detected, measured, and scientifically represented. The study proposes a conceptual framework of observability-driven structural bias to explain how interactions among indicators, methods, and climatic contexts systematically shape the representation of land degradation within the scientific literature.