Long-term and high-quality data are needed to assess the status, variability, change and drivers of dryland ecosystems, but these data are often lacking because dryland ecosystems are less studied without a consistent and systematic indicator framework. Moreover, these data are difficult to derive due to the highly fragmented spatial heterogeneity and temporal variability, which undermines our understanding of dryland ecosystems. This paper reviews the gaps in the existing essential variables of dryland ecosystems and proposes four criteria to identify essential dryland ecosystem variables (EDVs). The preliminary proposed EDVs have 6 dimensions and include 15 variables, which highlights the gaps in the current essential variables of thematic domains and focuses on the traits of dryland ecosystems. This study is only a first attempt, and a long-term process is required to establish the final EDVs and reach a consensus in the global dryland research community.