Incorporating ontological characteristics for global landform classification based on 30 meters DEM

Landform classification and mapping provide fundamental data for Earth science research, natural resource management, environmental monitoring, urban planning, and various other domains. Despite the availability of DEMs with 1-arc second resolution, global-scale studies on landform classification and mapping are inconsistent in terms of general classification systems and methods. Landforms represent not only assemblages of morphological characteristics but also encompass the human understanding of the Earth, which is constrained by the nature and scale of quantitative analysis. Here, we propose a novel framework for global landform mapping to significantly improve the quantitative evaluation of geomorphological features. The proposed framework incorporates geomorphological ontology that takes account of their conceptualization to construct classified objects. We propose the accumulated slope (AS) and mountain uplift index (MUI) to emphasize the integrity and continuity of geomorphological units, providing more precise results compared to traditional methods. Aggregating local terrain features into global metrics, AS effectively overcomes the potential negative influence of increased resolution on landform integrity. MUI aligns better with human perception of mountainous morphology and surpasses the limitations of window-based computing. In presenting the new framework, we have developed and made available a public dataset, Global Basic Landform Unit (GBLU), which incorporates a comprehensive set of objects that constitute the range of landforms on Earth. In emphasizing the integration of classification with quantitative analysis, GBLU highlights the connection between natural objects and human understanding in geomorphology and the Earth sciences. The GBLU outperforms previous datasets (the basic landform classification and global mountain assessment) in expressing landform details. GBLU can be downloaded at https://geomorph.deep-time.org. It serves as a valuable resource in facilitating a deeper understanding of landform spatial distribution and evolution, and supporting research in a diverse range of fields.