GoKawiilOur work relies on three datasets: (1) global digital elevation models and bathymetric data for both Earth and Mars; (2) maps of fluvial features on Earth (major global rivers and deltas) and Mars (valley networks, fluvial ridge systems, outlet canyons and interpreted deltas), along with maps of oceanic features on Earth (continental shelf, shelf break and ocean floor) and interpreted shorelines on Mars; and (3) results of elevation, slope, curvature and landscape classifications for each cell on both Earth and Mars. We describe the data and outline our methods for each dataset below and briefly explain the flowchart in Supplementary Fig. 1.
Global digital elevation data and bathymetric data
Earth
Three global digital elevation models integrate both land and ocean surfaces at different resolutions and levels of consistency: (1) the ETOPO Global Relief Model with a general average resolution of about 1.85 km per pixel63; (2) the SRTM30_PLUS Estimated Topography with a resolution of around 1 km per pixel64; and (3) the General Bathymetric Chart of the Oceans (GEBCO) with a resolution of approximately 500 m (ref. 65). For our global-scale topographic analysis, we used the ETOPO1 Global Relief Model because of its consistent pixel resolution across both terrestrial and oceanic regions. ETOPO1 provides a uniform 1 arcmin resolution (about 1.85 km per pixel), integrating satellite altimetry, shipboard echo-sounding and terrestrial measurements into a cohesive dataset63. This consistency is important for studies requiring seamless data across different terrains, ensuring that both land and ocean topography are represented with the same level of detail. By contrast, the SRTM30_PLUS Global Bathymetry and Topography dataset, while offering higher resolution for land areas (30 arcsec), lacks uniformity as it focuses primarily on terrestrial regions and provides less detailed coverage for the oceans64. Moreover, the GEBCO dataset, although detailed for ocean bathymetry, does not offer the same consistent pixel resolution for land topography, leading to potential discrepancies when integrating land and ocean data65. Therefore, ETOPO1 was selected to ensure uniform resolution and comprehensive coverage across both terrestrial and oceanic environments, addressing the need for consistent pixel data in our analysis.
Mars
We used the global Mars Orbiter Laser Altimeter (MOLA) gridded topography, which offers a pixel resolution of 463 m per pixel66. This dataset is derived from more than 600 million measurements covering the entire Martian surface. These measurements were meticulously adjusted to ensure consistency, providing a uniform pixel resolution across the entire terrain of Mars67,68.
Data resampling
We resampled both digital elevation models to multiple resolutions—2.5 km, 5 km and 10 km—for several key reasons: (1) Resampling the topographic data of Earth and Mars to a uniform resolution is essential to apply uniform analytical methods and enable direct comparison of topographic features across both planetary surfaces. (2) These specific resolutions were selected to intentionally exclude fine-scale landforms on Mars, as these are generally younger in age69. Our study, however, focuses on older, broader-scale topographic features that provide insights into ancient surface processes. The resampling was conducted using the ‘Resample’ tool in ArcGIS70,71, in which we applied the ‘nearest neighbour’ option to preserve the exact elevation values, minimizing significant interpolation and smoothing, and thereby ensuring the integrity of the original data72,73.
Maps of fluvial and oceanic features on both Earth and Mars
To identify a rough search zone on Mars for the transition from landscape to seascape, we used maps of the major world rivers and deltas on Earth40,41. These typically indicate where the terrestrial landscape ends and the oceanic zone begins. However, this assumption holds only if the rivers and deltas were active simultaneously. Changes in sea levels could alter this relationship, but the maps still provide a useful approximation of the extent of the transition zone. Moreover, we used an extensive dataset mapping seafloor geomorphic feature42, which not only helps to define the zone but also offers insights into how oceanic geomorphic features evolve spatially. We focused on the key geomorphic features that define the transition: the continental shelf, the shelf-break slope, the continental slope, the continental rise and the key ocean floor landforms (abyssal and hadal zones).
... continue reading