Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.22067338Supplementary Files for "Scale-dependent gravity-referenced slopes on Phobos using diffusion-smoothed shape models with recomputed effective gravity" Authors Ryodo Hemmi, PhDInstitute of Space and Astronautical Science, Japan Aerospace Exploration Agency (JAXA/ISAS) Hiroshi Kikuchi, PhDGakushuin University Overview This Zenodo record contains the supplementary data products and analysis software associated with the manuscript: Scale-dependent gravity-referenced slopes on Phobos using diffusion-smoothed shape models with recomputed effective gravity The study evaluates how gravity-referenced slope and downslope direction on Phobos depend on retained topographic wavelength and on the spatial support used to summarize facet-level quantities. The native input is the global stereophotoclinometric Phobos shape model v004 of Ernst et al. (2023), with an approximate mean facet-edge scale of 18 m. The native model is treated as the unsmoothed reference and has no assigned e-fold smoothing wavelength L. For compact labeling in filenames, figures, and tables, it is represented by 18 m; this value denotes its approximate mean facet-edge scale rather than an applied smoothing wavelength. Volume-preserved diffusion-smoothed models were generated independently from the native shape at e-fold wavelengths L = 80, 100, 125, 150, 175, 200, 250, 500, 1000, and 2000 m. Effective gravity was recomputed for each shape model from Phobos self-gravity, centrifugal acceleration, Euler acceleration, and the Martian tidal contribution. Sampling-window sensitivity was evaluated for W = 1, 2, 4, 6, and 10 degrees. Using the 11,080 m IAU_2015 Phobos reference-sphere radius, 1 degree corresponds to an arc length of approximately 193.4 m; the physical east-west width of a longitude interval decreases approximately as cos(latitude) on the reference sphere. To keep the archive compact while preserving reproducibility and reuse, the diffusion-smoothed OBJ models and large working GeoPackage files are not redistributed. The smoothed OBJ models can be regenerated from the publicly available native v004 shape model using the archived smooth_mesh.py script and the parameters reported in the manuscript. The final facet-level CSV products are provided directly so that users do not need to repeat the computationally expensive GFandSlope and effective-gravity calculations before reproducing or extending the statistical analyses. Original Phobos shape model The native input shape model is the global Phobos stereophotoclinometric shape model v004 of Ernst et al. (2023). Source:https://sbmt.jhuapl.edu/shared-files/ Input filename used in this study: phobos_g_018m_spc_obj_0000n00000_v004.obj The original native shape model is not redistributed in this Zenodo record. Users should obtain it from the original public source above. Archive contents 1. Individually compressed facet-level products (*.csv.gz) The final facet-level products are distributed as 11 separate gzip-compressed CSV files: v004_0018_surface_TA000.csv.gz v004_0080_surface_TA000.csv.gz v004_0100_surface_TA000.csv.gz v004_0125_surface_TA000.csv.gz v004_0150_surface_TA000.csv.gz v004_0175_surface_TA000.csv.gz v004_0200_surface_TA000.csv.gz v004_0250_surface_TA000.csv.gz v004_0500_surface_TA000.csv.gz v004_1000_surface_TA000.csv.gz v004_2000_surface_TA000.csv.gz They follow the naming convention v004_<scale4>_surface_TA000.csv.gz, where <scale4> is a zero-padded four-digit scale label in meters. The 0018 product is the native unsmoothed reference; 18 m denotes its approximate mean facet-edge scale and is not an assigned smoothing wavelength. The remaining labels correspond to the applied e-fold wavelengths L = 80, 100, 125, 150, 175, 200, 250, 500, 1000, and 2000 m. The files are provided separately rather than in a single approximately 8 GB combined archive, allowing users to download only the scale or scales needed. Each .csv.gz file decompresses to the correspondingly named .csv file. Each CSV contains commented metadata header lines followed by facet-level columns including: facet_id lat_deg lon_deg dynamic_slope_deg dynamic_slope_azimuth_deg facet_area g_total_x_m_s2 g_total_y_m_s2 g_total_z_m_s2 g_total_mag_m_s2 The header records the calculation epoch, adopted density, Martian geometry, angular velocity, and other calculation metadata. The products preserve a common facet ordering, allowing direct facet-to-facet comparison between the native model and all smoothing wavelengths. These CSV files are the principal derived source products used to construct the working analysis inputs and statistical products. Large GeoPackage working copies are not redistributed because they are derived from these CSV files and substantially increase the archive size. 2. phobos_slope_summary_products.zip Machine-readable summary products used in the manuscript: global_slope_summary.csv global_slope_histograms.csv roi_scale_response.csv roi_slope_histograms.csv grid_scale_response.csv grid_window_summary.csv vector_response_summary.csv analysis_manifest.json global_slope_summary.csv contains the global area-weighted slope statistics reported in Table 2. global_slope_histograms.csv contains the differential and cumulative global slope-frequency distributions used in Figure 4. roi_scale_response.csv and roi_slope_histograms.csv contain the representative-equatorial-region results used in Figure 5. grid_scale_response.csv contains the gridded global slope and azimuth statistics for all analyzed smoothing wavelengths and sampling-window widths. Each row represents one geographic grid cell and includes its latitude/longitude bounds and center, facet count, area weight, slope statistics, mean downslope azimuth, mean resultant length, and circular dispersion. These data can be directly reconstructed into global raster-like map products. grid_window_summary.csv contains the sampling-window sensitivity results for W = 1, 2, 4, 6, and 10 degrees, reported in Table 3 and plotted in Figure 8a of the manuscript. vector_response_summary.csv contains the surface-normal and effective-gravity vector-response statistics used in Figure 8. analysis_manifest.json records analysis settings, input conventions, scale values, and weighting choices. 3. phobos_slope_code.zip Principal analysis and smoothing scripts used in this study: smooth_mesh.pyPerforms diffusion smoothing of triangular OBJ shape models using a cotangent stiffness matrix, barycentric lumped mass matrix, implicit-Euler integration, e-fold wavelength calibration, and optional post-diffusion volume preservation. phobos_slope_analysis.pyCalculates global slope-frequency distributions, representative-region statistics, fixed-grid statistics, sampling-window sensitivity, and corresponding surface-normal/effective-gravity vector responses. The publication figures were assembled from the machine-readable products using standard plotting/GIS software. Figure-specific plotting scripts and GIS project files are not part of this record. 4. phobos_slope_gravity_reproduction.zip Scripts, configuration files, and notes required to reproduce the GFandSlope and subsequent effective-gravity processing stages. The archive includes: run_gfandslope.sh gfandslope.cfg.template readme_gravity.txt NoSpin_result_TA000_scripts_package/, containing the gravity wrapper, its launcher, requirements, documentation, and the SPICE kernels used to add centrifugal, Euler, and Martian tidal accelerations (the large public ephemeris kernel mar099s.bsp is not bundled; download it from NAIF and place it in spice/spk/ as described in readme.txt) and produce the final effective-gravity, slope, and azimuth products Phobos self-gravity was calculated using the external GFandSlope software: https://github.com/AiGIS-PyAiGIS/GFandSlope GFandSlope itself is not redistributed in this record. GFandSlope source revision used: c75f8dcafb38c08f6da74806adbb70f212bd9066 The GFandSlope stage used homogeneous bulk density 1862.297786 kg/m^3 and period: 0.0. The wrapper subsequently added centrifugal acceleration using |omega| = 2.2348606662528733e-4 rad/s, Euler acceleration with an angular-acceleration magnitude of approximately 1.0386e-11 rad/s^2, and the Martian tidal acceleration. The angular acceleration was calculated by a centered finite difference of the SPICE-derived angular velocity evaluated 1 s before and 1 s after the nominal epoch. The nominal geometry corresponds to true anomaly 0 degrees at epoch 2029 JAN 01 04:27:47.401 UTC. Calculation metadata are retained in the commented headers of the generated CSV files. The nominal epoch lies near the orbital minimum of the angular acceleration. Evaluated with the same IAU_PHOBOS rotation model, the maximum of the angular-acceleration magnitude over the orbit is approximately 1.04e-9 rad/s^2, about two orders of magnitude larger. Even then, the Euler acceleration at r = 11 km is about 1e-5 m/s^2, of order 0.2% of a representative surface effective-gravity magnitude, so the Euler term remains negligible at any orbital phase (see Section 2.2 of the manuscript). Some intermediate filenames retain the legacy string NoSpin. This naming is preserved for provenance; the final products used in the manuscript include centrifugal, Euler, and Martian tidal accelerations. See readme_gravity.txt in the gravity-reproduction archive for details. Coordinate and weighting conventions All vector comparisons are performed in a common Phobos-fixed Cartesian frame. The primary regional and global mapping analysis uses W = 2 degrees. Facets are assigned to sampling windows using native-model facet centroids, and the same facet assignments are retained across all smoothing wavelengths. The sampling-window sensitivity analysis additionally uses W = 1, 4, 6, and 10 degrees. Global slope-frequency and regional/grid statistics use facet-area weighting from the corresponding shape model. Cross-scale surface-normal and effective-gravity vector comparisons use native-model facet areas as fixed weights. Reproduction workflow A complete reproduction proceeds conceptually as follows: Obtain the native v004 Phobos OBJ shape model from the Small Body Mapping Tool archive. Run smooth_mesh.py independently for each target smoothing wavelength with volume preservation enabled. Convert each OBJ mesh to GFandSlope input format using util/Mesh2GFandSlopeInput.py from the GFandSlope repository. Run GFandSlope using run_gfandslope.sh and gfandslope.cfg.template to calculate shape-derived self-gravity. Run the accompanying gravity wrapper in NoSpin_result_TA000_scripts_package/ to reverse the outward-pointing GFandSlope GravAcc vector and add centrifugal, Euler, and Martian tidal accelerations, thereby generating the final facet-level effective-gravity, slope, and azimuth products. Use the resulting facet-level products to construct the working analysis inputs and run phobos_slope_analysis.py. Use the archived summary products to reproduce or independently replot the statistics presented in the manuscript. The smoothing sequence used in the study was generated with: python smooth_mesh.py phobos_g_018m_spc_obj_0000n00000_v004.obj \ -L 80 100 125 150 175 200 250 500 1000 2000 \ --preserve_volume --keep The principal statistical analysis was run with: python -u phobos_slope_analysis.py \ --gpkg-dir v004gpkg \ --native-scale 18 \ --native-obj phobos_g_018m_spc_obj_0000n00000_v004.obj \ --obj-pattern 'phobos_g_018m_spc_obj_0000n00000_v004_diffuse_efold_{L}_preserve.obj' \ --output-dir slope_analysis \ --chunk-size 200000 \ --obj-normal-chunk-size 500000 \ --regional-weight current \ --vector-weight native \ --global-hist-bin-width-deg 0.5 \ --roi-hist-bin-width-deg 0.5 \ --fine-quantile-bin-width-deg 0.001 \ --grid-sizes-deg 1 2 4 6 10 \ --grid-quantile-bin-width-deg 0.1 \ --gravity-mag-bin-width-pct 0.001 \ --gravity-mag-max-pct 100 \ --plots In the original workflow, working GeoPackage files were generated from the facet-level CSV products with GDAL for use by phobos_slope_analysis.py. These large intermediate GeoPackage files are not redistributed because the source facet-level products are included in this record as the 11 individually gzip-compressed CSV files listed above. External software GFandSlope:https://github.com/AiGIS-PyAiGIS/GFandSlope Small Body Mapping Tool:https://sbmt.jhuapl.edu/ mkgraticule_planet was used to generate the map graticules:https://doi.org/10.5281/zenodo.20397280https://github.com/ryodohemmi/mkgraticule_planet Related publication Hemmi, R., & Kikuchi, H."Scale-dependent gravity-referenced slopes on Phobos using diffusion-smoothed shape models with recomputed effective gravity."The Planetary Science Journal. Article DOI: [ARTICLE DOI WHEN AVAILABLE] Citation If you use these supplementary products or scripts, please cite both the associated journal article and this Zenodo record: Hemmi, R., & Kikuchi, H. (2026). Supplementary Files for "Scale-dependent gravity-referenced slopes on Phobos using diffusion-smoothed shape models with recomputed effective gravity." Zenodo. https://doi.org/10.5281/zenodo.22067338 Use of large language models The authors used ChatGPT (OpenAI) and Claude (Anthropic) for English-language editing and wording refinement of the associated manuscript, and for assistance in writing and debugging analysis code. All suggested changes and code were independently evaluated and tested by the authors, who take full responsibility for the archived scripts and data products. License Creative Commons Attribution 4.0 International (CC BY 4.0) GFandSlope is external software, is distributed separately under its own license, and is not redistributed as part of this Zenodo record. Contact Ryodo HemmiInstitute of Space and Astronautical Science, Japan Aerospace Exploration Agency (JAXA/ISAS)hemmi.ryodo@jaxa.jp
Keyword(s):
MeSH Term(s):
Grants and Funding:
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)