According to Kidner et al. (2001), “the viewshed is defined as the complete visible area of a viewing location”. Therefore, viewshed provides crucial information when planning Mars exploration using rovers or any other platforms. Many commercial or open source geospatial software provides viewshed analysis tools. Especially, GDAL has its own viewshed module – gdal_viewshed. Gdal_viewshed uses the method of Wang et al’s (2000) with the Digital Elevation Model (DEM) as input data. The delineation of viewshed is a quite straightforward task in a desktop environment if (1) GIS software is available; (2) DEM data is properly prepared; and (3) the performance of hardware is strong enough. However, it takes painstakingly long time often to download the DEM if the coverage area is large. Also, running viewshed delineation takes longer time if the tool runs with large size of input DEM on a consumer level desktop machine. Therefore, the research objectives is to build a web-based Mars viewshed delineation GIS application, which (1) prepares input Mars DEM data automatically; and (2) to delineate viewshed on powerful research grade server machine. The users of this tool won’t need to (1) download large size of Mars DEM data; or (2) use high performance research grade workstation. The only requirement for the users will be the access to the web browsers and internet connection.

Currently, the prototype of viewshed delineation tool is being developed in Geospatial Cyber Infrastructure (GCI) at Michigan Technological University (Michigan Tech). The geospatial cyber infrastructure (GCI) was built in Ubuntu Server. Django was used for web framework of the GCI since Django is known to be secure, scalable, versatile and fast to develop with (Django Software Foundation, 2025). Since Django requires database by default, PostgreSQL was used as database software considering future integration of PostGIS. Mapserver was used to serve Mars base layer (Mars Viking Colorized Global Mosaic 232m V2) (Williams, 2018). The GDAL and OGR were used as geoprocessing toolkits for each raster and vector data. Python was used as a server side language since it supports Django and GDAL/OGR. In the client side, OpenLayers was used as mapping toolkit to add input observation points and output viewshed areas.

In the prototype of viewshed delineation tool, users can mark multiple observing points on the base layer and can run viewshed delineation. These points can be assumed as observation points from exploration rovers. Finally, total viewable areas from observing points are displayed as layers of polygons. Currently, two types of DEMs are used as input data in the tool. The first input DEM is Mars HRSC MOLA BlendDEM Global 200m (Fergason et al., 2018), which was used for delineation viewshed for the entire surface of Mars. Since this DEM enables viewshed delineation for the entire Mars surface, the resolution of DEM is not quite high. Therefore, the result using this DEM can be used as viewshed area for small map scale (larger area). To provide higher resolution DEM, the clipped data from Mars MSL Gale Merged DEM 1m (Calef III et al, 2016) was added in the viewshed delineation tool as the first step. Since the resolution of this DEM is quite high, high precision result is expected for smaller area (larger map scale). As project progresses, additional high resolution Mars DEM data will be added in the Mars viewshed delineation tool to support high precision results in more exploration areas. In the case of high resolution DEM, the large data size tends to slow down the viewshed delineation process. Therefore, parallel computing (multithreading) was used to increase speed of viewshed delineation for multiple viewpoints. The output raster from viewshed delineation tool is converted to polygon vectors using gdal_polygonize.py.

As project evolves, user interface and output delineation results will be refined to provide more sophisticated user experience. In addition, more open source geospatial or software technologies will be integrated into GCI to add more geospatial analysis/processing capabilities. Currently, the tool is running in local development environment and being tested in a beta website served in Michigan Tech.

References
1. Calef III, F.J. and Parker, T., 2016, MSL Gale Merged Orthophoto Mosaic, PDS Annex, USGS, https://astrogeology.usgs.gov/search/map/mars_msl_gale_merged_dem_1m
2. Django Software Foundation (2025). Django Overview | Django, https://www.djangoproject.com/start/overview/
3. Fergason, R. L, et al. (2018). HRSC and MOLA Blended Digital Elevation Model at 200m v2. Astrogeology PDS Annex, USGS https://astrogeology.usgs.gov/search/map/mars_mgs_mola_mex_hrsc_blended_dem_global_200m
4. Kidner, D. B., et al. (2001). Visibility analysis with the multiscale implicit TIN. Transactions in GIS, 5(1), 19-37.
5. Wang, J., et al. (2000). Generating viewsheds without using sightlines. Photogrammetric engineering and remote sensing, 66(1), 87-90.
6. Williams, D. R. (2018). Viking Mission to Mars. https://nssdc.gsfc.nasa.gov/planetary/viking.html