All mosaics in this archive were developed in support of a USGS Scientific Investigation Map (SIM) and accompanying chronostratigraphy for Urvara crater, Ceres. Urvara is one of two large basins in the southern hemisphere of Ceres, with a diameter of 170 km. The map region spans -128 deg to -93 deg E longitude and -59 deg to -35 deg N latitude, slightly beyond the longitudinal range of the available high-resolution data. A single mosaic based on high-resolution images (3.5-20 m/px) acquired at the end of the Dawn mission forms the primary basemap for the SIM (Sizemore et al., 2025) with a previously produced Low Altitude Mapping Orbit (LAMO) mosaic (Roatsch et al., 2017; Nass et al., 2023) providing supplemental coverage. Mapping was carried out at a digitalization scale of 1:50,000 and a publication scale of 1:250,000. The basemap was also used to develop a crater database complete for craters larger than ~100 m in diameter inside the map region, totaling more than 20,800 craters. Crater density estimates (KDE) for crater diameters >400 m and >300 m and absolute model ages for 10 identified geologic units were also derived from the crater database (Sizemore et al., 2022; Sizemore et al., 2024).

The processing steps used in constructing the mosaics are as follows:

Select all clear filter FC2 level 1b calibrated images from the Ceres X2 Elliptical (C2E) mission phase covering Urvara region.  This gives approximately 1600 images in total of varying resolutions from a few m/pixel to around 40 m/pixel.

Load images into ISIS (USGS Integrated Software for Imagers and Spectrometers 3, 2024, DOI: 10.5066/P13YBMZA; Laura et al., 2024) using the dawnfc2isis routine, then use spiceinit and footprintinit to initialize those files with SPICE kernels, a DEM (Roatsch et al. 2016), and outline data.  The planetary constants kernel dawn_ceres_SPG20160107.tpc (Roatsch et al. 2016) is used, as is required for this DEM.  All subsequent processing is done using  routines, unless otherwise noted.

The findimageoverlaps routine is run on the full set of images to identify overlapping areas (stored in file overlaps.lis), and overlapstats is run to generate a file (overlap.stats) from which lists of all images overlapping a given image can be extracted.

Control points linking the images are generated using the findfeatures routine, which uses feature-based matching algorithms rather than the traditional approach of applying a grid of control points. For each individual image, a list of images that overlap it is generated from overlap.stats.  The findfeatures routine is then used to identify control points among that subset of images and generate a control network file. 

The control network files for each image are combined using cnetcombinept.  The control network is manually inspected using the interactive ISIS qnet program.  A few additional control points are added for several images near the edge of the mosaic that have little overlap.  The number of control points per overlap region is reduced to at most a few dozen using the cnetthinner routine. The result is a single control network file for the complete set of images. The control network for the full image set at this point consists of approximately 29,000 control points.

Ground control points are added manually using qnet. Approximately 40 ground control points are assigned, either by adding a new control point or by using an existing control point, and tying that point to the 35 m/pixel reference LAMO basemap (Roatsch et al., 2017).

Once the network of control points is defined, the jigsaw routine is run to perform a least-squares bundle adjustment that reduces mismatches between images and refines the image geometry information.  From the ISIS jigsaw documentation, "The jigsaw application attempts to minimize the reprojective error within the control network. For each control measure and its control point in the control network, the reprojective error is the difference between the control measure's image coordinate (typically determined from image registration) and the image coordinate that the control point projects to through the camera model." 

The jigsaw routine is first run without updating the geometry, and the resulting output is used to identify control points with large reprojective error.  These errant points are edited or removed in qnet and jigsaw is run again. This is repeated several times until there are no control points with reprojective errors larger than ~3x the mean value.  Approximately 1,000 control points are rejected this way.  The jigsaw routine is then run one final time to update the geometry information in the individual images and generate the final control network file containing approximately 28,000 points.  The sigma0 value from this final jigsaw run is 0.29 pixels (at our 5 m/pixel scale).  From the ISIS jigsaw documentation, sigma0 "is a complex value that includes many technical details outside the scope of [the] documentation. In a brief, it is a measurement of the remaining error in a typical observation."  A sigma0 value of less than 1 pixel indicates no significant registration issues between our final mosaic and the LAMO basemap.

Prior to map projection and mosaicing, a simple photometric correction to adjust for varying illumination conditions is performed using the photomet routine and clear-filter parameters from Schroder et al. (2017; 2018). The file photopar_c.txt contains these parameters.  Images are then map-projected to polar stereographic using cam2map and mosaiced together with the automos routine.

The main mosaic includes all 1583 images, re-scaled to 5 m/pixel resolution. As the data set includes images with pixel scales from a few m/pixel to around 40 m/pixel, mosaics with subsets of the images at different ranges of resolution are also generated. Finally, the ISIS .cub files are converted
to 8 bit GeoTIFF format using the GDAL (Geospatial Data Abstraction Library) package, and these are the files included in this bundle.

urvara_c2e_controlled_v2_8bit_geo.tif - All 1583 images, re-scaled to 5 m/pixel
urvara_c2e_controlled_v2_3_5to7mpp_8bit_geo.tif - 1003 images with native resolution of 3.5-7 m/pixel, rescaled to 5 m/pixel
urvara_c2e_controlled_v2_7to14mpp_8bit_geo.tif - 324 mages with native resolution of 7-14 m/pixel, rescaled to 10 m/pixel
urvara_c2e_controlled_v2_14to28mpp_8bit_geo.tif - 190 mages with native resolution of 14-28 m/pixel, rescaled to 20 m/pixel
urvara_c2e_controlled_v2_gt28mpp_8bit_geo.tif - 66 mages with native resolution over 28 m/ pixel, rescaled to 35 m/pixel


Included Files
--------------

Image mosaics:

urvara_c2e_controlled_v2_8bit_geo.tif
urvara_c2e_controlled_v2_3_5to7mpp_8bit_geo.tif
urvara_c2e_controlled_v2_7to14mpp_8bit_geo.tif
urvara_c2e_controlled_v2_14to28mpp_8bit_geo.tif
urvara_c2e_controlled_v2_gt28mpp_8bit_geo.tif

Image lists. First images listed are lowest on the stack, and subsequent ones are overlaid on top of them:

urvaramos.txt
urvaramos_3_5to7mpp.txt
urvaramos_7to14mpp.txt
urvaramos_14to28mpp.txt
urvaramos_gt28mpp.txt

Photometric correction parameters:

photopar_c.txt



References
----------
Laura, J., Acosta, A., Addair, T., Adoram-Kershner, L., Alexander, J., Alexandrov, O., Alley, S., Anderson, D., Anderson, J., Anderson, J., Annex, A., Archinal, B., Austin, C., Backer, J., Barrett, J., Bauck, K., Bauers, J., Becker, K., Becker, T., � Young, A. (2024). Integrated Software for Imagers and Spectrometers (8.0.3_RC1). Zenodo.

Nass, A and S. van Gasselt (2023). A Cartographic Perspective on the Planetary Geologic Mapping Investigation of Ceres. Remote Sensing, 15, 4209.

Roatsch, T., E. Kersten, K.-D. Matz, F. Preusker, F. Scholten, S. Elgner, S. E. Schroeder, R. Jaumann, C. A. Raymond, C. T. Russell, DAWN FC2 DERIVED CERES HAMO DTM SPG V1.0, DAWN-A-FC2-5-CERESHAMODTMSPG-V1.0, NASA Planetary Data System, 2016.

Roatsch, Th., E. Kersten, K.-D. Matz, F. Preusker, F. Scholten, R. Jaumann, C. A. Raymond, C. T. Russell (2017). High-resolution Ceres Low Altitude Mapping Orbit Atlas derived from Dawn Framing Camera images, Planetary & Space Science, 140, 74-79.

Schroder, S. E., and 11 colleagues (2017). Resolved spectrophotometric properties of the Ceres surface from Dawn Framing Camera images. Icarus 288, 201.

Schroder, S. E., and 14 colleagues (2018). Ceres' opposition effect observed by the Dawn framing camera. Astronomy and Astrophysics 620, A201.

Sizemore, H. G., D. A. Crown, J. E. C. Scully, D. C. Berman, A. Neesemann, D. L. Buczkowski, D. P. O'Brien (2022). High-resolution geologic mapping of Urvara Crater: Preliminary maps & crater counts. 53rd Lunar and Planetary Science Conference, abstract 2338.

Sizemore, H. G., D. A. Crown, D. C. Berman, J. E. C. Scully, A. Neesemann, D. L. Buczkowski, D. P. O'Brien (2024). High-resolution geologic mapping of Urvara Crater, Ceres: Chronology & Feature Analysis. Planetary Geologic Mappers 2024 (LPI Contrib. No. 3044), abstract 7023.

Sizemore, H. G., D. A. Crown, D. C. Berman, J. E. C. Scully, A. Neesemann, D. L. Buczkowski, D. P. O'Brien (2025). High-resolution geologic mapping of Urvara Crater, Ceres: Chronostratigraphy, features, and unit correlations. 56th LPSC, abstract 2450.