ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume V-3-2020
https://doi.org/10.5194/isprs-annals-V-3-2020-579-2020
https://doi.org/10.5194/isprs-annals-V-3-2020-579-2020
03 Aug 2020
 | 03 Aug 2020

A GLOBAL SHAPE MODEL FOR SATURN’S MOON ENCELADUS FROM A DENSE PHOTOGRAMMETRIC CONTROL NETWORK

M. T. Bland, L. A. Weller, D. P. Mayer, and B. A. Archinal

Keywords: Enceladus, Shape Model, Topography, Photogrammetry

Abstract. A planetary body’s global shape provides both insight into its geologic evolution, and a key element of any Planetary Spatial Data Infrastructure (PSDI). NASA’s Cassini mission to Saturn acquired more than 600 moderate- to high-resolution images (<500 m/pixel) of the small, geologically active moon Enceladus. The moon’s internal global ocean and intriguing geology mark it as a candidate for future exploration and motivates the development of a PSDI. Recently, two PSDI foundational data sets were created: geodetic control and orthoimages. To provide the third foundational data set, we generate a new shape model for Enceladus from Cassini images and a dense photogrammetric control network (nearly 1 million tie points) using the U.S. Geological Survey’s Integrated Software for Imagers and Spectrometers (ISIS) and the Ames Stereo Pipeline (ASP). The new shape model is near-global in extent and gridded to 2.2 km/pixel, ∼50 times better resolution than previous global models. Our calculated triaxial shape, rotation rate, and pole orientation for Enceladus is consistent with current International Astronomical Union (IAU) values to within the error; however, we determined a new prime meridian offset (Wo) of 7.063°. We calculate Enceladus’ long-wavelength topography by subtracting the best-fit triaxial ellipsoid from our shape model. The result is comparable to previous global models but can resolve topographic features as small as 5–7 km across in certain areas. To evaluate the spatially varying quality of the model, we calculate the point density (variable from 5 to more than 50 points per pixel), normalized median absolute deviation of the points within each pixel (typically less than 100 m), and the minimum expected vertical precision of each point (ranging from 29 m to 2 km).