Fully Controlled 6 Meters per Pixel Equatorial Mosaic of Mars From Mars Reconnaissance Orbiter Context Camera Images, Version 1. Issue 3 (15th March 2023)

• Record Type: Journal Article
• Authors: Robbins, Stuart J.; Kirchoff, Michelle R.; Hoover, Rachael H.
• Is Part Of: Earth and space science. Volume 10:Issue 3(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-03-15
• Subjects: Mars -- mosaic -- geology -- cartography; Space sciences -- Periodicals; Geophysics -- Periodicals; 500.5
• DOI: 10.1029/2022EA002443
• Languages: English
• ISSNs: 2333-5084

Abstract

Abstract: NASA's Mars Reconnaissance Orbiter ( MRO ) spacecraft has operated around Mars since March 2006. The Context Camera (CTX) aboard MRO has returned >125, 000 images of Mars, mostly at 5–6 m per pixel (mpp), providing ≈99% coverage with good‐quality images. Reconstruction of MRO 's orbit and camera pointing from ground tracking are offset from the Mars global coordinate system. This work focuses on correcting those data for the more than 50, 000 images in the equatorial region of Mars, between ±30° latitude, which comprises 50% of Mars' surface area. Determining and making these corrections in a relative control (image‐to‐image) and absolute or full control (image‐to‐ground reference) greatly improves the utility of the image data. Based on this work, the 95th percentile of tie points were offset from the Mars coordinate system by ≈155 m, corresponding to ≈28 CTX pixels. Controlling the more than 50, 000 images was accomplished through an efficient, automated approach with additional manual input and validation, described herein. The mosaic was generated with semi‐manual image order from these controlled data and is provided at both 6 mpp and 100 mpp. The mosaic is available to the community through NASA's Planetary Data System Imaging & Cartography Annex. Plain Language Summary: Creating a mosaic of a planetary surface from spacecraft images requires knowing a lot of geometry: Where the spacecraft was, how it was pointed, and how it was moving; and that same information is needed for the camera that took the images. This information is needed to properly place each pixel from each image at the correct location on a planet's surface. This information can be estimated based on tracking the spacecraft from Earth and knowing how it was commanded when images were taken. However, images projected onto the surface with that estimated information can have significant offsets relative to each other and relative to other data sets, limiting their usefulness. Software exists to match the same feature on multiple images and then to refine information about the spacecraft and camera geometry. We have improved our previous almost completely automated methods, and we have applied it to 50% of Mars' surface in a band around the equator. There are over 50, 000 images included in the mosaic, and the mosaic itself is 2.11 trillion pixels, rendered at 6 m per pixel. The average uncertainty of pixel location is now ≈5 m, about 100 times better than when based on Earth tracking. Key Points: We created a 6 mpp fully controlled Context Camera mosaic of Mars' equatorial region, within ±30° latitude (50% Mars' surface with 98.9% coverage) 69, 772 usable images were controlled via 2, 422, 635 tie points with 11, 906, 171 measures (1, 717 points constrained to THEMIS) Image data now have uncertainties of ≈5 m, a roughly 100× improvement over NASA‐supplied pointing data