PDS SBN review of the NH/KEM and R&A data by NAIF ================================================= By Marc Costa Sitja, Jan. 18, 2022. Based on the data provided in: https://sbnreview.astro.umd.edu/202201_NH_KEM/ on the hnh-a-lorri_mvic-5-geophys-v1.0 data set. Opening Remarks ------------------------- Only the data and documentation in accordance to NAIF's area of expertise have been reviewed. This mainly excludes the data/temperature and data/albedo collections. Items that are considered liens, and therefore required to be addressed to pass the review from NAIF side are marked with [LIEN], the other items are --strong-- suggestions. Documents ------------------------- 01-aareadme.txt Directory structure not outlined appropriately. Suggested updated file provided, aareadme_mcosta.txt 02-catalog/dataset.cat Minor typo in line 44: 1962 to 1,962. Suggested updated file provided, dataset_mcosta.cat 03-catalog/mvic.cat Minor update of a missing tab/spaces in lines 293 and 295. Suggested updated file provided, mvic_mcosta.cat data/shape_models ------------------------- The DSK shape models have been loaded and quickly assessed with SPICE Enhanced Cosmographia, the OBJ models have been loaded and quickly assessed with SBMT. In addition the DSK files have been checked using DSKBRIEF, converted to OBJ files with DSKEXP and the OBJ files have been converted to DSK files. Everything has been cross-checked, all the results match. 04-mu69_fr2kf_lopoly.obj The format of this OBJ file diverges from the other OBJs. Although this is documented appropriately in the label, it might lead to confusion. In fact the format provided for this file, without faucet normal vectors is adequate for MKDSK to be used, otherwise the input needs to be processed. 05-asp_ca04_06_icp_smooth_spi.bds [LIEN] This model is rotated -90 deg WRT the +X axis. The model should be aligned appropriately as the other models. See: cosmo_asp_ca04_06_icp_smooth_spi.png 06-Reference frame [LIEN] We cannot find a MU69 coordinate system document that the project agreed upon with SBN. If yes, it would be good to look at it in the context of this review. If not, are there any plans to produce such document, and most important of all, will it define the frame consistent with frames used in the shape model data provided in this data set? The DSKs have been generated using the reference frame 'MU69_FIXED'. This reference frame is not defined anywhere in the current NH SPICE data set. Our understanding is that this frame looks like something like: \begindata FRAME_MU69_FIXED = 2486958 FRAME_2486958_NAME = 'MU69_FIXED' FRAME_2486958_CLASS = 2 FRAME_2486958_CLASS_ID = 2486958 FRAME_2486958_CENTER = 2486958 OBJECT_2486958_FRAME = 'MU69_FIXED' \begintext Note that from the SPICE Toolkit N67 onwards (released on Jan 3 2022), the IAU_ARROKOTH frame is provided with the toolkit and its orientation data is provided via PCK from nh_stars_kbo_centaur_v003.tpc provided in the NH SPICE data set. This is further discussed below. 07-Missing DSK Surface names [LIEN] The Arrokoth shape models are stored in the DSKs as surfaces"; as any other SPICE "body", surfaces have a name and an ID. These surface names must be defined during the generation of the DSK and by an auxiliary Frames Kernel (FK) that defines the surfaces. The surfaces can be added in the MKSDSK setup file as indicated in the MKDSK User Guide: NAIF_SURFACE_NAME += 'surface name' NAIF_SURFACE_CODE += integer surface ID code NAIF_SURFACE_BODY += integer body ID code These assignments are needed if the surface name-ID association is not defined in a kernel listed in a KERNELS_TO_LOAD assignment. The same text can be included in the auxiliary FK. This latter solution is preferred. The OSIRSI-ReX Frames Kernel defining the Surface IDs can be used as a template: https://naif.jpl.nasa.gov/pub/naif/pds/pds4/orex/orex_spice/spice_kernels/fk/orx_shape_v03.tf Surfaces names are used by SPICE APIs to use the shape models as inputs. This auxiliary FK (e.g.: nh_shape_v01.tf) should be present in the data set and must be present in the SPICE data set. We recommend that the names of the surfaces are provided by this data set as defined by their authors. Each shape model provided by each DSK consists of a single surface; each surface should have a unique NAIF_SURFACE_NAME. Using the DSK filenam as the surface name without the *.bds extension and using capital letters is good enough. 08-Missing orientation information [LIEN] In order to use the shape models, especially the SPICE DSK files, the orientation model of MU69 is required. In addition DSK files usually require extra files to be used. Within this data set the rotation model of MU69 is only present in the refernced article [SPENCERETAL2020]: "Arrokoth’s rotational period is 15.92 ± 0.02 hours, with its rotational pole pointing to right ascension = 317.5 +- 1deg, declination = −24.9 +- 1deg, J2000 equinox" This article is not easy to access (requires at least a user at NATURE.) The current orientation information available from the SPICE data set https://naif.jpl.nasa.gov/pub/naif/pds/data/nh-j_p_ss-spice-6-v1.0/nhsp_1000/data/pck/nh_stars_kbo_centaur_v003.tpc is: ASTEROID 486958 (2014 MU69); 2486958 \begindata NAIF_BODY_NAME += ( 'ASTEROID 486958 (2014 MU69)' ) NAIF_BODY_CODE += ( 2486958 ) NH_TARGET_BODIES += ( 2486958 ) BODY2486958_POLE_RA = ( 317.4880752 0. 0. ) BODY2486958_POLE_DEC = ( -24.8876496 0. 0. ) BODY2486958_PM = ( 270. 0. 0. ) BODY2486958_RADII = ( 100. 100. 100. ) \begintext The file is: This model does not include the 15.92 hours rotation specified in [SPENCERETAL2020], to include it the BODY2486958_PM value should be corrected as in: BODY2486958_PM = ( 270, 542.71, 0 ) The RADII should be corrected as well to fit the ellipsoidal approximation of MU69. But this file is not referenced in this data set. The recommendation is to either explicitly cite the SPICE kernel data set and indicate that the information will be available there, to include the updated/corrected PCK in the document collection in a similar manner to the included IKs: document/nh_ralph_v100_ti.txt document/nh_lorri_v201_ti.txt or to include this information elsewhere within the data set. 09-File names * It is suggested to drop the *_spice, *_spi suffix. It is understood that this is to differentiate the DSK with the OBJ file, but since the DSK will also be present in the SPICE data set maybe a differentiating suffix could be added to the OBJ files instead. Also for the DSKs we recommend to version the file, to account for file updates and in accordance to the NAIF style for SPICE kernels names. The suggestion would be: asp_ca04_06_icp_smooth_v01.bds mu69_fr2kf_hipoly_v01.bds mu69_fr2kf_lopoly_v01.bds mu69_merged_v01.bds * It is also suggested to include mu69 in asp_ca04_06_icp_smooth_spi.bds for example: mu69_asp_ca04_06_icp_v01.bds 10-DSK comment area [LIEN] DSK binary files have a comment area accessible using NAIF's utility COMMNT. The comment area contains important information to understand the DSK file. We find that comment are included in the DSKs requires more work. Typically comment areas for DSKs have different sections: 1-Abstract 2-Objects 3-Pedigree 4-Related SPICE kernels 5-References 6-Contact Information 7-DSKBRIEF Summary 8-MKDSK Setup File and Run-time Traceback The current comments only contain section 8. An example of a complete DSK comment is attached: bennu_g_00880mm_alt_obj_0000n00000_v020.cmt The comment area should be extended to include all the sections mentioned above following the Bennu OSIRIS-ReX example. All the information required to fill the comment should be present elsewhere in the data set. We would like to emphasize that is very important to include in the comment area a complete list of all the kernels needed to use the shape model. This goes in section 4. A better example than the one for OSIRIS-ReX is provided hereunder: " Related SPICE kernels ------------------------------------------------------------------------ This file should be used together with the kernel set described below (*): Kernel Comments ------------------------------- ------------------------------------- nh_shape_v01.tf is a mission specific Frame Kernel that complements the mission's Frame Kernel to complete the DSK Surface ID Codes for the mission. nh_stars_kbo_centaur_v003.tpc is the Planetary Constants Kernel that provides the cartographic, physical constants data, and Name/ID mapping for bodies not covered by the project kernels (including Arrokoth.) nh_recon_arrokoth_od147_v01.bsp Contains ephemerides for the NH Spacecraft, the Pluto system bodies, and solar system bodies (including Arrokoth.) naif0012.tls is a Leapseconds Kernel file used in conversions between ephemeris time and UTC (coordinated universal time.) (*) or any higher version of the same kernels " Please note that the FK nh_shape_v01.tf does not yet exist so its generation should be considered, it is recommended to use the OSIRIS-ReX one as a template: https://naif.jpl.nasa.gov/pub/naif/pds/pds4/orex/orex_spice/spice_kernels/fk/orx_shape_v03.tf Note that as specified before, nh_stars_kbo_centaur_v003.tpc should be updated. In addition, either the 'MU69_FIXED' frame should be defined or the DSKs should be re-generated using 'IAU_ARROKOTH'. Since these prior items will have to be included in the SPICE data set please coordinate with Brian Enke (SWI). 11-PDS labels * The DESCRIPTION field of the label is too poor, it is recommended to use the description provided in the catalog/dataset.cat following the style of the OSIRIS-ReX DSKs the description of which is as follows: SPICE DSK file containing shape model data for the surface of asteroid (101955) Bennu, with Global coverage at .880 meter resolution, ALT-based, version 020, created by the ORX Altimetry Working Group (AltWG). The original name of this file was g_00880mm_alt_obj_0000n00000_v020.bds. From https://naif.jpl.nasa.gov/pub/naif/pds/pds4/orex/orex_spice/spice_kernels/dsk/bennu_g_00880mm_alt_obj_0000n00000_v020.xml * Note that the DSK labels that will be included in the SPICE data set will be different. This should be OK. End of document.