Chang'e MRM Calibrated Data (CALCE1, CALCE2)

CALCE1 – Chang'e-1 Microwave Radiometer Calibrated Data

CALCE2 – Chang'e-2 Microwave Radiometer Calibrated Data

Data Notice
The Chang'e-1 and Chang'e-2 Microwave Radiometer (MRM) calibrated data exhibit known calibration issues, including significant brightness temperature offsets between orbits (and sometimes within a single orbit) and elevated noise levels, particularly in CE-1 data. Users should apply appropriate calibration or modeling corrections before scientific analysis. Combining these data with complementary instruments such as Diviner, Mini-RF, or LROC is also recommended. For full details, see the Bundle README and Primary Bundle Documentation.

Instrument: Microwave Radiometer (MRM)

PDS4 Bundle:urn:nasa:pds:chang_e_microwave_processedDOI: 10.17189/9nfs-ed98

For more information about MRM calibrated products, see the Primary Documentation for chang_e_microwave_processed.

The CE-1 MRM L2C data set contains samples taken between 2007-11-27 and 2009-01-24. Most of these samples exhibit a regular observational cadence: 5 samples separated by 1.6-second intervals followed by 1 at a 3.6-second interval[#1]. The record has two gaps of ~3 months, several other gaps of 1-5 days, and 92 gaps of ~15 minutes. The data set contains ~6.3 million samples in total.

The CE-2 MRM L2C data set contains samples taken between 2010-10-15 and 2011-05-20. They exhibit the same observational cadence as the CE-1 data. The record has several time gaps ranging in size from ~15 minutes to ~7 days. There is also a ~2-week interval during a major orbital pattern change in February 2011 during which the mission flagged all data as off-nominal. (The calibration for the mission-flagged data appears to us as possibly compromised, perhaps along with geopositioning.) The data set contains ~8.7 million samples in total, and ~7.5 million excluding the 2-week period of mission-flagged data.

Both MRM data sets include samples of the entire lunar surface, and each MRM sampled most regions multiple times in adjacent orbital passes and in separate sets of passes at other times of day. Note, however, that because CE-1 held a roughly noon-midnight orbit throughout the observational phase of the mission, the CE-1 MRM data include few samples taken during the lunar “morning” or “evening,” particularly in lower latitudes. Conversely, because CE-2 changed its orbital pattern more dramatically over the course of its mission, moving from a noon-midnight orbit to a terminator orbit back to a noon-midnight orbit, the CE-2 MRM data include samples taken across the entire diurnal cycle (although sample density is greatest near midnight and noon).

MRM calibration relied on constants determined during ground testing along with regular warm and cold calibration measurements taken in flight (see Wang et al., 2010b for details). The CE-2 calibration system was identical in basic architecture, but ground calibration was significantly more rigorous and the space-pointing cold horn was tilted up approximately 15 degrees to address concerns about possible contamination (Feng et al., 2013; also compare Wang et al., 2010a, 2010b).

Both the CE-1 and CE-2 MRM data exhibit significant offsets in temperature between orbits (even adjacent orbits), and sometimes between groups of samples within the same orbit. The data are also quite “noisy”, especially the CE-1 data. The "noise" is interpreted as sample-to-sample variations, a number of obviously “bad” orbits, and many outlier points within otherwise “good” orbits. Even with best-effort data cleaning, many uncertainties remain, and it is recommended treating relative temperature values as more reliable in general than absolute values.

There are two types calibration data in this Bundle: calibrated L2C source data products and two concatenated table products.

All map products in this bundle’s data collection are based on the contents of the NAOC-held datasets `CE1_MRM_2C_20071127161816_20090114192053_B` and `CE2_MRM_2C_20101015085002_20110520125150_A. These datasets present the MRM L2C data as fixed-width ASCII tables, one table per orbit in which the MRM took observational data, each table in a separate file. These files have attached labels written in valid Parameter Value Language (PVL). Considered as data products, they are generally compliant with version 3.8 of the PDS Standards. The CE-1 dataset contains 1691 such files; the CE-2 dataset contains 2402. The Chang’e Project’s processing level definitions are similar to NASA EOSDIS definitions. Like EOSDIS L2 products — and most other L2 products across the Chang’e Project data corpus — the MRM L2C products consist of georeferenced time-series samples in physical units. PDS4-labeled versions of these source products have been included in this bundle’s data_source collection to help ensure the reproducibility of the research. Only a handful of “empty” products from the CE-1 dataset that contain a PVL label but no table have been omitted. All data files in the data_source collection are byte-level equivalent copies of the L2C files held by the NAOC. Their original filenames, including the unconventional “.2C” extension have been retained. However, they have been organized into separate volumes by month and year to make navigation easier.

Each concatenated table product consists of a PDS4-labeled FITS file. The TABLE HDU of each FITS file contains a binary table of annotated MRM data. Each row of these tables corresponds to one row from an L2C ASCII table. Equivalently, each row corresponds to a single MRM sample. The CE-2 table has ~8.7 million rows; the CE-1 table has ~6.3 million. They exclude only a handful of rows from the source data tables that contain special constants or out-of-bounds values (about 50 in total).

Note that these tables discard the solar azimuth and incidence fields from the source tables due to the presence of some implausible values. They replace them with SPICE-computed solar position vectors and local true solar time (LTST) at the boresight intercept point. They also do not directly copy the quality flag columns, as there is no public documentation on the meaning of the different flags; they instead propagate the presence of any nonzero quality flag value as an element of the FLAG column bitmask.

In addition to LTST and solar position vectors, the tables contain several other quantities that are not directly copied from the source files: surface normal vectors at boresight intercept point, time expressed as seconds since J2000 (for compatibility with SPICE), and a quality flag bitmask.

MRM calibrated products have the following file names:

Calibrated source data products

[orbiter]_BMYK_MRM-L_SCI_P_[start time]_[stop time]_[orbit number]_[letter].2C (with detached PDS label)

where:

orbiter = “CE1” or “CE2”

“start time” and “stop time”: provide nominal time bounds for the product in UTC time scale, expressed as YYYYMMDDHHMMSS. For instance, “20071127182559” is equivalent to “2007-11-27T18:25:59Z”. Note that these are not fixed to the first and last timestamps in the data tables (although they are usually within a few seconds), and should likely be understood as time bounds for the mission-designated orbit.

orbit number = the mission-designated orbit number as a 0-padded 4-digit number

letter” = “B” for CE-1 products and “A” for CE-2 products.

 

Two concatenated Table file names

ce1_mrm.fits/xml, which contains data from the CE-1 MRM, and

ce2_mrm.fits/xml, which contains data from the CE-2 MRM.

 

In ODE, MRM calibrated products have the following product IDs:

ce1_bmyk_mrm-l_sci_p_[start time]_[stop time]_[orbit number]_b

ce2_bmyk_mrm-l_sci_p_[start time]_[stop time]_[orbit number]_a

ce1_mrm_table

ce2_mrm_table

 

Reference and Footnotes

  • [#1]: The MRM’s observational cadence is clearly visible in some of the deconvolved maps, particularly in the CE-2 data at lower latitudes. Because the spatial offset “jumps” every sixth sample, the measured temperature gradient also “jumps” every sixth sample, creating evenly-spaced “blocks” along each orbital track.

  • Feng et al., 2013. Data processing and result analysis of CE-2 MRM. Earth SciJ China Univ Geosci 38, 898–906.

  • Wang, Z., Li, Y., Zhang, D., Jingshan, J., 2010a. Prelaunch Calibration of Chang’E-2 Lunar Microwave Radiometer, in: 2010 International Conference on Microwave and Millimeter Wave Technology. pp. 1551–1554.

  • Wang, Z., Li, Y., Zhang, X., JingShan, J., Xu, C., Zhang, D., Zhang, W., 2010b. Calibration and brightness temperature algorithm of CE-1 Lunar Microwave Sounder (CELMS). Sci. China Earth Sci. 53, 1392–1406. https://doi.org/10.1007/s11430-010-4008-x