LIME ASD spectral lunar reflectance dataset

Résumé

In the frame of the Lunar Irradiance Model of ESA (LIME - https://calvalportal.ceos.org/lime) project, hyperspectral irradiance measurements of the moon were taken with an ASD Field Spec 4 spectroradiometer from the Izaña Observatory (Spain) between April 2022 and September 2022. These measurements were processed into a spectral reference dataset of lunar reflectance for use in spectral interpolation within the LIME toolbox (LIME-TBX). The processed ASD dataset used within the LIME-TBX is provided here publicly. We note that this dataset has poor absolute calibration (systematic uncertainties >5%) but reasonably good relative calibration (uncertainties included in provided file).  We note that within the LIME-TBX, these ASD spectral reflectances are scaled to go through the LIME model reflectances (provided at CIMEL wavelengths), which have a much better absolute calibration. The ASD FieldSpec Pro 4 is a spectroradiometer available on the market that meets the set of general requirements and was chosen for this project. The set of laboratory tests were performed to verify some of the instrument’s characteristics prior to its field installation. These laboratory tests focused on the most significant instrument characteristics for lunar measurements and included spectral response function (SRF) test and field of view (FOV) and its homogeneity test and signal to noise ratio measurement. The adaptation of the ASD spectroradiometer for semi-autonomous in situ use consisted of: setting the fore-optics (1 ° lens + scrambler) in the holder and optic fibre protective conduit  building instrument weatherproof enclosure with thermal stabilization   development of LabVIEW software to enable automounts acquisition through the night.  The field campaign in Izaña started in April 2022 and lasted until end of June 2022, was extended for another one month of measurements in September 2022. The ASD data were then propagated to top of atmosphere (TOA) using Langley plots, and then calibrated and converted to lunar reflectance using the TSIS-1 solar irradiance. Next, measurements that are too noisy (e.g. in absorption features) were omitted. Specifically, wavelengths between 300 nm – 400 nm, 680 nm – 690 nm, 713 nm - 740 nm, 757 nm – 769 nm, 809 nm – 840 nm, 890 nm – 1000 nm, 1090 nm – 1181 nm, 1307 nm – 1540nm, 1740 nm – 2080 nm and 2345 nm – 2500 nm are masked. These wavelengths are masked and replaced using spectral interpolation. The TOA ASD data are then divided in bins of 10° phase angles spanning from -95° to 95°. The mean spectrum in each bin was then interpolated linearly to a phase angle grid with steps of 1 degree.  These interpolated lunar reflectances provide the reference ASD dataset. Uncertainties are also propagated through the derivation (i.e. from measurement through calibration and binning) of the reference ASD reflectances using the CoMet toolkit, and are provided within the data files following the digital effects table format (www.comet-toolkit.org). The data are provided in (self-describing) NetCDF format.  More information is available at https://calvalportal.ceos.org/lime.