Access

The LATMOS servers host the different ACS datasets presented below. It is possible to access the ACS directory from these servers and download the data with the SFTP secure file transfert protocole. We advise you to use FileZilla (https://filezilla-project.org/ ) to do so. After launching FileZilla, you will have to write the server host name, the login, the password and the port number (22 for SFTP) which are presented in the figure on the right:

Clique to edit

MIR Level 1A - Clean Data

MIR Level 1A files (L1A)  contain spectra corrected from electronic and acquisition artifacts and  derived from MIR Level 0B files (L0B)

 

L0B are a set of timely-ordered of full (640 X 512 pixels) or partial frames of raw data in ADU (acquisition numerical unit).  Theses frames are the averaging of 200 ,96 or 32 individual acquisitions done during the 3 ms exposure time. For each frame, the echelle orders are dispersed along the x-axis while each order is finely dispersed along the y-axis. There are 10 positions covering different groups of diffraction orders thus different wavelength ranges. The figure on the right presents two example of frames for two different positions with a cut along the y-axis highlighting the  echelle orders (the bump on the blue profiles).

Two examples of frame for two different positions with a cut along the y-axis

Two examples of frame for two different positions with a cut along the y-axis

L1A files are structured like the L0B files containing a set of timely-ordered of (640 X 512 pixels) frames of clean data in ADU (acquisition numerical unit) including several electronic and acqusition corrections: the subtraction of the dark, the compensation of the pixel to pixel non uniformity response, the flagging of hot pixels/cosmic rays and the correction of accumulation errors.

 

The dark is estimated from frames recorded when the instrument is not pointing at the Sun. Then, it is subtracted from the other frames. The figure on the right presents, for a given row, the electronic dark in blue, the thermal  dark in red and the signal from the Sun + dark in black. The subtracted dark is the sum of the electronic and thermal dark.

Electronic (blue) and thermal (red) dark and signal + dark (black)

Electronic (blue) and thermal (red) dark and signal + dark (black)

Acquisition errors are first detected, then classified in different category and corrected if possible. For instance, erroneous frames, copied row with an unknown factor and «garbage» rows are irreversible errors. But Copied-added rows, «scaled down rows» (with a known factor) and the glitch + offset are reversible. The figure below is the temporal first derivative between two frames. The black straight lines on the right side are typical acquisition errors.

First temporal derivative (between frames) for detecting acquisition errors

First temporal derivative (between frames) for detecting acquisition errors

The background stray light and thermal noise correction, the image registration & rectification and the multiple images correction is implemented in the MIR Level 1B (L1B).

 

(Credit for the first figure: Korablev et al 2017)

(Credit for the other figures: presentation given by L. Baggio for the ACS SWT in Tarusa, 27/09/2021)

MIR Level 1B - Rectified data

MIR Level 1B are produced from the MIR 1A files. MIR 1B contain a set of  (640 X 512 pixels) frames of rectified data in ADU (acquisition numerical unit). The data from the 1A are orthorectified first with 5.1° rotation to separate the spatial component form the spectral component. Then, a stretching is applied to have the same size for the vertical slit in all orders and positions.

In consequence, the previous 640*512 detector frame is projected in a new and larger 760 x 846 pixels reference frame. Part of the new frame without data are padded with nan (not-a-number).

The two figures on the right are an example of rotation applied on the detector frame. On the right, the frame before the 5.1° rotation, on the left after. Spectral and spatial components are now quasi-decoupled. From top to bottom: position 7, 13 and 3.

Rotation of the detector frame for positions 07, 13, 03

Rotation of the detector frame for positions 07, 13, 03

The figure on the right present two frames of two different positions (4,13) stretched to have solar absorption lines at the same vertical size ie to have the same size for the vertical slit which is fixed at 34.3 rows

Stretched position 04 and 13

Stretched position 04 and 13

The backgrund signal -the sum of the thermal radiation from the instrument and the stray light from the sun- is estimated from technical observations of the sun or from the observations themselves for certain positions and then is removed.

As the pixels in each frame are now registered along the y-axis in a spectral frame, the wavelenght calibration is easy to perform. The SOLAR-HRS high resolution disk-integrated solar spectrum (https://doi.org/10.3390/rs15143560) is used as a reference.

 

(Credit for the figures: presentation given by L. Baggio for the ACS SWT in Paris, 18/03/2024 )

MIR Level 2 - Transmissions

MIR Level 2 are produced from the MIR 1B files. MIR Level 2 contain a set of (760 X 846 pixels) frames of transmissions.

MIR 2 transmission are computed by dividing the absorption spectra by the spectrum observed beyond the martian atmosphere, i.e. the solar spectrum seen by ACS-MIR after additionnal corections including the removal of interference patterns.