OMI Solar Spectral Irradiance Data
                        ##################################

Authors:  Sergey Marchenko and Matthew DeLand

Contact:  sergey.marchenko@ssaihq.com

Last revision:  28 April 2025


Data Product
************

This file gives a basic description of the solar spectral irradiance data that 
we have produced from Aura OMI solar measurements.  Detailed discussion of the 
procedures used and validation steps can be found in the following papers:

S. V. Marchenko and M. T. DeLand (2014), "Solar spectral irradiance changes 
  during Cycle 24", Ap. J. 789:117, https://doi.org/10.1088/0004-637X/789/2/117.

S. V. Marchenko, M. T. DeLand, and J. L. Lean (2016), "Solar spectral irradiance 
  variability in Cycle 24:  Observations and models", J. Space Weather Space 
  Climate, 6, A40, https://doi.org/10.1051/swsc/2016036.

S. V. Marchenko, T. N. Woods, M. T. DeLand, S. Mauceri, P. Pilewskie, and M. 
  Haberreiter (2019), Improved Aura/OMI solar spectral irradiances: Comparisons
  with independent data sets and model predictions, Earth and Space Sci., 6, 
  2379-2396, https://doi.org/10.1029/2019EA000624.

We designate this dataset as Version 7 ('_v7') to recognize three changes from 
our previous Version 6 product:

   1. The analysis of the "common degradation" correction term, determined from
      the ratio of quartz volume diffuser and aluminum diffuser measurements, 
      uses a revised exposure time-dependent function.
   2. The Aura satellite drifted to beta angle values outside the normal 
      goniometric calibration range during October-December 2023 and September-
      December 2024. Empirical irradiance corrections were developed using 
      data from selected CCD rows that showed minimal changes during these 
      periods.
   3. The magnitude of the irradiance values at 279.5 nm and 280.5 nm, at the
      core of the Mg II line, is slightly supressed in late 2024 and early 
      2025. We believe that this is due to a non-linearity issue, and plan to 
      address it in our next data release.

Other changes from the Version 3 OMI SSI product described in our 2019 paper 
were incorporated previously, and include the following items:

   a. The instrument degradation model is based on data from all three OMI 
      channels (UV1, UV2, VIS) for better spectral consistency.
   b. The averaging period used to define our solar minimum reference spectrum 
      was shifted to cover 1 July 2008 - 30 April 2009. This provides 
      symmetric coverage around the sunspot minimum at ~2008.9 shown in NOAA 
      NGDC (National Geophysical Data Center) data.
   c. The filtering process to remove instrument noise was revised to improve
      flexibility.
   d. The SSI data are provided on a 0.5 nm wavelength grid for consistency 
      with the OMI spectral resolution (~0.4-0.6 nm).

More details will be given in an upcoming journal manuscript. The Version 7 OMI
SSI product is based on Collection 4 OMI L1B data, as described by Kleipool et 
al. [2022].

Kleipool, Q., et al. (2022). Ozone Monitoring Instrument (OMI) collection 4: 
  establishing a 17-year-long series of detrended level-1b data. Atmos. Meas. 
  Tech., 15, 3527-3553, https://doi.org/10.5194/amt-15-3527-2022.


Data Location
*************

The OMI data can be retrieved at the following web site:

   https://sbuv.gsfc.nasa.gov/solar/omi/


File Contents
*************

The OMI irradiance data, corrected for long-term degradation, are provided in a 
single IDL save set named 'OMI_SSI_coll4_v7_apr2025.sav'. All data are 
provided separately for each of three spectral channels, where the term 'CHN' 
in the variable name has the following definitions:

   'uv1'  =  UV1 channel, wavelength range = 265.0-309.5 nm
   'uv2'  =  UV2 channel, wavelength range = 310.0-364.5 nm
   'vis'  =  VIS channel, wavelength range = 365.0-500.0 nm

The irradiance data are provided on an 0.5 nm wavelength grid for each 
channel, i.e. 265.0, 265.5,...309.5 nm for UV1. A fill value of 0.0 is used to 
indicate that no valid irradiance data are available for a particular 
wavelength and date.

The first measurement date available in the OMI data product is 2 July 2006, and 
the last date included in the current data file is 3 March 2025. OMI SSI 
measurements are normally taken every 15 orbits, so the typical interval between
spectra is slightly longer than one day. The save set contains the following 
variables, where 'CHN' is defined as above for each channel:

   JUL_DATE          Observation date in adjusted Julian day format 
                       [JD-2450000.0]
   chn_REFERENCE     Reference irradiance spectrum created by averaging data 
                       between July 2008 and April 2009 [photons/cm^2/sec/nm]
   chn_BIN           Daily spectra, expressed as normalized irradiance relative 
                       to the reference spectrum irradiance value at each 
                       wavelength
   chn_BIN_STD       Normalized standard deviation within each spectral and 
                       daily bin
   LA_chn            Wavelength scale [nm], sample interval = 0.5 nm

======
NOTE: The file 'omi_solar_ref_jul2008_apr2009_coll4_v6_jan2025.sav' should be
read in AFTER opening the irradiance save set in order to create the 
"chn_REFERENCE" variables with corrected irradiance values.
======

For users who prefer to use data in HDF format, the file 
'omi_ssi_update_v7_20250428.h5' is also available.  It contains the same 
information as the IDL save set, with the following mapping of variables:

   JUL_DATE       -->  JulianDateAdj
   chn_REFERENCE  -->  IrradianceReference{chn}
   chn_BIN        -->  IrradianceNormalized{chn}
   chn_BIN_STD    -->  IrradianceStDev{chn}
   LA_chn         -->  Wavelength{chn}