OMSO2e_README_V003 by cuiliqing


									OMSO2e README File v1.0.0 Released May 16, 2011                                           1

This document describes the Ozone Monitoring Instrument (OMI) OMSO2e product,
which is the daily cloud screened and quality assured Level 3 (L3e) gridded data product
that corresponds to the improved Level 2 Planetary Boundary Layer (PBL) OMSO2
product. The latter is the U.S. OMI Science Team's total column sulfur dioxide orbital
Level 2 (L2) swath data product processed with Band Residual Difference (BRD)
algorithm [Krotkov et al 2006, 2008] and corrected according to Lee at al. [2009]. The
physical quantity is the total column density of SO2 in Dobson Units (1DU=2.69 ∙1016
molecules/cm2) assuming GEOS-Chem monthly profile shape climatology for year 2006
[Lee et al 2009]. The "e" at the end of "OMSO2e" represents "expanded" product, which
is based on selection of a single "best" OMI pixel to represent each grid-cell and therefore
includes ancillary parameters about ozone, clouds and observational geometry. The
ancillary parameters can be used to further improve the product (e.g., changing a-priori
SO2 profile shape assumption).

OMI was launched on July 15, 2004 on the NASA EOS Aura satellite, which is in a sun-
synchronous ascending polar orbit with 1:45pm local equator crossing time. The science
quality data collection started on October 1, 2004. The level 3 data are processed through
November 30 2010.

The L2 SO2 file, called a data granule, covers the sunlit portion of the orbit with an
approximately 2600 km wide swath containing 60 binned ground pixels or scenes per
viewing line (i.e., 60 scene numbers). During normal operations, 14 or 15 granules are
produced daily, providing fully contiguous asynoptic coverage of the globe. The adopted
L3e grid is a 0.25-degree by 0.25-degree grid in longitude and latitude, consistent with
the definition of OMI Grids for Level 3 and Level 4 Data Products [Veefkind et al. 2005].
The dimensions of this grid are 1440 by 720. The grid cell at coordinates (1, 1) is
centered at (longitude = -179.875, latitude = -89.875), and the grid cell at coordinates
(1440, 720) is centered at (longitude = 179.875, latitude = 89.875).

The L2 data are not averaged or weighted in any way in the L3e product. Each grid cell
in the L3e product contains only one "best pixel", selected from all ”good” L2
observations that overlap with the L3e grid cell, which has the shortest path length [path
length = 1/cos(solar zenith angle) + 1/cos(viewing zenith angle)]. An L2 observation can
be mapped onto more than one L3e grid cell, if the L2 observation overlaps with and has
the shortest path length for more than one L3e grid cell. A "good" OMSO2 L2 scene is
defined as one that has
 1) a solar zenith angle that is less than or equal to 70.0 degrees, and
 2) an SO2 column amount in Planetary Boundary Layer (ColumnAmountSO2_PBL) that
is not equal to the missing value, and
3) scene number range from 3 to 56 (1-based, inclusive), and
4) row anomaly flag not set (i.e., bit 11 of the L2 PBL quality flag is not set), and
5) terrain height that is less than or equal to 3.5km, and
6) RadiativeCloudFraction that is less than or equal to 0.2,
OMSO2e README File v1.0.0 Released May 16, 2011                                           2

The product is stored as one daily HDF-EOS 5 grid file, and has a size of 4.5MB.
Currently, OMSO2e product is not produced when OMI goes into the “zoom mode” for
one day every 452 orbits (~32 days). Since June 2007 signal suppression (i.e., Row
Anomaly) has been observed in Level 1B Earth radiance data at all wavelengths for
particular viewing directions (see OMI instrumental effects). The data affected by row
anomaly are excluded from OMSO2e product based on OMTO3 flag.

Algorithm Description
The SO2 columns are produced using the Band Residual Difference (BRD) algorithm
[Krotkov et al 2006, 2008] and post corrected according to recommendations in Lee et al
[2009]. The BRD algorithm uses a modified version (Version 8.5) of TOMS total ozone
algorithm (OMTO3) [Bhartia and Wellemeyer 2002] as a linearization step to derive an
initial estimate of total ozone assuming zero SO2. (See OMTO3 README file for more
detail). The residuals at the 10 wavelengths are then calculated as the difference between
the measured and computed N-values (N=-100*log10(I/F), I is Earth radiance and F is
solar irradiance ) using a vector forward model radiative transfer code that accounts for
multiple Rayleigh scattering, ozone absorption, Ring effect, and surface reflectivity, but
assumes no aerosols and no SO2. Cloudy scenes are treated as mixture of two opaque
Lamberian surfaces, one at the terrain pressure and the other at Radiative Cloud Pressure
(RCP) derived using OMI-measured Rotational Raman scattering at around 350 nm (see
OMCLDRR README file for more detail). In the presence of SO2, the residuals contain
spectral structures that correlate with the SO2 absorption cross-section. The residuals also
have contributions from other errors sources that have not yet been identified. To reduce
this interference, a median residual for a sliding group of SO2-free and cloud-free scenes
(OMTO3 radiative cloud fraction < 0.15) covering 15o latitude along the orbit track is
subtracted for each spectral band and cross-track position [Yang et al 2007].

The BRD algorithm [Krotkov et al 2006] uses differential residuals at the three
wavelength pairs with the largest differential SO2 cross-sections to maximize sensitivity
to anthropogenic emissions in the PBL. Each pair residual is converted to SO2 slant
column density (SCD) in Dobson Units (1DU=2.69 ∙1016 molecules/cm2) using
differential SO2 cross-sections data at constant temperature (275 K) [Bogumil et al 2003].
The SCDs of the three pairs are averaged and the average SCD is used in the following
steps. After the filtering process described above, only one “best” pixel is selected
representing each L3e grid cell, which has the shortest path length [path length =
1/cos(solar zenith angle) + 1/cos(viewing zenith angle)]. The selected “best”
representative SCD values are averaged for all L3e grid cells in Pacific sector for each
latitudinal bin (excluding Hawaii and transient volcanic SO2 signals above 6DU). The
Pacific average SCDs are subtracted from all SCDs for each latitude (i.e., “Pacific Sector
correction” [Lee at al 2009]) and the corrected SCDs are reported as the
“SlantColumnAmountSO2” data fields.

The corrected SCDs are converted to the SO2 vertical column density (VCD) using clear
sky monthly mean local air-mass factors (AMF): VCD = SCD_corrected/AMF and
reported in the “ColumnAmountSO2_PBL” data field. The AMFs are pre-computed for
OMSO2e README File v1.0.0 Released May 16, 2011                                          3

cloud-free conditions, using OMI measured ozone, observational geometry and a-priori
aerosols and SO2 profile shapes from GEOS-CHEM model [Lee et al 2009]. Daily AMF
were averaged for each month in 2006 and interpolated into the L3e grid. They can be
reconstructed as a ratio of the reported data fields: SlantColumnAmountSO2/

Data Quality Assessment
The OMSO2e errors are best described as pseudo-random (i.e. having different
systematic and random components depending on spatial and temporal scales) Gaussian-
like distribution with a large-scale mean of zero. The “sliding median” empirical residual
correction essentially acts as a high-pass filter reducing cross-track and low frequency
latitudinal biases, but allowing high frequency (i.e. “scene by scene”)noise in the
residuals to propagate into retrieved background SO2 data. The noise standard deviation
(sigma) is ~1.5 DU in the tropics, but increases with latitude, viewing and solar zenith
angles and total ozone [Krotkov et al 2008]. Given this large noise only pollution from
strong anthropogenic sources of SO2 (such as smelters and coal burning power plants)
and from strong regional pollution can be detected in daily data [Carn et al 2007; Krotkov
et al 2008]. Averaging over a larger area or for a longer time reduces the noise but slower
than the square root of the number of scenes averaged. The sigma reduces to ~0.8 DU
when 4 grid cells are averaged and approaches ~0.4 DU with increasing number of
averaged scenes. The SO2 detection limit in daily data is roughly twice of the 1 sigma

The Pacific sector correction reduces latitudinal biases compared to L2 product. Local
AMF correction further reduces background SO2 noise over oceans, where average AMF
is significantly larger than assumed in L2 product. However, the sliding median
correction results in the negative biases south and north of the broad polluted regions or
regions with persistent volcanic degassing. Such regional negative biases are not removed
by applying Pacific Sector correction. The known examples are summer negative biases
over Siberia and Canada and positive biases over North Atlantic and around ice shelf in
Antarctica. Due to remaining biases it is recommended to apply local bias correction to
the data.

Product Description
The OMSO2e product is written as HDF-EOS5 daily file ~5Mb. Data files are available
from Goddard Earth sciences Data and Information Services Center (GES DISC) web
site. For a list of tools that read HDF-EOS5 data files, please visit this link:

The information provided on these files includes: latitude, longitude, solar zenith angle,
OMI view angles, OMTO3 total ozone, radiative cloud fraction and SO2 slant and
vertical columns in Dobson Units. For a complete list of the parameters, please read the
OMSO2e file specification
OMSO2e README File v1.0.0 Released May 16, 2011                                       4

For general assistance with data archive, please, contact GES DISC. For questions and
comments related to the OMSO2 algorithm and data quality please contact Nickolay
Krotkov (, who has the overall responsibility for this

Krotkov, N.A., S.A. Carn, A.J. Krueger, P.K. Bhartia, and K. Yang (2006). Band residual
difference algorithm for retrieval of SO2 from the Aura Ozone Monitoring Instrument
(OMI). IEEE Trans.Geosci. Remote Sensing, AURA special issue, 44(5), 1259-1266,
doi:10.1109/TGRS.2005.861932, 2006

Krotkov, N. A., B. McClure, R.R. Dickerson, S.A. Carn, C. Li, P.K. Bhartia, K. Yang,
A.J. Krueger, Z. Li, P. Levelt, H. Chen, P. Wang, and D.R. Lu (2008), Validation of SO2
retrievals from the Ozone Monitoring Instrument over NE China, J. Geophys. Res., 113,
D16S40, doi:10.1029/2007JD008818.

Lee, C., R. V. Martin, A. van Donkelaar, G. O'Byrne, N. Krotkov, A. Richter, L. G.
Huey, and J. S. Holloway (2009), Retrieval of vertical columns of sulfur dioxide from
SCIAMACHY and OMI: Air mass factor algorithm development, validation, and error
analysis, J. Geophys. Res., 114, D22303, doi:10.1029/2009JD012123.

Veefkind, J.P., Definition of OMI Grids for Level 3 and Level 4 Data Products
  (OMI-Grids_L3L4, SD-OMIE-KNMI-443, 25 January 2005)

To top