Ozone Monitoring Instrument (OMI)

Here, below: Introduction, Mission objectives, Heritage, Instrument description, Atmospheric composition products, List of OMI data productsMore information?. This WebPage is mostly  based on Levelt et al., (2016) & Levelt et al., (2017) and other additional contributions.


The Ozone Monitoring Instrument (OMI) flies on the National Aeronautics and Space Adminsitration’ (NASA)’s Earth observing system Aura satellite launched on 15 July 2004 . The Aura satellite is focussed on observing atmospheric chemistry, in order to address the following major environmental questions:

  1. Is the ozone layer recovering as expected?
  2. What are the sources of tropospheric pollutants, their chemical transformation and their transport?
  3. How is Earth’s climate changing?

Mission objectives

At the start of the OMI project, the scientific objectives is directly related to these issues and can be summarized by 4 science questions to which the associated measurements largely contribute:

  • Is the O3 – Ozone layer recovering as expected ?
  • What are the sources of aerosols and trace gases that affect global air quality and how are they transported?
  • What are the roles of tropospheric O3 – Ozone and aerosols in climate change ?
  • What are the causes of surface UV-B change?

OMI allows the mapping of air pollution from an urban to super-regional scale.

A multi-year average tropospheric NO2 map from OMI satellite measurements: blue = low values, red = high values. WARNING:although Europe and China display a similar red colour, the values are actually not equal. The values beyond a threshold are in fact all saturated with dark red. NO2 pollution amount in China is actually larger than in Europe (Source: the Royal Netherlands Meteorological Institute – KNMI, OMI PI: Prof. Dr. Pieternel F. Levelt, extracted from Pr. Dr. P.F. Levelt Noble Lecture Series, September 28 – October 2, 2015, University of Toronto, Canada). See NO2 – Nitrogen dioxide WebPage here.


The heritage of OMI are:  NASA TOMS instrument (Heath et al., 1975; McPeters et al., 1998) that had only 8 wavelength bands, a ground-pixel size of 50 km × 50 km, and a daily global coverage; NASA Solar Backscatter Ultraviolet (SBUV) instrument (Cebula et al., 1988); and the European ESA instruments GOME (Burrows et al., 1999) and SCIAMACHY (Bovensmann et al., 1999; Noel et al., 2003), which introduced the concept of measuring the complete spectrum in the ultraviolet/visible/near-infrared wavelength range with a high spectral resolution. This enables to retrieve several trace gases from the same spectral measurement.

Instrument description

OMI is a nadir-looking push broom solar backscatter grating spectrometer that measures the Earth’s atmosphere and surface radiance spectrum over the entire wavelength range from 270 to 500 nm with a spectral resolution of about 0.5 nm. OMI has a wide swath of 2600 km on the surface, which enables measurements with a daily global coverage. The light entering the telescope is depolarised using a scrambler and then split into two channels: the ultraviolet (UV) channel (wavelength range 270 – 380 nm) and the visible  (VIS)channel (wavelength range 350 – 500 nm). The polarization scrambler is applied in the OMI telescope before entering the polarization sensitive spectrograph; this makes the instrument almost insensitive to the polarization of the incoming light.

In the normal global operation mode, the OMI pixel size is 13 km× 24 km at nadir (along x across track). In the zoom mode, the spatial resolution can be reduced to 13 km × 12 km. The high spatial resolution of OMI was one of the key technical achievements that enabled significant advances in air quality research and emission monitoring from space and what motivates future air quality missions like TROPOMI, on-board Sentinel-5 Precursor, to strive for even higher spatial resolution.

OMI combines the advantages of GOME, SCIAMACHY, and TOMS, measuring the complete spectrum in the UV-VIS wavelength range with a very high spatial resolution and a daily global coverage. This is possible by using a two-dimensional (2D) detector where on one axis of the detector the across-track ground pixels are imaged and on the other axis the spectral information is recorded. This sensing technique allows the simultaneous measurement of all the ground pixels in the swath; therefore, OMI doesn’t have a scan mirror. Another special feature of the OMI instrument is the type of diffuser used to observe the Sun. Such diffusers are required to reduce the intensity of the solar radiance. Because of its superior spectral behavior, the daily solar observations of OMI use the quatz volume diffuser (QVD) that has proven to be very stable. Therefore, TROPOMI only uses QVD solar diffusers.

OMI measurement principle (Levelt et al., 2006)

The one major anomaly of OMI is the so-called row-anomaly (Schenkeveld et al., 2017). A row anomaly is an anomaly  that affects the quality of the radiance data at all wavelengths for a particular viewing direction of OMI. This corresponds to a row on the 2D detectors, and hence the term ‘row anomaly’. The cause for the row anomaly is outside of the instrument; it is most likely caused by damage to the insolation blankets in which OMI is covered, blocking part of the field of view. Although early signs are observed starting in 2007, the main row-anomaly started in 2009. For TROPOMI, the lesson learned was to put an additional aluminum plate over the insolation blankets at the location where the field-of-view is close to the housing of the instrument.

OMI is a contribution from the Netherlands and Finland to NASA Aura. It was built by Dutch Space and TNO Science & Industry (formerly TNO-TPD), in co-operation with Finnish subcontractors VTT and Patria Finavitec. The instrument was financed by the Netherlands Agency for Aerospace Programmes (NIVR) and the Finnish Meteorological Institute (FMI). The Royal Netherlands Meteorological Institute (KNMI), via Prof. Dr. Pieternel Levelt, is the PI of OMI.

Atmospheric composition products

OMI data products include standard, near real-time (NRT) and very fast delivery (VFD). The standard products are available within two days after measurement. OMI also provides global NRT data for selected products that are available within three hours after measurement. The VFD products are available for a limited region1 covering most of Europe twenty minutes after measurement.

The O3 – Ozone layer monitoring (Science question 1) was the main objective at the start of the OMI project. OMI has turned out to be very stable and provides a long data record for monitoring the ozone layer, which is critical for the assessment of the Montreal Protocol and later modifications to it at Copenhagen and London. The OMI data record covers a period where the ozone depletion has stopped and where we probably observe the onset of recovery.

The second science question deals with air quality where OMI clearly exceeded the expectations. By its frequent observations of trace gases such as NO2 – Nitrogen dioxide, SO2 – Sulfur dioxide and HCHO – Formaldehyde, OMI contributed on research focused on the mapping of sources and transport of pollution, inversion modelling of emissions, as well as linking trends in air quality to policy measures. The OMI data show a steady decline in concentrations of NO2 in the United States, Europe and Japan, whereas in China first strong increases were observed, followed by decreases after 2014 (van der A et al., 2017; Liu et al., 2016). These improvements can all be linked to the success of policy measures.

The third science objective considers the contribution of OMI to climate research by observing tropospheric O3 greenhouse gas and aerosols, which mainly act as cooling agents, although OMI is best at detecting absorbing aerosol that can cause warming. Tropospheric O3 can be derived from the OMI data itself, or in combination with the Microwave Limb Sounder (MLS) and Tropospheric Emission Spectrometer (TES) instruments that are also manifested on the Aura platform as well as the Atmospheric Infra-Red Sounder (AIRS) on the EOS Aqua satellite that flies in formation with Aura. In both methods, it is important that a long-term data record on tropospheric ozone has been established. For aerosol, the focus has been on the absorption that can be derived in the UV. In combination with the TOMS, GOME and SCIAMACHY data, this is one of the longest aerosol data records available. Linked to aerosols are also the observations of SO2, an important precursor for aerosol particles. The observations show that in the many parts of the world SO2 is decreasing. However, in India we still observe strong increases due to the growing economy and the limited emissions control measures. Natural emissions of SO2 by volcanoes have also been monitored by OMI in great detail.

The last science question on the surface UV-B change is strongly linked to the long-term total ozone record. Research has focused on cases of high UV doses due to low total ozone (de Laat et al., 2010), showing a link in spring-time polar O3 loss with UV-B in the following summer in the extratropics (Karpechko et al., 2013) and on explaining the differences between UV dose derived from satellite and measured on ground (Bernhard et al., 2015). Whereas OMI was conceived as a research instrument, it also contributes to several operational applications. These applications make use of two data streams: the near real-time (NRT) data available within 3 hours of sensing and very-fast delivery (VFD) data available within 20 minutes of sensing via the direct readout capability. Although 5 these data streams were experimental, they turned out to be very successful. Operational users include the European Centre for Medium-range Weather Forecasts (ECMWF) and the US National Oceanic and Atmospheric Administration (NOAA) for ozone and air quality forecasts and the Volcanic Ash Advisory Centers (VAACs) for the rerouting of aircraft in case of a volcanic eruption. The NRT data are provided on the Tropospheric Emission Monitoring Internet Service (TEMIS) website for the scientific user community. The VFD images are distributed at SAMPO website.

A list is provided in the table at the bottom page. Here below is a (non-exhaustive) series of illustrations ordered by application types:

Air quality monitoring, air quality forecasting, pollution events and trends

OMI measures the key air quality components such as NO2 – Nitrogen dioxide, SO2 – Sulfur dioxide, BrO – Bromine oxide, OClO, and aerosol characteristics, all of which contribute to morbidity and mortality (WHO, 2014).

a) OMI tropospheric NO2 VCD data (x10^15 molecules/cm2) as an average for the ozone season (May-September) in 2005 (left) and 2012 (middle) over the eastern U.S. The difference (x10^15 molecules/cm2) between the two years is also shown (right). b) The same as a), but as an annual average (January-December). In the left and middle panels, the white areas indicate regions where at least one month has three or less days of data with which to create the monthly averages, such as in winter with persistent snow and/or cloud cover (Duncan et al., 2015).
OMI NO2 air pollution changes down to sub-urban scales in east Asia and Seoul region from 2005 to 2014 (Duncan et al., 2016)
OMI SO2 (DU) over Iraq in 2016.10.24 (Source: http://earthobservatory.nasa.gov/IOTD/view.php?id=8894). See “A large SO2 pollution in Iraq observed from space” WebPost here.

OMI aeorosol otical depth (AOD), single scattering albedo (SSA) and aerosol index (AI) are derived by the OMAERUV algorithm from the UV.  They are often used with SO2 and NO2 products to infer changes in technology and global economy.

Top-down emission estimates

OMI data have played a key role in the top-down estimation of NOx, SO2, and VOC emissions. Particulate matter (PM) emissions may be inferred via OMI AOD  measurements, but a direct relationship with PM emissions is still elusive (e.g. Hoff and Christopher, 2009). The high resolution OMI observations allow the emission sources to be resolved at a higher resolution than before, which is a distinct advantage for point sources of short-lived gases, including NO2 and SO2, since their sources can be derived with relatively simple methods based on mass balance (e.g. Duncan et al., 2013; de Foy et al., 2015; Fioletov et al., 2015, 2016; Liu et al., 2016; McLinden et al., 2016a). Complete emission maps from OMI observations have been derived using full inversion methods that involves the use of chemical transport models e.g. Qu et al., 2017. Streets et al. (2013) reviewed the current capability to estimate emissions from space, and in this section we highlight studies of emissions using OMI data that have been published subsequently.

NOx emissions in the Middle East in 2010 – left: bottom-up inventory EDGAR v4.3, right: the DECSO algorithm v3b applied to OMI NO2 observations (Ding et al., 2017b)

Volcanic monitoring

OMI can lay claim to being the first satellite instrument to be used for daily monitoring of volcanic emissions (e.g. Carn et al., 2008; Carn et al., 2013; McCormick et al., 2013; Flower and Carn, 2015), heralding a new era where satellite measurements have become an indispensable tool for volcanic gas monitoring in many regions. While instruments such as TOMS have been measuring SO2 and ash emissions by major eruptions since 1978 (e.g. Krueger, 1983; Carn et al., 2016), and GOME first demonstrated the potential for detection of tropospheric volcanic SO2 from space by hyperspectral UV 20 sensors (Eisinger and Burrows, 1998), the ‘volcano-scale’ pixel size (13×24 km at nadir) of OMI was a critical factor. OMI’s ability to detect volcanic SO2 at all levels from the planetary boundary layer (PBL) to the stratosphere, derived from volcanic activity of varying intensity from passive degassing to major stratospheric eruptions, has required the development of SO2 retrieval algorithms capable of spanning several orders of magnitude of SO2 column amount (e.g. from 0.2-2000 DU; Krotkov et al., 2006; Yang et al., 2007, 2009b, 2010; Li et al., 2013; Theys et al., 2014; Li et al., 2016) and direct retrieval of SO2 altitude from UV radiances (e.g. Yang et al., 2009a, 2010).

OMI SO2 Volc
Mean SO2 columns (in Dobson Units [DU]; 1 DU = 2.69 × 1016 molecules cm−2) for 2005–2007 over (a) the Aleutian Islands (USA) and (b) Indonesia. The volcanic SO2 sources (including paired sources) are labeled. The Aleutian map also shows locations of explosive eruptions since 2005 (red triangles), with symbol size proportional to total SO2 emission3,10. The Indonesian map also shows anthropogenic SO2 sources in Singapore and central Sulawesi, but does not show volcanic SO2 emissions from Sinabung, Rinjani and Sangeang Api, which first appeared after 2007. Maps were generated using Interactive Data Language (IDL) version 8.5.1 (http://www.harrisgeospatial.com/) (Carn et al., 2017).
Solar spectral irradiance monitoring

OMI collects solar spectral irradiance (SSI) data primarily to provide long-term on-orbit calibration, in particular for characterization of throughput degradation and wavelength calibration.

The Montreal Protocol, total O3 – Ozone, and UV radiation

The high spatial resolution measurements, first by TOMS and continued by OMI, have been particularly important in mapping the development of the Antarctic ozone hole each year. These current and recent results are a prominent aspect of the quadrennial ozone depletion assessment that is written for the Parties to the Montreal Protocol. The data record of total column ozone from OMI has proven to be very stable over the ten plus years of operation. Surface UV estimates based on OMI satellite data continue the long-term TOMS UV record (Eck et al., 1995; Krotkov et al., 1998, 2001, Tanskanen et al., 2006).

October 2016 monthly average OMI total O3 column over Antarctica
Left: 3-month mean UV index from OMVUB in the boreal fall season 2013 (September-November). Right: Global map of daily UV index on 16th October 2013 showing exceptional high UV index values in Patagonia due to the stretched ozone hole (Tankanen et al., 2006).

Tropospheric O3

OMI has fostered a large number of tropospheric ozone data products, both as ozone column amounts and ozone profiles. These products have been developed using either OMI measurements alone or in conjunction with other satellite measurements to improve sensitivity to near surface ozone (e.g. Bowman, 2013; Cuesta, et al., 2013; Hache et al., 2014) as summarized below. They have been used in tropospheric research (e.g., Sauvage et al., 2007; Ziemke et al., 2010; Cooper et al., 2014), for example to show evidence of decadal increases/trends in global tropospheric ozone, El Nino events during Aura (e.g. Chandra et al., 2009; BAMS State of the Climate report for year 2015), the 1-2 month Madden-Julian Oscillation (Ziemke et al., 2015, and references therein), and urban pollution (Kar et al., 2010).

Research data

Aerosol above cloud: The magnitude of direct radiative effects of aerosols above cloud directly depends on the aerosol loading, microphysical, and optical properties of the aerosol layer and the underlying cloud deck and geometric cloud fraction. The optical depth of carbonaceous and desert dust aerosol layers located above clouds (ACAOD) has been retrieved with OMI (Torres et al., 2012) leading to a global daily product spanning the OMI record (OMACA, OMI ACAOD).

H2O – Water vapor column: Water vapor has a set of absorption bands in the visible region of the spectra. Despite being much weaker than other bands at longer wavelengths, they can be used to retrieve water vapor from OMI. A new column water vapor product (OMH2O) has been developed, evaluated, and implemented (Wang et al., 2014, 2016). The product uses a spectral fitting window of 430-480 nm.

CHO-CHO – Glyoxal column: Glyoxal has been retrieved from OMI (Chan Miller et al., 2014, 2016) using wavelengths 435-461 nm. The retrieval of glyoxal is challenging due to its very weak absorption (optical depths on the order of 10-4-10-3). The OMI glyoxal research product is optimized to minimize interferences from stronger absorbers.

NO2 cloud slicing: The use of cloud pressure information from OMI has led to so-called cloud slicing approaches to retrieve profile information about trace gases (Belmonte Rivas et al., 2015).

Polar Mesospheric Clouds (PMCs): Backscatter ultraviolet (BUV) instruments such as OMI detect PMCs as an enhanced signal at short wavelengths. The broad cross-track coverage of OMI makes it possible to directly characterize local time variations in PMC occurrence frequency and intensity (DeLand et al., 2011).

Multi-platform Product & analyses using several instruments across platform(s)

The development of the so-called “A-train”, a constellation of satellites in a common afternoon orbit all flying within about 15 minutes of each other, has provided unique opportunities to combine data from different instruments into new products, to incorporate additional information to enhance existing OMI products, and to cross validate other products with OMI. In addition to the examples provided below, there are numerous works that employ cross platform comparisons for evaluation of OMI and other satellite data sets and algorithms

Field-of-View for collocation: In depth comparisons of collocated OMI and MODIS radiances have been conducted to estimate precisely its field-of-view (de Graaf et al., 2016; Sihler et al., 2016).

A-train collocated products: OMMYDCLD that contains both OMI cloud products as well as many Aqua MODIS statistical cloud parameters collocated to OMI footprints; OMMYDAGEO that collocates OMI geocoordinates (row and scan number) onto the MODIS granule at 3 and 10 km scales.

Aerosol products: Synergy with MODIS for aerosol layer height (Chimot et al., 2017 and my own research work); synergy OMAERUV with CALIOP measurements and CO – Carbon monoxide from AIRS (Torres et al., 2013); first global estimate of shortwave direct radiative effect of aerosols at the top of the atmosphere (TOA-DREA) over
land and ocean (Lacagnina et al., 2016).

Clouds & radiation: Radiative transfer calculations using collocated cloud extinction profiles from MODIS and CloudSat have been used to evaluate the OMI retrievals (Vasilkov et al., 2008). In addition, a third photon path length type measurement (from PARASOL measurements of oxygen absorption in the A-band) provided additional measurements for evaluation (Sneep et al., 2008).

Geometry-dependent Lambertian Equivalent Reflectivity (GLER): For most OMI algorithms, it is important to have accurate estimates of surface reflectance. Surface reflectance is complex because it varies with the sun-satellite viewing geometry as well as with time and space. Vasilkov et al. (2017) constructed a global time-varying geometry-dependent Lambert-equivalent reflectivity (GLER) product (i.e. for each OMI pixel) based on MODIS data and ocean models.

List of OMI data products

The table below lists the standard, near real-time (NRT) and very fast delivery (VFD) products. The standard products are available within two days after measurement.

OMI standard products along with their type (L1B: radiances and irradiances, L2: Orbital data, L3: Gridded data) delivery method (S: Standard, NRT, or VFD), and PI organization (the Royal Netherlands Meteorological Institute, KNMI, the Finnish Meteorological Institute (FMI), National Aeronautics and Space Administration (NASA) and Smithosonian Astrophyical Observatory (SAO)) (Levelt et al., 2017).

More information?

  • OMI website maintained by KNMI here
  • Tropospheric Emission Monitoring Internet Service (TEMIS) website here
  • SAMPO website here
  • Levelt et al., (2006): Levelt P.F., van den Oord G.H.J., Dobber M.R., Malkki A., Visser H., de Vries J., Stammes P., Lundell J.O.V., and Saari H., The Ozone Monitoring instrument, IEEE Transactions on Geoscience and Remote Sensing, vol. 44, NO. 5, May 2006.
  • Levelt et al. (2017): Levelt, P., Joiner, J., Tamminen, J., Veefkind, P., Bhartia, P. K., Stein Zweers, D., Duncan, B. N., Streets, D. G., Eskes, H., van der A, R., McLinden, C., Fioletov, V., Carn, S., de Laat, J., DeLand, M., Marchenko, S., McPeters, R., Ziemke, J., Fu, D., Liu, X., Pickering, K., Apituley, A., Gonzáles Abad, G., Arola, A., Boersma, F., Chan Miller, C., Chance, K., de Graaf, M., Hakkarainen, J., Hassinen, S., Ialongo, I., Kleipool, Q., Krotkov, N., Li, C., Lamsal, L., Newman, P., Nowlan, C., Suileiman, R., Tilstra, L. G., Torres, O., Wang, H., and Wargan, K.: The Ozone Monitoring Instrument: Overview of twelve years in space, Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2017-487, in review, 2017.
  • NO2 – Nitrogen dioxide WebPage here
  • Aerosols WebPage here
  • A large SO2 pollution in Iraq observed from space” WebPost here