The Berkeley High-Resolution NO2 Product



The Berkeley High-Resolution (BEHR) retrieval of NO2 vertical column density from OMI utilizes higher-resolution terrain and profile inputs compared to the operational retrievals in order to account for the spatial and seasonal variation of these parameters that have been shown to significantly effect the retrieved NO2 column densities.

BEHR relies on the NASA Standard Product for destriping and stratospheric subtraction methods, and thus we have adopted the following naming convention for BEHR versions: the version is described by a number, which corresponds to the version number of the Standard Product used, and a letter, which corresponds to the version of BEHR using stratospheric subtraction and destriping methods from that Standard Product version number. For example, the initial BEHR product described in more detail in Russell et al. (2011) is BEHR v1.0A, as it was the first BEHR retrieval and based on Standard Product v1.0. A subsequent version was prepared using the same Standard Product v1.0 and was named v1.0B.

Additionally, minor revisions to the format of the product that do not substantially change the data itself are identified with the suffix revN. Thus v1.0Brev1 would be a minor revision to v1.0B. v1.0B and v1.0Brev0 would be the same.  In the file names and attributes, the version number may be given as, e.g. v1-0B. This is the same as v1.0B, we use a hyphen instead of a period to avoid potential issues with OSes that expect a period to only separate the file extension from the file name.

The currently available version is BEHR v2.1C.  This was uploaded on 13 Jan 2017.  This changelog enumerates all the changes.

If you intend to utilize the scattering weights or averaging kernels, please read section 4.3 of the user guide first as it explains the consideration that went into their formatting.

BEHR is available for OMI observations since January 2005 to September 2016 over the continental United States (land only). We are in the process of updating to use version 3 of the NASA SP and expect to release a new product in summer 2017.

The following table provides a summary of the terrain and profile parameters used in the operational and BEHR products.

  Table 1: Comparison of parameters used in the BEHR product and those used in the operational products.


Data Description

The BEHR product is available in daily files (organized by individual overpasses) at the native OMI footprint resolution as .txt and .hdf files that include all of the variables in the following table. The product is also provided in a regridded format at 0.05° × 0.05° (125-65W, 25-50N) resolution as .hdf files; however, these files include a limited set of variables. Tools for working with .hdf files are available at Some people have had difficulty opening the BEHR .hdf files, especially when using older software; the files can be opened in MATLAB 2012a, and we have also included some Python code which can be used to open the files (please click here).  Please note that different functions or tools are often needed to open .hdf version 4 and version 5 files; all BEHR .hdf files are version 5 and need to be opened using the appropriate tools.

Table 2: Description of variables included in the BEHR product files.

 Variables SP BEHR Additional Description Units Range Fill
 Longitude  *      deg  [-180, 180]  
 Latitude  *      deg  [-90, 90]  
 Loncorn    *  Calculated corner longitudes  deg  [-180, 180]  
 Latcorn    *  Calculated corner latitudes  deg  [-180, 180]  
 AMFStrat  *    Stratospheric AMF  -  [0, Inf]  
 AMFTrop  *    Tropospheric AMF (standard product)  -  [0, Inf]  
 Areaweight    *  Reciprocal of pixel area, use to weight temporal average  km-2  [0, Inf]  -3.402e38
 BEHRAMFTrop    *  Tropospheric AMF (BEHR)  -  [0, Inf]  -32767
 BEHRColumnAmountNO2Trop    *  Tropospheric NO2 VCD (BEHR)  molec. cm-2  [0, Inf]  -1.267e30
 BEHRGhostFraction    *  Factor to estimate NO2 below cloud   -  [0, Inf]  -3.402e38
 BEHRNO2apriori    *  A priori NO2 profiles used  mixing ratio  [-Inf, Inf]  -3.402e38
 BEHRScatteringWeights    *  Vectors of BEHR scattering weights, weighted by clouds  -  [0, Inf]  -3.402e38
 BEHRAvgKernels    *  Vectors of BEHR averaging kernels (= sc.wt. / AMF)  -  [0, Inf]  -3.402e38
 BEHRPressureLevels    *  Vectors of pressure levels for a priori, s. weights, AKs  hPa  [0, Inf]  -3.402e38
 CloudFraction  *    OMI geometric cloud fraction  -  [0, 1]  -32767
 CloudPressure  *    OMI cloud centroid pressure  hPa  [0, Inf]  -32767
 CloudRadianceFraction  *    OMI cloud radiance (top of atmosphere light fraction)  -  [0, 1]  -32767
 ColumnAmountNO2  *    Total NO2 VCD  molec. cm-2  [0, Inf]  -1.267e30
 ColumnAmountNO2Trop  *    Tropospheric NO2 VCD  molec. cm-2
 [0, Inf]  -1.267e30
 ColumnAmountNO2TropStd  *    Standard deviation of SP tropospheric VCD  molec. cm-2
 [0, Inf]  -1.267e30
 ColumnAmountNO2Strat  *    Stratospheric NO2 VCD  molec. cm-2
 [0, Inf]  -1.267e30
 GLOBETerpres    *  Terrain pressure calculated from GLOBE database, avg. to pixel  hPa  [0, Inf]  -3.402e38
 MODISAlbedo    *  MODIS MCD43C3 product, avg. to pixel  -  [0, 1]  -3.402e38
 MODISCloud    *  MODIS MYD06_L2 product, avg. to pixel  -  [0, 1]  -3.402e38
 RelativeAzimuthAngle  *    Calculated azimuth angle between sun and satellite*  deg  [0, 180]  
 Row  *    Across track row number (0 based)  -  [0, 59]  
 SlantColumnAmountNO2  *    Total NO2 SCD  molec. cm-2
 [0, Inf]  -1.267e30
 SolarAzimuthAngle  *      deg  [-180, 180]  
 SolarZenithAngle  *      deg  [0, 90]  
 Swath  *    Swath number since OMI launch  -  [0, Inf]  
 TerrainHeight  *      m  [-Inf, Inf]  
 TerrainPressure  *      hPa  [0, Inf]  
 TerrainReflectivity  *    SP albedo value  -  [0, 1]  -32767
 Time  *    Time at start of scan (TAI93: seconds since Jan 1, 1993)  s  [0, Inf]  
 ViewingAzimuthAngle  *      -  [-180, 180]  
 ViewingZenithAngle  *      -  [-180, 180]  
 XTrackQualityFlags  *    Across track quality flags (for row anomaly)  -  N/A  
 vcdQualityFlags  *    Ground pixel quality flags  -  N/A  

SP: Variables taken directly from NASA's standard product (SP, collection 3, ver 2)

BEHR: Variables specific to the BEHR product described here

*The relative azimuth angle has an additional factor of 180 deg. to match a convention used in the scattering weight lookup table.

Why so many different fill vales?

Two come from the OMNO2 standard product: -1.267e30 and -32767 are used for long floats and integers, respectively. In BEHR fields, we use -3.402e38 to differentiate where fill values are added by our algorithm vs. the standard product. This has changed from previous version, in which we kept the NASA fill values in BEHRAMFTrop and BEHRColumnAmountNO2Trop, so please verify that your code is properly removing the new fills.



OMNO2 README File, available at

Russell, A. R., Perring, A. E., Valin, L. C., Bucsela, E., Browne, E. C., Min, K.-E., Wooldridge, P. J., and Cohen, R. C.: A high spatial resolution retrieval of NO2 column densities from OMI: method and evaluation, Atmos. Chem. Phys., 11, 8543-8554, doi:10.5194/acp-11-8543-2011, 2011. Article

Russell, A. R., Valin, L. C., and Cohen, R. C.: Trends in OMI NO2 observations over the United States: effects of emissions control technology and the economic recession, Atmos. Chem. Phys., 12, 12197-12209, doi:10.5194/acp-12-12197-2012, 2012. Article

Laughner, J.L., Zare, A. and Cohen, R.C.: Effects of daily meteorology on the interpretation of space-based remote sensing of NO2, Atmos. Chem. Phys., 16, 15247-15264, doi:10.5194/acp-16-15247-2016, 2016. Article