USGS EROS Archive - Vegetation Monitoring - EROS Visible Infrared Imaging Radiometer Suite (eVIIRS)
The Earth Resources Observation and Science (EROS) Center Visible Infrared Imaging Radiometer Suite (eVIIRS) collection is based on the Suomi National Polar-orbiting Partnership (S-NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) data acquired by the NPP, which is the result of a partnership between the National Aeronautics and Space Administration (NASA) and the National Oceanic and Atmospheric Administration (NOAA).
The EROS Visible Infrared Imaging Radiometer Suite (eVIIRS) Normalized Difference Vegetation Index (NDVI) collection uses the Visible Infrared Imaging Radiometer Suite (VIIRS) collection that is available at NASA’s Land Atmosphere Near-real-time Capability for EOS (LANCE) for the expedited product and Level-1 and Atmosphere Archive & Distribution System (LAADS) for the historical products.
Moderate resolution remote sensing provides a means for operational monitoring communities to develop historical trend information and use near-real-time deviations from temporal averages to identify areas of change. High quality, consistent and well- calibrated satellite measurements are needed to detect and monitor changes and trends, especially in vegetation patterns useful for drought, crop yield, phenology and fire potential studies.
The eVIIRS collection is based on the S-NPP VIIRS data acquired by the NPP. Even though VIIRS NDVI data are available as composites in the LP DAAC as 500-m and 1-km 8-day products, the eVIIRS product addresses the need for a 375-m dataset in the 7-, 14-, as well as 10-day dekadal datasets to match EROS Moderate Resolution Imaging Spectroradiometer (eMODIS).
The historical eVIIRS suite of products includes 7- and 14-day composites for the conterminous U.S. (CONUS) and Alaska and on a monthly 10-day dekadal schedule for the NDVI Famine Early Warning Systems (FEWS) regions including Central America/Caribbean and Mexico, Africa, Central Asia, South America, Australia and New Zealand. Expedited (near-real-time) production runs daily for the 7-day (CONUS) products. Each dataset delivers acquisition, quality, and NDVI information at 375-m spatial resolution. Because each of the composites are created from a varying number of images, eVIIRS composites include acquisition files to identify which of the inputs were used to populate the final composite. The metadata accompanying the data files summarize geographic bounds, projection parameters and product contact information.
eVIIRS currently produces NDVI and surface reflectance composites over CONUS, Alaska, and Central America/Caribbean and Mexico, whereas Africa, Central Asia, South America, Australia, and New Zealand contain NDVI composites only.
Composite Building Process
The first step in the eVIIRS composite building process is to create the various geotiff bands required for building an NDVI composite for each of the days and times in the compositing period. For example, the compositing period might be 7 or 10 days.
One of the GeoTIFF bands created is the NDVI band. NDVI is created from the known equation
NDVI=(NIR – red)/(NIR + red)
where red is VIIRS band I1 and NIR is band I2. These bands come from the 375-m surface reflectance bands of VNP09 granules.
The minimum NDVI values allowed are -0.1999, and all NDVI values (which normally fall between -1.0 and 1.0) are scaled by 10,000. Thus, the output NDVI values from this application fall between -2000 and 10,000 (with the exception of negative surface reflectance values) to match the LP DAAC MOD09 products.
If either of the red or NIR pixel values is background fill, then the output NDVI value is set to UNDEF (-2000). The same value applies if the red and NIR pixels are the same value. If either of the red or NIR pixels is negative, then the output NDVI value is set to NEG_SR (-3000).
Other bands created for the compositing process are Satellite Zenith Angle and Surface Reflectance Quality Bands 1, 2 and 4.
Composite Calculation
The software used to create 7-, 14- and 10-day dekadal NDVI composites was developed by the USGS EROS software development team. The software processes VIIRS 375-m swath granules to produce 375-m and/or 1-km composite products.
The software grids the required swath granules (VNP09 and NPP_IMFTS_L1) for the specified bounded region (using upper left and lower right corner coordinates). For each gridded granule, an NDVI band is generated using the red (band I1) and NIR (band I2) bands.
Once each granule has been gridded to the given coordinate extents and the NDVI has been built, then an enhanced maximum value composite (MVC) is generated. With a straight MVC algorithm, each pixel in the composite (for each band) would be filled using the pixel value of the granule with the highest NDVI for the current pixel. The MVC has been enhanced in the current eVIIRS algorithm by incorporating band quality, cloud mask, negative surface reflectance, and view angle into the algorithm for eVIIRS processing.
- The band quality information in the surface reflectance product is used to determine if a certain pixel is of “bad quality.” Bad quality pixels are not used in the final composite product.
- The NDVI algorithm flags negative surface reflectance values so that they will not be used in the output composite, if positive values are available. If either the red or NIR band surface reflectance values are negative, then the output NDVI value is flagged with a value of -3000. (NOTE: Undefined or background fill NDVI values are -2000 and that is specified as the lower limit of the “valid” NDVI values.) The composite algorithm then disregards granules with NDVI values of -3000 to omit negative surface reflectance values in the output composite.
- The cloud mask is used to minimize the presence of cloud pixels in the composite product. If the quality flag is probably clear and/or confident clear, then the pixel may be used to fill the current composite pixel.
- The view angle (distance from nadir) determines which of the two highest NDVI pixels is used in the composite. Of the non-cloudy, ideal-quality pixels, the two highest NDVI values are determined from the list of granules. Of these two granules, the pixel value which is closest to nadir is used to fill the composite.
The products are generated for S-NPP VIIRS over all areas including CONUS and Alaska and are projected to a regionally specific mapping grid and delivered in a compressed (zipped) GeoTIFF format.
Two spatial resolutions (375 m and 1,000 m) are available to download individually. These data are available in EarthExplorer as well Machine-to-Machine API. The download product is a .zip file containing 6 files, including three .tif, one jpeg, and two .txt files.
Following is a sample list of files from a CONUS 1-km NDVI composite .zip file named “US_eVSH_NDVI.2020.350-356.1KM.COMPRES.006.2020359170316.zip”:
US_eVSH_NDVI.2020.350-356.1KM.VI_ACQI.006.2020362063105.tif Acquisition geotiff
US_eVSH_NDVI.2020.350-356.1KM.VI_ACQT.006.2020362063105.txt Acquisition text file
US_eVSH_NDVI.2020.350-356.1KM.VI_META.006.2020362063314.met Metadata text file
US_eVSH_NDVI.2020.350-356.1KM.VI_NDVI.006.2020362063105.jpg Browse JPEG
US_eVSH_NDVI.2020.350-356.1KM.VI_NDVI.006.2020362063105.tif NDVI geotiff
US_eVSH_NDVI.2020.350-356.1KM.VI_QUAL.006.2020362063105.tif Quality geotiff
Acquisition Band (ACQI)
An acquisition band provides the user with a GeoTIFF image that can be overlayed onto the NDVI image so that the user can then identify each granule that provided the value for each NDVI pixel. The DOY and the acquisition number will be used to specify the acquisition. The acquisition number in this case represents not the time of the acquisition, but the order of capture. The output values are represented as an unsigned 16-bit integer using the equation
DOY * 1000 + acq_number
Thus, the first acquisition for DOY 117 would be 117001. The eleventh acquisition for DOY 117 would be 117011. To make it easier for the user to determine which acquisition integer value maps to which granule, an output acquisition table text file is written to match the acquisition values to the granule names.
Acquisition Text File (ACQT)
An acquisition text file provides the user a text file to accompany the Acquisition image, which gives the composite details of the data used to create the product. This list of the acquisition band values is used in the composite acquisition band product and the corresponding acquisition filename for that acquisition value.
160005 MA2RG_2020_160_2120_250m_NDVI.hdf
160004 MA2RG_2020_160_1940_250m_NDVI.hdf
160003 MA2RG_2020_160_1935_250m_NDVI.hdf
160002 MA2RG_2020_160_1800_250m_NDVI.hdf
160001 MA2RG_2020_160_1755_250m_NDVI.hdf
159007 MA2RG_2020_159_2035_250m_NDVI.hdf
159006 MA2RG_2020_159_2030_250m_NDVI.hdf
159005 MA2RG_2020_159_1900_250m_NDVI.hdf
Metadata (META)
The Metadata file provides the details about the composite image, including acquisition time period, publication dates, pixel and row counts, map projection information, datum, pixel resolution, satellite and platform, digital data type, fill values, scaling factors, and center and corner coordinates.
Browse JPEG
A full resolution browse image is provided in JPEG format and is zipped with the product and used to display the image on EarthExplorer. The color mapping is used to create a color image for the browse image and is not the same as the 16-bit single band data delivered in the NDVI GeoTIFF.
NDVI / Surface Reflectance GeoTIFF
The NDVI GeoTIFF is the product created in the compositing process detailed in the above section. All the other files including quality band, acquisition band, as well as the text files and browse are companion files for this NDVI product.
The Surface Reflectance composite is also created for CONUS, Alaska, and Central America/Caribbean and Mexico and is available as a separate download option in EarthExplorer.
Quality Band (QA)
A quality assurance (QA) band is 8 bits and produced by the composite software to identify the quality of each composite pixel. Most pixels will be filled with a pixel of good band quality. However, if none of the granules for a particular pixel are of good band quality or all of the pixels are cloudy, then the current pixel is filled with data from the best pixel possible where the preference order is valid NDVI over fill, good quality over bad, snow pixel over cloudy pixel. The following values are used in theQA band:
0 = good quality
1 = cloudy pixel
2 = bad band quality
4 = snow
10 = fill
Surface Reflectance Quality Band 1 is made up of the following bit values. Bit number 0 is the bit in the binary number which is of the lowest numerical value or Least Significant Bit (LSB).
| Bit No. | Parameter Name | Bit Combination | Definition |
|---|---|---|---|
| 0-1 | Cloud Mask Quality | 00 | Poor |
| 01 | Low | ||
| 10 | Medium | ||
| 11 | High | ||
| 2-3 | Cloud Detection and Confidence | 00 | Confident Clear |
| 01 | Probably Clear | ||
| 10 | Probably Cloudy | ||
| 11 | Confident Cloudy | ||
| 4 | Day/Night | 0 | Day |
| 1 | Night | ||
| 5 | Low Sun Mask | 0 | High |
| 1 | Low | ||
| 6-7 | Sun Glint | 00 | None |
| 01 | Geometry Based | ||
| 10 | Wind Speed Based | ||
| 11 | Geometry and Wind Speed Based |
Surface Reflectance Quality Band 2 is made up of the following bit values.
| Bit No. | Parameter Name | Bit Combination | Definition |
|---|---|---|---|
| 0-2 | Land/Water Background | 000 | Land and Desert |
| 001 | Land No Desert | ||
| 010 | Inland Water | ||
| 011 | Sea Water | ||
| 100 | --- | ||
| 101 | Coastal | ||
| 110 | --- | ||
| 3 | Shadow Mask | 0 | No Cloud Shadow |
| 1 | Shadow |
Surface Reflectance Quality Band 4 is made up of the following bit values. Bits 1 and 2 from quality band 4 are used to determine if the bits from a given gridded I1 (red) or I2 (near-infrared) band should be used in creating the NDVI band. Surface Directional Reflectance (SDR), Aerosol Optical Thickness (AOT), Ante Meridiem (AM), Precipitable Water (PW)
| Bit No. | Parameter Name | Bit Combination | Definition |
|---|---|---|---|
| 0 | BAD M11 SDR Data | 0 | No |
| 1 | Yes | ||
| 1 | Bad I1 SDR Data | 0 | No |
| 1 | Yes | ||
| 2 | Bad I2 SDR Data | 0 | No |
| 1 | Yes | ||
| 3 | Bad I3 SDR Data | 0 | No |
| 1 | Yes | ||
| 4 | Overall Quality of AOT | 0 | Good |
| 1 | Bad | ||
| 5 | Missing AOT Input Data | 0 | No |
| 1 | Yes | ||
| 6 | Invalid Land AM Input Data | 0 | Valid |
| 1 | Invalid AM Input Over Land or Over Ocean | ||
| 7 | Missing PW Input Data | 0 | No |
| 1 | Yes |
Additional Information
- Land Atmosphere Near real-time Capability for Earth Observing System (EOS) (LANCE)
- Level 1 and Atmosphere Archive and Distribution System (LAADS)
- Land Validation
- NOAA VIIRS
- JPSS VIIRS
Access Data
eVIIRS NDVI products held in the USGS archive can be searched using EarthExplorer. On EarthExplorer, eVIIRS NDVI products can be found under the Vegetation Monitoring category.
Digital Object Identifier (DOI)
The Earth Resources Observation and Science (EROS) Center Visible Infrared Imaging Radiometer Suite (eVIIRS) collection is based on the Suomi National Polar-orbiting Partnership (S-NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) data acquired by the NPP, which is the result of a partnership between the National Aeronautics and Space Administration (NASA) and the National Oceanic and Atmospheric Administration (NOAA).
The EROS Visible Infrared Imaging Radiometer Suite (eVIIRS) Normalized Difference Vegetation Index (NDVI) collection uses the Visible Infrared Imaging Radiometer Suite (VIIRS) collection that is available at NASA’s Land Atmosphere Near-real-time Capability for EOS (LANCE) for the expedited product and Level-1 and Atmosphere Archive & Distribution System (LAADS) for the historical products.
Moderate resolution remote sensing provides a means for operational monitoring communities to develop historical trend information and use near-real-time deviations from temporal averages to identify areas of change. High quality, consistent and well- calibrated satellite measurements are needed to detect and monitor changes and trends, especially in vegetation patterns useful for drought, crop yield, phenology and fire potential studies.
The eVIIRS collection is based on the S-NPP VIIRS data acquired by the NPP. Even though VIIRS NDVI data are available as composites in the LP DAAC as 500-m and 1-km 8-day products, the eVIIRS product addresses the need for a 375-m dataset in the 7-, 14-, as well as 10-day dekadal datasets to match EROS Moderate Resolution Imaging Spectroradiometer (eMODIS).
The historical eVIIRS suite of products includes 7- and 14-day composites for the conterminous U.S. (CONUS) and Alaska and on a monthly 10-day dekadal schedule for the NDVI Famine Early Warning Systems (FEWS) regions including Central America/Caribbean and Mexico, Africa, Central Asia, South America, Australia and New Zealand. Expedited (near-real-time) production runs daily for the 7-day (CONUS) products. Each dataset delivers acquisition, quality, and NDVI information at 375-m spatial resolution. Because each of the composites are created from a varying number of images, eVIIRS composites include acquisition files to identify which of the inputs were used to populate the final composite. The metadata accompanying the data files summarize geographic bounds, projection parameters and product contact information.
eVIIRS currently produces NDVI and surface reflectance composites over CONUS, Alaska, and Central America/Caribbean and Mexico, whereas Africa, Central Asia, South America, Australia, and New Zealand contain NDVI composites only.
Composite Building Process
The first step in the eVIIRS composite building process is to create the various geotiff bands required for building an NDVI composite for each of the days and times in the compositing period. For example, the compositing period might be 7 or 10 days.
One of the GeoTIFF bands created is the NDVI band. NDVI is created from the known equation
NDVI=(NIR – red)/(NIR + red)
where red is VIIRS band I1 and NIR is band I2. These bands come from the 375-m surface reflectance bands of VNP09 granules.
The minimum NDVI values allowed are -0.1999, and all NDVI values (which normally fall between -1.0 and 1.0) are scaled by 10,000. Thus, the output NDVI values from this application fall between -2000 and 10,000 (with the exception of negative surface reflectance values) to match the LP DAAC MOD09 products.
If either of the red or NIR pixel values is background fill, then the output NDVI value is set to UNDEF (-2000). The same value applies if the red and NIR pixels are the same value. If either of the red or NIR pixels is negative, then the output NDVI value is set to NEG_SR (-3000).
Other bands created for the compositing process are Satellite Zenith Angle and Surface Reflectance Quality Bands 1, 2 and 4.
Composite Calculation
The software used to create 7-, 14- and 10-day dekadal NDVI composites was developed by the USGS EROS software development team. The software processes VIIRS 375-m swath granules to produce 375-m and/or 1-km composite products.
The software grids the required swath granules (VNP09 and NPP_IMFTS_L1) for the specified bounded region (using upper left and lower right corner coordinates). For each gridded granule, an NDVI band is generated using the red (band I1) and NIR (band I2) bands.
Once each granule has been gridded to the given coordinate extents and the NDVI has been built, then an enhanced maximum value composite (MVC) is generated. With a straight MVC algorithm, each pixel in the composite (for each band) would be filled using the pixel value of the granule with the highest NDVI for the current pixel. The MVC has been enhanced in the current eVIIRS algorithm by incorporating band quality, cloud mask, negative surface reflectance, and view angle into the algorithm for eVIIRS processing.
- The band quality information in the surface reflectance product is used to determine if a certain pixel is of “bad quality.” Bad quality pixels are not used in the final composite product.
- The NDVI algorithm flags negative surface reflectance values so that they will not be used in the output composite, if positive values are available. If either the red or NIR band surface reflectance values are negative, then the output NDVI value is flagged with a value of -3000. (NOTE: Undefined or background fill NDVI values are -2000 and that is specified as the lower limit of the “valid” NDVI values.) The composite algorithm then disregards granules with NDVI values of -3000 to omit negative surface reflectance values in the output composite.
- The cloud mask is used to minimize the presence of cloud pixels in the composite product. If the quality flag is probably clear and/or confident clear, then the pixel may be used to fill the current composite pixel.
- The view angle (distance from nadir) determines which of the two highest NDVI pixels is used in the composite. Of the non-cloudy, ideal-quality pixels, the two highest NDVI values are determined from the list of granules. Of these two granules, the pixel value which is closest to nadir is used to fill the composite.
The products are generated for S-NPP VIIRS over all areas including CONUS and Alaska and are projected to a regionally specific mapping grid and delivered in a compressed (zipped) GeoTIFF format.
Two spatial resolutions (375 m and 1,000 m) are available to download individually. These data are available in EarthExplorer as well Machine-to-Machine API. The download product is a .zip file containing 6 files, including three .tif, one jpeg, and two .txt files.
Following is a sample list of files from a CONUS 1-km NDVI composite .zip file named “US_eVSH_NDVI.2020.350-356.1KM.COMPRES.006.2020359170316.zip”:
US_eVSH_NDVI.2020.350-356.1KM.VI_ACQI.006.2020362063105.tif Acquisition geotiff
US_eVSH_NDVI.2020.350-356.1KM.VI_ACQT.006.2020362063105.txt Acquisition text file
US_eVSH_NDVI.2020.350-356.1KM.VI_META.006.2020362063314.met Metadata text file
US_eVSH_NDVI.2020.350-356.1KM.VI_NDVI.006.2020362063105.jpg Browse JPEG
US_eVSH_NDVI.2020.350-356.1KM.VI_NDVI.006.2020362063105.tif NDVI geotiff
US_eVSH_NDVI.2020.350-356.1KM.VI_QUAL.006.2020362063105.tif Quality geotiff
Acquisition Band (ACQI)
An acquisition band provides the user with a GeoTIFF image that can be overlayed onto the NDVI image so that the user can then identify each granule that provided the value for each NDVI pixel. The DOY and the acquisition number will be used to specify the acquisition. The acquisition number in this case represents not the time of the acquisition, but the order of capture. The output values are represented as an unsigned 16-bit integer using the equation
DOY * 1000 + acq_number
Thus, the first acquisition for DOY 117 would be 117001. The eleventh acquisition for DOY 117 would be 117011. To make it easier for the user to determine which acquisition integer value maps to which granule, an output acquisition table text file is written to match the acquisition values to the granule names.
Acquisition Text File (ACQT)
An acquisition text file provides the user a text file to accompany the Acquisition image, which gives the composite details of the data used to create the product. This list of the acquisition band values is used in the composite acquisition band product and the corresponding acquisition filename for that acquisition value.
160005 MA2RG_2020_160_2120_250m_NDVI.hdf
160004 MA2RG_2020_160_1940_250m_NDVI.hdf
160003 MA2RG_2020_160_1935_250m_NDVI.hdf
160002 MA2RG_2020_160_1800_250m_NDVI.hdf
160001 MA2RG_2020_160_1755_250m_NDVI.hdf
159007 MA2RG_2020_159_2035_250m_NDVI.hdf
159006 MA2RG_2020_159_2030_250m_NDVI.hdf
159005 MA2RG_2020_159_1900_250m_NDVI.hdf
Metadata (META)
The Metadata file provides the details about the composite image, including acquisition time period, publication dates, pixel and row counts, map projection information, datum, pixel resolution, satellite and platform, digital data type, fill values, scaling factors, and center and corner coordinates.
Browse JPEG
A full resolution browse image is provided in JPEG format and is zipped with the product and used to display the image on EarthExplorer. The color mapping is used to create a color image for the browse image and is not the same as the 16-bit single band data delivered in the NDVI GeoTIFF.
NDVI / Surface Reflectance GeoTIFF
The NDVI GeoTIFF is the product created in the compositing process detailed in the above section. All the other files including quality band, acquisition band, as well as the text files and browse are companion files for this NDVI product.
The Surface Reflectance composite is also created for CONUS, Alaska, and Central America/Caribbean and Mexico and is available as a separate download option in EarthExplorer.
Quality Band (QA)
A quality assurance (QA) band is 8 bits and produced by the composite software to identify the quality of each composite pixel. Most pixels will be filled with a pixel of good band quality. However, if none of the granules for a particular pixel are of good band quality or all of the pixels are cloudy, then the current pixel is filled with data from the best pixel possible where the preference order is valid NDVI over fill, good quality over bad, snow pixel over cloudy pixel. The following values are used in theQA band:
0 = good quality
1 = cloudy pixel
2 = bad band quality
4 = snow
10 = fill
Surface Reflectance Quality Band 1 is made up of the following bit values. Bit number 0 is the bit in the binary number which is of the lowest numerical value or Least Significant Bit (LSB).
| Bit No. | Parameter Name | Bit Combination | Definition |
|---|---|---|---|
| 0-1 | Cloud Mask Quality | 00 | Poor |
| 01 | Low | ||
| 10 | Medium | ||
| 11 | High | ||
| 2-3 | Cloud Detection and Confidence | 00 | Confident Clear |
| 01 | Probably Clear | ||
| 10 | Probably Cloudy | ||
| 11 | Confident Cloudy | ||
| 4 | Day/Night | 0 | Day |
| 1 | Night | ||
| 5 | Low Sun Mask | 0 | High |
| 1 | Low | ||
| 6-7 | Sun Glint | 00 | None |
| 01 | Geometry Based | ||
| 10 | Wind Speed Based | ||
| 11 | Geometry and Wind Speed Based |
Surface Reflectance Quality Band 2 is made up of the following bit values.
| Bit No. | Parameter Name | Bit Combination | Definition |
|---|---|---|---|
| 0-2 | Land/Water Background | 000 | Land and Desert |
| 001 | Land No Desert | ||
| 010 | Inland Water | ||
| 011 | Sea Water | ||
| 100 | --- | ||
| 101 | Coastal | ||
| 110 | --- | ||
| 3 | Shadow Mask | 0 | No Cloud Shadow |
| 1 | Shadow |
Surface Reflectance Quality Band 4 is made up of the following bit values. Bits 1 and 2 from quality band 4 are used to determine if the bits from a given gridded I1 (red) or I2 (near-infrared) band should be used in creating the NDVI band. Surface Directional Reflectance (SDR), Aerosol Optical Thickness (AOT), Ante Meridiem (AM), Precipitable Water (PW)
| Bit No. | Parameter Name | Bit Combination | Definition |
|---|---|---|---|
| 0 | BAD M11 SDR Data | 0 | No |
| 1 | Yes | ||
| 1 | Bad I1 SDR Data | 0 | No |
| 1 | Yes | ||
| 2 | Bad I2 SDR Data | 0 | No |
| 1 | Yes | ||
| 3 | Bad I3 SDR Data | 0 | No |
| 1 | Yes | ||
| 4 | Overall Quality of AOT | 0 | Good |
| 1 | Bad | ||
| 5 | Missing AOT Input Data | 0 | No |
| 1 | Yes | ||
| 6 | Invalid Land AM Input Data | 0 | Valid |
| 1 | Invalid AM Input Over Land or Over Ocean | ||
| 7 | Missing PW Input Data | 0 | No |
| 1 | Yes |
Additional Information
- Land Atmosphere Near real-time Capability for Earth Observing System (EOS) (LANCE)
- Level 1 and Atmosphere Archive and Distribution System (LAADS)
- Land Validation
- NOAA VIIRS
- JPSS VIIRS
Access Data
eVIIRS NDVI products held in the USGS archive can be searched using EarthExplorer. On EarthExplorer, eVIIRS NDVI products can be found under the Vegetation Monitoring category.