Skip to main content
U.S. flag

An official website of the United States government

Landsat Science Products News Archive

By Landsat Missions

This page displays historical news items relating to new data products and/or services.

Return to Landsat Science Products Overview

 

April 1, 2020 - Landsat 8 Provisional Aquatic Reflectance Science Product Now Available

The provisional Aquatic Reflectance science product is currently available for Landsat 8 Operational Land Imager (OLI) acquistions only (2013-present) through the USGS EROS Center Science Processing Architecture (ESPA) on-demand interface.

The Aquatic Reflectance product measures the spectral distribution of visible solar-reflected radiation upwelling from the upper water column and is derived from Landsat Collection 1 Level-1 reflective bands. The product’s 30 meter spatial resolution  provides the aquatic remote sensing community the capability to map optically active components across inland and near-shore water bodies.

July 25, 2019 - Daily MODIS and VIIRS NDVI Products Now Available in ESPA

Daily NDVI products derived from NASA surface reflectance datasets are now available for processing from the USGS Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) On Demand Interface.

500-meter Moderate Resolution Imaging Spectroradiometer (MODIS)/Terra Surface Reflectance Daily (MOD09GA) Version 6 and MODIS/Aqua Surface Reflectance Daily (MYD09GA) Version 6 band 1 (red) and band 2 (NIR) data are used as input for Daily MODIS NDVI products delivered from ESPA.

 500-meter Visible Infrared Imaging Radiometer Suite (VIIRS)/NPP Surface Reflectance Daily (VNP09GA) Version 1 band I1 (red) and band I2 (NIR) data are used as input for Daily VIIRS NDVI products delivered from ESPA.

 These ordering options are visible in ESPA’s Additional MODIS/VIIRS Processing section. The EROS Science Processing Architecture On-Demand Interface User Guide provides additional technical details about the NASA Daily MODIS and VIIRS NDVI products. 

February 13, 2019 - New Landsat Level-3 Products Released

The USGS has developed research-quality, applications-ready, Landsat Level-3 Science Products that are now available for download from EarthExplorer. These products will reduce the amount of time users spend on data processing and provide research-quality, applications-ready information on burned areas, surface water extent and snow cover.

March 1, 2018 - ESPA Bulk Downloader Update

Users receiving unexplained 500 or 401 errors when using the ESPA Bulk Downloader are encouraged to download the recent update available at from the USGS GitHub location.

January 31, 2018 - Python requirement update

Effective Wednesday, January 31, 2018 Python Version 2.7.9 is the minimum software requirement on Windows systems to avoid known issues when communicating with the ESPA server.

November 29, 2017 - Delivery changes to Band 4 Solar/Sensor Zenith/Azimuth Angle Bands

A recent software release to the Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) on-demand interface changes the delivery to users of the Band 4 Solar/Sensor zenith/azimuth angle bands. While historically they were delivered with other ESPA products the user ordered, they will now only be included with the top of atmosphere reflectance product.

September 29, 2017 – Upcoming changes to Landsat 8 Surface Reflectance Code

On October 5, 2017, a number of modifications will be made to the Landsat Surface Reflectance Code (LaSRC) to optimize Landsat 8 surface reflectance processing and to improve Landsat 8 surface reflectance quality assessment (QA) information.

Optimizations to the LaSRC aerosol retrieval method greatly reduce Landsat 8 Surface Reflectance data processing times.  Instead of performing aerosol retrieval on a per-pixel basis, the new method performs the retrieval for a 3x3-pixel window. The aerosol value is computed for the center pixel of each 3x3 window. If the center pixel of the 3x3 window is found to be cloud, cloud shadow, or water, the algorithm will search the window for the closest non-cloudy, land pixel to compute the aerosol value. If no pixel within the 3x3 window is clear, the center pixel will be assigned an arbitrary value until the median aerosol value of all center pixels for each 3x3 window contained in the scene is computed. The arbitrary value is then replaced with the newly-determined median aerosol value. All remaining non-center pixels are then interpolated using the center pixel values of the closest 3x3 windows. In addition to the 3x3 window optimization, the aerosol computations are modified to utilize a set of polynomial coefficients to estimate the aerosols versus computing the aerosol for every pixel. The SR_AEROSOL_QA band will be updated with new bit descriptions.

Pixels within the PIXEL_QA band will be modified as follows:

  • Pixels flagged as high confidence cirrus (Bits 8-9) will no longer unset the clear bit (Bit 1).
  • Pixels flagged as terrain occluded (Bit 10) will also no longer unset the clear bit (Bit 1).

  • NOTE: Pixels flagged as high confidence cloud (Bits 6-7) will continue to unset the clear bit.

The Landsat 8 Surface Reflectance Code (LaSRC) Product Guide will be updated to reflect these changes.

June 29, 2017 – Useful updates made to ESPA

EROS Science Processing Architecture (ESPA) users now have the ability to cancel their order. This option is located on each order status page on the ESPA on-demand interface or via an http PUT method through the ESPA API. Additionally, a new option exists to allow users to select the Pixel QA band as an individual output product for each scene ordered.

Details about both of these options can be found in the ESPA On-demand User Guide.

March 17, 2017 – Processing System Updated to Support Landsat Collection 1 Science Processing

Recent algorithm updates to the Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) on-demand interface allow for the processing of Landsat science products (such as Surface Reflectance and surface reflectance-based Spectral Indices) for Landsat Level-1 Collection 1 scenes.

There are many differences between Landsat Pre-Collection science data and Landsat Collection 1 science products generated by ESPA. Users should consult the respective product guides for complete information:  

Landsat 4-7 Surface Reflectance Product Guide

Landsat 8 Surface Reflectance Product Guide

October 14, 2016 – Upgrades to Landsat Science Data Processing System

On October 13, 2016, the following update was made to the Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) On-Demand interface, the processing system that is used to create Landsat science products:

The ESPA On-Demand Interface User Guide provides more details about the NetCDF option, along with the addition of the ESPA Application Programming Interface (API) information and access location.

June 10, 2016 - New Landsat 8 Surface Algorithm Coming Soon

Users familiar with Landsat 8 Surface Reflectance (L8SR) data are aware of the artifacts in the current provisional data products. After extensive research, redevelopment, and analysis, an updated version of the algorithm is ready for release, and it will improve L8 Surface Reflectance data products. By July 1, 2016, the “Land Surface Reflectance Code (LaSRC) will replace the existing L8SR algorithm used to produce L8 Surface Reflectance data products, and will be implementing the following changes:

December 16, 2015 – Updates to ESPA ordering interface

Updates been made to the ESPA processing system include:

The ESPA On Demand Interface User Guide provides more information on these changes.

September 28, 2015 – Update to Provisional Landsat 8 Surface Reflectance (L8SR), Landsat 4-7 Surface Reflectance (LEDAPS) and CFmask shadow prediction

Provisional Landsat 8 Surface Reflectance (L8SR) version 0.3.1 was implemented on September 28, 2015, and included changes to how the climate modeling grid (CMG) is accessed at scene edges, as well as at land/water boundaries. 

Landsat 4-7 Surface Reflectance (LEDAPS) version 2.3.1 was implemented on September 28, 2015, which includes bug fixes to saturation flagging. Saturation was previously incorrectly flagged as 16000, but is now flagged correctly at 20000. 

For both LEDAPS and L8SR, the CFmask was modified to correctly select cloud heights for shadow projection. Previously, the ideal cloud height was not being selected.

May 28, 2015 - Surface Reflectance Processing of new Landsat 8 scenes available

Surface Reflectance processing has resumed for Landsat 8 scenes acquired after December 18, 2014. The processing was suspended during the thermal sensor (TIRS) side-b electronics swap and Level 1 reprocessing effort, as described in the May 8, 2015 Calibration Notice.  As noted on the calibration notice webpage, there may be missing data, or scenes that were processed to OLI-only; these scenes will not process to Surface Reflectance.

May 13, 2015 - Update to Provisional Landsat 8 Surface Reflectance (L8SR) Output & CFmask Cloud Confidence Band Addition

Provisional Landsat 8 Surface Reflectance (L8SR) version 0.3.0 was implemented on May 13, 2015 and includes improved land/water masking at coastal boundaries, greatly reducing the number of artifacts found at the edges of coastal features. These improvements are detailed in the “Known Issues” section of the Landsat 8 Surface Reflectance Product Guide.

For Landsat 4-7 Surface Reflectance and Landsat 8 Surface Reflectance products, a provisional CFmask cloud confidence band is now being generated for each scene. The thresholds and attributes of the new band are detailed in the “CFmask Cloud Confidence Band” section found in the Landsat 4-7 Surface Reflectance Product Guide and the Landsat 8 Surface Reflectance Product Guide.

Both LEDAPS and L8SR algorithms now use the thermal constants K1 and K2, earth-sun distance, radiometric gain/bias, and reflectance gain/bias values from the Level 1 metadata file (.MTL) file for computing the Top of Atmosphere (TOA) reflectance and TOA brightness temperatures where hard-coded values were previously used. Prior to certain March 17, 2015, the ETM+ .MTL file will not contain these fields, so the hard-coded values will continue to be used for TOA calculation.

April 16, 2015 - Landsat 8 TIRS Reprocessing Plans

Currently, surface reflectance (SR), top-of-atmosphere (TOA) reflectance, and SR-based spectral indices can be processed for Landsat 8 scenes acquired April 11, 2013 (DOY 101) through December 18, 2014 (DOY 352) only, due to changes made to Thermal Infrared Sensor (TIRS) data. (See the December 22, 2014 and December 24, 2014 Landsat Headlines for more details about the issue.)

After extensive investigations, the TIRS Mechanism Control Electronics was switched from the primary to redundant side on March 2, 2015. The sensor began collected data on March 4, however data collected from this date until March 12 were collected for calibration purposes only. Data collected from March 13 forward have valid TIRS data.

Beginning April 30, 2015, Landsat 8 scenes acquired from December 19, 2014 to March 13, 2015 will be reprocessed to include valid TIRS data. The reprocessing effort is expected to complete by May 18, 2015, after which Surface Reflectance and other high level processing for these scenes will resume. A new announcement will be made on this page when this processing will again become available.

More recent details can be found in the April 16, 2015 Headline. 

January 21, 2015 - Landsat 4-7 Surface Reflectance Fixes

Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) version 2.2.1 was implemented on January 21, 2015 and includes the correction of two issues introduced by LEDAPS 2.2.0 on December 23, 2014:

1) HDF Scientific Data Sets (SDS) Attributes for QA Bands - The fill value for QA bands was eliminated in LEDAPS 2.2.0, but an erroneous fill value was still present in the SDS attributes as -3333, which is used as an internal flag to signify that the fill value does not actually exist. The "fill" attribute has been removed for QA bands in HDF SDS in the new release.

2) Reversed Class Values in QA Bands- Many of the class numbers in the .XML files and the class descriptions in the HDF SDS files were erroneously reversed. These have now been corrected such that “0” definitively indicates a condition has not been met, and “255” indicates that a condition has been met for fill, dark dense vegetation, cloud, cloud shadow, snow, and adjacent cloud.

Note that these changes do not impact the data values, only the attribute descriptions listed within the files and their metadata.

January 15, 2015 – Known Issues Reporting for Provisional Landsat 8 Surface Reflectance Data Available

User feedback regarding the usability of Provisional Landsat 8 Operational Land Imager (OLI) Surface Reflectance (SR) data products is now available in the “Known Issues” section of the Landsat 8 Surface Reflectance Product Guide.

December 23, 2014 - Provisional Landsat 8 Surface Reflectance Data Available

Provisional Landsat 8 Operational Land Imager (OLI) Surface Reflectance (SR) data products are now available through the Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) On Demand Interface and EarthExplorer*.

As currently produced for Landsat 4-5 and Landsat 7 scenes, product options such as top-of-atmosphere (TOA) reflectance, surface reflectance (SR), and SR-based spectral indices are available for Landsat 8 scenes. Please Note: currently, only Landsat 8 scenes acquired from April 11, 2013 (DOY 101) through December 18, 2014 (DOY 352) can be processed to Surface Reflectance, due to the recent change of the Thermal Infrared Sensor (TIRS) acquired data. Please see the December 22, 2014 Headlines for more details.

* Surface Reflectance (SR) data only for individual scenes can be selected using EarthExplorer, under the Landsat CDR Data Sets. These requests will be sent to the ESPA On-Demand interface for processing and data delivery.

It is important to note that Landsat 8 scenes will not be processed with the same algorithm used for Landsat 4, 5 and 7 scenes. A new algorithm has been developed to take advantage of Landsat 8’s unique characteristics.

The L8SR algorithm currently uses the scene center for the sun angle calculation and then hard-codes the view zenith angle to 0. The solar zenith and view zenith angles are used for calculations as part of the atmospheric correction. Future plans may modify both LEDAPS and L8SR to calculate sun angles on a per-pixel basis.

Users are encouraged to review the Landsat 8 Surface Reflectance Product Guide for more details and important information.

December 3, 2014 - Landsat 8 Surface Reflectance Climate Data Record Development Nears Completion

By the end of 2014, Provisional Landsat 8 Operational Land Imager (OLI) Surface Reflectance (SR) Climate Data Records (CDRs) will be available through the Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) On Demand Interface. As currently produced for Landsat 4-5 TM and Landsat 7 ETM+ scenes, product options such as top-of-atmosphere (TOA) reflectance, surface reflectance (SR), and SR-based spectral indices will be available for Landsat 8 scenes.

It is important to note that Landsat 8 scenes will not be processed with the same algorithm used for Landsat 4, 5 and 7 scenes. A new algorithm has been developed to take advantage of Landsat 8’s unique characteristics.   The L8SR algorithm currently uses the scene center for the sun angle calculation and then hardcodes the view zenith angle to 0. The solar zenith and view zenith angles are used for calculations as part of the atmospheric correction. Future plans may modify both LEDAPS and L8SR to calculate sun angles on a per-pixel basis.

Users are encouraged to review the Landsat 8 Surface Reflectance Product Guide for more details and important information.

November 3, 2014 - Processing resumes for data acquired after October 17, 2014

Recently, the National Centers for Environmental Prediction (NCEP) climate grid data [which is used by the Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) in deriving surface reflectance (SR) data for Landsat TM (L4-5) and Landsat ETM+ (L7) scenes] changed product formats from netCDF3 to netCDF4.

Due to these changes, LEDAPS was unable to process SR for data acquired after October 17, 2014 (DOY 290). Modifications to the LEDAPS code have been made to accommodate the new NCEP products, and data processing has now resumed. Please note: the reflectance corrections and format of the surface reflectance data products is not changing, and no reprocessing of data products before DOY 290 2014 will be necessary.

October 10, 2014 – Upcoming Landsat 8 Surface Reflectance Data Products

Prototype Landsat 8 Surface Reflectance product options will be available to all users via the Earth Resources Observation and Science (EROS) Center Science Processing Architecture (ESPA) On Demand Interface in the upcoming months.

Landsat 8 will not be processed with the Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) that is used for Landsat 4, 5 and 7 SR data. A new algorithm has been developed to take advantage of Landsat 8’s unique characteristics.

The original L8 SR code was written in FORTRAN and output only SR. The USGS EROS translation to a C version adds options for TOA reflectance and SR-based spectral indices.

All Landsat 8 scenes will be processed to SR products, including pre-WRS2; however, scenes with a solar zenith angle greater than 76 degrees can be corrected only to TOA reflectance.

More details about the upcoming L8 SR data, as well as a better known timeframe of availability will be provided on this site. Relevant documentation will also be made available before public release.

August 19, 2014 – Upcoming changes to ESPA output file names

By the end of August, the "fmask" band's suffix will permanently change to "cfmask" to accurately reflect the algorithm used in ESPA (for example, the cloud mask for the scene LE70390372008210EDC00 will be named "LE70390372008210EDC00_cfmask" instead of "LE70390372008210EDC00_fmask".) Please update any programs or scripts you may be using to reflect this change.

Subsequent changes will be carried out in the XML metadata files:

August 15, 2014 – Addition of MODIS data products to ESPA

By the end of August 2014, the following MODIS data will be available via the ESPA on-demand interface:

MOD and MYD 09 - all Surface Reflectance products (Version 005)

MOD and MYD 13 – all Normalized Difference Vegetation Index (NDVI) products (Version 005)

ESPA’s existing output format, reprojection, extent and pixel resizing options will become available for these products, and can be applied simultaneously with available Landsat 4-5 and Landsat 7 scenes. Production of spectral indices and climate data records (CDRs) cannot be produced from MODIS scenes.

July 21, 2014 – Changes to ESPA Processing and Output options of Surface Reflectance Climate Data Record (CDR) data products

On July 29, 2014, the EROS Science Processing Architecture (ESPA) On-demand interface will be upgraded to incorporate more high-level processing options for Landsat 4/5 TM and Landsat 7 ETM+ Surface Reflectance products. Interface updates include:

HDF: A separate HDF file for each product type requested in the order (i.e. Source Products, Top of Atmosphere Reflectance, etc.) will be returned. Each HDF file will contain a number of science data sets (SDS), which represent the individual bands in that product. Each band is stored as an external SDS, and there will be .img files for each SDS, in addition to the product’s HDF file.

In the case of ETM+ level 1 products, the HDF file will contain 2 grids. The first grid includes the 30-meter Bands 1-7 data. The second grid includes the 15-meter Band 8 data. Because of the differences in spatial resolution, an ENVI header file cannot be written and delivered.

An XML file will be delivered containing global metadata and band metadata for the desired bands.

GeoTIFF: A single GeoTIFF file for each band requested in the order will be returned. An XML file will be delivered containing global metadata and band metadata for the desired bands.

Binary: A single file (.img file and associated .hdr file) for each band requested in the order will be returned. The .img file contains the raw binary image data and the .hdr file contains the ENVI header for that band. An XML file will be delivered containing global metadata and band metadata for all desired bands.

Metadata will be delivered in an Extensible Markup Language (.XML) file. The top of atmosphere (TOA) brightness temperature delivered as Band 6 in the TOA reflectance product will be labeled in Kelvin instead of currently-noted Celsius.

NOTE: Surface Reflectance data products requested from EarthExplorer will be delivered in GeoTIFF format only.

June 26, 2014 – ESPA Login Changes

Users are reminded that as of today, June 26, 2014, the ESPA On-Demand interface now uses the USGS login credentials - the same used to access EarthExplorer.

May 27, 2014 – Upcoming Login Changes to ESPA

During the week of June 23, 2014, the username and password needed to request Landsat Surface Reflectance and Spectral Indices products through the ESPA On-Demand interface will be disabled. Users may instead enter the site with USGS authenticated registration credentials (the same used to access EarthExplorer).

March 28, 2014 – Upcoming Changes to Landsat Surface Reflectance Climate Data Records, Sample Data Available

In the near future, several changes will be made to the processing and output options of SR CDR products. One change will allow users to select from three output formats: HDF, GeoTIFF, or Binary. (Surface Reflectance data products requested from EarthExplorer will be delivered in GeoTIFF format only.) The metadata will be delivered in an Extensible Markup Language (.XML) file. The top of atmosphere (TOA) brightness temperature delivered as Band 6 in the TOA reflectance product will be labeled in Kelvin instead of currently-noted Celsius.

November 30, 2013 –Landsat surface reflectance-derived spectral indices (NDVI, EVI, SAVI, MSAVI, NDMI, NBR, NBR2) are available on demand from ESPA.

July 15, 2013 - Landsat QA Tools Updated

The Landsat QA Tools have been updated to include a bit extraction function to interpret Landsat 8 Operational Land Imager (OLI) Quality Assurance (QA).

July 8, 2013 – Surface Reflectance Available for Landsat 4 TM Scenes

Surface Reflectance products for Landsat 4 Thematic Mapper (TM) are now available from the ESPA On-Demand Interface or EarthExplorer.

July 8, 2013 – Updates to Landsat Surface Reflectance Climate Data Records

The Landsat Surface Reflectance Climate Data Record (CDR) now includes an alternative cloud and cloud shadow mask layer, generated from a C version of Boston University’s Function of Mask (Fmask). 

May 17, 2013 – Landsat Surface Reflectance Climate Data Record Available through EarthExplorer

Landsat Surface Reflectance Climate Data Record (CDR) products are available on-demand for many Landsat 5 and all Landsat 7 scenes through EarthExplorer.

April 30, 2013 – Updated LEDAPS Processing Algorithm 

The final fiscal year 2013 version of the Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS 1.2.1) is available for processing Landsat Surface Reflectance Climate Data Record (CDR) products on-demand. This new version will process Landsat scenes acquired over Antarctica, handling the Polar Stereographic map projection. 

March 27, 2013 – Additional Options Added to EROS Science Processing Architecture On-Demand Production Services

EROS Science Processing Architecture (ESPA) on-demand production services now include options to select certain products individually or in packages. Options include surface reflectance, top of atmosphere reflectance, source data, source metadata, and surface reflectance-derived normalized difference vegetation index (NDVI). 

March 27, 2013 – Landsat Ecosystem Disturbance Adaptive Processing System Version 1.2.0 Available

Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) version 1.2.0 is available for on-demand production of surface reflectance data. Major changes include: Polar Stereographic support, application of auxiliary files for atmospheric correction based on scene acquisition time read from the source metadata, and a fix for a 0.5 degree rounding error in sun angle calculations. 

January 16, 2013 – Source metadata included with Surface Reflectance Products

All source information is now included in Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) surface reflectance packages. .

January 3, 2013 –Surface Reflectance Collections Available on EarthExplorer

The Global Land Survey (GLS) collections from 2000, 2005, and 2010 have been processed to surface reflectance and are now accessible through EarthExplorer, from the Landsat CDR/ECV data set.

December 6, 2012 - Surface Reflectance Production beyond April 2012

An updated version of Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS 1.1.1) includes the most recent ancillary data available for atmospheric correction. Requests can now be made for data within one week of acquisition.

November 27, 2012 - LEDAPS 1.1.0 Released

A new version of Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) code has been implemented for Landsat Surface Reflectance processing. Most noticeable will be a new metadata file name and corrected fill/gap labels in the quality assurance (QA) layers. 

October 25, 2012 - New Surface Reflectance Algorithms Released

Landsat Ecosystem Disturbance Adaptive Processing System has been base-lined as LEDAPS 1.0.0. All surface reflectance produced from this date forward will be generated using updated software.

October 15, 2012 - Global Land Survey Collections Processed to Surface Reflectance

Global Land Survey (GLS) collections from 2010 and 2005 have been processed to surface reflectance and are available for discovery and download from the ESPA GLS Visualization Interface. No registration is required.

Optimizations to the LaSRC aerosol retrieval method will reduce Landsat 8 Surface Reflectance data processing times.  Instead of performing aerosol retrieval on a per-pixel basis, the new method performs the retrieval for a 3x3-pixel window. The aerosol value is computed for the center pixel of the window and is used to interpolate the aerosol values for the remaining clear pixels in the window. If the center pixel of the 3x3 window is found to be cloud or water, the algorithm will search the window for the closest non-cloudy, land pixel to compute the aerosol value for the 3x3 window. Any cloud, cloud shadow, or water pixels are assigned the median aerosol value of all 3x3 windows. The SR_AEROSOL QA band has been updated with new bit description .  

Pixels within the PIXEL_QA band will be modified as follows:

  • Pixels flagged as high confidence cirrus (Bits 8-9) will no longer unset the clear bit (Bit 1).
  • Pixels flagged as terrain occluded (Bit 10) will also no longer unset the clear bit (Bit 1).
  • NOTE: Pixels flagged as high confidence cloud (Bits 6-7) will continue to unset the clear bit.
  • Pixels flagged as high confidence cirrus (Bits 8-9) will no longer unset the clear bit (Bit 1).
  • Pixels flagged as terrain occluded (Bit 10) will also no longer unset the clear bit (Bit 1).
  • NOTE: Pixels flagged as high confidence cloud (Bits 6-7) will continue to unset the clear bit.

Optimizations to the LaSRC aerosol retrieval method will reduce Landsat 8 Surface Reflectance data processing times. Instead of performing aerosol retrieval on a per-pixel basis, the new method performs the retrieval for a 3x3-pixel window. The aerosol value is computed for the center pixel of the window and is used to interpolate the aerosol values for the remaining clear pixels in the window. If the center pixel of the 3x3 window is found to be cloud or water, the algorithm will search the window for the closest non-cloudy, land pixel to compute the aerosol value for the 3x3 window. Any cloud, cloud shadow, or water pixels are assigned the median aerosol value of all 3x3 windows. The SR_AEROSOL QA band has been updated with new bit descriptions.

  • Added the option of NetCDF output file format
  • The "blockiness" or image artifacts resulted from inputs to the aerosol retrieval which remained at the global climate model grid resolution versus being interpolated to the Landsat pixel resolution. These updates will eliminate the blockiness issue. A second level of blockiness occurred over areas where aerosol retrieval was not successful. The interpolation of those aerosols resulted in blocks of the same aerosol value. This has also been fixed with the implementation of a better aerosol interpolation method.
  • Image extents are now restricted to 200,000,000 pixels, doubling the previous limit
  • This new version of LaSRC tries to retrieve aerosols over all pixels, except cirrus pixels. The retrievals are then tested based first on the model residual and then on the NDVI combined with Band 5 reflectance. If these tests fail, then the pixel is marked as failed for aerosol retrieval. Any pixel which failed retrieval is ultimately attempted to be interpolated. Water pixels (flagged if the NDVI < 0.01), cloud pixels, and cirrus pixels are not interpolated. The final reflectance corrections are applied using the retrieved/interpolated aerosols.
  • Previous interpolation issues along coastal water bodies led us to implement a land/water mask to better identify coastal waters, since aerosols were not being retrieved over coastal waters and this resulted in significant blockiness along the coastal areas. Given the change in the new version of LaSRC to attempt aerosol retrieval over all pixels, the higher resolution land/water mask has been removed.
  • While interpolation solves the data continuity issues, it introduces additional uncertainty to the reflectance values in areas where aerosols were interpolated. Therefore, we are adding an additional band (unsigned 8-bit integer) with flags to indicate where aerosol measurements were successful, failed (likely water or clouds), or interpolated.
    • CFMask may now be ordered as a standalone product
    • Image extents are now restricted to 200,000,000 pixels, doubling the previous limit
    • Original Input Products (untouched from the source) have been removed as a product option. Input products (formerly Customized Input Products) are still available
  • The "name" variable will be "cfmask" instead of "fmask"
  • The "short_name" variable will be "LXXCFMASK" instead of "LXXFMSK", where LXX could be LT4, LT5 or LE7.
  • The "long_name" variable will be "cfmask_band" instead of "fmask_band"
  • Re-projection options for Polar Stereographic
  • Re-projection and subset functionality for L1T products
  • User-defined output formats for HDF-EOS2 (.hdf) , GeoTIFF (.tif), and Binary (.img):