Processing Baseline

The processing baseline GIPP (PROBAS- L1C, PROBA2 - L2A) are configuration files that evolve over the Mission lifetime. This evolution is done to manage the introduction of new product features or processing steps (e.g. introduction of the geometric refinement), and the correction of errors in the processing chain.

The evolutions of the processing baseline are tracked and justified in the 'Processing Baseline Status' section of the Data Quality Report that is published on a monthly basis by the MPC/CC and available from the Sentinel-2 Document Library.

The relationship between each PROBAS, the XSD, and the Product Specification Document (PSD) is set out in the S2 User Guide.

Level-1C processing
Processing Version	Date Introduced	Reason
02.00	23 November 2015	Initial product Baseline
02.01	15 December 2015	updated to accommodate changes in the Tiling Grid configuration file
02.02	3 May 2016	to correct an issue identified in which anomalous pixels with incorrect value are present in the overlap of datastrips. This occurred in only a few products.
02.03	9 June 2016	Affected by a strong spectral mis-registration anomaly, and withdrawn.
02.04	15 June 2016	To replace 02.03
02.05	27 April 2017	A fix introduced to the satellite ancillary metadata, to correct instances where the value of Instrument Measurement Time (IMT) is not represented correctly.
02.06	23 October 2017	An update of the ‘Product Discriminator’ component of the product name and filename, and the product-level Metadata file in the structure holding the quality control report summary references.
02.07	6 November 2018	The crenellation effect at the edge of images was significantly reduced by estimating the inverse locations for grid points lying just after the edge of the acquisition. The accuracy of the tile sensing time reported in the tile metadata was improved by computing for each tile the average acquisition time of all relevant pixels. The new tile sensing time will be reported in the field General_Info/ SENSING_TIME of the Tile Metadata without format change. The detector footprint vector files provide the precise location of the boundary between detectors, without overlap. This improvement will enable Users to have an accurate estimation of the viewing direction for each pixel. The correlated across-track noise on very dark images was significantly reduced thanks to an improved radiometric correction algorithm.
02.08	8 July 2019	Correction of the formatting to the instrument telemetry of the Datastrip metadata (FEE Temperature Value and GPS_Time date).
02.09	4 February 2020	Improvement of the NO DATA mask accuracy Improvement ECMWF GRIB reference time accuracy Correction of a sporadic anomaly on product footprint Removal of the spurious viewing angle information for detectors not present in the image.
03.00	30 March 2021	Activation of the Geometric Refinement. Use of the new Copernicus Digital Elevation Model (COP-DEM) at 90 m, with corresponding improvement of the geolocation performance over mountainous areas in particular, replacing the Planet DEM90. Improvement of the Level-1C cloud detection algorithm over high-altitude terrains Improvement of the Level-1C Technical Quality (TECQUA) mask accuracy Additional On-Line Quality Control checks applicable to the geometrically refined products.
03.01	30 June 2021	Increment to introduce: 1. A minor modification of the Datastrip metadata, to harmonise them with Refined products, by adding processing_specific_parameters fields for unrefined products. 2. Update to the algorithm used to derive the checksum - from MD5 to the more resilient SHA3 (Secure Hash Algorithm 3) with SHA3-256 instance hash codes.
04.00	25 January 2022	1. Correction of the radiometric bias between SENTINEL-2A and SENTINEL-2B (keeping SENTINEL-2A as reference). 2. Generation of the Level-1C quality masks in raster format. The masks will be grouped in 3 categories as defined as follow: MSK_CLASSI: 60 m classification masks (i.e., the cirrus and opaque clouds). MSK_QUALIT: For each spectral band, and at the same spatial resolution as that band, a raster file composed of all the radiometric and technical quality masks. MSK_DETFOO: for each spectral band, and at the same spatial resolution as that band, a 1-byte raster file using 4 bits to encode the 12 detector footprints. Rasterization of MSK_CLASSI and MSK_QUALITY will be implemented as single bit “multi-band” raster files. 3. Addition of ECMWF (European Centre for Medium-Range Weather Forecasts) parameters: u/v wind components (10u and 10v) and relative humidity (r). The format of the AUX_ECMWFT file will remain unchanged. 4. AUX_CAMSFO: In AUX_DATA folder embedded in the Level-1C products, in a single dedicated GRIB-formatted file named as AUX_CAMSFO, the following CAMS (Copernicus Atmosphere Monitoring Service) forecast data for that specific Tile: Total Aerosol Optical Depth at 550 nm (aod550), Surface Geopotential (z), Black Carbon Aerosol Optical Depth at 550 nm (bcaod550), Dust Aerosol Optical Depth at 550nm (duaod550), Organic Matter Aerosol Optical Depth at 550 nm (omaod550), Sea Salt Aerosol Optical Depth at 550 nm (ssaod550), Sulphate Aerosol Optical Depth at 550 nm (suaod550), Total Aerosol Optical Depth at 469 nm (aod469), Total Aerosol Optical Depth at 670 nm (aod670), Total Aerosol Optical Depth at 865 nm (aod865) and Total Aerosol Optical Depth at 1240 nm (aod1240). 5. Addition of Level-1C snow/ice mask: A new SNOICE raster mask in JPEG2000 format. The snow/ice mask will be included in the MSK_CLASSI. The snow percentage will also be added into the Quality_Indicators_Info section of the L1C Tile metadata. 6. Provision of negative radiometric values (implementing an offset):A radiometric offset will be added up to the image reflectance at Level-1C. The dynamic range will be shifted by a band-dependent constant, i.e. RADIO_ADD_OFFSET. From the user’s point of view, the L1C Top of Atmosphere (TOA) reflectance (L1C_TOA) shall be retrieved from the output radiometry as follows: Digital Number DN=0 will remain the “NO_DATA” value For a given DN in [1;2¹⁵-1], the L1C TOA reflectance value will be: L1C_TOAi = (L1C_DNi + RADIO_ADD_OFFSETi) / QUANTIFICATION_VALUEi The offset will be reported in a new field in the General_Info/Product_Image_Characteristics section of the Datastrip and User Product Metadata. This evolution allows avoiding the loss of information due to clamping of negative values in the predefined range [1-32767] occurring over dark surfaces. 7. Addition of the DOI (Digital Object Identifier) in the Level-1C metadata: A DOI is a string of numbers, letters and symbols used to permanently identify a document and/or a dataset and to create a link to it on the web. The DOI URL will be reported in a new field in the General_Info section of the Datastrip and User Product Metadata.
05.00	Sentinel-2 Collection-1	The Processing Baseline identifier 05.00 tags every Sentinel-2 Collection-1 product issued from the reprocessing activity. As part of the Sentinel-2 reprocessing campaign, the improvements introduced in recent processing baselines 04.00 deployed on 25 January 2022 is generalised to the historical archive. Furthermore, the PB 05.00 uses of the new 30 m Copernicus Digital Elevation Model (DEM), with corresponding improvement of the geolocation performance over mountainous areas in particular. More details can be found here.
05.09	06 December 2022	Improvement of quality mask for radiometric saturation. The Processing Baseline identifier 05.09 tags the operational products generated with the same operational processor used to generate the Copernicus Sentinel-2 Collection-1 but with the 90 m version of the Copernicus DEM (the ‘9’ suffix expresses here the use of the 90m DEM).
05.10	11 December 2023	Use of the new Copernicus https://doi.org/10.5270/ESA-c5d3d65 (COP-DEM) at 30 m, with corresponding improvement of the geolocation performance over mountainous areas in particular, replacing the Copernicus DEM at 90m.

The Processing Baseline identifier 05.00 tags every Sentinel-2 Collection-1 product issued from the reprocessing activity. As part of the Sentinel-2 reprocessing campaign, the improvements introduced in recent processing baselines 04.00 deployed on 25 January 2022 is generalised to the historical archive. Furthermore, the PB 05.00 uses of the new 30 m Copernicus Digital Elevation Model (DEM), with corresponding improvement of the geolocation performance over mountainous areas in particular. More details can be found here.

Level-2A processing
Processing Version	Date Introduced	Reason
02.07	26 March 2018	Initial product Baseline
02.08	23 May 2018	Update of the quality control reports embedded in the L2A products (referring to ‘OLQC' reports in the product specifications). Change in the coding of the section of the Product Level Metadata, in order to group all band image files of the same tile in the same record.
02.09	8 October 2018	Improvement of the accuracy of Scene Classification over water, urban and bare areas thanks to the usage of the Climate Change Initiative (CCI) Land Cover data. Improvement of the spatial homogeneity of surface reflectance between adjacent tiles.
02.10	6 November 2018	See L1C 02.07 (above)
02.11	21 November 2018	Improvement of the terrain correction algorithm in mountainous areas, in order to avoid artefacts for pixels under semi-transparent clouds.
02.12	6 May 2019	Improvement of the terrain correction algorithm, in order to avoid adjacent tile discontinuities in terrain correction for seemingly flat areas with gentle slopes.
02.13	8 July 2019	Correction of the formatting to the instrument telemetry of the Datastrip metadata (FEE Temperature Value and GPS_Time date).
02.14	4 February 2020	See L1C 02.09 (above)
03.00	30 March 2021	Correction of a light halo observed along Level-2A image boundaries, either at the edge of the swath or at the end of data-strips Improvement of the Level-2A cloud probability layer near the boundary of the swath Improvement of the topographic correction thanks to the use of the Copernicus 90 m DEM
03.01	30 June 2021	Increment to introduce: 1. Minor modification of the Datastrip metadata, to harmonise them with Refined products, by adding processing_specific_parameters fields for unrefined products. 2. Update to the algorithm used to derive the checksum - from MD5 to the more resilient SHA3 (Secure Hash Algorithm 3) with SHA3-256 instance hash codes.
04.00	25 January 2022	In addition to the first five above-mentioned evolutions inherited from the corresponding Level-1C product, the following L2A evolutions will be included: Provision of negative radiometric values (implementing an offset): the dynamic range will be shifted by a band-dependent constant, i.e. BOA_ADD_OFFSET. From the user’s point of view, the L2A Surface Reflectance (L2A SR) shall be retrieved from the output radiometry as follows: Digital Number DN=0 remains the “NO_DATA” value For a given DN in [1;2¹⁵-1], the L2A Surface Reflectance (SR) value will be: L2A_SRi = (L2A_DNi + BOA_ADD_OFFSETi) / QUANTIFICATION_VALUEi The radiometric offset value will be consistent with the one applied to the Level-1C product and will be reported in a new field in the General_Info/ Product_Image_Characteristics section of the Datastrip and User Product Metadata. It is also noted that the percentage of negative surface reflectance pixels per band will be also reported in the L2A_QUALITY report in the QI_DATA folder of the tile. Provision of Band 01 at 20 m spatial resolution: Band 01 will be added with 20 m resolution (in the 20 m resolution image folder) in addition to the native 60 m resolution. Addition of Level-2A Quality Indicators (new fields in the tile metadata and additional xml L2A_QI_report in the QI_DATA folder): New Quality Indicators CLOUDY_PIXEL_OVER_LAND_PERCENTAGE, AOT_RETRIEVAL_METHOD, GRANULE_MEAN_AOT, GRANULE_MEAN_WV, OZONE_SOURCE, and OZONE_VALUE, will be added in the Level-2A tile metadata in the Quality_Indicators_Info section. In addition, Quality Indicators providing information on the Scene Classification, Atmospheric Correction and Auxiliary Data will be reported in a new L2A_QUALITY.xml report in the QI_DATA folder of the tile. Aerosol correction using CAMS auxiliary data: The Level-2A processor will use the CAMS Total Aerosol Optical Depth (AOD) at 550 nm when Dark Dense Vegetation (DDV) pixels are missing in the image. This improvement will mitigate the chess-board effect that is sometimes visible between tiles. Improvement of the bright target classification over coastal areas: The cloud detection over coastal areas of bright targets (e.g. buildings and bright sand) is improved to mitigate against false detections. Improvement of the topographic and casted shadows: The class “dark features and shadows” (SCL = 2) is replaced by “topographic and casted shadows” (keeping code SCL = 2), while dark features pixels will be re-assigned to the “unclassified” class (SCL = 7). Improvement of clouds and cloud shadows detection: Thanks to the use of the Sentinel-2 MSI parallax properties between bands B08 (resampled at 20m) and B8A (20m native resolution), the accuracy of the cloud detection algorithm has been improved, reducing the false clouds detection of bright targets - especially in urban areas - as well as reducing the snow false detection in high altitude clouds. Also, the cloud shadow algorithm benefits from using the parallax properties, by retrieving an estimate of the cloud top-height to improve the localization of clouds shadows on ground. Addition of the DOI (Digital Object Identifier) in the Level-2A metadata: Similarly, to Level-1C product, a DOI will be reported in a new field in the General_Info section of the Datastrip and User Product Metadata.
05.00	Sentinel-2 Collection-1	The Processing Baseline identifier 05.00 tags every Sentinel-2 Collection-1 product issued from the reprocessing activity. As part of the Sentinel-2 reprocessing campaign, the improvements introduced in recent processing baselines 04.00 deployed on 25 January 2022 is generalised to the historical archive. Furthermore, the PB 05.00 uses of the new 30 m Copernicus Digital Elevation Model (DEM), with corresponding improvement of the geolocation performance over mountainous areas in particular. More details can be found here.
05.09	6 December 2022	Identification of defective pixels from missing instrument source packets in L2A Scene Classification map The Processing Baseline identifier 05.09 tags the operational products generated with the same operational processor used to generate the Copernicus Sentinel-2 Collection-1but with the 90 m version of the Copernicus DEM (the ‘9’ suffix expresses here the use of the 90m DEM).
05.10	11 December 2023	Use of the new Copernicus https://doi.org/10.5270/ESA-c5d3d65 (COP-DEM) at 30 m, with corresponding improvement of the geolocation performance over mountainous areas in particular, replacing the Copernicus DEM at 90m.

News

Menu Display

Go to