MODAPS PCR VNP 22-005

PCR Number VNP 22-005
Date 2022-08-26
Initiator Sudipta Sarkar, Carol Davidson
Abstract All VIIRS products--
For all science changes, refer to section 8 above.
Problem Statement Generate improved version of the C2 SNPP and JPSS1 (J1) VIIRS land products through Collection 2 reprocessing, using the C2 version of L1B/Geolocation products as input (C2.1 for J1), along with land product specific changes recommended by the VIIRS land science teams. This particular PCR is for Tier1 set of products that are scheduled to go in first for reprocessing.
Description of Change Key C2 specific land product science changes and updates, other than routine bug fixes, are summaried below:
  • Implementation of the L1B cross-calibration for selected RSB VIS/NIR and SWIR bands to mitigate the bias observed in the TOA reflectance values between SNPP and J1 and to constrain them within 1%, using MODIS Aqua as the reference.
  • Use these new cross-calibrated (CC) LIB inputs for all downstream L2 products.
  • Start using the 8-calss land-water-mask (LWM) from the V**03 geolocation products, instead of the 5-class LWM from the cloud mask. This will be in line with what has been done for MODIS processing all along.
  • Cloud Mask - Improve cloud mask performance thorugh the use of 1km rolling gridded snow inputs and replcing the 5km static VI with 1km VI files. Also made some algorithm improvements for better delineation of snow/sea-ice over the higher latitudes.
  • LSR -
  • Propagate the LWM from the upstream geolocation and create a new ?land_water_mask? layer in the L2 product for possible use by the downstream products.
  • Use improved version of the algorithm for flagging of aerosol impact, a change implemented in the C61 MODIS reprocessing
  • Addition of a new secondary daynight flag to be used by the downstream LSR CMG code and for better retrieval of CMG values over all daytime pixels. Also updated some SDS attributes to make them more consistent with other data products.
  • Snow -
  • Reduced snow commission errors in L2 snow, through the use of a new SWI (Snow Water Index) and addition of a new basic QA value to mark pixels that were screened for false snow.
  • Addition of daily snow CMG product.
  • Fire - Bug fixes in the 375m snow algorithm to avoid exceeding array bounds and causing segmentation faults. Some other bug fixes were also made along with this.
  • LST -
  • Introduction of the daily, 8-day and monthly LST CMG products.
  • Switch from using MERRA2 in C1 to GEOS5 inputs in C2.
  • Retrieve LST over inland water bodies by including pixels designated as inland water bodies.
  • VI -
  • Correct QA VIUsefulness bits to ignore BRDF flag.
  • Use a VI only dedicated LWM developed by SCF.
  • IST - Addition of new daily V[N|J][P|1]30P1D/N daily global products.
  • Sea-Ice - Addition of new daily V[N|J][P|1]29P1D daily global products.
  • VNP04ANC - Change the format of the VNP04ANC products from HDF4 to NetCDF4/HDF5.
Products Affected VIIRS land products
Software Affected VIIRS land products
Processing String to Receive the Change S-NPP Forward Processing. S-NPP Reprocessing. J1 Forward Processing. J1 Reprocessing.
Downstream Product Effects All VIIRS land products
Data Granules to be Used for Science Testing Series of product-specific science tests and land chain-tests have been run generating global data from winter and summer period and have been reviewed by LDOPE and the respective science teams.
Justification Transitioning to C2 variations will help address several critical issues like:
1. Benefit from all the improvments made in the C2 NASA L1B/Geolocation products.
2. Making J1 land data products available to the users for the first time since the launch of mission.
3. Ensuring NASA EOS land data continuity through MODIS Aqua comparable SNPP and J1 land products.
4. Use of a consistent and improved MODIS compliant 8-class LWM across all land data products, except VI.
5. Uses improved cloud mask from use of daily VIIRS snow and high resolution climatology VI replacing the 25km NISE data and 5km Climatology VI used in C1 reprocessing that resulted in blocky artifact with noticeable performance degradation in downstream products.
6. Make a number of product specific improvements and bug fixes across the board for different land products
Effective Date for Implementation of Change PGEs noted in attached document are expected to go for C2 reprocessing for both SNPP and J1, from 2018001, by 08/31/2022 on Ops7. A second stream of reprocessing, with the same set of PGEs and from the beginning of the SNPP mission will follow this on Ops4, by September, 2022.
Status Major algorithm development completed. Sadashiva Devadiga approved 09/14/2022. Sudipta Sarkar approved 09/14/2022.