MODAPS PCR 15-001

PCR Number 15-001
Date 2015-01-12
Initiator Sadashiva Devadiga
Abstract C6 Land PGEs Global PCR.

Product Suite: Surface Reflectance --Polarization correction, de-trending and gain adjustments of Terra and Aqua L1B --Improvement to the aerosol retrieval and correction algorithm, and use of new aerosol retrieval LUTs--Refinements to the internal snow, cloud, and cloud shadow detection algorithms. Uses BRDF database to better constraint the different thresholds used.--Process ocean bands and provide QA datasets for these bands--Improved discrimination of salt pans from cloud and snow and flag salt pan in band QA.

Product Suite: Thermal Anomalies and Active Fire--Refinements to internal cloud mask, which sometimes flags heavy smoke as clouds--Fix for frequent false alarms in the Amazon that are caused by small (~1 km2) clearings within forests--Fix to correct a bug that causes incorrect assessment of cloud and water pixels adjacent to fire pixels near the scan edge--Detect small fires using dynamic thresholding--Process ocean and coastline pixels to detect fire from oil rigs.

Product Suite: Sea-ice --No change to the science algorithm.

Product Suite: Land Surface Temperature and Emissivity--Remove cloud-contaminated LSTs not only from L3 LST products but also from L2 LST products.--Update the coefficient LUT for the split-window algorithm with comprehensive regression analysis of MODIS simulation data in bands 31 and 32 over wide range of surface and atmospheric conditions, especially extending the upper boundary for (LST -- Ts-air) in arid and semi-arid regions and increasing the overlapping between various sub-ranges in order to reduce the sensitivity of the algorithm to the uncertainties in the input data (i.e., column water vapor and air surface temperature from MxD07).--Make minor adjustments in the classification-based surface emissivity values, especially for land-cover type of bare soil and rocks.--Tune the day/night algorithm by adjusting weights to improve its performance in desert regions where the incorporated split-window algorithm may not work well.--Generate new LST products for 8-day and monthly at 6 km grids (in response to user community requests).

Product Suite: Vegetation Indices--Generates 16-day composite VI using two 8-day composite reflectance product (MxD09A1) in the 16-day period.--This Surface Reflectance Input is based on the Minimum Blue compositing approach used to generate the 8 day LSR product--Product format is consistent with the C5 version of the product generated using the L2G daily surface reflectance product.--Update to the Long term Global CMG Average Vegetation Index Product database used to fill the gaps in the CMG product suite

Product Suite: BRDF/Albedo--Generated daily, still using the 16-day composite algorithm.--Observations are weighted to estimate the BRDF/Albedo at the first day of the 2nd 8-day period of the 16-day period.--Uses improved back up database -- pixel based updated from latest full inversion as opposed to the land cover based DB used in C5.--Improved quality and more retrieval at high latitudes from use of all available observations. C5 used only four observations per day.--Uses current day snow status instead of the majority snow/no-snow status from the 16-day period.--CMG Albedo is estimated through retrieval of albedo using all the clear sky observations within 1 km grid for MCD43D as opposed to aggregating from 500 m albedo--Uses L2G-lite surface reflectance as input

Product Suite: LAI/FPAR--Uses daily L2G-lite surface reflectance as input as opposed to MODAGAGG used in C5--Products are generated at native resolution of 500 m--Uses improved multi-year land cover product

Product Suite: GPP/NPP--Uses updated BPLUT and updated version of the daily GMAO--Generate product at native resolution of 500 m using the 8-day composite LAI/FPAR at 500 m resolution

Product Suite: Gridding and L2G products--Better geolocation accuracy

Product Suite: Snow Cover--Generates C5 snow as input to the downstream processes: L2 snow for LST and L3 daily snow for BRDF/Albedo

Product Suite: Burned Area Intermediate--Improvement to the quality filtering algorithm and additional quality flags

Product Suite: VCF Intermediate--No changes

Product Suite: L1 Subsets--AERONET and FLUXNET subset updated to include additional sites--New set of subsets generated for over 400 sites called the "BELMANIP" sites.

Product Suite: GIBS and WMTS Imagery--New in C6 to support generation of imagery for GIBS and WMTS--L2 granules and L3 tiles are projected into tiles in geographic projection, before converting to images using the color LUTs provided by the science team.

Problem Statement Need to update the land PGEs to include code changes to address known software bugs and algorithm issues in the operational C5 version of the science processing algorithms, and to implement additional algorithm changes and improvements proposed by the science team for use in the C6 reprocessing of the MODIS Terra and Aqua land data products.
Description of Change Update the land PGEs in the first tier of reprocessing to include changes to fix known issues from the C5 version of the algorithms and implement the C6 improvements to the generating algorithms proposed by the science team -- see the attached document for product list, PGE versions generating the product and summary of algorithm changes.
Products Affected All products in the first tier of reprocessing-- see that attached document for list of products
Software Affected PGE: All products in Tier 1 Block. Process: All processes. Production Rules. LUTS: BP-LUT: where applicable. DAAC Ingest. Loader: All. Input File.
Processing String to Receive the Change Terra Forward. Terra Reprocessing. Aqua Forward. Aqua Reprocessing
Downstream Product Effects Most products will change due to changes in their inputs and/or their science
Data Granules to be Used for Science Testing Ran science tests to verify product specific changes and chain tests to test and evaluate the downstream effect from changes to one or more upstream land PGEs, generating global products for desired period of time such as 8 day, 16-day, months and a year. Science tests were also run from two different seasons, summer and winter.2010-059
Justification List of changes and improvements to the suite of the MODIS Land products to be generated in the C6 reprocessing was compiled based on the review of the C5 operational products by the members of the MODIS Land science teams and LDOPE, and feedback from the land data product user community. These algorithms changes include changes in scientific approach resulting in significant improvement in product quality, change in temporal and spatial resolution of the product, and change in the product format and sometime even addition of new products to the product suite. Implementation of these changes in the operational processing is expected to introduce spatiotemporal changes in the dynamic range of the product suite and may result in minor impacts on prior results based on C5.
Effective Date for Implementation of Change Completed.
Status Major algorithm development completed. Michael King approved on 02/11/2015.