Surface Type
Team Lead: Xiwu (Jerry) Zhan
- Surface Type ATBD, (PDF, 783 KB)
- EDR Long Term-Monitoring: Surface Type
Background
The accurate description of actual terrestrial surface types from regional to global scales is essential information or input for many science and societal applications: land surface models for numerical weather, hydrological and ecological predictions, land and water resources management, implementation of national and international policies related to bio-diversity and climate change, and global change studies as input to climate and biogeochemical models.
Product History and Product Access
Since launch of S-NPP in October 2011, the official JPSS IDPS ST EDR was archived at CLASS Simultaneously, the JPSS VIIRS Surface Type Team at STAR and University of Maryland produced a global surface type intermediate product (global land cover map) from VIIRS, using the MODIS heritage C5.0 decision tree classification algorithm. In February 2013, the ST EDR product has reached beta maturity in IDPS. In July and December 2014, the product has reached provisional and validated 1 maturity, respectively. The most recent version of the Global Surface Type Map is available from S-NPP_VIIRS_GST_IGBP_2014. For more information please contact Rui Zhang or Xiwu Zhan.
Users
Earth System Modelers: Land surface parameterization for GCMs (e.g.: Global Forecast System (GFS) and North America Mesoscale (NAM) Model of NWS-NCEP-EMC; Goddard Earth Observing Model (GEOS) system of NASA; The Weather Research and Forecasting (WRF) Model and Community Earth System Model (CESM) of National Center for Atmospheric Research (NCAR)) and for Hydrological processes (e.g. the Community Hydrologic Prediction System – CHPS of NWS Office of Hydrological Development)
Carbon and ecosystem studies: Carbon stock, fluxes, biodiversity
Downstream products in IDPS of JPSS: Directly - Land surface temperature; Indirectly - Cloud mask, aerosol products, other products requiring global land/water location information
Example of ST EDR product
Example of Global ST Intermediate Product (IP)
Validation
The objective of ST validation is to ensure the overall classification accuracy of intermediate product aka global land cover map to meet the level 1 product requirement. The validation also checks the implementation of the included snow/ice, fire and vegetation quality flags in the EDR data. The initial validation was performed using visual comparisons against high resolution images by Boston University team. The validation results suggested the VIIRS surface type intermediate product is compatible with the heritage MODIS land cover product, and the classification accuracy is very similar to the result of MODIS land cover. A recent validation was conducted by the University of Maryland team for the validated 1 stage maturity review. The validation results concluded that 73.92% of classification accuracy was obtained in the intermediate product, which exceeded the required 70% threshold in the level 1 requirement document. More details about the recent validation efforts could be found here.
Ongoing improvements
The Surface Type team is tasked to generate the Global Surface Type Map annually so that Surface Type EDR can describe both long term and short term changes of the land surface using the most recent S-NPP and future JPSS observations. New improvements for the global surface type map, including new support vector machine based classification algorithm, new training samples, and new annual classification metrics, will be implemented for next delivery. Surface Type EDR will be improved with new algorithm to identify short time surface type changes such as flooded areas, burned areas and deforested areas as well as the active fire and snow covered areas. More comprehensive validation of the Global Surface Type Map will be continuously conducted.
