Why we need to validate land surface phenology products: an update from the CEOS Land Product Validation subgroup

Currently, the values of land surface phenology matrices derived from satellite data are highly variable. Two major reasons for the observed differences are the satellite-based sensor used and the algorithm employed to extract the phenological matrices. However, there are some specific accuracy requirements for the use of these land surface phenology products in ecosystem models. This implies the importance of a user understanding the uncertainty and discrepancies between different phenology products in relations to that user’s intended use of the product(s). In this talk we will review the list of currently available satellite-based phenology products as well as a summary of primary applications. We will then describe the current work of the phonology working group within the Committee on Earth Observing Satellites (CEOS) Land Product Validation (LPV) sub-group. The mission of CEOS-LPV is to foster quantitative validation of higher-level global land products derived from remote sensing data and relay results that are relevant to users. There is now an international consensus to establish a set of “core sites” where future remote sensing phenology validation efforts will focus. The talk will present information on these sites, initial data for these sites, and plans on utilizing these sites to initiate an international validation-based remote sensing phenology product inter-comparison. The talk will conclude with a description of how such an inter-comparison can help inform the land surface phenology user community and would lead to quantifying uncertainty in these phenological matrices.