Authors: Xiaoyang Zhang*, , Jiamin Wang, South Dakota State University
Topics: Land Use and Land Cover Change, Remote Sensing, Biogeography
Keywords: Land Surface Phenology, Multiple Spatial Resolution, satellite data
Session Type: Paper
Start / End Time: 1:10 PM / 2:50 PM
Room: Balcony B, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Land surface phenology (LSP) has been widely characterized AVHRR, SPOT-VEGETATION, and MODIS data. Due to the coarse spatial resolution, the remote sensing monitors the seasonal dynamics of the land surface that often consists of multiple types of vegetation mixed with other scene objects, such as soil, water, and human structures. It is hypothesized that LSP can only reflect the phenological dates associated with a certain amount of plants within a coarse pixel and that different LSP events throughout a growing season reflect the progress of different vegetation types within a pixel. To explore this hypothesis, we first detect LSP from time series of Landsat 8 and Sentinel-2 data in 2017 at a 30m pixel for three ecosystems that are forests, croplands, and shrublands across the United States. Specifically, a time series of two-band enhanced vegetation index (EVI2) is calculated from a consistent, harmonized surface reflectance product generated from Landsat-8 Operational Land Imager (OLI) and Sentinel-2 Multi Spectral Instrument (MSI) observations. The EVI2 time series are then used to detect four phenological transition dates (greenup onset, maturity onset, senescence onset, and dormancy onset). The 30m phenological dates are further spatially matched with 500m LSP derived from the Visible Infrared Imaging Radiometer Suite (VIIRS) in 2017. Furthermore, 30m LSP timings in a 500m VIIRS pixel are stratified for different vegetation types based on 30m national land cover dataset. Finally, the variation of 30m phenological dates within 500m pixels across various vegetation types is explored for the four phenological transition dates, respectively.