Authors: Michael Li*, University of Illinois at Urbana Champaign, Chunyuan Diao, University of Illinois at Urbana Champaign
Topics: Remote Sensing
Keywords: Remote Sensing, Crop Phenology, Agriculture
Session Type: Virtual Paper
Start / End Time: 3:05 PM / 4:20 PM
Room: Virtual 17
Presentation File: No File Uploaded
Crop phenology plays a critical role in understanding crop seasonal growing dynamics in response to environmental and management factors. Recent developments in remote sensing have created new possibilities to study crop phenology using various types of remote sensing data. Particularly, the availability of near-surface and high-resolution satellite imageries facilitate the characterization of crop phenological developments for precise agricultural development. The objective of this research is to investigate the capabilities of both near-surface and high-resolution satellite time series in monitoring crop phenological developments at fine spatial scales. The study sites include all the corn sites and three non-corn sites from the PhenoCam network. The high-resolution satellite images were obtained through Planet, and the near-surface images were obtained through the PhenoCam network. Based on our previously developed crop phenological monitoring framework, we conducted time series phenological filtering, time series phenological modeling, and time series phenological characterization to detect two critical phenophases (i.e., start of season and end of season) of several crop species using both PhenoCam and Planet time series. We also evaluated the consistency between the detected phenophases using the visual phenological observations. The results indicated that there are strong relationships between visually assessed phenophases with both PhenoCam and Planet derived phenophases of crops, which validated the feasibility of using both near-surface and high-resolution satellite remote sensing to study crop phenology at fine spatial scales. The results show great promise to advance precise and site-specific agricultural monitoring, particularly in spatially heterogeneous agricultural landscapes and small-holder agricultural systems.