DATimeS

DATimeS reconstructs discontinuous optical remotely sensed time series and quantifies vegetation phenology, including Leaf Area Index (LAI), by combining time series interpolation methods and advanced machine learning such as Gaussian Process Regression (GPR) for gap-filling.

Key Features:

• Advanced fitting algorithms: Incorporates advanced fitting algorithms, including Gaussian Process Regression (GPR), to reconstruct vegetation temporal patterns and provide uncertainty estimates.
• Integration of interpolation and machine learning: Integrates advanced machine learning algorithms into established time series interpolation methods for robust gap-filling.
• Cloud-free composite maps: Generates cloud-free composite maps from satellite time series to provide uninterrupted observations of vegetation dynamics.
• Regular and irregular time series support: Processes both regular and irregular satellite time series datasets.
• Phenological indicators (LAI): Quantifies Leaf Area Index (LAI) fluctuations across multiple crop seasons to derive phenological indicators per crop type.

Scientific Applications:

• Crop phenology characterization: Enables characterization of crop phenology using optical remotely sensed data through accurate gap-filling and temporal reconstruction.
• Agricultural research: Supports agricultural research by quantifying LAI dynamics and seasonal phenological indicators.
• Environmental monitoring and ecology: Supports environmental monitoring and ecological studies by providing continuous time series and cloud-free composites.
• Sentinel-2 LAI case study: Demonstrated on Sentinel-2 Leaf Area Index time series over a site in Spain to reconstruct vegetation patterns and phenological metrics.

Methodology:

Uses time series interpolation methods and advanced machine learning algorithms, including Gaussian Process Regression (GPR), for gap-filling and reconstruction; generates cloud-free composite maps; processes regular and irregular satellite time series; and quantifies Leaf Area Index (LAI) across seasons.