PaddockTS: paddock-level satellite time series analysis of agroecosystem dynamics
Year: 2026
Type: preprint
Abstract: PaddockTimeSeries (PaddockTS) is an open-source data analysis pipeline that aligns Earth observation data with agricultural fields, or paddocks—the core land management units where agricultural decisions are made and outcomes such as yield, phenology, productivity, and soil condition are recorded. PaddockTS enables scalable agricultural monitoring and research on how land management and environmental variation shape agroecosystem dynamics. For a user-defined location and date range in Australia, PaddockTS delineates paddock boundaries from Sentinel-2 time-series imagery using geospatial segmentation or accepts user-provided boundaries, aggregates Sentinel- 2 satellite time series within them, and returns analysis-ready datasets and graphics describing vegetation dynamics and derived seasonal features. It also retrieves environmental covariates, including terrain, soil, weather, and water-balance variables. By summarising vegetation dynamics at the paddock level, PaddockTS makes Earth observation data easier to integrate with management records, environmental conditions, and agricultural and ecological outcomes. Its outputs enable comparisons across paddocks, years, and environmental gradients and provide analysis-ready inputs for agroecosystem modelling and machine-learning prediction tasks. PaddockTS can be used as a programmable Python workflow for reproducible research or through a web interface for exploratory analysis without coding.
DOI: 10.31223/x5821z
Language: English
Subfield: Ecology
Field: Environmental Science
Domain: Physical Sciences
SDGs: Life on land, Zero hunger
Open Access status: green
OpenAlex ID: W7205709378