AGRI-PHENOPY: AN ALGORITHM FOR EXTRACTING AGRICULTURAL CROPS PHENOLOGY METRICS FROM VEGETATION INDEX TIME SERIES

We developed a Python-based algorithm to extract phenological metrics from vegetation index time series obtained from areas of different agricultural crops. The algorithm can be used with vegetation index time series from various sensors, with different spatial and temporal resolutions. It corrects abrupt drops in the time series, smooths the time series using a double logistic function, and extracts four phenological metrics: Start of Season, Peak of Season, End of Season, and Length of Season. In this article, we describe the algorithm and use Normalized Difference Vegetation Index as an example to compare the phenological metrics, although it can be applied to different indices. The results highlight the efficiency of our algorithm in extracting phenological metrics from vegetation index time series of various agricultural crops. Our algorithm is designed to simplify the extraction of phenological metrics and enable the fast and efficient processing of large data volumes.