High-Quality Crop Phenology NDVI Time Series from Copernicus HR-VPP

This dataset contains high-quality NDVI time series derived from the “Copernicus High Resolution Vegetation Phenology and Productivity” (HR-VPP) product for extensive irrigated crops located in two major Mediterranean irrigation areas in northeastern Spain (Monegros and Zaidín, Aragón).
The dataset covers six growing seasons (2018–2023) and includes 1,673 agricultural plots representing six cropping systems: barley, wheat, peas, maize monoculture, maize double cropping, and sunflower double cropping.
NDVI observations were extracted from Sentinel-2 HR-VPP products with a spatial resolution of 10 m and a temporal frequency of 5 days. A rigorous cloud-masking strategy based on the QFLAG2 quality layer was applied to retain only high-quality observations. Missing observations were reconstructed using a Kalman filter-based approach, evaluated in controlled simulations of missing data.
In addition, the dataset incorporates a multi-level quality control workflow that combines clustering based on the dynamic time warping (DTW) technique, K-means clustering, and outlier detection methods based on functional depth (Fraiman-Muniz, h-modal, random projection, and Tukey depth) to remove anomalous phenological trajectories.
Two complementary datasets are provided in CSV format:
• Imputed NDVI time series that include both original and reconstructed observations.
• Filtered and labeled NDVI time series that contain only representative phenological trajectories and cluster assignments obtained using the DTW method.
The dataset is intended for use in crop phenology, remote sensing, precision agriculture, irrigation management, machine learning, time series analysis, and the validation of phenological methodologies or data imputation techniques.
These data were generated as part of the doctoral thesis: “Modelización y teledetección para la actualización de la demanda hídrica de cultivos en grandes zonas regables ante el reto de mejorar la gestión del agua en un contexto de cambio climático”, which is part of the project “Ajuste de ciclos y coeficientes de cultivos para la optimización de la gestión del agua en grandes zonas regables en un contexto de cambio climático (LAIKcA; PID2021-124029OR-I00). Alexey Valero-Jorge sincerely thanks the PRE2022-102328 grant funded by MICIU/AEI/10.13039/501100011033 and FSE+.