Soil Moisture, Irrigation Actuator and Weather Dataset from a Multi-Sector Precision-Irrigation
Description
This dataset documents a full growing season (February–September 2025) of an IoT-instrumented farm in Arnesano (Lecce, Apulia, Italy), where five cultivated sectors, open-field tomato (×2), potted tomato, zucchini and blueberry, were monitored with soil-moisture, pH and electrical-conductivity probes and served by individually controlled irrigation and fertigation valves. It is organised in three layers: - The raw layer contains untouched exports of the IoT platform: 486,894 sensor measurements and 3,055,944 valve state records at native acquisition rate. - The processed layer contains 789 irrigation and fertigation events reconstructed and agronomically validated from the noisy valve signals, five per-sector time series on a 10-minute grid (33,259 rows each) combining sensors, reconstructed applied water volume and ERA5 reanalysis weather, and five analysis-ready tables with engineered features and 24-hour-ahead soil-moisture targets. - The code layer is the Python preprocessing pipeline that produces the processed layer from the raw exports, depending only on pandas, numpy and requests. Unlike most public soil-moisture datasets, this one publishes actuator telemetry alongside sensor readings, so applied water is a measured and actionable variable rather than an assumption. Five sectors with four crop/substrate configurations share the same weather forcing, supporting cross-sector generalisation studies. The raw layer is published unmodified, including sensor dropouts, flatlines and spikes, so alternative cleaning strategies can be benchmarked against the one implemented here. See README.md for the full description, methods, reproduction instructions and limitations, and DATA_DICTIONARY.md for a column-by-column reference. How to cite If you use this dataset, please cite the following publications: - Adamo, T., Colizzi, L., Dimauro, G., Guerriero, E., & Lomonte, N. (2026). Empirical decision model learning for multi-sector greenhouse irrigation under water supply restrictions. Smart Agricultural Technology, 15, 102558. https://doi.org/10.1016/j.atech.2026.102558 - Colizzi, L., Dimauro, G., Guerriero, E., & Lomonte, N. (2025). Artificial intelligence and IoT for water saving in agriculture: A systematic review. Smart Agricultural Technology, 11, 101008. https://doi.org/10.1016/j.atech.2025.101008 - Colizzi, L., Dimauro, G., Lomonte, N., & Mastronardi, L. (2026). Predicting soil moisture in an urban garden using sensor data and machine learning techniques. In J. Bravo, J. Fontecha, & J. Ballesteros (Eds.), Proceedings of the International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI 2025), Volume 2 (pp. 190–202). Springer Nature Switzerland. ISBN 978-3-032-16995-2. https://doi.org/10.1007/978-3-032-16995-2_18
Files
Institutions
- University of Bari Aldo MoroApulia, Bari
- University of SalentoApulia, Lecce