Mascherpa et al. (2025) SmartWT: An open IoT sensor, datalogger and GPRS data transmission device for monitoring water levels in rice fields, with application to AWD irrigation

Identification

• Journal: Computers and Electronics in Agriculture
• Year: 2025
• Date: 2025-12-13
• Authors: Pietro Mascherpa, Michele Rienzner, Darya Tkachenko, Ramiro González Garza, Fabio Brandalese, Ezio Naldi, Claudio Gandolfi, Arianna Facchi
• DOI: 10.1016/j.compag.2025.111324

Research Groups

• Department of Agricultural and Environmental Sciences (DiSAA), University of Milan, Italy
• Department of Biology, University of Konstanz, Germany
• Centre for the Advanced Study of Collective Behaviour, University of Konstanz, Germany
• IMPRS, Max Planck Institute of Animal Behavior, University of Konstanz, Germany
• Department of Information Engineering (DEI), University of Padova, Italy

Short Summary

This study introduces SmartWT, an open-source IoT system for remote, continuous monitoring of ponding water levels in rice fields using an ultrasonic sensor within a Water Tube. It demonstrates the device's robustness, accuracy (typically less than 0.01 m error), and long battery life (120 days) in challenging paddy environments, thereby facilitating the Alternate Wetting and Drying (AWD) irrigation technique.

Objective

• To present SmartWT, an Arduino-based IoT system for continuous water level monitoring in Water Tubes (WTs) throughout the rice growing season, aiming to reduce farmer effort in implementing AWD irrigation, increase Water Productivity (WP), and reduce greenhouse gas emissions and arsenic content in rice grains.
• To rigorously evaluate the device's accuracy through laboratory tests, assess the impact of omitting a temperature sensor on measurement error, and demonstrate its robustness, reliability, and accuracy during a full cropping season field trial in a rice field environment.

Study Configuration

• Spatial Scale:
  • Laboratory tests: Cylindrical test chamber (0.35 m diameter, 1.20 m high) with PVC pipes of 0.07 m, 0.095 m, and 0.135 m inner diameters.
  • Field tests: Nine SmartWT devices installed at three experimental sites in the Pavia province, Italy (Zeme, Castel D’Agogna, Robbio), within the main Italian rice-growing area. Devices were mounted on pipes with 0.135 m internal diameter and 0.50 m length, perforated along 0.30 m.
• Temporal Scale:
  • Laboratory tests: Linearity tests (approximately 40 hours per experiment), accuracy tests (multiple measurements), battery consumption tests (25 minutes, 11 transmissions).
  • Field tests: Full cropping season (26 May to 19 September 2024), spanning 120 days, with data transmission every 2 hours.

Methodology and Data

• Models used:
  • Hardware: Arduino Nano (ATMEGA328P processor), Shield Internet GPRS SIM900, JSN-SR20 waterproof ultrasonic sensor, Recom R-78E5.0-0.5 voltage regulator, TPL5110 timer module. A DS18B20 temperature sensor was used for some laboratory and field temperature monitoring.
  • Software/Data Management: Custom open-source firmware (Arduino IDE), Supabase (edge functions for data storage, processing, and visualization), Telegram Bot API (chat-based user interface), ThinkSpeak (optional traditional plotting interface).
  • Analysis: Linear mixed-effect models (LM models) using MATLAB's fitlme routine for accuracy assessment.
• Data sources:
  • Ultrasonic sensor measurements (JSN-SR20) for water level.
  • Manual distance measurements (using a pointed pin) for reference in accuracy tests.
  • DS18B20 temperature sensor for air temperature inside the WT.
  • Current consumption measurements (Peak Tech 1356 oscilloscope and multimeter) for battery life assessment.
  • Daily rainfall data for field test context.

Main Results

• The SmartWT system, based on open-source architecture (Arduino Nano, GPRS board, ultrasonic sensor), provides continuous remote monitoring of ponding water levels in rice fields.
• Laboratory tests revealed erratic fluctuations in measurements at distances shorter than 0.25-0.30 m and a systematic deviation proportional to the measured distance. These issues were addressed by applying a correction factor (Eq. 4) in the firmware and recommending a minimum measurement distance of 0.30 m.
• With temperature compensation, the device achieved a measurement error of less than ±0.004 m (95th percentile). Without a dedicated temperature sensor, the maximum expected error is approximately +0.018 m to -0.024 m (at 0.80 m distance and extreme temperatures of 14 °C or 40 °C), which is considered acceptable for AWD irrigation management.
• The system is optimized for minimal power consumption, running on two 6 V lead-acid batteries (4.0 A·h total capacity). Laboratory tests estimated a battery life of approximately 200 days with data transmission every 2 hours.
• Field tests over a 120-day cropping season confirmed the device's reliability in harsh paddy environments, with all nine devices operating autonomously without battery recharging or replacement.
• The data transmission success rate, after protocol adjustments, improved to over 99%.
• The estimated total cost of a SmartWT device is approximately €200.

Contributions

• Development of a novel, open-source, and cost-effective (approximately €200) IoT device (SmartWT) specifically designed for continuous water level monitoring in rice field Water Tubes (WTs).
• Addresses the significant labor and cost barriers associated with traditional and existing AWD monitoring methods.
• Demonstrates high robustness and reliability in challenging paddy environments, achieving a measurement error typically less than 0.01 m (without a dedicated temperature sensor) and an extended autonomous operating time (120+ days) without requiring solar panels or battery replacement.
• Offers direct data transmission via GPRS, simplifying installation and eliminating the need for external gateways or complex Wireless Sensor Networks (WSN) infrastructure.
• Provides a practical and farmer-friendly solution to facilitate the implementation of the AWD irrigation technique, contributing to water savings, reduced greenhouse gas emissions, and lower arsenic content in rice grains.
• The study includes a rigorous evaluation methodology, combining controlled laboratory experiments with full-season field testing, enhancing the credibility and applicability of the findings.

Funding

• RISO-SOST (Regione Lombardia, RDP 2014-2020, Operation 16.2.01, n. 18638)
• PROMEDRICE (PRIMA-Section2 2022, funded by MUR for the Italian research group activity, n. 4696)

Citation

@article{Mascherpa2025SmartWT,
  author = {Mascherpa, Pietro and Rienzner, Michele and Tkachenko, Darya and Garza, Ramiro González and Brandalese, Fabio and Naldi, Ezio and Gandolfi, Claudio and Facchi, Arianna},
  title = {SmartWT: An open IoT sensor, datalogger and GPRS data transmission device for monitoring water levels in rice fields, with application to AWD irrigation},
  journal = {Computers and Electronics in Agriculture},
  year = {2025},
  doi = {10.1016/j.compag.2025.111324},
  url = {https://doi.org/10.1016/j.compag.2025.111324}
}