
Relative humidity in arid soils – system development and application
Raz Balas (1)(2), Noa Balaban (2), Elad Levintal (3).
(1) Other Institute (insert manually)
(2) Nuclear Research Center Negev (NRCN) ; Address: Beer Sheva Po box 9001
(3) Zuckerberg Institute for Water Research, Ben Gurion University of the Negev, Sde Boker, 84990, Israe
Unit of Energy Engineering, Ben Gurion University of the Negev, Beer Sheva 84105
In arid environments, characterized by low near-surface water content, non-rainfall water inputs (vapor, dew, and fog) significantly impact soil moisture dynamics. As direct water content measurement is challenging, soil relative humidity (RH) is key to analyzing in-situ water and vapor fluxes. High-resolution in-situ measurements are essential but challenging, as standard RH sensors are designed for air rather than soil. This discrepancy causes frequent failures and limits long-term field monitoring due to high moisture sensitivity.
This research addresses this gap by developing and validating an innovative, low-cost, open-source RH sensing system, designed specifically for long-term operation in arid soils.Methodologically, the system integrates SHT31-D sensors, protectively coated with Plasti Dip to prevent moisture corrosion. Data is aggregated via a multiplexer for local storage and real-time transmission, ensuring continuous autonomous operation powered by solar energy. The system was installed in the Yamin Plain, Negev Desert, Israel, anarid region with a mean annual rainfall of ~72 mm and potential evapotranspiration of ~2,600 mm. The ten sensors in the system are based on SHT31-D, six of them is in house built, and four commercial reference sensors - three from Dracal Technologies and one from DFRobot. The sensors were installed in the atmosphere and at three soil depths (two sensors in the air, three at 2.5 cm, three at 10 cm, and two at 50 cm) to capture continuous, real-time RH and temperature data.
Prior to installation, the system underwent calibration against reference sensors, temperature and RH data from a close meteorological station. Preliminary results show a correlation between diurnal adsorption and evaporation cycles of non-rainfall water inputs (NRWI), compared to soil water dynamics during rainfall events. This study provides essential high-resolution data for advancing process-based understanding of moisture transport in dryland environments, and the basis for future wide-spread installation in arid soil environmnets.



