Development and Application of

Cell-Phone-Based Internet of Things (IoT)

Systems for Soil Moisture Monitoring

José O. Payero1*, Michael W. Marshall1 , Bhupinder S. Farmaha1, Rebecca Hitchcock Davis1,

Ali Mirzakhani Nafchi2

Clemson University Edisto Research and Education Center, Blackville, USA

1

Cornell University, New York, USA

2

How to cite this paper: Payero, J.O., Abstract

Marshall, M.W., Farmaha, B.S., Davis, R.H.
and Nafchi, A.M. (2021) Development and Active soil moisture monitoring is an important consideration in irrigation
Application of Cell-Phone-Based Internet water management. A permanent and readily accessible record of changes in
of Things (IoT) Systems for Soil Moisture
soil moisture can be used to improve future water management decision-
Monitoring. Agricultural Sciences, 12, 549-
564.
making. Similarly, accessing stored soil moisture data in near-real-time is
https://doi.org/10.4236/as.2021.125035 also essential for making timely farming and management decisions, such as
where, when, and how much irrigation to apply. Access to reliable communi-
Received: April 15, 2021
cation systems and delivery of real-time data can be affected by its availability
Accepted: May 17, 2021
Published: May 20, 2021
near production fields. Therefore, soil moisture monitoring systems with real-
time data functionality that can meet the needs of farmers at an affordable
Copyright © 2021 by author(s) and cost are currently needed. The objective of the study was to develop and field-
Scientific Research Publishing Inc.
test affordable cell-phone-based Internet of things (IoT) systems for soil mois-
This work is licensed under the Creative
Commons Attribution International
ture monitoring. These IoT systems were designed using low-cost hardware
License (CC BY 4.0). components and open-source software to transmit soil moisture data from
http://creativecommons.org/licenses/by/4.0/ the Watermark 200SS or ECH2O EC-5 sensors. These monitoring systems
Open Access utilized either Particle Electron or Particle Proton Arduino-compatible de-
vices for data communication. The IoT soil moisture monitoring systems
have been deployed and operated successfully over the last three years in
South Carolina.

Keywords
Soil Moisture, Internet of Things, IoT, Arduino, Sensors, Irrigation

1. Introduction
Soil moisture is an essential component of the hydrological, agricultural, ecolog-

DOI: 10.4236/as.2021.125035 May 20, 2021 549 Agricultural Sciences

J. O. Payero et al.

ical, and environmental cycles. It affects land-atmosphere interactions, the ex-

change of water and energy fluxes, rainfall-runoff processes, net ecosystem ex-
change, and food security [1]. It also plays a significant role in affecting critical
physical processes in numerical weather prediction, climate modeling, agricul-
tural crop growth modeling, and flood forecasting [2]. Soil moisture is an essen-
tial component of the hydrologic cycle and a critical variable for optimizing effi-
cient irrigation water management in agriculture [3]. Therefore, obtaining accu-
rate real-time estimates of spatial and temporal variations in soil moisture are
critical [4]. Significant effort has been devoted to soil moisture measurement
using various techniques, including manual in-situ estimation, automated in-
situ sensing networks, numerical modeling, and remote sensing applications [1].
Researchers have used in-situ sensing sites or networks to measure, record,
and transmit soil moisture data from many locations worldwide [1]. For exam-
ple, traditional sensing technologies that measure soil moisture include the neu-
tron probe [5], time-domain reflectometry (TDR) [6] [7], frequency-domain
reflectometry (FDR) [8], capacitance [2] [9] [10], and heat dissipation [11]. How-
ever, researchers have used other emerging in-situ and proximal sensing tech-
niques, such as cosmic-ray neutrons [1] [12], global positioning system (GPS)
signals [1] [13], Global Navigation Satellite System (GNSS) [3], and distributed
temperature sensing (DTS) [1] [14] [15] [16].
There have been significant advances in the large-scale estimation of soil mois-
ture from remote sensing using measurements from optical, thermal, passive, and
active microwave sensors [4]. Satellite-based large-scale soil moisture monitor-
ing includes the Soil Moisture and Ocean Salinity Mission (SMOS), Soil Mois-
ture Active Passive Mission (SMAP), Airborne Microwave Observatory of Sub-
canopy and Subsurface Mission (AirMOSS) [1], Sentinel-1 [3] [17], and Mod-
erate-Resolution Imaging Spectroradiometer (MODIS) [18]. However, at the lo-
cal field level, there are still limitations to using satellite-based remote sensing
soil moisture data for irrigation management, including limited surface penetra-
tion, cloud interference, perturbation by meteorological conditions and vegeta-
tion and surface roughness, and low spatial and temporal resolution [1] [4].
Despite the existence of in-situ soil moisture monitoring networks in some
areas and the promise of new soil moisture products derived from remote sens-
ing, farmers need information at the field scale with near-real-time data for daily
irrigation decisions. There is also a need for simple and affordable systems to
collect and transmit soil moisture data from agricultural fields. Recent advances
in low-cost open-source electronics, wireless communication, and the Internet
of things (IoT) technologies provide new opportunities for agricultural soil mois-
ture monitoring [19]. Many researchers from around the world are currently de-
veloping systems to address these shortcomings. For example, researchers have
created different IoT systems for soil moisture monitoring in China [20] [21],
India [22] [23] [24], Japan [25], and the United States [26] [27]. Therefore, the

DOI: 10.4236/as.2021.125035 550 Agricultural Sciences

 J. O. Payero et al.

objective of this study was to develop and field-test affordable cell-phone-based

Internet of things (IoT) systems for soil moisture monitoring. These end-use IoT
systems would be accurate, affordable for small farmers, robust under normal
field conditions, reliable, and easy to use.

2. Methods
2.1. Research Site Description
Field and laboratory studies were conducted at the Clemson University Edisto
Research and Education Center (EREC) near Blackville, SC (33˚21'51.21N,
81˚19'45.74W). The research site is located in the humid southeast USA, which
receives an average annual precipitation of 1198 mm [28]. Despite the high
rainfall, irrigation is an essential component of agricultural production in South
Carolina due to the uneven rainfall distribution during the growing season and
coarse-textured soils with low water-holding capacities. The EREC contains
around 953 ha, including native forest, pasture, commercial crops, and agricul-
tural research plots. The surrounding region produces numerous crops, includ-
ing forages, fruit, vegetables, and row crops (cotton, corn, soybeans, and pea-
nuts). These high-value fruits and vegetables are typically irrigated, while the low-
value forages and agronomic crops can be irrigated or rainfed.

2.2. Prototype Development

The IoT system prototypes were developed in 2017 with the capability to read
and transmit data from either the Watermark 200SS (The Irrometer Company,
Inc., Riverside, CA) or the ECH2O EC-5 (METER Environment, Pullman, WA)
soil moisture sensors. The IoT systems were assembled using prototyping circuit
boards and included inputs for up to four Watermark or EC-5 sensors (Figure
1). The main component of the two IoT system prototypes was the Particle

Figure 1. The printed circuit board prototypes are shown for the (a) Watermark and (b)
EC-5 sensors. The Particle Electron device included a microcontroller and a cellular
modem.

DOI: 10.4236/as.2021.125035 551 Agricultural Sciences

J. O. Payero et al.

Electron device (Particle Industries, Inc., San Francisco, CA), which was part of
a Particle Electron 2G/3G (EMEA) Starter Kit
(https://store.particle.io/collections/dev-kits). These IoT system prototypes also
included an antenna for a cellular modem, a LiPo battery to power the micro-
controller, and sensor input connectors.
The Particle Electron device included an STM32F205 ARM Cortex M3 mi-
crocontroller with a 1 MB flash, 128 MB RAM, and a cellular modem [u-blox
SARA U-series (3G)]. The microcontroller had a total of 36 pins, including 28
GPIOs (D0-D13, A0-A13), TX/RX, 2 GNDs, VIN, VBAT, WKP, 3V3, and RST.
The microcontroller operated at 3.3VDC with 12-bit Analog-to-Digital (A/D)
inputs. The 12-bit resolution means that when using the analog inputs to make
measurements, the output would result in an integer output (ranging from 0 to
4095). The cellular modem operated at frequencies of 850 and 1900 MHz, suita-
ble for use in the Americas, Australia, and New Zealand. The Particle Electron
kit also came with a SIM card with access to the 2G/3G cellular network service
(available in over 100 countries). The microcontroller was Arduino-compatible
and could be programmed using the Arduino language and libraries via the Par-
ticle Web Integrated Development Environment (Web IDE)
(https://www.particle.io/).
The Watermark and EC-5 sensors measure soil moisture in different ways.
The Watermark sensor measures soil water tension using a solid-state electrical
resistance sensor. Soil water tension changes with soil water content, and the
Watermark sensor measures these changes using electrical resistance. The Wa-
termark sensors produce an output ranging from 0 to 200; the 0 kPa represents
wet (saturated) soil and 200 kPa, air-dry soil. In this study, the sign of the output
was reversed so that the output readings would make more sense for end-users.
The range of output values from the sensor would then be −200 kPa (air-dry
soil) to 0 kPa (saturated soil). Therefore, as the soil dries, the sensor output
readings from the Watermark would become more negative.
In contrast, The EC-5 sensor determines volumetric water content (VWC) by
measuring the soil dielectric constant using capacitance/frequency domain
technology. The EC-5 produces an analog signal output, which is correlated to
the current soil VWC. Therefore, to read the EC-5 sensor using a microcontrol-
ler or other datalogger, the procedure is to apply a known and regulated voltage
to the sensor and then measure the output voltage [26]. The Watermark sensor
can be read in a similar manner using a microcontroller. Similar to the EC-5, a
known, regulated voltage is applied to the sensor and then the output voltage is
measured. However, since output resistance is required, the input voltage must
first pass through a voltage divider circuit [27]. In Figure 1(a), the sensor pro-
totype for the Watermark sensors with the voltage divider is shown. The voltage
divider circuit was constructed using a custom-made printed circuit board
(PCB). This voltage divider PCB was built to accommodate four Watermark
sensors.

DOI: 10.4236/as.2021.125035 552 Agricultural Sciences

 J. O. Payero et al.

2.3. Sensor Calibration

As indicated earlier, when using the analog inputs from the microcontroller to
make measurements, the output of the analog-to-digital converter (ADC) needs
to be an integer in the range of 0 to 4095. However, when measuring the Wa-
termark sensors, the output is required in units of soil water tension (kPa), and
when measuring with the EC-5 sensors, the result is required in units of VWC
(m3∙m−3). Therefore, lab calibrations were conducted to convert the ADC output
to soil water tension or VWC. For the Watermark calibration, sensors were in-
stalled inside soil containers with varying water content, ranging from air-dry to
saturated soil (Figure 2(a)). Watermark sensor readings were taken using the
microcontroller and a Watermark Handheld Meter (The Irrometer Company,
Inc., Riverside, CA). For the EC-5 calibration, four EC-5 sensors were inserted
inside a wide mouth Erlenmeyer flask (Figure 2(b)). The water level was
changed in each flask to produce VWC outputs to simulate the range between
air-dry to saturated soil. At each water level, the analog outputs from the EC-5
sensors were measured using the microcontroller and a ProCheck handheld me-
ter (Meter Group, Pullman, WA).

2.4. Laboratory Testing of Prototypes

Once the prototypes were constructed, and the sensors were calibrated, a labor-
atory test was conducted to evaluate if the system was sensitive to soil moisture
changes and assess the data communication system’s reliability. Four Water-
mark sensors and four EC-5 sensors were placed into 250 mL beakers filled with
soil (Figure 2(a)). The sensors were connected to the prototypes, and sensor
readings were collected every hour. The soil containers were stored in an indoor
laboratory and allowed to dry at ambient room temperature. When the soil was
dry, water was added to each container bring it back to saturation. This process
was repeated to create three soil drying cycles between early October and the end
of November 2017. The data collected on an hourly basis during that time-period
was transmitted to the ThingSpeak website (ThingSpeak.com) for storage and
visualization. The data stored on the ThingSpeak website data could also be vi-
sualized on a cell phone using the ThingView App [26] [27].

Figure 2. The (a) Watermark and (b) EC-5 sensors calibration procedure in the labora-
tory.

DOI: 10.4236/as.2021.125035 553 Agricultural Sciences

J. O. Payero et al.

2.5. Development of PCB

Once the laboratory experiments were conducted, the first step towards moving
from the prototypes to mass production and field deployment was designing and
fabricating PCBs to integrate the microcontroller with the sensors. The PCB for
the Watermark and the EC-5 sensors were designed and manufactured in early
2018 using the Pad2Pad (https://www.pad2pad.com/) system (Figure 3). Pad2Pad
provides free PCB design software and online ordering and manufacturing of the
designed PCB boards. The Watermark and EC-5 sensors PCB boards were con-
structed with connectors for up to four sensors. The voltage divider circuit was
then integrated into the Watermark sensor PCB board. The Particle Electron de-
vice could be soldered directly to the PCB, or two female 18-pin headers could
be soldered to the PCB to allow easy removal and replacement of the Particle
Electron device. The unit price of the PCB manufacturing decreases with quan-
tity. For this study, an initial batch of 50 boards was ordered for a reasonable
cost.

2.6. Setup for Field Deployment

The IoT soil moisture monitoring system component are shown in Figure 4.
Since the system was intended to be installed in remote locations, the system
contained an internal power supply of a 3.7 VDC, 2000 mAh Lithium-Ion (LiPo)
battery, recharged using a 6 V, 2 W externally mounted solar panel (Adafruit
Industries, Inc., NY). The internal power supply system also included a
USB/DC/Solar Lithium Ion/Polymer charger (Adafruit Industries, Inc., NY).
This device is a solar charger controller that automatically uses the solar panel
power when available to maintain or recharge the LiPo battery level.
The IoT soil moisture monitoring systems were housed in an enclosure for
protection of all the electronic components. The enclosure was constructed from
a 1.5 in (38 mm) PVC pipe. A PVC cap was glued to the back of the solar panel,
which served to close the pipe while supporting the solar panel in a horizontal
position. A ¾ in (19 mm) PVC pipe was attached at the other end of the enclo-
sure to keep the electronics above the crop canopy. The sensor wires could be
inserted inside the vertical pipe to protect the wires. Figure 4(b) shows that the
Watermark sensors were attached to a ½ in (13 mm) PVC pipes of different
lengths, which facilitated field installation and removal of the sensors out of the
field at the end of the growing season.

Figure 3. Printed circuit board for (a) the Watermark sensors and (b) the EC-5 sensors.

DOI: 10.4236/as.2021.125035 554 Agricultural Sciences

 J. O. Payero et al.

Figure 4. Soil moisture monitoring system (a) the components of the data logging and
data transmission system and (b) the Watermark sensors attached to PVC pipes. The Par-
ticle Electron device includes a microcontroller and a cellular modem.

2.7. Field Deployment

In 2018, 2019, and 2020, cell-phone-based soil moisture monitoring systems
were constructed and installed in approximately twenty fields in South Carolina.
These fields were planted to various crops in commercial farms and research ap-
plications at the EREC. For example, Figure 5(a) shows the soil moisture moni-
toring system installed in a peanut field in Blackville, SC, in 2018. Due to cost,
the IoT soil moisture monitoring systems installed in the field used the more
economical and durable Watermark sensors. The Watermarks were also easier
to install and remove from the field than the EC-5 sensors. The sensors were in-
stalled in the soil profile at the following depths of 15, 30, 45, and 60 cm.

2.8. System Updates and Modifications

In 2019, although the soil moisture monitoring system was working as expected,
there were concerns about the long-term availability of the 2G/3G cellular net-
works the Particle Electron device was using in the United States. Therefore, the
Particle Electron component was replaced with a Boron device (Particle Indus-
tries, Inc., San Francisco, CA). The Boron device supported Long Term Evolu-
tion (LTE) cellular technology and also offered legacy support for 2G/3G cellular
networks. However, the Boron device pin layout on the microcontroller and
footprint were different compared to the Particle Electron device. The Water-
mark and EC-5 sensor PCBs (similar to the PCB shown in Figure 3) were mod-
ified and fabricated to accept the new Boron device.
Also, other changes and enhancements were made to enhance its affordability
and ease of fabrication and installation. For example, the 2 W solar panel was
replaced with a larger capacity 3.5 W, 6 V 583 mA Mini Solar Panel Module
(Sunnytech, Amazon.com). The solar charger controller was also replaced with a
DFR0264 solar charger (DFRobot, Digi-Key Part Number 1738-1177-ND). In-
stead of housing the electronics in a PVC pipe, a NEMA economy electronics

DOI: 10.4236/as.2021.125035 555 Agricultural Sciences

J. O. Payero et al.

Figure 5. (a) The IoT soil moisture monitoring systems installed in a peanut field in 2018
and the updated (b) IoT soil moisture monitoring system installed in a cotton field in
2019.

box with a solid, weather-resistant door was added (BUD Industries, Ama-
zon.com). The NEMA box was mounted to a 2 in (50 mm) vertical PVC pipe
and the solar panel was glued to a cap which was placed on top of the vertical
pipe. The updated automated IoT soil moisture system was then installed in a
cotton field in 2019 (Figure 5(b)).

2.9. Cost of Monitoring System Components

The unit prices and costs of the soil moisture monitoring system are shown in
Table 1. These prices and costs were divided into the following components:
original Particle Electron design, modified Boron design, sensors, data commu-
nication, and data storage/visualization. The parts needed to build the field setup
(excluding the sensors) using the original Particle Electron design or the mod-
ified Boron design resulted in a similar cost of around US$130.00. For both the
original Particle Electron and the Boron design, the Particle Electron Starter Kit
or Boron Starter Kit represented 54% of the total cost. The cost of four Water-
mark sensors was around US$200.00, while four EC-5 sensors were approx-
imately US$500.00.
A cellular data plan was obtained and used to transmit the data from one field
setup to the Internet was around US$5.00/month (US$60.00/year). The cost de-
pended on the volume of data transmitted. The cost of data storage/visualization
using ThingSpeak also depended on the number of channels and the volume of
data sent to the system. For this study, ThingSpeak offered a free account, which
came with some limitations to the number of channels and data volume, which
would be sufficient for most users. They also provided educational and commer-
cial licenses that offer more data storage and flexibility. An educational license
was obtained for this project at a cost of US$250/year. In this case, using
ThingSpeak would be a developer cost and not a cost to the final user of the soil
moisture monitoring system. Currently, there is no cost to visualizing the data
on a cell phone using the ThingView App.

DOI: 10.4236/as.2021.125035 556 Agricultural Sciences

 J. O. Payero et al.

Table 1. Cost of components of the watermark and EC5 soil moisture monitoring sys-
tems.

Unit Price Total Cost

Component Item Quantity
(US$) (US$)

Particle Electron Starter Kit 1 $71.35

Solar panel (6 V, 2 W) 1 $29.00

Particle Electron design Solar Charger Controller 1 $17.50

PCB fabrication 1 $4.00

PVC pipe & Fittings 1 $10.00 $131.85

Boron Starter Kit 1 $67.21

LiPo battery 1 $12.50

Enclosure 1 $12.70
Boron design Solar panel (6 V, 3.5 W) 1 $12.99
Solar charger controller 1 $4.90
PCB fabrication 1 $4.00
PVC pipe & Fittings 1 $10.00 $124.30
Watermark 4 $50.00 $200.00
Sensors
EC-5 4 $125.00 $500.00

Data communication Cell phone data plan 1 $60/year $60/year

Data storage/ ThingSpeak annual fee 1 0 - $250 0 - $250

Visualization ThingView App annual fee 1 Free Free

3. Results and Discussion

3.1. Laboratory Sensor Calibration
The results of laboratory sensor calibration for the Watermark and EC-5 sensors
are shown in Figure 6 and Figure 7, respectively. For the Watermark sensors,
relationship between ADC output of the Particle Electron device and soil water
potential was explained by a 4th degree non-linear polynomial (R2 = 0.997, p <
0.01). The soil water potential measured by the Watermark Handheld Meter had
a range of −200 to 0 kPa (Figure 6(a)). For the same range of soil water potential
values, there was a strong linear relationship (R2 = 0.988, p < 0.01) between the
resistance measured with the microcontroller and the soil water potential meas-
ured with the Watermark Handheld Meter (Figure 6(b)). Figure 6 also showed
that despite the Watermark Handheld Meter range is limited 0 to 200 kPa (the
range is 0 to −200 kPa in Figure 6), the ADC output or resistance measured by
the microcontroller registered values were sensitive even under dry soil condi-
tions. However, to maintain consistency with the Watermark sensor manufac-
turers recommendations, the output of the microcontroller was limited to the
range between 0 and −200 kPa.
Similarly, there was an excellent linear relationship (R2 = 0.994, p < .01) be-
tween the ADC output of the microcontroller and the VWC measured with the

DOI: 10.4236/as.2021.125035 557 Agricultural Sciences

J. O. Payero et al.

Figure 6. Relationship between (a) analog to digital converter (ADC) output and soil wa-
ter potential, and (b) resistance and soil water potential measured using the Watermark
sensors.

Figure 7. The relationship between particle electron Analog to Digital Converter (ADC)
output and Volumetric Water Content (VWC) reading using the EC-5 sensors.

ProCheck Handheld Meter (Figure 7). The linear equation was then used to
convert the microcontroller’s output (resistance, kOhm) to either soil water po-
tential (kPa) or VWC values (Figure 6).

3.2. Particle Electron Device Prototype Laboratory Test

The study results conducted with the Particle Electron device prototype during
three soil drying cycles are shown in Figure 8 for the Watermark sensors and
Figure 9 for the EC-5 sensors. The screenshots from the ThingSpeak website (a
single channel is shown), which illustrated sensor data visualization (Figure 8
and Figure 9). The soil moisture readings from the Watermark and EC-5 sen-
sors followed a similar but the expected trend during the soil drying cycles fol-
lowed by wetting after water was added to the soil. For the duration of the test,
the cellular data connection with the sensors was very reliable, and the hard-
ware/software components operated as expected. Afterwards, the PCBs were fa-
bricated and assembled for field deployment.

DOI: 10.4236/as.2021.125035 558 Agricultural Sciences

 J. O. Payero et al.

Figure 8. The Volumetric Water Content (VWC) of soil collected during 3 drying cycles
in the laboratory using four Watermark sensors.

Figure 9. The Volumetric Water Content (VWC) of soil collected during 3 drying cycles
in the laboratoryusing four Decagon EC-5 sensors.

3.3. Field Deployment and Testing of Soil Moisture Monitoring

System
As indicated earlier, in 2018, 2019, and 2020, cell-phone-based soil moisture
monitoring systems were constructed and installed in twenty fields with varying
cropping systems in South Carolina. The objective of this study was to determine

DOI: 10.4236/as.2021.125035 559 Agricultural Sciences

J. O. Payero et al.

system reliability and performance under field conditions over time. The intent
of this part of the study was to demonstrate the systems long-term reliability in
the field; therefore, soil moisture readings were collected from a grapevine orc-
hard site in Pickens County, SC over three growing seasons (Figure 10). In the
orchard site, a soil moisture monitoring system with four Watermark sensors
was installed at depths of 15, 30, 45, and 60 cm in July 2018. The hourly VWC
data collected during the crop growing season is shown in Figure 10. These re-
sults showed that the soil moisture monitoring system operated robustly and
collected data, without loss, for nearly three years, demonstrating accuracy and
reliability of the system.
Similarly, Figure 11 showed the VWC collected using the four EC-5 sensors
installed in a rye cover crop in Barnwell County, SC. Figure 11(a) shows a sea-
sonal trend of data collected every two hours from Dec 2017 to March 2018.
TheVWC data collected using the EC-5 sensor was eratic compared to the Wa-
termark sensor data. The EC-5 sensor VWC data over a 5-day period showed
diurnal trend for all soil depths with VWC increasing in the morning hours and
decreasing later in the day (Figure 11(b)). This diurnal effect on the EC-5 soil
moisture readings suggested a potential temperature sensitivity of the sensors.
However, the pattern was similar across all depths. The trend could have also
resulted from other environmental issues affecting the actual VWC. Other po-
tential factors affecting VWC could include presence of early morning dew (typ-
ical in the study area during the growing season) or upward movement of water

Figure 10. Hourly soil water potential collected in 2018, 2019, and 2020 with four Wa-
termark sensors installed in a grapevine orchard in South Carolina.

DOI: 10.4236/as.2021.125035 560 Agricultural Sciences

 J. O. Payero et al.

Figure 11. The Volumetric water content (VWC) collected using four EC-5 sensors in-
stalled at 15, 30, 45, and 60 cm depths in a rye cover crop in South Carolina. The seasonal
trend of VWC collected during the cover crop cycle from Dec 2017 to March 2018 are
shown in (a) and the diurnal effects on VWC (five day period) are also shown in (b). (a)
EC-5 sensors, 2017-2018; (b) EC-5 sensors.

in the soil profile due to a fluctuating soil water table. However, the EC-5 sensor
measurements accurately represented the day-to-day changes in VWC which is
adequate for scheduling day-to-day irrigation decisions.

4. Conclusion
In this project, the IoT systems for soil moisture monitoring were designed, fa-
bricated and deployed to the field. The systems developed utilized either the Par-
ticle Electron or Particle Proton devices for data storage and transmission (Ar-
duino-compatible microcontroller and a cellular modem). Based on the results
from this research, these IoT systems can be used effectively to collect and
transmit soil moisture data from either a Watermark or EC-5 sensors. The Wa-
termark and EC-5 are commercially available soil moisture sensors typically
used in irrigation scheduling. The IoT systems also allowed data visualization in
real-time on a computer or cell phone. The cost of components for building the
IoT system (minus sensor cost) was around US$130.00. Over the last three years
(2018, 2019, and 2020), the IoT systems have been deployed successfully in the
field and used to monitor and collect soil moisture data several research field tri-
als and commercial farms in South Carolina. The IoT systems have shown to be
robust and reliable under South Carolina field conditions. However, the IoT

DOI: 10.4236/as.2021.125035 561 Agricultural Sciences

J. O. Payero et al.

systems depend on accessible and reliable cell phone service near the site of
monitoring which could be a significant limitation in remote areas. The IoT
systems developed and deployed in the project are being demonstrated across
commercial farms in South Carolina to promote the practice of using real-time
soil moisture data for irrigation scheduling decisions.

Acknowledgements
Technical Contribution No. 6979 of the Clemson University Experiment Station.
This material is based upon work supported by NIFA/USDA project SC-1700593
and SC-1700592. USDA-NRCS Projects 69-3A75-17-274 and NR203A750013G010
also provided additional funding.

Conflicts of Interest
The authors declare no conflicts of interest regarding the publication of this pa-
per.

References
[1] Ochsner, T.E., Cosh, M.H., Cuenca, R.H., Dorigo, W.A., Draper, C.S., Hagimoto,
Y., et al. (2013) State of the Art in Large-Scale Soil Moisture Monitoring. Soil
Science Society of America Journal, 77, 1888-1919.
https://doi.org/10.2136/sssaj2013.03.0093
[2] Smith, A.B., Walker, J.P., Western, A.W., Young, R.I., Ellett, K.M., Pipunic, R.C., et
al. (2012) The Murrumbidgee Soil Moisture Monitoring Network Data Set. Water
Resources Research, 48, Article No. W07701.
https://doi.org/10.1029/2012WR011976
[3] Martín, A., Ibáñez, S., Baixauli, C., Blanc, S. and Anquela, A.B. (2020) Mul-
ti-Constellation GNSS Interferometric Reflectometry with Mass-Market Sensors as
a Solution for Soil Moisture Monitoring. Hydrology and Earth System Sciences, 24,
3573-3582. https://doi.org/10.5194/hess-24-3573-2020
[4] Wang, L. and Qu, J.J. (2009) Satellite Remote Sensing Applications for Surface Soil
Moisture Monitoring: A Review. Frontiers of Earth Science in China, 3, 237-247.
https://doi.org/10.1007/s11707-009-0023-7
[5] Kodikara, J., Rajeev, P., Chan, D. and Gallage, C. (2014) Soil Moisture Monitoring
at the Field Scale Using Neutron Probe. Canadian Geotechnical Journal, 51,
332-345. https://doi.org/10.1139/cgj-2012-0113
[6] Herkelrath, W.N., Hamburg, S.P. and Murphy, F. (1991) Automatic, Real-Time
Monitoring of Soil Moisture in a Remote Field Area with Time Domain Reflecto-
metry. Water Resources Research, 27, 857-864. https://doi.org/10.1029/91WR00311
[7] Skierucha, W., Wilczek, A., Szypłowska, A., Sławiński, C. and Lamorski, K. (2012) A
TDR-Based Soil Moisture Monitoring System with Simultaneous Measurement of
Soil Temperature and Electrical Conductivity. Sensors, 12, 13545-13566.
https://doi.org/10.3390/s121013545
[8] Berthelin, R., Rinderer, M., Andreo, B., Baker, A., Kilian, D., Leonhardt, G., et al.
(2020) A Soil Moisture Monitoring Network to Characterize Karstic Recharge and
Evapotranspiration at Five Representative Sites Across the Globe. Geoscientific In-
strumentation, Methods and Data Systems, 9, 11-23.

DOI: 10.4236/as.2021.125035 562 Agricultural Sciences

 J. O. Payero et al.

https://doi.org/10.5194/gi-9-11-2020
[9] Fares, A. and Alva, A.K. (2000) Evaluation of Capacitance Probes for Optimal Irri-
gation of Citrus through Soil Moisture Monitoring in an Entisol Profile. Irrigation
Science, 19, 57-64. https://doi.org/10.1007/s002710050001
[10] Osenga, E.C., Arnott, J.C., Endsley, K.A. and Katzenberger, J.W. (2019) Bioclimatic
and Soil Moisture Monitoring Across Elevation in a Mountain Watershed: Oppor-
tunities for Research and Resource Management. Water Resources Research, 55,
2493-2503. https://doi.org/10.1029/2018WR023653
[11] Wyatt, B.M., Ochsner, T.E., Fiebrich, C.A., Neel, C.R. and Wallace, D.S. (2017)
Useful Drainage Estimates Obtained from a Large-Scale Soil Moisture Monitoring
Network by Applying the Unit-Gradient Assumption. Vadose Zone Journal, 16,
1-15. https://doi.org/10.2136/vzj2017.01.0016
[12] Köhli, M., Schrön, M., Zreda, M., Schmidt, U., Dietrich, P. and Zacharias, S. (2015)
Footprint Characteristics Revised for Field-Scale Soil Moisture Monitoring with
Cosmic-Ray Neutrons. Water Resources Research, 51, 5772-5790.
https://doi.org/10.1002/2015WR017169
[13] Qiao, X., Khalilian, A., Payero, J.O., Maja, J.M., Privette, C. and Han, Y.J. (2016)
Evaluating Reflected GPS Signal as a Potential Tool for Cotton Irrigation Schedul-
ing. Advances in Remote Sensing, 5, 157-167. http://doi.org/10.4236/ars.2016.53013
[14] Sayde, C., Gregory, C., Gil-Rodriguez, M., Tufillaro, N., Tyler, S., Van de Giesen, N.,
et al. (2010) Feasibility of Soil Moisture Monitoring with Heated Fiber Optics. Wa-
ter Resources Research, 46, Article No. W06201.
https://doi.org/10.1029/2009WR007846
[15] Striegl, A.M. and Loheide, S.P. (2012) Heated Distributed Temperature Sensing for
Field Scale Soil Moisture Monitoring. Groundwater, 50, 340-347.
https://doi.org/10.1111/j.1745-6584.2012.00928.x
[16] Sourbeer, J.J. and Loheide, S.P. (2016) Obstacles to Long-Term Soil Moisture Mon-
itoring with Heated Distributed Temperature Sensing. Hydrological Processes, 30,
1017-1035. https://doi.org/10.1002/hyp.10615
[17] Bauer-Marschallinger, B., Freeman, V., Cao, S., Paulik, C., Schaufler, S., Stachl, T.,
et al. (2019) Toward Global Soil Moisture Monitoring with Sentinel-1: Harnessing
Assets and Overcoming Obstacles. IEEE Transactions on Geoscience and Remote
Sensing, 57, 520-539. https://doi.org/10.1109/TGRS.2018.2858004
[18] Zhang, F., Zhang, L.W., Shi, J.J. and Huang, J.F. (2014) Soil Moisture Monitoring
Based On Land Surface Temperature-Vegetation Index Space Derived from MODIS
Data. Pedosphere, 24, 450-460. https://doi.org/10.1016/S1002-0160(14)60031-X
[19] International Telecommunication Union (2013) Overview of the Internet of Things.
ITU-T Y-Series Recommendations (Y.2060), Global Information Infrastructure, In-
ternet Protocol Aspects and Next-generation Networks. International Telecommu-
nication Union, Geneva, 22 p.
[20] Chung, W.Y., Villaverde, J.F. and Tan, J. (2013) Wireless Sensor Network Based Soil
Moisture Monitoring System Design. 2013 Federated Conference on Computer
Science and Information Systems, Kraków, 8-11 September 2013, 79-82.
[21] Zhang, X., Zhang, J., Li, L., Zhang, Y. and Yang, G. (2017) Monitoring Citrus Soil
Moisture and Nutrients Using an IoT Based System. Sensors, 17, Article No. 447.
https://doi.org/10.3390/s17030447
[22] Athani, S., Tejeshwar, C.H., Patil, P., Patil, M.M. and Kulkarni, R (2017) Soil Mois-
ture Monitoring Using Iot Enabled Arduino Sensors with Neural Networks for Im-

DOI: 10.4236/as.2021.125035 563 Agricultural Sciences

J. O. Payero et al.

proving Soil Management for Farmers and Predict Seasonal Rainfall for Planning
Future Harvest in North Karnataka—India. Proceedings of the International Con-
ference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC 2017),
Palladam, 10-11 February 2017, 43-48.
https://doi.org/10.1109/I-SMAC.2017.8058385
[23] Kodali, R.K. and Sahu, A. (2016) An IoT Based Soil Moisture Monitoring on Losant
Platform. Proceedings of the 2016 2nd International Conference on Contemporary
Computing and Informatics (IC31), Noida, 14-17 December 2016, 764-768.
https://doi.org/10.1109/IC3I.2016.7918063
[24] Ezhilazhahi, E.M. and Bhuvaneswari, P.T.V. (2017) IoT Enabled Plant Soil Moisture
Monitoring Using Wireless Sensor Networks. Proceedings of the 2017 IEEE Third
International Conference on Sensing, Signal Processing and Security (ICSSS),
Chennai, 4-5 May 2017, 345-349. https://doi.org/10.1109/SSPS.2017.8071618
[25] Nagahage, E., Nagahage, I. and Fujino, T. (2019) Calibration and Validation of a
Low-Cost Capacitive Moisture Sensor to Integrate the Automated Soil Moisture
Monitoring System. Agriculture, 9, Article No. 141.
https://doi.org/10.3390/agriculture9070141
[26] Payero, J.O., Nafchi, A.M., Davis, R. and Khalilian, A. (2017) An Arduino-Based
Wireless Sensor Network for Soil Moisture Monitoring Using Decagon EC-5 Sen-
sors. Open Journal of Soil Science, 7, 288-300.
https://doi.org/10.4236/ojss.2017.710021
[27] Payero, J.O., Nafchi, A.M., Khalilian, A., Qiao, X. and Davis, R. (2017) Development
of a Low-Cost Internet-Of-Things (IoT) System for Monitoring Soil Water Poten-
tial Using Watermark 200SS Sensors. Advances in Internet of Things, 7, 71-86.
https://doi.org/10.4236/ait.2017.73005
[28] U.S. Climate Data (2020) Climate Blackville—South Carolina.
https://www.usclimatedata.com/climate/blackville/south-carolina/united-states/ussc
0025

DOI: 10.4236/as.2021.125035 564 Agricultural Sciences