Smart Agricultural Technology 2 (2022) 100032

Contents lists available at ScienceDirect

Smart Agricultural Technology

journal homepage: www.elsevier.com/locate/atech

Automation of soil moisture sensor-based basin irrigation system

Monalisha Pramanik a,∗, Manoj Khanna a, Man Singh a, D.K. Singh a, Susama Sudhishri a,
Arti Bhatia b, Rajeev Ranjan c
a
Water Technology Center, Indian Agricultural Research Institute, New Delhi 110012, India
b
Division of Environment Science, Indian Agricultural Research Institute, New Delhi 110012, India
c
Division of Agricultural Physics, Indian Agricultural Research Institute, New Delhi 110012, India

a r t i c l e i n f o a b s t r a c t

Keywords: Irrigation is one of the largest consumers of fresh water in the agricultural sector. Surface irrigation is one the
Basin irrigation most widely adopted method of irrigation due to its simplicity and low energy requirement. Although, it is mostly
Automation associated with poor irrigation application efficiency mainly due to deep percolation and uneven distribution of
Internet of things (IoT)
irrigation water. A study was conducted in closed-end level basin layout in loamy soil to establish and test an
Soil moisture sensor
Internet of Things (IoT) based system through a wireless connection between the automatic check gate and
Wireless communication
LoRa soil moisture sensors which can be operated remotely based on real-time soil moisture status. An attempt had
been made to identify the suitable location of the sensor in the basin layout for improving irrigation application
efficiency. An aluminium automatic-check gate fitted in an iron frame was placed in the concrete water supply
channel to control the water flow. Three capacitance-based soil moisture sensors were placed at 25%, 50% and
75% of field length (i.e., 15 m, 30 m and 45 m) at 37.5, 15 cm and 7.5 cm depth, respectively. A wireless
communication network was established between capacitance-based soil moisture sensors and an automatic check
gate through the LoRa module and GSM module. The software used for automatic control of the check gate in this
study was Croplytics® by Agsmartic. The system was tested for nine irrigation events under bare soil conditions.
Three different operational schedules based on different soil moisture sensor placement had been tested under
40%, 30% and 20% soil moisture deficit conditions. The results suggested that the best location of sensors for
closing the system would be at 37.5 cm depth placed at 25% length from the inlet in higher soil moisture deficit
condition and at 7.5 cm depth placed at 75% length in low moisture deficit condition. Overall, the irrigation
application efficiency was improved up to 86.6% with automation.

1. Introduction Well-designed and managed surface irrigation systems have the po-
tential to deliver high application efficiencies [7]. Despite that, most sur-
Water scarcity makes its management one of the most critical chal- face irrigation systems have been operating with significantly lower and
lenges faced in arid and semi-arid regions. Water management and in- highly variable application efficiencies. Surface irrigation will remain a
creased water use productivity are among the most effective manage- dominant method for the foreseeable future internationally as well as in
ment options for conserving water resources, especially in irrigated agri- India [8]. Hence, it is expected that the quest for the improvement of the
culture. Agriculture is the largest consumer of freshwater [1,2,]. En- application efficiency of surface irrigation will continue. The variables
hancing irrigation efficiency could save a large amount of water which or parameters that affect the performance of irrigation can be grouped
could be used to bring an additional area under irrigation. Surface ir- into three general categories: field geometry (length, width, slope, fur-
rigation remains the most commonly used method for irrigating crops row cross-section, etc.), field conditions (infiltration and roughness rela-
and pastures in India and the world due to its low cost and low energy tions), and management variables (flow rate, cut-off time, soil moisture
requirement [3]. Surface irrigation plays a significant role in the provi- deficit at the time of irrigation etc. [9]. Raising application efficiency
sion of food and fodder in India. For instance, the system was used on further would only come about by managing each irrigation through
about 80 % of the total irrigated land in 2016-2017 [4]. However, these varying inflow rates and time of irrigation cut-off in real-time to pro-
systems are often associated with low water application efficiency and vide the best performance for the prevailing conditions. Many studies
high labor requirements [5,6]. have reported that the application efficiency could be increased by 20%

∗
Corresponding author.
E-mail address: [email protected] (M. Pramanik).

https://doi.org/10.1016/j.atech.2021.100032
Received 3 August 2021; Received in revised form 6 December 2021; Accepted 13 December 2021
2772-3755/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

by merely doubling the inflow rate in a basin (or border check) irrigation
for pasture and fodder crops in northern Victoria, Australia [10,11].
Automation through an IoT system could be an effective approach
to improve a conventional surface irrigation system operation. An au-
tomated surface irrigation system refers to its operation with timers,
sensors or computers or mechanical appliances with minimal manual
involvement. Many researchers have reported that automation in irri-
gation projects using an intelligent irrigation controller and wireless
sensor network could save water up to 38% [12,13]. Automation is a
smart technique to deal with the problem of high labor requirements
and low water application efficiency of surface irrigation systems. Many
soil moisture sensors such as tensiometer, gypsum block, granular ma-
trix sensor, time-domain reflectometer, dielectric probe are commer-
cially available for soil moisture measurement and they could gener-
ally be used for manual or integrated with automatic irrigation control
systems through an IoT system [14–16]. The sensor senses the water
advance front and gives a signal to cut-off the flow. Application effi-
ciencies above 90% could be achievable in basin irrigation by choosing
the correct irrigation duration which reduces the tailwater runoff [17].
The automated system with a wireless soil moisture sensor network was Fig. 1. Location map (Google Earth) of the study area at the research farm,
tested in a sage crop field for 136 days, showed water savings of up to IARI, New Delhi.
90% compared with the agricultural zone’s traditional irrigation prac-
tices [18]. Automation of irrigation based on one Bay Drive (a remotely
operated field input gate actuator) gate helps to reduce the time spent ature at regular interval to schedule irrigation through a mobile-based
by workers for water level control and flow regulation in paddy [19]. app in several crops [26,27]. IoT and WSN has been effectively exe-
In India, most of the canal irrigation system has faced inequitable water cuted on a few farming tasks, for example, water quality administra-
supply from head to the tail end user. Conventionally, there is high dis- tion, smart irrigation system, data-driven analysis, pest control and dis-
crepancies in supply and demand due to human involvement in outlet ease monitoring [28–36]. A wide range of irrigation systems like trickle,
gate opening and closing for water flow [20]. Automation of water flow sprinkler or surface irrigation systems could be automated by utilizing
check gates at the field level may help to reduce the inequitable water clocks, sensors or mechanical devices. In this automation framework,
supply on a real-time basis [21]. some standard information concerning the soil moisture level of a field
IoT is the network of physical objects or "things" embedded with is predefined. The sensor initially sends the information through a mi-
electronics, software, sensors and network connectivity, which enables crocontroller to the server at a periodic interval. In the event, when soil
these objects to gather and exchange data. IoT system empowers the moisture in the field was below the threshold value of the information
farmers to use modern technology to beat huge difficulties of monitoring base, the microcontroller triggered the irrigation of plants till the pre-
and labor crisis. New innovative and price effective IoT applications are defined soil moisture level is reached [37]. Considering the above chal-
helping the farming sector to improve the standard, quantity, sustain- lenges and prospect of IoT based automatic irrigation system, a proto-
ability and price effectiveness of agricultural production. An IoT based type automated water flow check gate with communication system and
automation system sensing, processing, communicating and actuating an actuator/motor to operate it on a real-time basis for basin irrigation
the physical devices which perform the task i.e. closing and opening of system was developed and tested on the bare soil. The application of
valves/gates as per demand or turn [22]. Sensors and actuators are the this automated check gate may be at outlet in canal irrigation network.
backbones of an IoT system. In an automated architecture, the sensor Soil moisture sensor placement and the optimum number of sensors
node sends data to the actuator and the actuator perform the task as are also important factors that significantly affect sensor-based surface
per the prescribed task while in semiautomated the sensor node sends irrigation automation, performance and cost-effectiveness. This is still a
data to the sink (central controller) and the sink command the actua- researchable question that needs to be evaluated. Therefore, an experi-
tor to perform the task. Less energy consumes in automatic than semi- ment was laid out to develop an IoT based wireless automatic soil mois-
automatic architecture but the addition of more sensing nodes is easier ture sensor-based basin irrigation system to enhance field application
in semi-automatic architecture as there is no need to develop a separate efficiency. Also, an attempt had been made to identify suitable location,
algorithm for communication and coordination [23]. Many factors af- depth and number of soil moisture sensors coupled with an automatic
fect the actuator performance i.e. communication, coordination between check gate in the basin layouts field. This crucial information may help
sink and actuator and most importantly sufficient energy to perform the in upscaling of automation water distribution network in canal based
task. There should be proper synchronisation among the actuators in irrigation systems.
the case of the multi actuator. Many commercial sensors, actors, proces-
sors and communication components are available but they should be
self-sufficient in terms of energy and cost-effective especially in the agri- 2. Material and methods
culture sector where electricity availability is a challenge [24]. The sys-
tem should use low-cost and easily available components which allows 2.1. Experimental site
fast application and easy maintenance. Various combinations of differ-
ent components of communication and processing known as modules An experiment was conducted at the Research Farm of ICAR-Indian
were tested based on data communication, cost-effectiveness, power Agricultural Research Institute (IARI), New Delhi from September to
management and it found that LoRa, communicate up to 1000 m dis- December of the year 2020 in bare soil conditions (Fig. 1). The exper-
tance compared to 150 m by Wifi transmission module while Arduino imental field was situated at 28o 37′ 55" N latitude and 77o 09′ 36" E
Uno consumes less energy than raspberry pi processing unit [25]. Many longitude with an average elevation of 230 m above mean sea level. The
researchers have reported that IoT system with wireless sensor network experimental field’s soil texture was loam with a bulk density of 1.55 Mg
(WSN) with Zigbee or LoRa forms of communications system has suc- m−3 . The field capacity and wilting point of the soil were 35.8 % (v/v)
cessfully communicated the sphere data i.e., soil moisture, soil temper- and 15.5% (v/v), respectively. The basic infiltration rate is 6 mm/h. The

2
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Fig. 2. Specifications of check gate and different components inside the control unit box.

infiltration rate was determined using a double-ring infiltrometer. The 2.2.2. Software component
basin irrigation layout (blocked end) was of the size of 60 m × 14 m. The software used in the study for automation is Croplytics® de-
veloped by Agsmartic and customized according to our research need,
which consists of web and mobile-based interface so that users can ac-
cess it from anywhere. It consists of an interface that offers sensor lo-
2.2. Components of automatic irrigation system
cation on Google Maps, real-time soil moisture status and gate config-
uration for ON and OFF according to the predefined threshold value of
2.2.1. Hardware components
soil moisture. By entering the sensor’s global positioning system (GPS)
The flow of water to the field is controlled by the solar-powered
location in the field, users can easily locate the sensor in a large area.
automatic-check gate, which is wirelessly linked with soil moisture sen-
The programme code is written in C and C++ language. At every 3 min
sors placed in the basin. The gate automatically opens and closes based
interval, the soil moisture data is stored on a dedicated server using
on prefixed soil moisture sensor observations for real-time operation
the GSM module. However, the time interval can be customized accord-
of an irrigation event. The check gate is made up of an aluminium
ing to the need. Less time interval leads to more power consumption.
sheet (7.4 mm thickness) fitted with an iron frame is installed at the
The embedded GSM module helps to access the system information from
inlet of the experimental field in the water supply channel. Rubber
anywhere anytime through the web or mobile apps. A user needs a login
gaskets were provided to prevent water leakage from the sides of the
and password to access the system information (Fig. 3).
aluminium sheet. The height of the opening of the gate is 30 cm. The
Users can switch ON and OFF the automatic check gate either
gate was designed for completely open and close operational mode. The
through the web or mobile app. The soil moisture sensors detail and
detailed specifications of the gate are shown in Fig. 2a. A mini solar
gate configuration can be accessed under the soil moisture and irriga-
panel, control box and solar charger unit were mounted on the gate. The
tion tab, respectively. The graphical representation of daily soil moisture
power unit was connected with a 12V DC motor /actuator and controller
variation and downloading of data could be done through web access
unit. The controller unit consists of a voltage converter, motor driver
for any period. The input parameters are minimum, and a maximum
module, distance measurement ultrasonic module and Arduino nano-
threshold value of soil moisture needs to enter for gate irrigation oper-
microcontroller with global system for mobile (GSM) module (Fig. 2b).
ation (Fig. 4). Users can set the value according to the crop water re-
The specification and description of hardware inside the controller unit
quirement and management allowed a deficit for a specific crop. As the
are presented in Table 1. The control box was kept inside a metal box
soil moisture reached or exceeded a pre-set threshold value, the motor
to protect it from rain and any other mishandling. An ultrasonic sen-
driver module activated and opened the gate. Similarly, when soil mois-
sor has been incorporated in the control box to avoid the gate being hit
ture came up to 90% of field capacity, the gate started to close. Users can
by the ground and top while closing and opening the gate respectively.
define the maximum allowable depletion of soil moisture stress accord-
A metal plate is placed just beneath the ultrasonic sensor to create an
ing to crop conditions. The block diagram and flow chart of different
obstacle and measure the opening and closing height of the gate. Ul-
steps in the automation process is shown in Fig. 5.
trasonic sensing elements emit short, high-frequency sound pulses at
regular intervals associate degreed once they hit an object, then they’re
mirrored back as echo signals to the detector, which itself computes the 2.2.3. Capacitance based soil moisture sensor
gap to the target supported the time-span between emitting the signal The capacitance-based soil moisture sensor operates based on the
and receiving the echo. The metal plate created an obstacle in which dielectric constant of soil and water. The electrodes do not contact
sound pulse hit and return to the sensor through which distance of the soil directly, which makes it corrosion resistant compared to the resis-
plate can be calculated. The system was programmed to open and close tance type sensor. The capacitance sensor consists of three main com-
the gate up to 20 cm and 50 cm distance from the ultrasonic sensor. The ponents: the positive plate, negative plate and the space between the
control box has three buttons, one for power off and on, the second for plates known as a dielectric. The considerable difference between soil
operational mode (automatic/manual) and the third to move the gate up and water dielectric constant helps to identify water molecules present
and down manually. The whole system was powered with a 12 V 7 Ah in the soil mass. The sensor’s capacitance was measured using a 555
battery recharged by a 10 W solar panel. 10 W, 12 V polycrystalline time-based circuit that produces a voltage proportional to the capaci-
solar panel sufficient to charge 12 V rechargeable battery. The charger tor inserted in the soil (Fig. 6). The measured voltage is converted to a
controller indicates the overload, charging status and low battery for digital number through an analogue to digital converter (ADC). The soil
proper operation of the system. moisture content and voltage have an inverse relationship, i.e. higher the

3
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Table 1
Specifications, description and function of the hardware placed inside the controller unit.

S.No. Sensor Module Model Specification Function

1 Motor drive A4988 driver Supply voltage (V): 8 to 35 To drive motor gears
Current (without Hestsink): 1 A
Current (with Hestsink): 2 A
Logic Input: 3–5.5 V
2 Ultrasonic sensor JSN-SR04T JSN-SR04T-2.0 Working voltage: 3.0–5.5 V DC To measure the depth of
opening and closing of gate
Working current: < 8 mA
Accuracy: ±1 cm
Acoustic emission frequency: 40 khz
Angle: less than 75°
Working temperature: -20–70 ◦ C
3 Arduino Nano Arduino Nano V3.0 Microcontroller Atmel ATmega328 SMD Package To control gate opening and
closing based on moisture data
Operating voltage (logic level): 5 V
Digital I/O Pins:14 (of which 6 provide PWM output)
Analog Input Pins: 8
DC current per I/O Pin: 40 mA
Flash memory: 32 KB (of which 2KB used by bootloader)
SRAM: 2 KB
EEPROM: 1 KB
Clock speed: 16 MHz
4 LoRa HPD Tek HPD13A1.1V Frequency range: 868 MHz To send soil moisture data to
gateway
Modulation: FSK/GFSK/MSK/LoRa
SPI Data Interface
Sensitivity: -136 dBm
Output power: +20 dBm
Data rate: <300 kbps
5 Voltage Converter LM2596 DC-DC Buck Input voltage - 3.2 V–40 V DC To convert voltage from 12 V to
Converter 5V
Output voltage - 1.25 V–35 V DC
6 Solar charger module for CN3065 18650 Li-Ion Mini Input: 4.4-6 V To recharge Li-Ion battery
soil moisture sensor Solar Charger Board Module
Current: 500 mA
Interface: 2-pin JST connectors (or PH2.0)
Short circuit protection
continuous charge current: up to 500 mA
7 Microcontroller ATMEGA328P- U Minimum supply voltage: 2.7 V To process different algorithm
Maximum supply voltage: 5.5 V
No. of I/Os 23
No. of Timers/Counter: 3
8-bit Microcontrollers – MCU
Program memory size: 32 kB
Data RAM size and type: 2kB SRAM
Data ROM Size and type: 1024 B EEPROM
Interface type: I2C, SPI, USART
8 Solar Panel Green Power Power:10 W To recharge 12 V battery
Output: 12 V
9. GSM Module Simcom SIM800A Quad Band GSM Module To connect with the cloud
server
Bands: GSM 850 MHz, EGSM 900 MHz, DCS 1800 MHz,
PCS 1900 MHz
5 V interface
Voltage Supply Required- 9VDC to 12VDC with at least
2A Peak Current Capability
Operating temperature: -40 C to +85 C

Fig. 3. Screenshot of Croplytics® (a) web

and (b) mobile-based app.

4
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Fig. 4. Soil moisture tab and gate configuration tab in Croplytics® web system.

Fig. 5. Block diagram and flow chart of the automatic surface irrigation system.

voltage value lower the moisture content. This type of capacitor-based an analogue type that was compatible with Arduino Nano with an ana-
soil moisture sensor is low cost and has low energy consumption with logue to digital converter (ADC). A 3.7 V Li-Ion battery is embedded in
reasonable accuracy and is therefore widely used in smart and auto- the control box with a 5 V solar panel on the top for an uninterrupted
mated irrigation systems. The sensor comprises a PVC pipe, solar panel power supply. The circuit box consists of a microcontroller, LoRa mod-
and circuit box. The capacitive soil moisture sensor V1.2 operates at ule (for wireless communication) and solar charger module with Li-Ion
low voltage 3.3 to 5 V, which is compatible with a low voltage micro- rechargeable battery of 3.7 V 8800 mAh. A microcontroller is a small
controller. It supports three-pin sensor interfaces. The sensor’s output is and low-cost microcomputer consisting of the processor, memory (ROM,

5
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Fig. 6. Capacitance based soil moisture sensor (a) external component and (b) internal components of the circuit box.

a low rate through an unlicensed radio spectrum. In India, 865 MHz to

868 MHz frequency radio spectrum is used in LoRa communications.
Long-distance (15–20 km) and low power consumption make it a suit-
able option for Internet of Things (IoT) application in agriculture. It
uses spread-spectrum technology with a broader band. Its frequency-
modulated chirp utilizes coding gain for increased receiver sensitivity.
The gateway, cloud server and the automatic check gate were connected
through a global system for mobile (GSM) module. The SIM800A Quad-
Band GSM/GPRS Module with RS232 Interface was used due to its wider
application and low power consumption. SIM800A support Quad-band
850/900/1800/1900 MHz, can transmit voice, SMS and data informa-
tion with low power consumption.
The gateway received soil moisture data at every 3 min interval
through the LoRa module. The system went to sleep mode in between
intervals to minimize power consumption. Further, the data is stored
to the dedicated cloud server and communicated to the gate through
Fig. 7. Wireless communication between soil moisture sensors (SMS) and check GSM. Users could schedule the automatic check gate on predefined max-
gate. imum and minimum soil moisture conditions according to crop condi-
tion/demand.

RAM and EPROM), serial ports and peripherals (timer) to perform spe- 3. Experimental layout
cific tasks. According to accuracy, speed, memory and task, the micro-
controller was selected. The size of the experimental basin layout is 60 m × 14 m with a
0.0005 m/m slope. Three capacitance-based soil moisture sensors were
2.3. Wireless communication network placed at 25%, 50% and 75% of field length at 37.5 cm (Sensor number
1), 15 cm (Sensor number 2) and 7.5 cm (Sensor number 3) soil depth,
One of the main and important aspects of any IoT device is the ability respectively. The inlet of the field was equipped with a solar-powered
to communicate to transfer and exchange data between other devices. automatic-check gate placed in a water supply channel to start and stop
IoT devices primarily use wireless communication to communicate with inflow to basin layout. An ultrasonic flow meter Model Unidata 6526E)
other devices. Wireless communication network refers to wireless trans- was placed at the inlet to measure the flow rate. The layout of the ex-
mission of data over the air through electromagnetic waves line Infrared perimental plot is presented in Fig. 9.
(IR), radio frequency (RF), satellite without physical connection with
wire. IoT devices could be connected through internet and non-internet 3.1. Calibration of soil moisture sensor
protocol based on area coverage, power consumption and memory. The
non-internet protocol requires less memory and consumes less power in The soil moisture sensors were calibrated before installation in the
a limited range. Many network protocols are available according to their field. Thirteen soil samples were randomly collected from different soil
use such as ZigBee, Hypertext Transfer Protocol (HTTP), Wi-fi, Bluetooth moisture conditions ranging from very dry to wet concurrently with
and Long-Range Wide Area Network (LoRaWAN). In this study, a wire- soil moisture sensor reading. The soil moisture was determined using
less communication network was established among the soil moisture the gravimetric method. After weighing the fresh weight, samples were
sensors, gateway, cloud server and automatic check gate, as shown in oven-dried at 105 ºC for 24 h. The soil moisture content was determined
Fig. 7. A gateway is a passage between two different networks with a dif- using Eq. (1).
( )
ferent transmission protocol. It can be unidirectional and bidirectional 𝑚𝑤𝑒𝑡 − 𝑚𝑑𝑟𝑦
based on the direction of data flow. It comprehends the sensors data and 𝜃𝑔 = × 100 (1)
𝑚𝑑𝑟𝑦
translate it to a compatible mode before sending it to the cloud network.
The gateway was installed at 100 m apart from the soil moisture sen- Where, ɵg is the gravimetric soil moisture content (%), mwet and mdry
sors and powered through AC power with a 6 A adapter. The gateway is is the fresh and dry weight of soil (g), respectively.
comprised of a microcontroller, LoRa module, global system for mobile The gravimetric soil moisture content was converted to volumetric
(GSM) module and buck converter (Fig. 8). LoRa module is a wireless soil moisture content using Eq. (2).
communication module used for wireless communication between sen- 𝜃𝑔 × 𝜌𝑠𝑜𝑖𝑙
sors and gateway over a long-range distance. It enables sending data at 𝜃𝑣 = (2)
𝜌𝑤𝑎𝑡𝑒𝑟

6
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Fig. 8. Gateway and its different component.

Fig. 9. Layout of the experimental field with the placement of soil moisture sensor and check gate.

Fig. 10. Relationship between volumetric soil mois-

ture content and the digital value of the soil moisture
sensor.

Where, ɵv is the volumetric soil moisture content (%), 𝜌soil and 𝜌water 3.2. Soil moisture sensor placement
are soil and water density, respectively.
The sensor’s observed digital value varied between 620 for dried soil The soil moisture sensor (SMS) placement was done based on two
(𝜃𝑣 =15.5 %) and 384 for saturated soil (𝜃𝑣 =42%). An exponential rela- factors one is crop root zone depth and another is water flow reach
tionship has been found between volumetric soil moisture content and which is categorised as head, mid and end reach. The depth of sensors
the sensor’s digital value with a very good coefficient of determination varied according to the depth of crop root zone and reach of the flow.
(R2 = 0.827) (Fig. 10). The system was designed with intention of growing wheat crops in the

7
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Fig. 11. Soil moisture sensor (SMS) placement in the

field.

Table 2 3.4.2. Irrigation application efficiency (Ea )

Placement of soil moisture sensors in the field at different depth. Irrigation application efficiency is the ratio of the quantity of water
Soil Moisture Sensor Depth, z (cm) Distance, x (m) stored into the root zone of the crops to the quantity of water delivered
into the field. Total water applied during each event was measured using
SMS 1 37.5 15 (25%L)
an ultrasonic flow meter (Model Unidata 6526E) installed in the channel
SMS 2 15.0 30 (50%L)
SMS 3 7.5 45 (75%L) just after the gate. Irrigation application efficiency was calculated using
Eq. (3) [38].

Table 3 𝑉𝑠
𝐸𝑎 = × 100 (3)
Description of different operational schedules of automatic check gate. 𝑉𝑎

Operational Where, Ea is irrigation application efficiency, Vs is the amount of

schedule (OS) Description water stored in the root zone and Va is the volume of applied irrigation.
OS1 Gate close according to soil moisture sensor 1 at 37.5 cm ∑3
Vs was calculated before each irrigation using formula 𝑉𝑠 = 𝜃𝑖 × 𝜌𝑖 ×
OS2 Gate close according to soil moisture sensor 2 at 15 cm 𝑖=1
OS3 Gate close according to soil moisture sensor 3 at 7.5 cm 𝑍𝑖 where 𝜃𝑖 is mass water content before irrigation, 𝜌𝑖 is the bulk density
and 𝑍𝑖 is the depth of soil layer in cm for ith layer.

rabi season (December to April). Different combinations of depth (z) 3.4.3. Distribution uniformity (DUlq )
and distance (x) of the sensors were tested to identify the best location Distribution uniformity is a measure of how uniformly water is ap-
of the sensor in automated surface irrigation (Table 2). Three SMSs were plied to the target area. The low quarter DU (DUlq ) is a measure of the
placed at three different soil depths i.e., 37.5 cm, 15 cm, and 7.5 cm at average of the lowest quarter of samples, divided by the average of all
25%, 50% and 75% of the length (L) of field, respectively to monitor soil samples. The higher the DUlq , the better the coverage of the area mea-
moisture at the head, middle and end reach of the basin layout (Fig. 11). sured. The soil samples were collected after each irrigation event from
a 10 m × 4.5 m grid to determine the distribution uniformity. DUlq was
3.3. Irrigation management calculated using Eq. (4) [39].
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑙𝑜𝑤𝑒𝑠𝑡 1∕4
The operation of the check gate is based on the real-time soil mois- 𝐷 𝑈𝑙 𝑞 = × 100 (4)
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑎𝑙𝑙
ture status. The check gate opened as the pre-set threshold values of soil
moisture sensor number 2 reached or exceed corresponds to 40%, 30% 3.4.4. Water requirement efficiency (WRE)
and 20% soil moisture depletion of field capacity. The check gate was Water requirement efficiency (WRE) is intended to measure the de-
closed as the soil moisture content reached 90% of the field capacity of gree to which the field has been under irrigated. The value of this pa-
the respective soil moisture sensor. The description of the three different rameter is always 100% when the entire field has been fully irrigated.
operational schedules for the closing of the gate is given in Table 3. WRE was calculated by Eq. (5) [40].
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑑𝑒𝑝𝑡ℎ 𝑜𝑓 𝑟𝑜𝑜𝑡 𝑧𝑜𝑛𝑒 𝑠𝑡𝑜𝑟𝑎𝑔𝑒
3.4. Irrigation system performance 𝑊 𝑅𝐸 = × 100 (5)
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑑𝑒𝑝𝑡ℎ 𝑜𝑓 𝑝𝑜𝑡𝑒𝑛𝑡𝑖𝑎𝑙 𝑠𝑡𝑜𝑟𝑎𝑔𝑒

The overall irrigation system performance was assessed by irrigation 4. Results and discussion
performance indicators and the performance of an automatic check gate
to maintain the pre-set water condition in the field. 4.1. Irrigation performance

3.4.1. Irrigation performance indicators The automatic irrigation system was tested in bare soil with differ-
Three well-known performance indicators such as irrigation appli- ent threshold values of soil moisture. Three operational schedules were
cation efficiency, low quarter distribution efficiency and water require- monitored for evaluation of the system. For each operational sched-
ment efficiency were used to evaluate irrigation performance. ule, three irrigation events were monitored in the bare field condition.

8
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Table 4
The details of the irrigation events under bare soil condition.

Volumetric moisture
Operational Date of irrigation Soil moisture content (%) Irrigation Distribution Water
schedule (OS) (dd-mm-yy) Depletion (%) application uniformity requirement
Gate Open Gate Close efficiency (%) (DUlq) efficiency (%)

OS 1 04.09.20 ≥40 27.6 34.1 93.0 0.96 100

20.09.20 ≥30 29.6 33.4 92.8 0.92 100
10.09.20 ≥20 31.6 33.2 83.3 0.82 100
OS 2 05.10.20 ≥40 27.6 35.0 89.8 0.87 94.1
22.10.20 ≥30 29.6 35.3 84.7 0.84 91.2
27.10.20 ≥20 31.6 34.9 83.3 0.82 89.5
OS 3 20.11.20 ≥40 27.6 34.3 81.6 0.78 86.4
04.11.20 ≥30 29.6 35.5 82.3 0.84 89.2
11.11.20 ≥20 31.6 34.3 88.6 0.91 97.1

Fig. 12. Hydraulic parameters of nine irrigation events.

The various hydraulic parameters of irrigation events are presented in ture depletion whereas the maximum water requirement efficiency was
Table 4 and Fig. 12. The gate was completely opened and closed as the observed in 40 % soil moisture depletion.
threshold value reached the pre-set value during each irrigation event. The time difference between the water front to reach to a sensor and
The system took 80 s to open the gate and 70 s to completely close the time of cut-off was 22.3, 10.6 and 3.5 min in OS 1, OS 2 and OS, 3,
the gate. During each irrigation event check, the gate was successfully respectively. Since the soil moisture sensor was placed at 37.5 cm depth,
opened and closed automatically as per the pre-set value of soil moisture the time taken by water to reach the depth was maximum in OS 1. It
reached or exceeded. was observed that at 40% and 30% soil moisture depletion conditions,
In the first operational schedule (OS 1), irrigation was started when OS1 performed better than other operational schedules, however, low
soil moisture depletion reached or exceeded 40%, 30% and 20% of field distribution efficiency and water requirement efficiency were recorded.
capacity (FC). The flow rate varied from 12.7 to 18.2 lps. The waterfront This might have happened due to the early reach of a threshold value of
advance time to reach the soil moisture sensor (SMS) number 1 was soil moisture for closing in 20% soil moisture depletion condition (soil
4.8, 6.2 and 5.7 min and the cut-off time of cut-off was 30, 33 and moisture sensor at 37.5 cm depth at 15% length in the field). The low
21 min in 40%, 30% and 20% of FC, respectively. The highest irrigation distribution uniformity showed that a quarter of the field was under
application efficiency, distribution uniformity and water requirement irrigated. A similar trend was observed in OS3 at 40% soil moisture de-
efficiency were observed 93.1 %, 0.96 and 100 %, respectively in 40 % pletion. Low distribution efficiency and water requirement showed that
depletion of FC (Table 4). the pre-set value of soil moisture (90% of field capacity) reached early
In the second operational schedule (OS 2), the flow rate and time even before the entire field got irrigated (soil moisture sensor located at
of cut-off varied from 10.7 to 15.7 lps and 24 to 42 min, respectively. 7.5 cm depth and 75% length distance).
The waterfront advance reached at soil moisture sensor (SMS2) in 25.9, The irrigation performance indicators i.e., irrigation application effi-
28.6 and 15.6 min at 40%, 30% and 20% soil moisture depletion, respec- ciency, distribution efficiency and water requirement efficiency were
tively. The highest application efficiency was observed at 40% depletion found to be higher in 40% soil moisture depletion in OS1 and OS2
followed by 30% and 20% depletion of FC. In the third operation sched- compared to OS3. It indicates that the larger soil moisture deficit im-
ule (OS 3), the flow rates and time of cut-off varied from 10.2 to 13.6 proved surface irrigation performance. Overall, it was found that OS
lps and 21 to 45 min, respectively. The highest irrigation application 1 performed best among three operational schedules. The performance
efficiency and distribution uniformity were found in 20 % soil mois- indicators suggested that the best location of soil moisture sensor for

9
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

Table 5 a 12 V solar rechargeable battery. No power issue has emerged during

Cost of different components of automated irrigation system. the evaluation period.
S. No. Component Cost (US$)

1. Check gate and controller unit 4.2.2. Cost of the system

a. Gate with frame 40.0 The cost of an automated irrigation system comprising four modules,
b. GSM module 10.7 check gate and controller unit, solar charging module, gateway and soil
c. Motor drive 1.3 moisture sensor module are presented in Table 5. The total approximate
d. Ultrasonic sensor 9.3
e. Arduino Nano 3.9
cost of an automated irrigation system was US$160.
f. Voltage converter 0.7
Sub Total 65.9 5. Conclusions
2. Solar charging module
a. 10-W solar panel 10
b. Solar charger controller 3.3 An automatic check gate has been developed for the basin and con-
c. 12 V battery 9.2 nected to IoT through the wireless soil moisture sensor network using
Sub Total 22.5 GSM with the aim that farmers/users can operate the gate opening and
3. Gateway
closing remotely with minimum manual interventions. Farmer can get
a. LoRa module 13.3
b. GSM module 10.7 the real-time moisture status of the field remotely and can make a de-
c. Micro controller 3.2 cision of irrigation start and end on the basic real-time moisture status.
d. Buck converter 0.7 The system was successfully tested in bare loamy soil. The study also
e. Antenna 10.7 attempted to identify the suitable location and number of sensors in
Sub Total 38.6
4. Soil moisture sensor
the field to increase irrigation application efficiency. Three operational
a. Capacitance sensor 2.5 schedules based on a different combination of depth and length along
b. LoRA module 13.3 the flow had been evaluated, which indicated that at least two soil mois-
c. Micro controller 3.2 ture sensors are required for the efficient operation of the system. One
d. 5 V Li-Ion battery 9.3
for the opening which is at 15 cm depth and 50% length of the field and
e. Solar charger module 3.2
e. PVC pipe 2.7 another for closing the system. The best location of the sensor for clos-
f. Solar panel 0.9 ing the system would be at 37.5 cm depth placed at 25% length from
Sub Total 32.0 the inlet in higher soil moisture deficit condition and at 7.5 cm depth
Total 159.5 or say US$160 placed at 75% length in low moisture deficit condition. It indicated that
the location of the sensor for closing the system depends upon the soil
moisture depletion. Overall, the average application efficiency had been
closing the gate should be at head reach i.e. 25% L from the field in- improved up to 86.6% through automation. The hardware and software
let and at 37.5 cm depth in higher soil moisture deficit condition while of the system have performed well in different scenarios. The automatic
at low soil moisture deficit scenario, the best location of the sensor for check gate is quite robust in different weather conditions (rainy, hot,
closing the system should be at 7.5 cm depth and 75% length from the in- winter) with minimal maintenance. The web/ mobile-based application
let. The maximum deep percolation occurred at head reach in the basin helped to maintain desired soil moisture remotely with minimal manual
[41]. Thus, by reducing deep percolation at head reach, application effi- intervention. The power consumption was also sustainable with using
ciency could be improved significantly [5]. A similar line of results was solar energy. Automation significantly improves application efficiency,
also reported by Amiri et al., [42] indicating that the suitable depth of reduces labor requirement, remotely accessed, and operates based on a
sensors should be in the range of 10–20 cm in an automatic drip irriga- real-time basis. Overall, the results indicated that the automatic check
tion system under maize crop. Overall, an average of 86.6% application gate has performed well with wireless sensors and irrigation applica-
efficiency, 0.90 distribution uniformity and 94.2% water requirement tion efficiency has improved significantly through IoT based systems
efficiency were observed in nine irrigation events. It showed that au- using wireless soil moisture based automatic basin irrigation system.
tomation could be an effective approach to improve the irrigation ap- The results showed a bright prospect of automation in surface irrigation
plication of surface irrigation systems. The soil moisture deficit greatly systems to improve irrigation efficiency.
affects the system performance and eventually sensor placement.
Declaration of Competing Interest
4.2. Assessment of automatic check gate
The authors declare that they have no known competing financial
The automatic check gate was assessed based on its opening and clos- interests or personal relationships that could have appeared to influence
ing according to predefined soil moisture status, its robustness, power the work reported in this paper.
and cost-effectiveness.
Acknowledgment
4.2.1. Power requirement
The power requirement of the system was met out by solar energy The authors would like to thank ICAR- Indian Institute of Soil and
through a solar photovoltaic module. The system was quite energy- Water Conservation, Dehradun for granting study leave. We are thank-
efficient, as it required less maintenance and a sufficient power supply ful to Water Technology Center, Farm Operation Service Unit (FOSU),
to the system was available even on cloudy days. A fully charged bat- ICAR-Indian Agricultural Research Institute, New Delhi and M/S Ags-
tery was sufficient to operate the check gate in three consecutive days martic for providing all the research and experimental support.
without charging in case of cloud cover. The power required by the actu-
ator i.e. 12 V gearbox 70 rpm DC motor required 2500 mA. The voltage References
requirement for soil moisture sensor varied from 3.7 to 5.5 V (5 mA)
and for gateway 12 V power was required. GSM required a minimum of [1] V. Dhawan, Water and agriculture in India. Federal Ministry of Food and Agriculture,
1 A to operate while the microcontroller required 26 μA. The check gate (2017) 1-28.
[2] FAOWater for Sustainable Food and Agriculture: A report produced for the G20
controller has different components which required a minimum of 5 V Presidency of Germany, Food and Agriculture Organization of the United Nations
and 1 A to operate. Although the check gate controller was powered by Rome, 2017.

10
M. Pramanik, M. Khanna, M. Singh et al. Smart Agricultural Technology 2 (2022) 100032

[3] D.L. Bjorneberg, Irrigation Methods, Reference Module in Earth Sys- [23] I.F. Akyildiz, I.H. Kasimoglu, Wireless sensor and actor networks: re-
tems and Environmental Sciences, Elsevier, 2013 ISBN 9780124095489, search challenges, Ad Hoc Netw. 4 (2) (2004) 351–367 ISSN 1570-8705,
doi:10.1016/B978-0-12-409548-9.05195-2. doi:10.1016/j.adhoc.2004.04.003.
[4] GoIPocket book of Agricultural Statistics, Government of India, Ministry of Agricul- [24] J. Rodríguez-Robles, Á. Martin, S. Martin, J.A. Ruipérez-Valiente, M. Castro, Au-
ture & Farmers Welfare, Department of Agriculture, Cooperation & Farmers Welfare, tonomous sensor network for rural agriculture environments, low cost, and energy
Directorate of Economics & Statistics, New Delhi, 2017. self-charge, Sustainability. 12 (15) (2020) 5913, doi:10.3390/su12155913.
[5] R.J. Smith, S.R. Raine, J. Minkovich, Irrigation application efficiency and deep [25] D. Loukatos, K.G. Arvanitis, in: Multi-Modal Sensor Nodes in Experimental Scalable
drainage potential under surface irrigated cotton, Agric. Water Manag. 71 (2) (2005) Agricultural IoT Application Scenarios, Springer, Berlin, Germany, 2021, pp. 101–
117–130. 128, doi:10.1007/978-3-030-71172-6_5.
[6] V. Dhawan. Water and agriculture in India, background paper for the South Asia [26] J. Muangprathub, N. Boonnam, S. Kajornkasirat, N. Lekbangpong, A. Wanichsombat,
expert panel during the global forum for food and agriculture (GFFA) (2017) P. Nillaor, IoT and agriculture data analysis for smart farm, Comput. Electron. Agric.
[7] R.K. Sivanappan, Status, scope and future prospects of micro-irrigation in India, in: 156 (2019) 467–474, doi:10.1016/j.co mpag.2018.12.011.
Proceedings of the Workshop on micro-irrigation and sprinkler irrigation system, [27] J. Lloret, S. Sendra, L. Garcia, J.M. Jimenez, A wireless sensor network deploy-
28-30, 2008, pp. 1–7. CBIP New DelhiApril. ment for soil moisture monitoring in precision agriculture, Sensors 21 (2021) 7243,
[8] S.R. Raine, The future of irrigation technologies and practices is here now, keynote doi:10.3390/s21217243.
address, in: Proceedings of the Irrigation Australia 2006 Conference, Brisbane, Aus- [28] A.N. Prasad, K.A. Mamun, F.R. Islam, H. Haqva, Smart quality monitoring system,
tralia, 2006. in: Proceedings of the 2nd Asia-Pacific World Congress on Computer Science Engi-
[9] A.J. Clemmens, Feedback control of basin irrigation system, J. Irrig. Drain. Eng. 118 neering, 2015, pp. 1–6.
(3) (1992) 481-196, doi:10.1061/(ASCE)0733-9437(1992)118:3(480). [29] G. Parameswaran, K. Sivaprasath, Arduino based smart drip irrigation system using
[10] R.J. Smith, M. Uddin, M.H. Gillies, P.M. Clurey, Evaluating the per- internet of things, Int. J. Eng. Sci. Comput. 6 (5) (2016) 5518–5521.
formance of automated bay irrigation, Irrig. Sci. 34 (2016) 175–185, [30] S.J. Patil, A. Patil, Precision agriculture for water management using IoT, Int. J.
doi:10.1007/s00271-016-0494-8. Recent Innov. Trends Comput. Commun. 5 (12) (2017) 142–144.
[11] R.M. Michael, A. Hussain, M.H. Gillies, J. Nicholas, Inflow rate and [31] Y. Shekhar, E. Dagur, S. Mishra, R.J. Tom, M. Veeramanikandan, S. Sankara-
border irrigation performance, Agric. Water Manag. 155 (2015) 76–86, narayanan, Intelligent IoT based automated irrigation system, Int. J. Appl. Eng. Res.
doi:10.1016/J.AGWAT.2015.03.017. 2 (18) (2017) 7306–7320.
[12] H.M. Al-Ghobari, F.S. Mohammad, M.S.A. El Marazky, A.Z. Dewidar, Automated [32] M. Mehra, S. Saxena, S. Sankaranarayanan, R.J. Tom, M. Veeramanikandan, IoT
irrigation systems for wheat and tomato crops in arid regions, Water SA 43 (2017) based hydroponics system using deep neural networks, Comput. Electron. Agric.
354–364, doi:10.4314/wsa.v43i2.18. 155 (2018) 473–486, doi:10.1016/j.compag.2018.10.015.
[13] A.P. Bowlekar, S.T. Patil, U.S. Kadam, M.S. Mane, S.B. Nandgude, N.K. Palte, [33] S. Nath, J.K. Nath, P.K.C. Sarma, IoT based system for continuous measurement and
Performance evaluation of real time automatic irrigation system on the yield monitoring of temperature, soil moisture and relative humidity, Int. J. Electron. Eng.
of cabbage (Brassica oleracea L.), Int. J. Pure Appl. Biosci. 7 (2019) 160–165, Technol. 9 (3) (2018) 106–113.
doi:10.18782/2320-7051.7433. [34] M.C. Vuran, A. Salam, R. Wong, S. Irmak, Internet of underground things: sensing
[14] S. Millán, J. Casadesús, C. Campillo, M.J. Moñino, M.H. Prieto, Using soil moisture and communications on the field for precision agriculture. IEEE World Forum Inter-
sensors for automated irrigation scheduling in a plum crop, Water 11 (2019) 2061, net of Things, WF-IoT (2018) 586-91.
doi:10.3390/w11102061. [35] S.S. Singh, Smart irrigation system using IoT, Int. J. Innov. Technol. Explor. Eng. 8
[15] M. Hardie, Review of novel and emerging proximal soil moisture sensors for use in (12) (2019) 183–186 DOI, doi:10.35940/ijitee. L1054.10812S19.
agriculture, Sensors 20 (2020) 6934, doi:10.3390/s20236934. [36] E.A. Abioye, M.S.Z. Abidin, M.S.A. Mahmud, S. Buyamin, A. Rahman, A.O. Otuoze,
[16] J. Vera, W. Conejero, A.B. Mira-García, M.R. Conesa, M.C. Ruiz-Sánchez, Towards M.S.A. Ramli, O.D. Ijike, IoT-based monitoring and data-driven modelling of drip
irrigation automation based on dielectric soil sensors, J. Hortic. Sci. Biotechnol. irrigation system for mustard leaf cultivation experiment, Inf. Process. Agric. 8 (2)
(2021) 1–12. (2021) 270–283, doi:10.1016/j.inpa.2020.05.004.
[17] R.J. Smith, J.M. Uddin, M.H. Gillies, Evaluating the performance of automated bay [37] In: A. Barman, B. Neogi, S. Pal, Solar-powered automated IoT-based drip ir-
irrigation, Irrig. Sci. 34 (2016) (2016) 175–185, doi:10.1007/s00271-016-0494-8. rigation system, in: P. Pattnaik, R. Kumar, S. Pal, S. Panda (Eds.), IoT and
[18] J. Gutiérrez, J.F. Villa-Medina, A. Nieto-Garibay, M.A. Porta-Gándara, Automated Analytics for Agriculture. Studies in Big Data, 63 Springer, Singapore, 2020,
irrigation system using a wireless sensor network and GPRS module, IEEE Trans. doi:10.1007/978-981-13-9177-4_2.
Instrum. Meas. 63 (2013) 166–176, doi:10.1109/TIM.2013.2276487. [38] C.M. Burt, A.J. Clemmens, K.H. Solomon, R.D. Bliesner, L.A. Hardy, T.A. Howell,
[19] D. Masseroni, P. Moller, M.Romani R.Tyrell, A. Lasagna, G. Sali, A. Facchi, C. Gan- Irrigation performance measures: efficiency and uniformity, ASCE J. Irrig. Drain.
dolfi, Evaluating performances of the first automatic system for paddy irrigation in Eng. 123 (6) (1997) 423–442, doi:10.1061/(ASCE)0733-9437(1997)123:6(423)11.
Europe, Agric. Water Manag. 201 (2018) 58–69, doi:10.1016/j.agwat.2017.12.019. [39] J.L. Merrian, J. Keller, Farm Irrigation System Evaluation, 3rd ed., Agricultural and
[20] A.B. Mandavia, Modernization of irrigation system operational management by way Irrigation Engineering Department, Utah State University, Logan, Utah, 1978.
of canal automation in India, in: Proceedings of the Modernization of irrigation sys- [40] W.R. Walker, G.V. Skogerboe Surface Irrigation: Theory and Practice, Prentice-Hall
tem operations: 5th ITIS network international meeting, Aurangabad, 1998 28-30 Inc, Englewood Cliffs, NJ, USA, 1987.
October. [41] R.K. Sharma, H.R. Singh, Deep percolation losses in border-checks, J. Soil Water
[21] K.S. Shingote, P. Shahane, Microcontroller based flow control system for canal Conserv. 8 (2) (2009) 65–74.
gates in irrigation canal automation, in: Proceedings of the IEEE 6th In- [42] Z. Amiri, M. Gheysari, M.R. Mosaddegh, S. Amiri, M.S Tabatabaei, An attempt to
ternational Conference on Advanced Computing (IACC), 2016, pp. 796–800, find a suitable place for soil moisture sensor in a drip irrigation system, Info. Process.
doi:10.1109/IACC.2016.152. Agric. (2021), doi:10.1016/j.inpa.2021.04.010.
[22] N. Sales, O. Remédios, A. Arsenio, Wireless sensor and actuator system for smart
irrigation on the cloud, in: Proceedings of the IEEE 2nd World Forum on Internet of
Things (WF-IoT), 2015, pp. 693–698, doi:10.1109/WF-IoT.2015.7389138.

11