An internet of things labelled dataset for aquaponics fish pond water quality monitoring system

Abstract

Aquaculture, which is the breeding of fishes in artificial ponds, seems to be gaining popularity among urban and sub-urban dwellers in Sub-Saharan Africa and Asia. Tenant aquaculture enables individuals irrespective of their profession to grow fishes locally in a little space. However, there are challenges facing aquaculture such as the availability of water, how to monitor and manage water quality, and more seriously, the problem of absence of dataset with which the farmer can use as a guide for fish breeding. Aquaponics is a system that combines conventional aquaculture with hydroponics (the method of growing plants in water i.e. soilless farming of crops). It uses these two technologies in a symbiotic combination in which the plant uses the waste from the fish as food while at the same time filtering the water for immediate re-use by the fish. This helps to solve the problem of frequent change of water. An Internet of Things (IoT) system consisting of an ESP-32 microcontroller which controls water quality sensors in aquaponics fish ponds was designed and developed for automatic data collection. The sensors include temperature, pH, dissolved oxygen, turbidity, ammonia and nitrate sensors. The IoT system reads water quality data and uploads the same to the cloud in real time. The dataset is visualized in the cloud and downloaded for the purposes of data analytics and decision-making. We present the dataset in this paper. The dataset will be very useful to the agriculture, aquaculture, data science and machine learning communities. The insights such dataset will provide when subjected to machine learning and data analytics will be very useful to fish farmers, informing them when to change the pond water, what stocking density to apply, provide knowledge about feed conversion ratios, and in predict the growth rate and patterns of their fishes.

Full text

Data in Brief 43 (2022) 108400
Contents lists available at ScienceDirect
Data in Brief
journal homepage: www.elsevier.com/locate/dib
Data Article
An internet of things lab elle d dataset for
aquaponics fish pond water quality
monitoring system
C.N. Udanor
a , ∗, N.I. Ossai
b
, E.O. Nweke
a
, B.O. Ogbuokiri
a
,
A.H. Eneh
a
, C.H. Ugwuishiwu
a
, S.O. Aneke
a
, A.O. Ezuwgu
a
,
P.O. Ugwoke
a
, Arua Christiana
c
a
Department of Computer Science, University of Nigeria Nsukka, Nigeria
b
Department of Environmental Biology and Zoology, University of Nigeria Nsukka, Nigeria
c
ICT Unit, University of Nigeria Nsukka, Nigeria
a r t i c l e i n f o
Article history:
Received 20 December 2021
Revised 12 June 2022
Accepted 16 June 2022
Available online 20 June 2022
Keywo rds:
Aquaculture
Sensors
Microcontroller
Catfish
aquaponics
IoT
a b s t r a c t
Aquaculture, which is the breeding of fishes in artificial
ponds, seems to be gaining popularity among urban and sub-
urban dwellers in Sub-Saharan Africa and Asia. Tenant aqua-
culture enables individuals irrespective of their profession to
grow fishes locally in a little space. However, there are chal-
lenges facing aquaculture such as the availability of water,
how to monitor and manage water quality, and more se-
riously, the problem of absence of dataset with which the
farmer can use as a guide for fish breeding. Aquaponics is a
system that combines conventional aquaculture with hydro-
ponics (the method of growing plants in water i.e. soilless
farming of crops). It uses these two technologies in a symbi-
otic combination in which the plant uses the waste from the
fish as food while at the same time filtering the water for im-
mediate re-use by the fish. This helps to solve the problem of
frequent change of water. An Internet of Things (IoT) system
consisting of an ESP-32 microcontroller which controls water
quality sensors in aquaponics fish ponds was designed and
developed for automatic data collection. The sensors include
∗Corresponding author.
E-mail address: collins.udanor@unn.edu.ng (C.N. Udanor).
https://doi.org/10.1016/j.dib.2022.108400
2352-3409/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license
( http://creativecommons.org/licenses/by/4.0/ )

2 C.N. Udanor, N.I. Ossai and E.O. Nweke et al. / Data in Brief 43 (2022) 108 40 0
temperature, pH, dissolved oxygen, turbidity, ammonia and
nitrate sensors. The IoT system reads water quality data and
uploads the same to the cloud in real time. The dataset is
visualized in the cloud and downloaded for the purposes of
data analytics and decision-making. We present the dataset
in this paper. The dataset will be very useful to the agricul-
ture, aquaculture, data science and machine learning com-
munities. The insights such dataset will provide when sub-
jected to machine learning and data analytics will be very
useful to fish farmers, informing them when to change the
pond water, what stocking density to apply, provide knowl-
edge about feed conversion ratios, and in predict the growth
rate and patterns of their fishes.
©2022 The Authors. Published by Elsevier Inc.
This is an open access article under the CC BY license
( http://creativecommons.org/licenses/by/4.0/ )
Specifications Tabl e
Subject Aquaculture
Specific subject area Aquaponics –Aquaculture (fish breeding) and Plant breeding (hydroponics)
Type of data Tabl e,
Chart
How the data were acquired The datasets were acquired using water quality sensors such as submersible
temperature sensor, dissolved oxygen , turbidity, pH, ammonia, and nitrate
sensors connected to ESP 32 [4] , a 32-bit microcontroller. The microcontroller
has an in-built WiFi module which enables the sensors’ data to be
automatically uploaded to a cloud computing platform, known as Thingspeak
IoT cloud, through an Internet of Things (IoT) network. C programming
language was used to code the software program written to control the
sensors
with the Arduino 1.8 . 4 integrated development environment (IDE)
known as sketch. The code was embedded in the microcontroller.
The sensors used include:
• 1. DF Rob ot pH sensor probe for Arduino version 2.0
• 2. DF Robot Dissolved Oxygen sensor probe for Arduino
• 3. DF Robot DC 5V TS-300B Turbidity Sensor Module Mixed Water
Detection Module Water Quality Tes t Turbidity Transducer For Arduino
• 4. Dallas DS18B20 temperature sensor.
• 5. Ammonia detection sensor NH
3
gas sensor module MQ137
• 6. Nitrate detection sensor NO
3
gas sensor module MQ135
The dataset were cleaned, annotated and uploaded to the Kaggle Data
Repository [13] for machine learning, which is licensed and free to download.
Data format Raw (in .csv format)
Description of data collection The IoT system was programmed to automatically read six water quality
parameters for
each of the 12 fish ponds and transmit them to the cloud storage
hosted by Thingspeak ( https://thingspeak.com/channels/14140 62/ ) every 5 s .
Besides the data collected by the IoT system, measurements were made
fourth-nightly of the fish’s weight and length by random sampling. About 10 to 15
samples were taken.
Data source location • Institution: Universit y of Nigeria Nsukka
• City/Town/Region: Nsukka/Nsukka/Enugu State
• Country: Nigeria
• Latitude and longitude (and GPS coordinates, if possible) for collected
samples/data: Latitude: 6.85813, Longitude: 7.3968
Data accessibility Repository name: Kaggle
Direct link to the dataset: https://www.kaggle.com/dataset/
e81da8b7666dc7af41cdc3aa5ef96c5547e4f412598a030f40d4 4 4550965e34f
Data identification number (DOI):
10.34740/kaggle/dsv/2681778
No access control is required to view or download the datasets.

C.N. Udanor, N.I. Ossai and E.O. Nweke et al. / Data in Brief 43 (2022) 108 40 0 3
Value of the Data
• The sensor parameter readings from the fish pond gives realtime situation report on the
condition of the water in the fish pond.
• Agricultural consultants, policy makers, and government agencies such as the ministry of
agriculture will need the dataset to advise farmers and governments both for planning and
predicting the performance of the fish.
• With the dataset being publicly available researchers, lecturers, and the machine learning
community can use it to develop models that predict the growth rate of the fish, feed in-
take and feed conversion rates.
1. Data Description
The data files are 12 comma separated values (.csv) format files, each representing an
aquaponics fish pond as shown in Table 1 . The files are named serially as in IoT_pond1.csv to
IoT_pond12.csv
The table contains the date and time the data were collected, the entry id of each data from 1
to n. Then the next six columns contain the IoT water quality parameters. The last two columns
were manually measured values as described above.
Fig. 1 shows a screen capture of the dataset on Kaggle website. Kaggle provides a graphical
view of each of the dataset. The figure shows a section of the dataset as displayed on the Kaggle
data repository. The columns in the figure represent the sensors such as the temperature, tur-
bidity, dissolved oxygen, etc. The respective values of the sensors are also plotted in the figure
using histograms.
Tabl e 1
Dataset table.
created_at entry_id Temperature
(C)
Turbidity
(NTU)
Dissolved
Oxygen
(mg/l)
pH Ammonia
(mg/l)
Nitrate
(mg/l)
Length
(cm)
Weig ht
(g)
Fig. 1. Screenshot of the pond1 aquaponics IoT datasets.

4 C.N. Udanor, N.I. Ossai and E.O. Nweke et al. / Data in Brief 43 (2022) 108 40 0
2. Experimental Design, Materials and Methods
Wate r quality impacts on fish growth rate, feed consumption, and their general wellbeing
[1 , 2] . Farmers’ ignorance of how to manage pond water has resulted to the death of fishes [3] .
Six water quality sensors were used during the experiment for the data collection. These
included temperature, turbidity, pH, dissolved oxygen, ammonia, and nitrate sensors. Each of
the sensors was first of all calibrated according to industry and manufacturers’ specifications
before being programmed and tested. The process of calibration involved using some calibration
solutions, in some cases (such as sodium hydroxyl), as in the cases of the pH and dissolved
oxygen sensors. The gas sensors were also calibrated with the relevant gases, after which the
calibration codes were run. At the end of the modular programming, all the codes for the six
sensors were integrated and the system was tested until the output was satisfactory.
The IoT system was programmed to automatically read the six water quality parameters for
each of the 12 aquaponics fish ponds and transmit them to the cloud storage hosted by Things-
peak ( https://thingspeak.com/channels/1414062/ ) every 5 s. Besides the data collected by the
IoT system, measurements were made fourth-nightly of the fish’s weight and length by random
sampling. About 10 to 15 samples were taken from each pond, weighed on an electronic scale
and measured with a measuring tape. The length and weight of the fish were then added as the
7th and 8th columns, respectively to the downloaded IoT water parameters tables for each of
the 12 ponds. Thus, a complete dataset table was built having the IoT-generated water quality
parameters and the manually measured length and weight parameters. IoT terminals may be in-
corporated with intelligence to take decisions, invoke actions and provide amazing services and
improve quality of life [5 , 6] .
Experimental Setup : The experiment was carried out in mobile aquaponics tarpaulin ponds
measuring 2 ×1 ×1 m
3
. The ponds were used in an indoor production system. The volume of
water was 1 ×2 ×0.9(m
3
). The ponds were filled at 0.9 m
3 level of water and covered with a
mosquito net. Seed beds of a chosen leguminous plant (amarantus spp.), the African spinach
(Amaranthus hybridus), were constructed and fixed on top of the pond cover of the mobile
tarpaulin pond. Each pond was connected to a 0.5 horsepower water pump for recirculation.
Sensors of the different parameters were submerged into pond water following the methods of
the pond controller system design described in sections above.
Physico-chemical parameters remain key factors in sustainable aquaculture production. These
parameters include: Temperature, pH, dissolved oxygen, ammonia, turbidity, nitrate, and nitrite
among others.
i. Temperature : The Dallas DS18B20 waterproof temperature sensor was submerged in each of
the 12 aquaponics ponds, being connected to the ESP 32 microcontroller. It senses the water
temperature in degrees Celsius and returns the value in realtime to the microcontroller which
in turns uploads the values to the cloud repository in the appropriate data table as shown in
Table 1 . The values were monitored to ensure they do not exceed the acceptable temperature
for aquaculture species ranges between 25.5 °C and 30.5 °C.
ii. pH : DF Robot pH sensor probe for Arduino version 2.0 was used to measure the degree of
acidity or alkalinity in fish ponds to see if the values lie within the acceptable range of pH
for tropical aquaculture is 6.5–8.2.
iii. Dissolved Oxygen (DO) : DO has relatively lower solubility and availability in aquatic life than
in terrestrial environments [7 , 8] . It is also required for several processes in the aquatic envi-
ronment such as; oxidation, nitrification, and decomposition [9] . DF Robot Dissolved Oxygen
sensor probe for Arduino is the sensor interfaced with the ESP 32 microcontroller for col-
lecting the dissolved oxygen in the fish pond. The amount of dissolved oxygen in water is
measured in milligrams per litre (mg/l).
iv. Ammonia : Ammonia accumulates in fish ponds due to the breakdown of the protein rich
fish feeds [10] . Ammonia detection sensor NH
3
gas sensor module MQ137 was interfaced
with the ESP 32 microcontroller. The Ammonia sensor was suspended above the pond water

C.N. Udanor, N.I. Ossai and E.O. Nweke et al. / Data in Brief 43 (2022) 108 40 0 5
and senses the concentration and toxicity in ponds in realtime. The amount of ammonia is
measured in milligrams per litre (mg/l).
v. Turbidity : Turbidity measures the clarity or otherwise of a water sample. The turbidity unit is
measured in NTU “Nephelometric Turbidity Units”. Turbidity sensors measure reflected (IR)
light at 90 degrees to the source. The infrared phototransistor [11] will have a change in
resistance itself and the change of voltage sensor will be obtained. This enables the DF Robot
DC 5V TS-300B Turbidity Sensor Module Mixed Water Detection Module Water Quality Tes t
Turbidity Transducer For Arduino connected to the ESP 32 microcontroller to measure the
pond water turbidity.
vi. Nitrate : is a by-product of nitrite oxidation during the latter stages of the nitrogen cycle and
is present to some degree in all fish ponds [12] . Nitrate detection sensor NO
3
gas sensor
module MQ135 interfaced with the ESP 32 microcontroller was suspended above the pond
water to measure the nitrate concentration in milligrams per litre (mg/l).
Experimental Animals : 10 0 0 numbers of five (5) week old fingerlings (mean initial weight: 2
±1.1 g) of the African catfish ( Clariasgariepinus ) were procured from a reputable hatchery and
acclimatized for two weeks. During the acclimatization, the fish were fed with diets containing
42% crude protein at the rate of 7% of their body weights in divided rations of morning (8:00 h)
and evening (4:00 h).
Ethics Statements
The experiment complied with the ARRIVE guidelines and were carried out according to the
UK Animals (Scientific Procedures) Act, 1986 and associated guidelines; EU Directive 2010/63/EU
for animal experiments.
Funding
This work was supported by the Meridian Institute under the Lacuna Fund for Agricultural
Dataset 2020 [Grant No.: 0326-S-001, 2020 ], at 105 Village Place, Dillon, Colorado 80435, United
States.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal rela-
tionships that could have appeared to influence the work reported in this paper.
The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:
Data Availability
Sensor Based Aquaponics Fish Pond Datasets (Original data) (Kaggle).
CRediT Author Statement
C.N. Udanor: Conceptualization, Methodology, Software, Supervision, Writing – original draft;
N.I. Ossai: Methodology, Data curation; E.O. Nweke: Data curation, Writing –review & editing;
B.O. Ogbuokiri: Software, Validation, Visualization, Data curation; A.H. Eneh: Software, Valida-
tion, Visualization, Data curation; C.H. Ugwuishiwu: Data curation; S.O. Aneke: Data curation;
A.O. Ezuwgu: Data curation; P.O . Ugwoke: Data curation; Arua Christiana: Data curation.

6 C.N. Udanor, N.I. Ossai and E.O. Nweke et al. / Data in Brief 43 (2022) 108 40 0
Acknowledgments
We acknowledge the support of the Department of Zoology and Environmental Biology, Uni-
versity of Nigeria in providing the enabling ground for this research work.
References
[1] C.N. Udanor , N.I. Ossai , B.O. Ogbuokiri , O.E. Nweke , P.O . Ugwoke , U.K. Ome ,A Pilot Implementation of a Remote IoT
Sensors for Aquaponics System Datasets Acquisition, Springer Nature, 2021 In Press .
[2] H.A. Mohammed, I. Al-Mejibi, Smart monitoring and controlling system to enhance fish production with minimum
cost, J. Theor. Appl. Inf. Technol. 98 (10) (2018) 2872–2885 Vol. ISSN: 1992-8645, E-ISSN:1817-3195 Ava ila ble: https:
//www.jatit.org/volumes/vol96No10/12vol96No10.pdf. . Accessed October 28, 2019 .
[3] PenState Extension, Wa ter Quality Concerns for Ponds, PenState Extension. Available at: https://extension.psu.edu/
water- quality- concerns- for- ponds (Retrieved: 7th March, 2021)
[4] Expressif Systems, ESP32 WROOM32 Datasheet, Expressif
Systems, Available at: https://www.espressif.com/sites/
default/files/documentation/esp32- wroom- 32 _ datasheet _ en.pdf
[5] V. Rozsa , M. Denisczwicz , M.L Dutra , P. Ghodous , C.F. da Silva , N. Moayeri , F. Biennier , N. Figay , An application
domain-based taxonomy for IoT sensors, in: Proceedings of the 23rd ISPE International
Conference on Transdisci-
plinary Engineering: Crossing Boundaries, Curit i ba, Brazil, 2016, pp. 249–258. Oct .
[6] R. Khan , S.U. Khan , R. Zaheer , S. Khan , Future internet: the internet of things architecture, possible applications and
key challenges, in: Proceedings of the 10 th International Conference on Frontie rs of Information Technology, IEEE,
2012, pp. 257–260 .
[7] A. Bhatnagar , P. Devi , Water quality guidelines for the management of pond fish culture, Int. J. Environ. Sci. 3 (6)
(2013) 1980–2009 .
[8] G. Delince , in: The Ecology of the Fish Pond Ecosystem with Special Reference to Africa,
Kluwer Ac ademic Publish-
ers, Dordrecht, 1992, p. 230pp .
[9] J. Erez , M.D. Krom , T. Neuwirth , Daily oxygen variatio ns in marine fish ponds, Elat, Israel, Aquaculture 84 (3-4)
(1990) 289–305 .
[10] Southern Regional Aquaculture Centre (SERAC)Managing ammonia. The Fish Site Newsletter, University of St. An-
drews Scotland, 2013 .
[11] L.W. Hakim , L. Hasanah , B. Mulyanti , A. Aminudin , Characterization of turbidity water sensor SEN0189 on the
changes of total suspended solids in the water, J. Phys.: Conf. Ser. 1280 (2) (2019) 1280–022064 .
[12] S. Sharpe , How to lower nitrate in the aquarium, (2020) Available at:  13:it al ic  https://www.thesprucepets.com/
nitrates- in- the- aquarium- 1381883
 /13:italic 
[13] U. Collins, B. Ogbuokiri, N. Onyinye, Sensor Based Aquaponics Fish Pond Datasets, Kaggle, 2022 Data set, doi: 10.
34740/KAGGLE/DSV/3748790 .