The Use of Big Data in Water Resources Management

Sara Bouziane

, Badraddine Aghoutane

, Aniss Moumen

, Ali Sahlaoui

, Anas El Ouali

Informatics and Applications Laboratory, Science Faculty of Meknes, Moulay Ismail University, Meknes, Morocco

Ibn Tofail University, National School of Applied Sciences, Kenitra, Morocco

Laboratory of Geo-Engineering and Environment, Faculty of Sciences, Moulay Ismail University, Meknes, Morocco

Department of Environment, Functional ecology and environmental engineering laboratory, Faculty of Sciences and

Technology, Sidi Mohamed Ben Abdellah University, Fez, Morocco

Keywords: Big Data Analytics, Water Resources Management, Information System, IoT.

Abstract: Water management has become an essential vector in the Moroccan government policy since independence.

The volume of the water data, and the diversity of the actors, lead us to think about new methods of analysis

to manage water resources with efficiency. The development of new technologies, such as Big Data, is an

essential tool for assuring this management. Researchers have used these technologies to develop several

architectures and algorithms to deal with the water scarcity concern. In this paper, we identify research works

related to the topics of water and Big Data, and we discuss the different proposed architectures, according to

a review of systemic exploratory literature. In the end, we draw up our perspectives.

1 INTRODUCTION

We found in recent years much talk in public debate

and the social sciences about Big Data and IoT

technologies (Elhassan et al., 2020). That provides

methods to collect, manage, and process a large

volume of data that ensures the management of large

and diversified information systems such as water

data.

In Morocco, water has been considered an

essential component of the economy since

independence (Hafed et al. 2018). In this perspective,

the Water Basin Agency (ABH), as an independent

public organization, is implementing a water

management strategy with the various actors of the

water sector at the level of the watersheds following

the water strategy coordinated and monitored by the

Directorate General of Water (Moumen et al. 2016).

And all over the world, due to increasing water

demand, climate, and hydrological deficit, water

resource managers have started to look for practical

strategies for water resource management.

2 LITERATURE REVIEW

2.1 Methodology

The literature review refers to a process and reporting

structure to classify and identify research and results

published to date for a given topic (in our case Big

Data in the field of water resources) (Moumen et al.

2016).

To conduct our Literature Review, as explained in

Figure 1, we have identified 217 references in the

various databases and have stored them in the Zotero

library.

Figure 1: Working methodology.

Constitution of the

ZOTERO library

RIS File Export

NVIVO treatment

Analysis

Bouziane, S., Aghoutane, B., Moumen, A., Sahlaoui, A. and El Ouali, A.

The Use of Big Data in Water Resources Management.

DOI: 10.5220/0010728000003101

In Proceedings of the 2nd International Conference on Big Data, Modelling and Machine Learning (BML 2021), pages 38-44

ISBN: 978-989-758-559-3

We have created this number of references using

the method shown in Figure 2.

Figure 2: Literature review process.

2.2 Meta Analyzes

2.2.1 Analyze of References

Table 1 below represents the number of references by

bibliographical category.

Table 1: References categories.

Type of references Number

Journal articles 110 (57,9%)

Conference papers 76 (40%)

Chapte

s & books 4 (2,1%)

Total 190

We have collected these references from various

databases, as shown in Figure 3 below. Scopus

occupies first place with 79 articles, followed by the

Institute of Electrical and Electronics Engineers

(IEEE) with 64 references, and ScienceDirect with 40

documents. We have collected the other documents (7

references) from the Google Scholar database.

Figure 3: Percentage of references by databases.

Table 2 below presents the complete list of

journals containing articles. We can notice that most

articles are published in the journals “Advances in

Intelligent Systems and Computing” and

“Environmental Modelling & Software”.

Table 2: Research papers per journals.

Journal

Nb.

Articles

Advances in Intelligent Systems and Computing 6

Environmental Modelling & Software 5

Procedia Computer Science 4

Procedia Engineering 4

Water (Switzerland) 4

Communications in Computer and Information

Science

Journal of Cleaner Production 3

Arabian Journal of Geosciences 2

Computers and Electronics in Agriculture 2

IEEE Access 2

Journal of Petroleum Science and Engineering 2

Sustainable Computing: Informatics and Systems 2

Unassigne

Water Research 2

Advances in Water Resources 1

Agricultural Systems 1

Agricultural Water Managemen

Applied Geography 1

Applied Sciences (Switzerland) 1

Array 1

Biosystems Engineering 1

Computational Geosciences 1

Computer Communications 1

Computer Networks 1

Computers and Geosciences 1

Computers, Materials and Continua 1

Concurrency Computation 1

Current Opinion in Biotechnology 1

Dili Xuebao/Acta Geographica Sinica 1

E3S Web of Conferences 1

Environmental science & technology 1

Environmental Science and Technology 1

Environmental Technology & Innovation 1

Environmental Technology and Innovation 1

Field Crops Research 1

Future Generation Computer Systems 1

Groundwate

Handbook of Environmental Chemistry 1

Hydrology and Earth System Sciences 1

HydroResearch 1

IEEE Latin America Transactions 1

IEEE Sensors Journal 1

IEEE Transactions on Big Data 1

IEEE Transactions on Industrial Informatics 1

IEEE Transactions on Sustainable Computing 1

International Journal of Embedded Systems 1

Irrigation and Drainage 1

Jilin Daxue Xuebao (Diqiu Kexue Ban)/Journal of

Jilin University (Earth Science Edition)

Journal of Ambient Intelligence and Smart

Environments

Journal of Coastal Research 1

Journal of Environmental Managemen

The Use of Big Data in Water Resources Management

Table 2: Research papers per journals (cont.).

Journal

Nb.

Articles

Journal of Hohai University 1

Journal of Marine Systems 1

Journal of Sustainable Development of Energy,

Water and Environment Systems

Journal of the Chinese Institute of Civil and

Hydraulic Engineering

Lecture Notes in Electrical Engineering 1

Lecture Notes of the Institute for Computer

Sciences, Social-Informatics and

Telecommunications Engineering, LNICST

Linye Kexue/Scientia Silvae Sinicae 1

Meitan Xuebao/Journal of the China Coal Society 1

Modeling and Optimization in Science and

Technologies

Nongye Jixie Xuebao/Transactions of the Chinese

Society for Agricultural Machinery

Open Agriculture 1

Remote Sensing 1

Remote Sensing of Environmen

Research of Environmental Sciences 1

Science of the Total Environment 1

Sensors 1

Sensors (Switzerland) 1

Sensors and Actuators B: Chemical 1

Shuili Fadian Xuebao/Journal of Hydroelectric

Engineering

Shuili Xuebao/Journal of Hydraulic Engineering 1

SN Applied Sciences 1

Sustainability (Switzerland) 1

Sustainability Science 1

Sustainable Cities and Society 1

Utilities Policy 1

Water Resources and Industry 1

Water Resources Management 1

Water Science and Technology 1

Water Security 1

Wiley Interdisciplinary Reviews: Wate

Total 110

The following Figure 7 shows that most of the

sources in our dataset were published in the years

2020 and 2021. Also, the year 2015 was the start point

of publishing works relating to water and Big Data.

Figure 7: Number of sources per year.

Table 3 classifies the number of papers in our

dataset by language, and we can see that 188

references are written in English, and only two

sources are written in French.

Table 3: Research sources per language.

Language Number

lish 188

French 2

Total 190

2.2.2 Word Analysis

We present the most commonly used and treated

words in our documents in a word cloud as a visual

representation demonstrated in Figure 4.

At first glance, we note that ‘Water’, ‘data’, ‘big’,

‘management’, and ‘system’ are the most common

words in our references. That refers to the central

issue of this paper, which is the use of a Big Data

system in water resources management.

Figure 4: Word cloud.

That is also illustrated in Figure 5 next, which

represents the most frequent word occurrences. We

notice that the predominantly used word are ‘water’,

‘data’, ‘big’, ‘system’, and ‘management’, which

highlights the context of our paper.

BML 2021 - INTERNATIONAL CONFERENCE ON BIG DATA, MODELLING AND MACHINE LEARNING (BML’21)

Figure 5: Most frequent words.

Figure 6 shows the least important commonly

used words, which are ‘IoT’, ‘Lake’ (57), ‘Applying’,

‘Indices’ (56), ‘Groundwater’, ‘Objective’,

‘Conservancy’, ‘Demand’ (52), and finally

‘Challenges’ and ‘Implementation’ (51).

Figure 6: Least frequent word occurrences.

2.2.3 Grid Analysis

This thematic analysis provides a basis for data

comparison to find out approaches in behaviour like

domain, topic, technology, etc.

Our analysis allowed us to have 18 topics, as

described in Table 4.

Table 4: List of relevant topics and associated terms.

Topic Words

Big Data

"Big Data" processing analysis

storage analytics "data

management" visualization

inte

ration

redictin

IoT

IoT sensor "internet of things"

"remote sensin

" real-time

Decision

decision "decision tool" "decision-

makers" DDS SDSS "business

intelli

ence" "decision makin

BDA tools

implementation spark apache

Hadoop NoSQL Map Reduce

framewor

Analysis

Statistic exploratory analysis

orithm "machine learnin

Information

ste

"information system"

Cloud "cloud computing"

Architecture Architecture

Algorithm Algorithm Algorithms

Computing platform system computing

Water

water Groundwater hydrological

wastewate

Climate Climate

Agriculture Agriculture

Urban Urban

Energy Energy

Food Food

Analysis of the domains of application in Table 5

next shows that 180 documents deal with the topic of

water, 47 documents treat the subject of decision-

making, and the domains of agriculture (22), urban

(26), energy (21), food (7), and climate (13).

Table 5: Domain of application.

Water 180

Decision 47

Agriculture 22

Urban 26

Energy 21

Food 7

Climate 13

As shown in Table 6, the commonly treated

technologies are those associated with “Big Data”

(178), followed by “Computing” (154),

“Architecture” (25), “Algorithm” (32), “IoT” (92),

“BDA Tools” (93), “Information System” (10),

“Analysis” (100), and “Cloud computing” cited in 12

references.

The Use of Big Data in Water Resources Management

Table 6: Technology and statistics.

Big Data 178

Computing 154

Architecture 25

Algorithm 32

IoT 92

BDA Tools 93

Information System 10

Analysis 100

Cloud 12

Table 7 below represents the crossover of the most

cited technologies with fields of application. We

notice that Big Data, Computing, IoT, and BDA

Tools are the most generally cited technologies

compared to others in the water area. That proves that

we can move considerably in our research work in the

Big Data application. We also notice that 32

references are talking about algorithms in the water

field and 23 present architectures in this field of

application.

Table 7: Grid of cited technologies and domains of

application.

lgorithm

nalysis

rchitecture

BDA

Tools

Big

Data

Cloud

Computing

Information

system

IoT

Agriculture

3 7 3 9 19 1 17 1 11

Climate

1 4 2 5 11 1 12 0 6

Decision

10 27 7 19 45 6 39 4 25

Energy

5 12 6 10 21 1 19 1 9

Food

0 2 0 3 5 0 5 0 2

Urban

3 14 1 16 24 0 25 1 11

Water

32 97 23 87 169

10 145

10 88

3 DISCUSSION

At the end of this review, and based on the above

crossover, we discuss and compare the different

architectures applied in water management.

3.1 Theoretical Architectures

Using topics Architecture/Water, we found articles

presenting Big Data architectures for water

management. Following, we describe the three must-

referenced results.

3.1.1 Big Data Open Platform for Water

Resources Management

In this paper, the authors present the first part of a Big

Data Open Platform that can manage and process a

massive volume of water data collected from different

sources and anticipate and avoid catastrophic

situations such as floods and droughts. This part is

developed and based on J2EE technology and

provides better decision-making relating to the

management of natural resources.

The presented model is composed of 9 blocks, as

shown in Figure 7 below.

Figure 7: Big Data Open Platform for Water Resources

Management proposed by Ridouane Chalh et al. (Chalh et

al. 2015).

3.1.2 Application of Big Data in Water

Ecological Environment Monitoring

In this paper, the authors present a Big Data system

for China's water ecological data management, which

BML 2021 - INTERNATIONAL CONFERENCE ON BIG DATA, MODELLING AND MACHINE LEARNING (BML’21)

the primary services are: the control of measurement

of the water environmental data so we can use them

in environmental protection projects; the government

of the nutritional value of water bodies, so it helps in

fisheries; and also it serves in the domains of the

water conservancy engineering and coastal

engineering by providing data support for

environmental protection strategies of water

conservancy projects and coastal projects.

Figure 8 next represents the presented model and

shows the processing of the local data based on cloud

computing and artificial intelligence that provides

classification, count, and visualization of these data.

Figure 8: Flow Chart of Data Processing presented by

Huanchun Ma et al. (Huanchun Ma et al. 2020).

The authors also presented some problems of

water environmental data and the corresponding

countermeasures.

3.1.3 A Pilot Infrastructure for Searching

Rainfall Metadata and Generating

Rainfall T Product using the Big Data

of NEXRAD

In this paper, Bong-Chul Seo et al. (Moumen et al.

2016) present the pilot framework developed by The

Iowa Flood Center (IFC) and known as IFC-Cloud-

NEXRAD. The main objective of this infrastructure

is to explore rainfall metadata and generate rainfall

products over the Iowa domain. The framework is

based on the NEXRAD radar Level II and data are

accessible via cloud storage, which allows

researchers to ready access to the radar data.

Figure 9 below presents the architecture of the

IFC-Cloud-NEXRAD system.

Figure 9: IFC-Cloud-NEXRAD framework architecture.

The Use of Big Data in Water Resources Management

3.2 Comparison of Proposed

Architectures

Table 8.

Modules Statut Advantages Limits

Architecture 1

Chalh et al. (2015)

 Users

 Interfaces

 Search Engine

 Decision

Module

 Knowledge-

Based System

 Big Data

Analysis

module

 Geographic

Information

System (GIS)

 Simulation

Models

 Communicatio

n module

 Computing

and processing

module

The platform is in the design stage.

 Presence of a

decision-

making

module

 Integration of

the

geographic

module

 Presence of a

search engine

 Architecture

does not

follow a

process of

data flow.

Architecture 2

Huanchun Ma et al. (2020)

 Big Data

Analysis

Module

 Geographic

Information

System (GIS)

 Data

warehouse

 Visualization

module

 Processing

module

 User interface

 Decision-

making



The system is available.

 Neural

network

system and

decision-

making

 Data mining

 The

automation

of data

acquisition

still needs to

be developed

 Data

management

is not secure

enough

 The data

mining

scheme

needs to be

improved

Architecture 3

Bong-Chul Seo et al. (2019)

 Computing

and processing

module

 Users interface

 Space-time

research

 Big Data

Analysis

module

 Decision and

estimation

module

 Geographic

Information

System (GIS)

 Radar

 Visualization

Module

The framework is developed and could be

transferrable to other geographic regions and

li i

 Ready access

to NEXRAD

radar data

 Provides

rainfall

products

 Based on

cloud

computing

 Estimation of

rainfall

metadaa

 The

maximum

length of

period for

requesting

products is

limited to 30

days.

4 CONCLUSION

This work summarizes our exploratory study by

identifying and discussing the different research

works in the water domain and Big Data technologies.

In terms of perspectives, the next step is to study

different proposed machine learning algorithms and

models for collecting data relating to water and utilize

these results to carry out the prototyping of a Big Data

system for water management, according to users'

expectations.

REFERENCES

Jamal Elhassan et al., 2020. Big Data Analytic Architecture

for Water Resources Management: A Systematic

Review, In Proceedings of the 4th Edition of

International Conference on Geo-IT and Water

Resources 2020, Geo-IT and Water Resources 2020

(GEOIT4W- 2020). Association for Computing

Machinery, New York, NY, USA, Article 19, 1–5.

Hafed, K. et al. 2018. "Decisional Information Systems of

the Public Actors in Moroccan Oasis Zones: Case

Study Draa-Tafilalet Region", International

Conference on Intelligent Systems and Computer

Vision (ISCV 18), doi:10.1109/isacv.2018.8354047.

Moumen, A. et al., 2016. Web Services for Sharing and

Reuse Water Data In Morocco. Proceedings of the

Mediterranean Conference on Information &

Communication Technologies 2015, 297–302.

doi:10.1007/978-3-319-30298-0_31

R.Chalh et al., June 2015. Big Data Open Platform for

Water Resources Management, International

Conference on Cloud Technologies and Applications

(CloudTech), Marrakech, Morocco.

Huanchun Ma et al., 2020. IOP Conf. Ser.: Mater. Sci. Eng.

750 012044.

Bong-Chul Seo et al., 2019. A Pilot Infrastructure for

Searching Rainfall Metadata and Generating Rainfall

T Product using the Big Data of NEXRAD,

Environmental Modelling and Software 117 (2019) 69–

75.

BML 2021 - INTERNATIONAL CONFERENCE ON BIG DATA, MODELLING AND MACHINE LEARNING (BML’21)