Environment and Renewable Energy: Proceedings of the 2023 9th International Conference on Environment and Renewable Energy (Springer Proceedings in Earth and Environmental Sciences) 9819700558, 9789819700554

This book presents peer-reviewed papers of the 2023 9th International Conference on Environment and Renewable Energy (IC

150 78 5MB

English Pages 164 [156] Year 2024

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Conference Committees
Preface
Contents
Heat Transfer, Heat Recovery, and Energy Conservation
Study on Soil Temperature Recovery Prediction Model of Ground Source Heat Pump
1 Introduction
2 Prediction Model Establishment
3 Experimental System
4 Model Verification
5 Conclusion
References
A Comparative Study of the Transcritical CO2 Cycle and the Organic Rankine Cycles Using R245fa and Low GWP Refrigerants in Low Temperature Geothermal Utilization
1 Introduction
2 Method
2.1 Cycle Descriptions
2.2 Energy Analysis
2.3 Simulation Process
3 Results
4 Conclusions
References
Numerical Simulation of the Influence of Air Space Thickness in Heat Transfer of a High-Performance Glazing System
1 Introduction
2 Model Definition
2.1 Definition of the High-Performance Glazing System
2.2 Governing Equations
2.3 Model Solution
3 Results and Discussion
3.1 The Thermal Conductance Values of Entire Selection (L2D)
3.2 The Influence of the Air Space Thickness in the Internal Heat Flux
3.3 The Total Heat Losses of the Glazing System in Wintertime
4 Conclusion
References
Renewable Energy and Clean Energy Technology
Impact of Location on Hybrid Renewable Energy System Design Optimization: A Case Study for Two Cities of USA
1 Introduction
2 Modeling of the System
2.1 Modeling of Photovoltaic Module
2.2 Modeling of Wind Turbine Generators
2.3 Modeling of the Battery
2.4 Modeling of Diesel Generator
3 Formulation of the Optimization Problem
4 Methodology
4.1 Algorithm Formulation
5 Data and Resource
6 Result Analysis
7 Conclusion and Future Work
References
Stability of a Hydroxyapatite-Supported Nickel Catalyst in Dry Reforming of Methane
1 Introduction
2 Materials and Methods
3 Results
3.1 Catalyst Stability
3.2 Catalyst Regeneration
4 Conclusion
References
Renewable-Based Energy Mix Optimization for Weak Interconnected Communities
1 Introduction
2 Materials and Methods
3 Case Study and Results
4 Conclusion
References
Pyrolysis Kinetics of Microalgae Residues—A Comparative Study on DAEM Using Different Distribution Functions
1 Introduction
2 Theoretical Background
2.1 Distributed Activation Energy Model
2.2 Continuous Distributed Activation Energy Function F(E)
3 Methodology
4 Results and Discussion
4.1 Kinetic Modelling with Gaussian Distributed Activation Energy Model
4.2 Kinetic Modelling with Logistic Distributed Activation Energy Model
4.3 Kinetic Modelling with Weibull Distributed Activation Energy Model
4.4 Comparison Between Three Distributed Activation Energy Models
5 Conclusion
References
Low-Cost Anodic Material Made of Rice Husk Charcoal and Acrylic Paint for Soil-Based Microbial Fuel Cells
1 Introduction
2 Method and Experiment
2.1 Electrode Fabrication Method
2.2 SMFC Setup and Measurement
2.3 Observation of Electrode Surface
3 Result and Discussion
3.1 SEM Image of the Anode
3.2 Anode Performance
4 Conclusion
References
Water Pollution Control, Water Resource Management and Air Quality Assessment
Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum, and Musa Paradisiaca Species as Natural Coagulants in the Treatment of the Cunas River—Peru
1 Introduction
2 Material and Methods
2.1 Water Sampling
2.2 Obtaining Natural Coagulants
2.3 Stock Solution
2.4 Dosages of Natural Coagulants
2.5 Jar-Test
2.6 Parameter Analysis
3 Result
3.1 Pre-test Results
3.2 Post-test Results
3.3 Effectiveness of Natural Coagulants
3.4 Optimal Doses
4 Discussion
5 Conclusion
References
Geological Factors Influencing River Morphological Changes: Implications in the Agricultural Sector
1 Introduction
2 Methodology
3 Results and Discussion
3.1 Geological Factors Influencing River Morphological Changes
3.2 The Implications of the River Morphological Changes to the Agricultural Sector
4 Conclusion
References
Assessment of Particle Filter Technique for Data Assimilation in the Forecasting of Streamflows for the Tocantins River Basin in Brazil
1 Introduction
2 Methodology
2.1 Semi-Distributed MGB Hydrological Model
2.2 Particle Filter (PF)
2.3 Uncertainty in Rainfall Forcing
3 Experimental Model
3.1 Area of Study and Available Data
3.2 Integration of Particle Filter with the MGB Hydrological Model
3.3 Parameters of the PF Scheme
4 Discussion of Results
4.1 Sensitivity Test
4.2 Streamflow Forecasting
5 Conclusion
References
Development of a Biodegradable Detergent Based on Quinoa as an Alternative to Minimize Eutrophication
1 Introduction
2 Materials and Methods
2.1 Biodegradable Detergent Formulations
2.2 Composition of the Biodegradable Detergent
2.3 Inputs and Quantities for Making the Biodegradable Detergent
2.4 Procedures to Elaborate the Biodegradable Detergent
2.5 Conventional Detergent Formulations
2.6 Efficacy of the Biodegradable Detergent
2.7 Sample Collection
3 Results and Discussions
3.1 Assay Methodology
3.2 Analysis of Analyzed Parameters
4 Conclusion
References
Author Index
Recommend Papers

Environment and Renewable Energy: Proceedings of the 2023 9th International Conference on Environment and Renewable Energy (Springer Proceedings in Earth and Environmental Sciences)
 9819700558, 9789819700554

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Springer Proceedings in Earth and Environmental Sciences

Pen-Chi Chiang   Editor

Environment and Renewable Energy Proceedings of the 2023 9th International Conference on Environment and Renewable Energy

Springer Proceedings in Earth and Environmental Sciences Series Editors Natalia S. Bezaeva, The Moscow Area, Russia Heloisa Helena Gomes Coe, Niterói, Rio de Janeiro, Brazil Muhammad Farrakh Nawaz, Institute of Environmental Studies, University of Karachi, Karachi, Pakistan

The series Springer Proceedings in Earth and Environmental Sciences publishes proceedings from scholarly meetings and workshops on all topics related to Environmental and Earth Sciences and related sciences. This series constitutes a comprehensive up-to-date source of reference on a field or subfield of relevance in Earth and Environmental Sciences. In addition to an overall evaluation of the interest, scientific quality, and timeliness of each proposal at the hands of the publisher, individual contributions are all refereed to the high quality standards of leading journals in the field. Thus, this series provides the research community with well-edited, authoritative reports on developments in the most exciting areas of environmental sciences, earth sciences and related fields.

Pen-Chi Chiang Editor

Environment and Renewable Energy Proceedings of the 2023 9th International Conference on Environment and Renewable Energy

Editor Pen-Chi Chiang Graduate Institute of Environmental Engineering National Taiwan University Taipei, Taiwan

ISSN 2524-342X ISSN 2524-3438 (electronic) Springer Proceedings in Earth and Environmental Sciences ISBN 978-981-97-0055-4 ISBN 978-981-97-0056-1 (eBook) https://doi.org/10.1007/978-981-97-0056-1 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore Paper in this product is recyclable.

Conference Committees

Conference General Chair Nguyen Thi Anh Tuyet, Hanoi University of Science and Technology, Vietnam Dean of School of Environmental Science and Engineering

Conference Chair Zoltán Pásztory, University of Sopron, Hungary

Conference Program Chair Dimitrios Karamanis, University of Patras, Greece

Conference Publication Chair Pen-Chi Chiang, National Taiwan University, Taiwan

Conference Local Chairs Nguyen Thuy Chung, Hanoi University of Science and Technology, Vietnam Vu Dinh Tuan, VNU-University of science, Hanoi, Vietnam

v

vi

Conference Committees

International Steering Committee Haliza Abdul Rahman, Universiti Putra Malaysia, Malaysia Doan Pham Minh, Centre Rapsodee, IMT Mines Albi, France Siroux Monica, INSA Strasbourg (National Institute of Applied Sciences), France

International Technical Committee Amit Kumar, Malaviya National Institute of Technology Jaipur, Jaipur, India Nakamoto Trang, Ritsumeikan University, Japan Zbigniew Leonowicz, Wrocław University of Science and Technology, Poland Mohammad Arif Kamal, Aligarh Muslim University, India Muhammad Imran Khan, Hamad Bin Khalifa University, Qatar Karena Quiroz Jiménez, Peruvian University of Applied Sciences, Peru Massimo Andretta, Alma Mater Studiorum-Università di Bologna, Italy Enrique Romero-Cadaval, University of Extremadura, Spain Siroux Monica, INSA Strasbourg (National Institute of Applied Sciences), France Eduard Krylov, Kalashnikov Izhevsk State Technical University, Russia Nga P. Le, Vietnam National University, Vietnam Quoc Dung Trinh, Hanoi University of Science and Technology, Vietnam Asher Brenner, Ben-Gurion University of the Negev, Israel K. S. K. Rao Patnaik, Andhra University, India Tiago Sequeira, CEFAGE-UBI, Universidade da Beira Interior, Portugal Susana Garrido Azevedo, CEFAGE-UBI, Universidade da Beira Interior, Portugal Muslum Arici, Kocaeli University, Turkey Jacek Dach, Poznan University of Life Sciences, Poland Arkadiusz Piwowar, Wroclaw University of Economics, Poland Khairil, Universitas Syiah Kuala, Indonesia Berislav Andrlic, Polytechnic in Pozega, Croatia Imre Czupy, University of Sopron, Hungary Haliza Abdul Rahman, Universiti Putra Malaysia, Malaysia Amares Singh, University Tunku Abdul Rahman, Malaysia Doan Pham Minh, Centre Rapsodee, IMT Mines Albi, France Le Kieu Hiep, Hanoi University of Science and Technology, Vietnam Hung Shangchao, Fuzhou Polytechnic, China B. Shahul Hamid Khan, Indian Institute of Information Technology, Kancheepuram, India Thusitha Sugathapala, University of Mpratuwa, Sri Lanka Aria Gusti, Universitas Andalas, Indonesia Rubén Mogrovejo, Universidad Peruana de Ciencias Aplicadas, Peru Yu Feng, Tsinghua University, Beijing, China

Preface

With great pleasure we are introducing the proceedings of 2023 9th International Conference on Environment and Renewable Energy. 2023 9th International Conference on Environment and Renewable Energy (ICERE 2023) was held successfully during February 24–26, 2023, Hanoi, Vietnam. This event is expected to encourage the passion for researching Environment and Renewable Energy. The event was graced by three Keynote speakers and three invited speakers. Prof. Nguyen Thi Anh Tuyet, the Dean of the School of Environmental Science and Technology, Hanoi University of Science and Technology, Vietnam; Prof. Marco Abbiati from University of Bologna, Italy; Prof. Dimitrios Karamanis from University of Patras, Greece; Assoc. Prof. Bashir Ahmmad ARIMA from Yamagata University, Japan; Assoc. Prof. Doan Pham Minh from Deputy Director of RAPSODEE Research Center, France; Dr. Nguyen Thuy Chung from Hanoi University of Science and Technology, Vietnam. Each keynote speech took 30 minutes and invited speech took 20 minutes for presentation and discussion while normal speakers in technical sessions were given 12 minutes and 3 minutes for presentation and discussion, respectively. Conference is organized in 3 sub-sessions under the topics of: Environmental Pollution Control and Resource Management; Renewable Energy, Fuel Cell Technology and Energy Management; Food Processing, Food Chemistry, and Agroforestry Ecological Analysis. All participants actively participated in the meeting and had a heated discussion. All accepted papers presented at the ICERE 2023 were included in this volume, which contained three chapters with topics: (1) Heat Transfer, Heat Recovery, and Energy Conservation (2) Renewable Energy and Clean Energy Technology (3) Water Pollution Control, Water Resource Management, and Air Quality Assessment. The conference series is a result of selection of the most important articles. All articles were peer reviewed through process administrated by the Editors.

vii

viii

Preface

We would like to thank our colleagues who have organized the conference. Indeed, it was a successful conference. Last but not least, we also would like to express our gratitude to all authors who showed keen interest in ICERE 2023. Taipei, Taiwan June 2023

Pen-Chi Chiang

Contents

Heat Transfer, Heat Recovery, and Energy Conservation Study on Soil Temperature Recovery Prediction Model of Ground Source Heat Pump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Longyue Du, Jinhua Chen, Maochuan Tang, and Ruoen Xu A Comparative Study of the Transcritical CO2 Cycle and the Organic Rankine Cycles Using R245fa and Low GWP Refrigerants in Low Temperature Geothermal Utilization . . . . . . . . . . . . . Kun Hsien Lu, Hsiao Wei Chiang, and Pei Jen Wang Numerical Simulation of the Influence of Air Space Thickness in Heat Transfer of a High-Performance Glazing System . . . . . . . . . . . . . . Hung Anh Duong Le and Zoltán Pásztory

3

17

27

Renewable Energy and Clean Energy Technology Impact of Location on Hybrid Renewable Energy System Design Optimization: A Case Study for Two Cities of USA . . . . . . . . . . . . . . . . . . . Md. Arif Hossain, Taiyeb Hasan Sakib, Saad Mohammad Abdullah, Ashik Ahmed, and Ishtiza Azad Stability of a Hydroxyapatite-Supported Nickel Catalyst in Dry Reforming of Methane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Thanh Son Phan and Doan Pham Minh Renewable-Based Energy Mix Optimization for Weak Interconnected Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Giovanni Brumana, Elisa Ghirardi, Giuseppe Franchini, and Silvia Ravelli

41

55

65

ix

x

Contents

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study on DAEM Using Different Distribution Functions . . . . . . . . . . . . . . . . . . . . Khanh-Quang Tran, Hau-Huu Bui, Wei-Hsin Chen, Salman Raza Naqvi, Thuat T. Trinh, and Jo-Shu Chang Low-Cost Anodic Material Made of Rice Husk Charcoal and Acrylic Paint for Soil-Based Microbial Fuel Cells . . . . . . . . . . . . . . . . . Trang Nakamoto, Soichiro Hirose, Dung Nakamoto, Keisuke Nishida, and Kozo Taguchi

75

93

Water Pollution Control, Water Resource Management and Air Quality Assessment Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum, and Musa Paradisiaca Species as Natural Coagulants in the Treatment of the Cunas River—Peru . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Francis Eduardo Ambrosio Rosales, Ingrid Lucia Artica Cardenas, Leslie Alison Vargas Cordova, and Steve Dann Camargo Hinostroza Geological Factors Influencing River Morphological Changes: Implications in the Agricultural Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Akhmad Zamroni and Decibel V. Faustino-Eslava Assessment of Particle Filter Technique for Data Assimilation in the Forecasting of Streamflows for the Tocantins River Basin in Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Karena Quiroz Jiménez Development of a Biodegradable Detergent Based on Quinoa as an Alternative to Minimize Eutrophication . . . . . . . . . . . . . . . . . . . . . . . . 139 Karina Yupanqui Pacheco, José Vladimir Cornejo Tueros, and Fiorella Milagros Pacheco Dawson Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Heat Transfer, Heat Recovery, and Energy Conservation

Study on Soil Temperature Recovery Prediction Model of Ground Source Heat Pump Longyue Du, Jinhua Chen, Maochuan Tang, and Ruoen Xu

Abstract Establishing an accurate model and studying the soil temperature recovery characteristics of the ground heat exchangers under intermittent operation is the key to formulate the operation strategy of a ground source heat pump system under intermittent operation or composite ground source heat pump systems. In this paper, based on the analogy idea of the thermal resistance–capacitance model, the heat balance network equations between buried pipe, backfill material, and soil are constructed, and a model that can quickly and easily predict the recovery of soil temperature is proposed, and the soil temperature change during the intermittent operation of the system is predicted. By comparing the predicted value of the soil temperature change in each layer during the intermittent period of the system under the model proposed in this paper with the simulated value of the stratified heat transfer model which has been verified by the actual engineering measured data, it is concluded that in the 150 h intermittent recovery period, the relative error of the soil temperature recovery degree between the two is less than 10% in most of the time. Therefore, the model proposed in this paper can accurately predict the change of soil temperature in the intermittent period of buried pipe heat exchanger and provide a reference for the design of operation control strategy of buried pipe ground source heat pump system. Keywords Ground-coupled Heat Pump · Earth temperature prediction · Intermittent operation · Recovery model

1 Introduction Ground source heat pump has been paid more and more attention because of its advantages of high efficiency and energy saving. It has been applied in a large number of engineering projects and has become one of the important choices for a large number of new projects. In different regions, whether the intermittent and zonal operation of ground source heat pump system or the operation strategy of composite system L. Du · J. Chen (B) · M. Tang · R. Xu Chongqing University, Chongqing, China e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 P.-C. Chiang (ed.), Environment and Renewable Energy, Springer Proceedings in Earth and Environmental Sciences, https://doi.org/10.1007/978-981-97-0056-1_1

3

4

L. Du et al.

combined with other energy forms are all affected by soil temperature recovery. Therefore, accurate and rapid prediction of soil temperature changes is of great significance for the use of ground source heat pump systems and the optimization of control strategies. The soil temperature recovery characteristics of the ground heat exchanger during intermittent operation are mainly affected by factors such as system operation mode, geological conditions, and climatic conditions of the project location. Scholars at home and abroad have also conducted a lot of research from different perspectives. Yan Shang et al. [1] studied the effects of soil thermal conductivity, porosity, backfill material type, outdoor temperature, solar radiation intensity, and wind speed on soil temperature recovery, and concluded that the physical parameters of soil had a significant effect on its temperature recovery, while the effects of outdoor climatic conditions and other factors were not obvious. Sofyan et al. [2] proposed a new method (internal source term method) to evaluate the seasonal fluctuation of soil temperature by establishing the energy balance equation. Rees [3] studied the temperature fluctuation of the fluid in the U-tube flow process in the ground heat exchanger in a short period of time, and established a two-dimensional finite volume model of the ground heat exchanger suitable for medium and short time scale thermal response. Based on the moving finite line heat source model and superposition principle, Zhang et al. [4] proposed an analytical solution model of soil temperature during the intermittent period, and concluded that after 5 days of operation, the soil temperature rise under intermittent operation was 0.8 °C lower than that under continuous operation. Jin Guang [5] studied the heat and moisture migration of the soil during the heat transfer process of the buried pipe through the sandbox experiment, and concluded that the soil can be restored to the initial low temperature in the 18th hour after the shutdown of the system, and different start-stop ratios will affect the operation efficiency of the unit. Liu et al. [6] found that the temperature recovery of circulating water of single U-tube heat exchanger is faster than that of double U-tube ground heat exchanger by measuring the temperature variation of circulating medium during intermittent operation of ground source heat pump under heating conditions. Li et al. [7] tested the intermittent operation of heat pump and simulated the change of soil temperature under the operation and shutdown of heat pump. They believed that the recovery of soil temperature was not uniform with time, which mainly occurred in the early stage of the interval, and it was of little significance to wait for the absolute recovery of soil temperature for little significance. Liu et al. [8] simulated a practical project in Shanghai and analyzed the influence of the imbalance rate of cold and heat load on the ground temperature recovery. Huo et al. [9] established a soil layered heat transfer model based on the natural stratification of soil geology at different depths and the measured data of temperature measurement points at different depths, and verified its accuracy by Fluent simulation, which can be used for the verification of the model in this paper. Many scholars [10–14] use Fluent to simulate the recovery of soil temperature. In this paper, based on the Fluent simulation, the initial temperature of each heat exchange area in the intermittent period of the system is obtained. Combined with the heat balance network analysis method [15–19], a simplified numerical model is established to analyze the

Study on Soil Temperature Recovery Prediction Model of Ground …

5

soil temperature recovery. Soil temperature recovery can also be calculated easily and quickly to provide guidance for intermittent operation strategy.

2 Prediction Model Establishment The thermal resistance–capacitance model is usually used to predict the response process of fluid temperature and underground temperature field during the operation of the unit. By dividing the nodes inside and outside the borehole, the heat balance network is constructed to solve the problem. However, the model is usually only applicable to the case where the input cold and heat of the unit is constant. For the dynamic change of cold and heat input in actual operation, the calculation difficulty and calculation amount are greatly increased. When the unit is shut down, due to the disappearance of dynamic cold and heat input, the temperature recovery process of most soil is mainly determined by the initial temperature of each underground heat exchange node and the thermal resistance-heat capacity characteristics in its control area, except that the temperature recovery of the surface soil is affected by external climatic conditions. Based on the thermal resistance–capacitance model, the heat transfer medium of the ground heat exchanger is noded and the corresponding heat balance network is constructed. A prediction of soil temperature recovery process suitable for relatively constant boundary conditions is proposed. The soil temperature recovery model established in this paper mainly includes two parts. Firstly, the fluid (water supply, backwater part), backfill material, borehole wall, and other media and boundaries are discretized and simplified as heat exchange nodes. At the same time, the soil outside the borehole is divided into 10 control areas according to the radial control distance and simplified as heat exchange nodes, as shown in Fig. 1. Secondly, according to the heat transfer process between the nodes, the equivalent heat network is constructed as shown in Fig. 2, and the heat balance equation of each heat exchange node is listed accordingly. To simplify the model, make the following assumptions: (i) ignore the internal axial heat transfer of the same layer. That is, the heat transfer medium in the same layer is simplified into one node, the temperature of the regional node is the average volume temperature of the corresponding heat transfer medium, and the temperature of the boundary node is the average wall temperature of the corresponding boundary; (ii) Ignore the axial heat transfer between the layers. In addition, due to the different soil depths of each layer, in order to facilitate the calculation, the thermal resistance and heat capacity of the axial unit length of the heat exchange medium are used as the solution parameters in the model. The calculation formula symbols and units are shown in Table 1. Heat balance equation of fluid node in pipe: Cf

Tf2 − Tf1 dT f 1 Tg − T f 1 = + dt Rff R fg

(1)

6 Fig. 1 Diagram of interior and exterior nodes of borehole

Fig. 2 Heat balance network diagram

L. Du et al.

Study on Soil Temperature Recovery Prediction Model of Ground …

7

Table 1 Symbol description Variables

Subscripts

C

Heat capacity per unit length [J/ (m K)]

f

Fluid

T

Temperature [°C]

t

Time

R

Thermal resistance [(m K)/W]

g

Backfill material node

r

Radius [m]

b

Borehole boundary node

λ

Thermal conductivity [W/(m K)]

s, j

Radial jth layer node

k

Relative error

s, m

Axial layer m node

D

Distance [m]

Ground

Soil outer boundary node

e

Equivalent radius

bt

Summation

p

Pipe

po

Outer diameter of pipe

pi

Inner diameter of water pipe

Cf

Tf1 − Tf2 dT f 2 Tg − T f 2 = + dt Rff R fg

(2)

Heat balance equation of backfill node: Cg

T f 1 − Tg dTg T f 2 − Tg Tb − Tg = + + dt R fg R fg Rgb

(3)

Heat balance equation of borehole wall node: Tg − Tb Ts,1 − Tb + =0 Rgb Rs,1

(4)

First soil node heat balance equation: Cs,1

Tb − Ts,1 dTs,1 Ts,2 − Ts,1 = + dt Rbs Rs,1

(5)

Heat balance equation of middle layer soil node: Cs, j

Ts, j+1 − Ts, j dTs, j Ts, j−1 − Ts, j = + dt Rs, j Rs, j−1

(6)

Outermost node heat balance equation: Cs,10

TGr ound − Ts,10 dTs,10 Ts,9 − Ts,10 = + dt Rs,10 Rs,9

(7)

8

L. Du et al.

The thermal resistance R f g between the fluid node and the backfill node, and the thermal resistance Rgb between the backfill node and the borehole wall node will affect the total thermal resistance of the borehole during the intermittent period. The calculation formula of R f g is as follows: Rgb

( ) 1 rb = ln 2π λg rg

(8)

The position of the backfill node is the key to determining the thermal resistance Rgb . Considering the equivalent radius re and the drilling radius rb , the calculation formula of the equivalent circular area radius r g where the backfill node is located is as follows: / re2 + rb2 rg = (9) 2 √ For a single U buried pipe, the equivalent radius re = 2r po , r po is the inner diameter of the pipe (m). The calculation formula of the thermal resistance R f g between the fluid node and the backfill node in the tube is as follows: ) ( R f g = 2 Rbt − Rgb Rbt = Rb +

Rp 2

(10) (11)

Since there is no convective heat transfer during the soil temperature recovery process after the unit shutdown, the total thermal resistance of the borehole during the recovery process does not include the convective thermal resistance. The calculation formula of the thermal resistance R p of the pipe wall is: Rp =

r po 1 ln 2π λ p r pi

(12)

The method of solving the thermal resistance Rb of the backfill is complicated. In order to simplify the calculation, this paper selects the two-dimensional model of heat transfer in the borehole [20] to solve the problem. The calculation formula is as follows: ( ) rb4 rb λb − λs,m 1 2r b ln (13) + Rb = + ln ln 4π λb r po 2D λb + λs,m rb4 − D 4 The interference thermal resistance R f f between the supply and return pipes refers to the calculation method proposed by Louis Lamarche [21]. After solving the ' equivalent thermal resistance Ra in the borehole, the calculation formula of R f f is:

Study on Soil Temperature Recovery Prediction Model of Ground …

Rff =

4Ra 'Rb 4Rb − Ra '

9

(14)

In addition, the solution of the soil thermal resistance Rs, j outside the borehole is carried out without heat input after the heat pump is shut down. The soil thermal conductivity and the initial temperature of each node of the soil are known, and the steady-state heat conduction formula of the cylinder wall under the first boundary condition in the time step is used to solve, that is: Rs, j =

r j+1 1 ln 2π λs rj

(15)

After obtaining the above parameters, 5 s is taken as the time step, that is, the heat transfer between nodes within 5 s is considered to be steady-state heat transfer. The average initial temperature of each control area (i.e., heat exchange node) during the shutdown of the heat pump is obtained by the stratified heat transfer model, and the axial heat transfer is ignored. Substitute it into Formulas (1)–(7) for iterative calculation, and the temperature change data of each node after the shutdown of the heat pump can be obtained.

3 Experimental System In this paper, the predicted value of the soil restoration model is verified by the simulated value of the layered heat transfer model [9] verified by the measured data of a project in Sha Ping Ba District of Chongqing. Figure 3 is the experimental system of the project, which is mainly composed of ground heat exchanger, ground source heat pump unit, temperature monitoring, recording, acquisition system, and other supporting equipment. Through geological exploration, it is known that the geology of the project site is divided into 4 layers. In this project, temperature sensors with a spacing of 5 m or 10 m are arranged along the depth direction of the buried pipe wall. Combined with the engineering geological stratification and the arrangement of temperature measuring points, the underground rock and soil mass is divided into 12 layers, which are 0–3 m, 3–5 m, 5–10 m, 10–15 m, 15–20 m, 20–30 m, 30–36 m, 36–40 m, 40–60 m, 60–68 m, 68–80 m, 80–100 m.

4 Model Verification Due to the lack of measured data after unit shutdown, this paper uses the layered heat transfer model which has been verified to simulate the soil temperature recovery data after the actual project shutdown to verify the accuracy of the model.

10

L. Du et al.

Fig. 3 Schematic diagram of experimental system

Through actual measurement and analytical solution, the change rate η of comprehensive thermal conductivity between adjacent soil layers is controlled not to exceed 3%. Finally, the soil is divided into 7 layers, and ( the initial)boundary conditions |Rs,n −Rs,n−1 | × 100%, Rs,n is the of each layer are shown in Table 2. Where η = Rs,n−1 comprehensive thermal resistance of the nth layer of soil in the axial direction, and Rs,n−1 is the comprehensive thermal resistance of the n-1 layer of soil in the axial direction. Table 2 Initial boundary conditions of each layered soil Layer Depth number range

Nature of soil

Density Specific Thermal Conductive Initial (kg/m3 ) heat J/ conductivity resistance temperature (kg K) W/(m K) (m K)/W °C

1

0–3 m

Silty clay

1940.2

1755

2.383

0.139

23.03

2

3–15 m

Mudstone

2166.8

1194

2.069

0.165

19.68

3

15–30 m

Mudstone

2166.8

1110

1.925

0.178

19.35

4

30–36 m

Mudstone

2166.8

1162

2.015

0.170

19.25

5

36–40 m

Sandstone, 2290.6 mudstone

1599

2.600

0.128

19.26

6

40–68 m

Sandstone, 2290.6 mudstone

1557

2.533

0.131

19.21

7

68–100 m Siltstone, mudstone

997

2.145

0.164

19.18

2365.0

Study on Soil Temperature Recovery Prediction Model of Ground …

11

In addition, the soil belongs to the heat transfer medium with great heat capacity, and the soil temperature changes very slowly during the intermittent period of the heat pump. Therefore, if the absolute value of the soil temperature is used for comparative analysis, the accuracy of the analytical solution model cannot be well reflected. Therefore, this paper adopts the relative error of the soil temperature recovery degree to verify: k=|

|T ' − T0 | − |T − T0 | | × 100% |T − T0 |

(16)

In the formula: k represents the relative error of soil temperature recovery degree; T0 represents the initial soil temperature when the heat pump stops, °C; T represents the Fluent simulation soil recovery temperature based on the layered heat transfer ' model at a certain time, °C; T represents the soil recovery temperature of the model at a certain time, °C. Because the surface soil temperature is greatly affected by the external environment, it is difficult to use the numerical model to accurately predict. Therefore, this paper selects the 2nd to 7th layer of soil in the range of 3–100 m for temperature recovery analysis. The recovery model was verified by calculating the average soil temperature of each axial layer and the overall average soil temperature. In addition, in the case of ignoring the axial heat transfer, the heat transfer inside the soil diffuses from the borehole wall to the far boundary during the intermittent period of the heat pump. Therefore, the temperature change of the heat transfer area of each part of the soil can directly reflect the characteristics of this heat transfer process. Taking the fourth layer of soil (30–36 m) as an example, as shown in Fig. 4. It can be seen from Fig. 4 that the temperature of the first layer of soil in the radial direction adjacent to the borehole wall is significantly higher than that in other areas from the time of shutdown of the heat pump unit, while the soil temperature decreases rapidly with time and then decreases slowly. The main reason is that because the first layer of soil in the radial direction is the closest to the borehole at the time of shutdown, the temperature is the highest, and heat is continuously released to the far boundary, so the temperature has been decreasing. But with the passage of time, the temperature of the first layer of radial soil is getting lower and lower, and the temperature difference with other layers of soil is getting smaller and smaller, so the temperature drop rate gradually slows down. Finally, the radial first layer of soil gradually reaches a thermal equilibrium with the outside soil, and its temperature remains almost constant. The radial 2–6 layers of soil release heat to the soil in the near and far boundary areas while accepting the soil heat in the near borehole wall area, so the temperature change shows a trend of increasing first and then decreasing, and as the position gradually moves away from the borehole wall, the temperature inflection point gradually moves backward. The radial 7–10 layers mainly accept the heat transfer of the soil near the borehole wall, so the temperature generally shows an upward trend, and as the position gradually moves away from the borehole wall, the temperature rise rate gradually slows down.

12

L. Du et al.

Fig. 4 Comparison between simulated and predicted values of soil temperature recovery in radial direction

Based on the layered heat transfer model, the average initial temperature of 2–7 layers of soil during the shutdown of the heat pump unit was simulated by Fluent software to be 20.154 °C, 19.852 °C, 19.747 °C, 19.691 °C, 19.646 °C, 19.676 °C. With the increase of recovery time, the radial volume weighted average temperature of 2–7 layers of soil gradually decreased (Fig. 5), and the temperature decline rate gradually accelerated. This is due to the high temperature of the fluid in the tube when the heat pump stops, and the heat dissipation to the soil is still maintained at the initial stage of temperature recovery. However, because the heat capacity of the fluid in the tube is much lower than that of the soil, the temperature drop rate is faster, so as time goes on, the heat dissipation effect of the fluid in the tube on the soil gradually decreases, and the soil temperature drop rate increases. The simulated soil temperature is slightly larger than the model solution temperature at the beginning of the calculation. As time goes on, the difference between the simulated temperature and the model solution temperature becomes smaller and smaller, and except for the second layer of soil (3–15 m), the simulated temperature of the other five layers of soil

Study on Soil Temperature Recovery Prediction Model of Ground …

13

is lower than the model solution temperature. After error calculation, it can be seen that within about 23 h of temperature recovery, the relative error of soil temperature recovery degree. The absolute error of soil simulation temperature and model solution temperature is very small, and the relative error is also very small. However, due to the soil temperature recovery value is also very small, and it may be that the axial heat transfer effect in the early stage of temperature recovery is greater than the radial heat transfer, so the relative error of soil recovery temperature is larger, which is higher than 10%. However, with the passage of recovery time, although the absolute error of soil simulation temperature and model solution temperature increases, the difference is very small. At the end of the 150-h interval, except that the relative error of the soil temperature recovery degree of the 7th layer was 10.73%, the relative errors of the other 2–6 layers were 3.53%, 4.74%, 5.48%, 9.06%, and 7.85%, respectively, all less than 10%. Therefore, it can be considered that the model proposed in this paper is more accurate in solving the average recovery temperature of soil layers in 150 h of recovery period. After obtaining the average recovery temperature of each layer of soil, the overall average value of the soil is further solved according to the depth weighting, and the comparison between the simulated soil temperature and the model solution temperature is shown in Fig. 6. The comparison between the average simulated soil temperature and the model solution temperature is similar to that of 3–7 layers of soil. The simulated temperature is slightly higher than the model solution temperature at the beginning and gradually decreases to lower than the model solution temperature. At the beginning of the intermittent period, the relative error of the soil temperature recovery degree is large, but the absolute error and relative error between the simulated value and the predicted value of the soil recovery temperature have been maintained at a very small level, and the relative error of the soil temperature recovery degree has not exceeded 10% after about 24 h. Therefore, it can also be considered that the recovery model proposed in this paper has a high accuracy for the prediction of the overall soil temperature recovery during the 150-h intermittent period.

Fig. 5 Comparison and verification of average soil temperature of 2–7 layers in axial direction

14

L. Du et al.

Fig. 6 Comparison of soil average recovery temperature direction

5 Conclusion In this paper, the heat transfer medium is nodeized and the heat balance network is established based on the thermal resistance and capacitance heat transfer model. A model that can be used to predict the temperature recovery of the soil during the intermittent operation of the ground source heat pump system is proposed. At the end of the 150-h recovery period, except for the first layer of soil (0–3 m) which is greatly affected by the external environment, the relative error of the average recovery temperature of the other layers of soil is small, and the relative error of the degree of soil temperature recovery is 7.19%. The predicted value of soil recovery temperature during the whole intermittent period is very close to the simulated value, and the relative error of soil temperature recovery degree in most of the time is less than 10%, which has high accuracy for the recovery prediction of soil temperature during the intermittent period.

References 1. Shang, Y., Li, S., Li, H.: Analysis of geo-temperature recovery under intermittent operation of ground-source heat pump[J]. Energy Build. 43(4), 935–943 (2011) 2. Sofyan, S.E., Hu, E., Kotousov, A., et al.: A new approach to modelling of seasonal soil temperature fluctuations and their impact on the performance of a shallow borehole heat exchanger[J]. Case Studies Thermal Eng. 22, 100781 (2020) 3. Rees, S.J.: An extended two-dimensional borehole heat exchanger model for simulation of short and medium timescale thermal response[J]. Renew. Energy Oxford: Pergamon-Elsevier Sci. Ltd. 83, 518–526 (2015) 4. Analyses on soil temperature responses to intermittent heat rejection from BHEs in soils with groundwater advection[J]. Energy Build Elsevier 107, 355–365 (2015) 5. Jin, Chen, Guo, et al.: Experimental study on soil heat and moisture transfer under intermittent operation of ground source heat pump [J]. J. Sola Energy 42(2), 254–259 (2021)

Study on Soil Temperature Recovery Prediction Model of Ground …

15

6. Liu, Tang, Weei et al.: Experimental study on intermittent heating operation of ground source heat pump [J]. Renew. Energy (1), 59–61 (2008) 7. Shang, Li, Dai, et al. Study on the variation characteristics and recovery characteristics of ground temperature in intermittent operation of ground source heat pump [J]. J. Dalian Univ. Technol. 52(3), 350–356 (2012) 8. Liu, Zhang, Gao, et al.: Study on soil temperature recovery characteristics of ground source heat pump [J]. Hvac (11), 147–150 (2008) 9. Huo: Study on layered heat transfer model of vertical ground heat exchanger [D]. Chongqing University (2016) 10. Cui, P., Yang, H., Fang, Z.: Numerical analysis and experimental validation of heat transfer in ground heat exchangers in alternative operation modes[J]. Energy Build. 40(6), 1060–1066 (2008) 11. Gao, Q., Li, M., Yu, M.: Experiment and simulation of temperature characteristics of intermittently-controlled ground heat exchanges[J]. Renew. Energy Oxford: PergamonElsevier Sci. Ltd 35(6), 1169–1174 (2010) 12. Xiaoling, C., Yanping, Y., Liangliang, S., et al.: Restoration performance of vertical ground heat exchanger with various intermittent ratios[J]. Geothermics Oxford Pergamon-Elsevier Sci. Ltd 54, 115–121 (2015) 13. Qian, H., Wang, Y.: Modeling the interactions between the performance of ground source heat pumps and soil temperature variations[J]. Energy Sustain. Dev. 23, 115–121 (2014) 14. Jalaluddin, M.A.: Thermal performance investigation of several types of vertical ground heat exchangers with different operation mode[J]. Appl. Therm. Eng. 33–34, 167–174 (2012) 15. A computational capacity resistance model (CaRM) for vertical ground-coupled heat exchangers[J]. Renew. Energy Pergamon 35(7), 1537–1550 (2010) 16. Zarrella, A., Scarpa, M., De Carli, M.: Short time step analysis of vertical ground-coupled heat exchangers: the approach of CaRM[J]. Renew. Energy Oxford: Pergamon-Elsevier Sci. Ltd. 36(9), 2357–2367 (2011) 17. Bauer, D., Heidemann, W., Müller-Steinhagen, H., et al.: Thermal resistance and capacity models for borehole heat exchangers[J]. Int. J. Energy Res. 35(4), 312–320 (2011) 18. Bauer, D., Heidemann, W., Diersch, H.-J.G.: Transient 3D analysis of borehole heat exchanger modeling[J]. Geothermics 40(4), 250–260 (2011) 19. Minaei, A., Maerefat, M.: A new analytical model for short-term borehole heat exchanger based on thermal resistance capacity model[J]. Energy Build. 146, 233–242 (2017) 20. Minaei, A., Maerefat, M.: Thermal resistance capacity model for short-term borehole heat exchanger simulation with non-stiff ordinary differential equations[J]. Geothermics, Oxford: Pergamon-Elsevier Sci. Ltd 70, 260–270 (2017) 21. Lamarche, L., Kajl, S., Beauchamp, B.: A review of methods to evaluate borehole thermal resistances in geothermal heat-pump systems[J]. Geothermics, Oxford: Pergamon-Elsevier Sci. Ltd 39(2), 187–200 (2010)

A Comparative Study of the Transcritical CO2 Cycle and the Organic Rankine Cycles Using R245fa and Low GWP Refrigerants in Low Temperature Geothermal Utilization Kun Hsien Lu, Hsiao Wei Chiang, and Pei Jen Wang

Abstract Low temperature geothermal energy has high potential in clean and sustainable energy applications though it is difficult to be efficiently applied to conventional power units due to the low enthalpy. This study compares transcritical CO2 (T-CO2 ) cycle and organic Rankine cycles (ORCs) in terms of heat recovery efficiency (ηHR ) and system efficiency (ηsys ), using 90 °C–150 °C (Th,in ) of hot water as the heat source. The analysis was conducted via MATLAB and NIST REFPROP database. The working fluid selected for ORCs are R245fa and low global warming potential refrigerants, i.e. R152a, R600a, R1234ze(E) and R1234yf. The results show that the T-CO2 cycle with lower Th,in has dominant advantage in terms of ηHR , and its ηsys is comparable with the ORCs. Along with Th,in increases, the ηHR of the ORCs increase more rapidly; hence, the thermodynamic advantages of the T-CO2 cycle would be diminished. However, the floor-layout costs of the T-CO2 cycle is lower than the ORCs due to its small volumetric flow rate at the expander outlet, especially with high Th,in . In comparisons between the ORCs, R1234yf and R1234ze(E) can provide better system performance than the others, while R152a performs relatively inferior. Keywords Geothermal energy · Transcritical CO2 cycle · Organic rankine cycle

1 Introduction As there are more energy demands than ever owing to the increasing global population and industrial activities, it is inevitable to explore alternative energy sources to provide clean and sustainable energy. Geothermal energy has vast amount in the nature and it has no intermittent availability issue as may be encountered when using solar or wind energy. Therefore, applying a heat-to-power unit to harvest the K. H. Lu (B) · H. W. Chiang · P. J. Wang National Tsing Hua University, Hsinchu 30013, Taiwan e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 P.-C. Chiang (ed.), Environment and Renewable Energy, Springer Proceedings in Earth and Environmental Sciences, https://doi.org/10.1007/978-981-97-0056-1_2

17

18

K. H. Lu et al.

geothermal energy is an attractive way to meet the soaring energy demands. However, most of the geothermal sources are available at low temperature levels ( 0. Where η is called scale parameter; β, the shape parameter; U, location parameter characterizing the distribution. If the value of β = 1, the distribution equivalent to the exponential distribution. The Weibull distribution approximates to the normal distribution when β > 4 [22]. The value of U implies that the reactions with activation energy less than this value do not take place [1]. The mean activation energy is calculated through gamma function, given by Eq. (13) ) ( 1 +1 (13) μ = γ + η[ β The standard deviation is described by Eq. (14) / ( ) ) ( 1 2 + 1 − η2 [ 2 +1 σ = η2 [ β β

(14)

80

K.-Q. Tran et al.

3 Methodology The pyrolysis kinetic of biomass materials can be simulated by employing the single DAEM, based on the assumption the overall pyrolysis process is just a single reaction to produce volatile matter and char. However, this assumption cannot characterize precisely the thermal-decomposition behaviour of biomass due to its variable composition. Recently, the parallel-reaction model has been proven as an accurate approach [23]. Since the microalgae residues contain the remaining lipid, protein and the cell wall including hemicellulose, cellulose and lignin, it is reasonable to assume five pseudo-components for the pyrolysis of microalgae residues [15]. Following this assumption, the pyrolysis kinetic can be interpreted by Eq. (15), in with ci is the contribution factor of ith component. dα ∑ dαi ci = i = 1, 2, 3, 4, 5 dt dt 1 5

(15)

The non-linear least square is employed for the optimization of kinetic simulation which aims to minimize the objective function as described in Eq. (16) [( ]2 ) ) ( n ∑ dαi dαi S= − (16) dt exp dt model i=1 ( i) ( i) where dα and dα are the experiment and modelled conversion rate, n dt ex p dt model is the number of experimental points. The quality of curve fitting is evaluated by Eq. (17) [23] / ⎛ ⎞ S

N Fit(%) = ⎝1 − ( dαi ) [ dt exp ]

⎠.100%

(17)

max

The five pseudo-components model was simulated for both cases of the reaction order equal one (n = 1) or different than one (n # 1).

4 Results and Discussion 4.1 Kinetic Modelling with Gaussian Distributed Activation Energy Model Figures 1 and 2 respectively represent graphically the results obtained from the simulation for C. sp. JSC4 and C. sorokiniana CY1, assuming the five pseudo-components model. From Figs. 1 and 2, the kinetic parameters are extracted and presented in Tables 1 and 2, accordingly.

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study …

81

0.3

(a)

dα/dt (mg/min)

0.25

da/dt (exp) Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid Protein

0.1

model 0.05 0 300

400

500

600

700

800

900

1000

Temperature (K) 0.3

(b)

dα/dt (mg/min)

0.25

da/dt (exp) Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid Protein

0.1

model 0.05 0 300

400

500

600

700

800

900

1000

Temperature (K) Fig. 1 Simulation and curve fitting of C. sorokiniana CY1 with Gaussian distributed activation energy, a n = 1; b n # 1

It is observed from Figs. 1 and 2 that among five pseudo-components hemicellulose is the most reactive since it started decomposing first, which was followed by cellulose and lignin. On the other hand, hemicellulose and cellulose decomposed within a relative narrow temperature window, while the decomposition of lignin took place within a wider range of temperature, lasting till 850 K approximately. In addition, for both C. sp. JSC4 and C. sorokiniana CY1, the thermal decomposition of protein started before lipid. Overall, the observed trends are in good agreement with the literature [15, 23]. As shown in Table 1, the calculated kinetic parameters of hemicellulose, cellulose and lignin are in agreement with data reported in the literature [24, 25]. The mean activation energy of C. sorokiniana CY1 being 281.08 kJ/mol and 295.25 kJ/mol for n = 1 and n#1, respectively, was somehow high compared to the typical activation energy of biomass pyrolysis. However, these values are still reasonable in comparison with the literature of investigation on cellulose [26]. On the other hand, for both cases

82

K.-Q. Tran et al. 0.3

(a) dα/dt (mg/min)

0.25

da/dt (exp) Hemicellulose Cellulose Lignin Lipid Protein model

0.2 0.15 0.1 0.05 0 300

400

500

600

700

800

900

1000

Temperaute (K) 0.3

(b) 0.25

dα/dt (mg/min)

da/dt (exp) Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid 0.1

Protein model

0.05 0 300

400

500

600

700

800

900

1000

Temperatura (K)

Fig. 2 Simulation and curve fitting of C. sp. JSC4 with Gaussian distributed activation energy; a n = 1; b n # 1

of n = 1 and n#1, it appeared that the mean activation energy of protein and lipid of C. sorokiniana CY1 was higher than this of C. sp. JSC4. However, the fit quality of the case of n#1 was better than that for the case of n = 1. This observation is valid for both samples of microalgae residues.

4.2 Kinetic Modelling with Logistic Distributed Activation Energy Model Figures 3 and 4 demonstrate graphically the results obtained from the simulation for C. sp. JSC4 and C. sorokiniana CY1, assuming the five pseudo-components model. Tables 3 and 4 present the kinetic parameters extracted from the modelling

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study …

83

Table 1 Extracted kinetic data with Gaussian distributed activation energy, n = 1 Sample C. sp. JSC4

Hemicellulose

A (min−1 )

c

Fit (%)

90.61

16.24

1.36E + 06

0.17

98.07

211.57

37.92

3.13E + 14

0.37

Lignin

71.44

12.81

1.27E + 03

0.16

Lipid

108.59

19.46

8.72E + 05

0.12

Cellulose

C. sorokiniana CY1

σ (KJ/mol)

E a (KJ/mol)

Protein

90.46

6.21

1.71E + 05

0.18

Hemicellulose

88.14

15.80

4.74E + 05

0.12

281.08

50.38

2.30E + 19

0.35

Lignin

42.53

25.45

4.20E + 03

0.20

Lipid

168.01

30.11

5.46E + 09

0.13

Protein

119.03

21.33

2.07E + 07

0.18

Cellulose

99.10

Table 2 Extracted kinetic data with Gaussian distributed activation energy, n#1 E a (KJ/ mol)

σ (KJ/ mol)

A (min−1 )

c

n

Fit (%)

Hemicellulose

90.60

16.24

9.21E + 05

0.16

1.02

98.45

Cellulose

224.05

40.16

2.75E + 15

0.36

1.18

Lignin

61.74

11.07

2.46E + 02

0.16

1.36

Lipid

101.07

18.12

2.82E + 05

0.14

1.12

Protein

79.78

14.30

3.30E + 04

0.19

1.25

Hemicellulose

78.90

14.15

9.75E + 04

0.12

1.01

Cellulose

295.25

52.92

2.41E + 20

0.34

1.11

Lignin

53.90

32.25

9.08E + 04

0.20

1.60

Lipid

158.78

28.46

1.57E + 09

0.14

1.12

Protein

113.92

20.42

1.10E + 07

0.19

1.12

Sample C. sp. JSC4

C. sorokiniana CY1

99.35

and simulation. As can be in the figures, hemicellulose started decomposing first, followed by cellulose and then lignin. The decomposition process of protein started earlier than lipid which can be explained by the relative stability between them [27, 28]. On the other hand, the modelling curve fits well to the experiment curve for both samples with high fit qualities, being about 99% (Tables 3 and 4). According to a literature of kinetic study of cellulose by applying logistic distribution energy model, the mean activation energy was 258.57 kJ/mol [10]. This is compatible to 195.55 kJ/mol for C. sp. JSC4 and 253.68 kJ/mol for C. sorokiniana CY1, obtained from the present study. Indeed, the mean activation energy of lipid and protein were in consonance with the activation energy of our previous study by employing five pseudo-components [15].

84

K.-Q. Tran et al. 0.3

(a) dα/dt (mg/min)

0.25

da/dt (exp) Hemicellulose Cellulose Lignin Lipid Protein model

0.2 0.15 0.1 0.05 0 300

400

500

600

700

800

900

1000

Temperaute (K) 0.3

dα/dt (mg/min)

0.25

(b)

da/dt (exp) Hemicellulose Cellulose Lignin Lipid Protein model

0.2 0.15 0.1 0.05 0 300

400

500

600

700

800

900

1000

Temperature (K)

Fig. 3 Simulation and curve fitting of C. sorokiniana CY1 with Logistic distributed energy, a n = 1; b n # 1

4.3 Kinetic Modelling with Weibull Distributed Activation Energy Model Figures 5 and 6 respectively show the graphical data obtained from the simulation for C. sp. JSC4 and C. sorokiniana CY1. The extracted kinetic parameters from the modelling and simulation are shown in Tables 5 and 6, accordingly. Overall, the fit quality was high, being approximately 99%. This suggests that Weibull distributed activation energy model is applicable to pyrolysis kinetic modelling of lignocellulosic biomass materials. The thermal behaviour of every component coincided with the observed trends from Gaussian model and Logistic model. Certainly, the decomposition process of hemicellulose occurred first, which is then followed by cellulose and lignin, whereas protein started decomposing before lipid. The location factor was examined with the value of 0, resulting in good fit quality. This is understandable because the value of mean activation energy less than 0 does

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study …

85

0.3

(a) dα/dt (mg/min)

0.25 da/dt (exp) Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid Protein

0.1

Model

0.05 0 300

400

500

600

700

800

900

1000

Temperature (K) 0.3

(b)

0.25

da/dt (exp)

dα/dt (mg/min)

Hemicellulose Cellulose

0.2

Lignin Lipid

0.15

Protein Model

0.1 0.05 0 300

400

500

600

700

800

900

1000

Temperature (K)

Fig. 4 Simulation and curve fitting of C. sp. JSC4 with Logistic distributed activation energy, a n = 1; b n # 1 Table 3 Extracted kinetic data with Logistic distributed activation energy, n = 1 E a (KJ/ mol)

σ (KJ/ mol)

A (min−1 )

c

Fit (%)

Hemicellulose

100.50

24.11

5.10E + 07

0.17

98.31

Cellulose

194.98

46.78

8.79E + 14

0.34

Lignin

43.73

10.49

3.77E + 01

0.17

Lipid

Sample C. sp. JSC4

C. sorokiniana CY1

101.09

24.25

1.24E + 06

0.14

Protein

91.13

21.87

1.04E + 06

0.18

Hemicellulose

94.53

22.68

7.56E + 06

0.13

233.27

55.97

8.65E + 17

0.36

Lignin

47.12

23.79

3.90E + 03

0.20

Lipid

148.51

35.63

4.03E + 09

0.12

Protein

107.82

25.87

2.19E + 07

0.17

Cellulose

98.77

86

K.-Q. Tran et al.

Table 4 Extracted kinetic data with Logistic distributed activation energy, n#1 Sample C. sp. JSC4

σ (KJ/ mol)

A (min−1 )

c

n

Fit (%) 98.62

93.00

22.31

1.46E + 07

0.17

1.64

195.55

46.91

1.05E + 15

0.35

1.02

Lignin

63.51

15.24

1.25E + 03

0.16

1.76

Lipid

Hemicellulose Cellulose

C. sorokiniana CY1

E a (KJ/ mol)

110.03

26.40

9.35E + 06

0.13

1.64

Protein

91.00

21.83

1.12E + 06

0.19

1.91

Hemicellulose

93.50

21.47

8.91E + 06

0.12

1.02

253.64

60.85

3.67E + 19

0.35

1.12

Lignin

55.27

27.91

3.05E + 04

0.20

1.46

Lipid

162.71

39.04

4.88E + 10

0.13

1.26

Protein

118.81

28.50

2.50E + 08

0.20

1.29

Cellulose

99.09

0.3

(a) 0.25

da/dt (exp)

dα/dt (mg/min)

Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid Protein

0.1

model

0.05 0 300

400

500

600

700

800

900

1000

Temperature (K) 0.3

(b)

0.25

da/dt (exp)

dα/dt (mg/min)

Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid Protein

0.1

model

0.05 0 300

400

500

600

700

800

900

1000

Temperature (K)

Fig. 5 Simulation and curve fitting of C. sorokiniana CY1 with Weibull distributed activation energy, a n = 1; b n # 1

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study …

87

0.3

(a)

da/dt (exp)

0.25

Hemicellulose

dα/dt (exp)

Cellulose 0.2

Lignin Lipid

0.15

Protein Model

0.1 0.05 0 300

400

500

600

700

800

900

1000

Temperatue (K) 0.3

(b) 0.25

dα/dt (exp)

da/dt (exp) Hemicellulose

0.2

Cellulose Lignin

0.15

Lipid Protein

0.1

Model 0.05 0 300

400

500

600

700

800

900

1000

Temperature (K) Fig. 6 Simulation and curve fitting of C. sp. JSC4 with Weibull distributed activation energy, a n = 1; b)n # 1

not have any significance. The shape parameter, β, was between 6 and 7; then Weibull distribution turned to be normal distribution in this singular case [22]. Indeed, a similar observation was reported earlier by Lakshmanan et al. [1].

4.4 Comparison Between Three Distributed Activation Energy Models As can be seen from the data presented in Figs. 1, 2, 3, 4, 5 and 6, the overall simulated curve as well as the five pseudo-component curves obtained from the modelling using the three different distribution functions are quite similar to one

88

K.-Q. Tran et al.

Table 5 Extracted kinetic data with Weibull distributed activation energy, n = 1 Sample C. sp. JSC4

σ (KJ/ mol)

A (min−1 )

c

Fit (%) 98.43

92.00

17.33

2.04E + 06

0.17

224.38

42.26

5.23E + 15

0.35

Lignin

57.98

10.92

2.16E + 02

0.16

Lipid

Hemicellulose Cellulose

C. sorokiniana CY1

E a (KJ/ mol)

110.30

20.77

1.27E + 06

0.14

Protein

91.07

17.15

2.46E + 05

0.18

Hemicellulose

99.75

16.52

2.17E + 06

0.12

275.24

44.53

1.94E + 18

0.36

Lignin

65.88

12.37

9.44E + 02

0.20

Lipid

157.12

28.67

1.47E + 09

0.15

Protein

127.23

22.69

8.41E + 07

0.17

Cellulose

98.86

Table 6 Extracted kinetic data with Weibull distributed activation energy, n # 1 Sample C. sp. JSC4

σ (KJ/ mol)

A (min−1 )

c

n

Fit (%) 98.58

90.63

17.07

1.69E + 06

0.17

1.29

223.57

42.10

4.48E + 15

0.35

1.01

Lignin

58.90

11.09

2.16E + 02

0.16

1.06

Lipid

Hemicellulose Cellulose

C. sorokiniana CY1

E a (KJ/ mol)

109.52

20.63

1.47E + 06

0.14

1.25

Protein

97.14

18.29

6.94E + 05

0.18

1.52

Hemicellulose

97.40

4.47

3.59E + 04

0.12

1.23

266.24

50.14

4.46E + 18

0.35

1.01

Lignin

64.44

12.14

7.25E + 02

0.19

1.19

Lipid

183.48

34.57

8.48E + 10

0.15

1.42

Protein

123.75

22.18

5.31E + 07

0.19

1.16

Cellulose

99.16

another. The kinetic parameters extracted for hemicellulose, cellulose and lignin from the simulation assuming the three different distributed activation energy models were in good agreement with the literature [15, 23, 29]. Among these, the mean activation energy for cellulose of C. sorokiniana CY1 obtained from the Gaussian model was somehow high compared to the typical activation energy of cellulose between, being within 195–213 kJ/mol [24]. However, the activation energy obtained from DAEM has the mean value that its standard deviation, σ, is also taken into account for further considerations. In general, among the pseudo-components, the standard deviation of cellulose is highest, which indicates that the actual activation energy varies in a wider range. In addition, the obtained results are also consistent with our earlier study [15]. On the other hand, it was observed that lipid started decomposing after protein for both samples within a temperature window of about 500–700 K for protein and about

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study …

89

560–740 K for lipid. These observations were confirmed by another investigation on pyrolysis of protein and lipid extracted from green microalgae [16]. Furthermore, the protein and lipid simulated curves of C. sp. JSC4 were more flattened in the case of n#1 compared to the case of n = 1 for the three different models. Noticeably, the pyrolysis of lipid and protein of C. sorokiniana CY1 gave higher values of mean activation energy than those of C. sp. JSC4. As the matter of fact, the mean activation energy as well as its standard deviation varied slightly among three models. However, the fit quality was high enough to satisfy the accuracy requirement of kinetic modelling with the assumption of five pseudo-components. Moreover, it was not clear which model was better than the others since none of the models exhibited predominantly higher fit quality than the others. For instance, the Gaussian distributed activation energy model seemed to be the best option for kinetic modelling of C. sorokiniana CY1, n#1; but the logistic distributed activation energy model gave the highest fit quality in the case of C. sorokiniana CY1, n = 1. On the other hand, the Weibull distributed activation energy model was considered as the best choice for the kinetic modelling of C. sp. JSC4, n = 1. Figure 7 demonstrates graphically the influence of the mean activation energy and its standard deviation on the probability function, f (E), of hemicellulose for the case of n = 1. In this case, the “bell” shape of the logistic was bigger than that of the Gaussian and Weibull due to its higher standard deviation, 24.11 kJ/mol, compared to 16.24 kJ/mol of the Gaussian and 17.33 kJ/mol of the Weibull. In the other words, the actual activation energy, obtaining from logistic model spreads within a wider range than those of the Gaussian and Weibull, as can be seen from the graph. The

Fig. 7 Probability density function curves of hemicellulose, n = 1

90

K.-Q. Tran et al.

results revealed that the reaction order of every pseudo-component varied between one and two, which is realistic. The fit quality for the cases of n#1 was always higher, though slightly, than that of n = 1 and the kinetic parameters were more reasonably realistic.

5 Conclusion The kinetic modelling and analysis for pyrolysis of two microalgae residues was performed assuming different distributed activation energy models with different mathematical distribution functions Gaussian, Logistic and Weibull. It appeared that all the three models described closely the experimental data with similar and high accuracies. The extracted kinetic parameters were within the acceptable range, compared to the literature. Regarding the fit quality, the results from modelling and simulations with n #1 were slightly better than that of n = 1. There was no clear evidence to conclude one model better than the others. All the three distributed activation energy models are suitable to kinetic modelling of lignocellulosic biomass pyrolysis processes using TGA.

References 1. Lakshmanan, C.C., White, N.: A new distributed activation energy model using Weibull distribution for the representation of complex kinetics. Energy Fuels 8(6), 1158–1167 (1994) 2. Papadikis, K., Gu, S., Bridgwater, A.V., Gerhauser, H.: Application of CFD to model fast pyrolysis of biomass. Fuel Process. Technol. 90(4), 504–512 (2009) 3. Zhang, J., Chen, T., Wu, J., Wu, J.: A novel Gaussian-DAEM-reaction model for the pyrolysis of cellulose, hemicellulose and lignin. RSC Adv. 4(34), 17513–17520 (2014) 4. Seo, D.K., Park, S.S., Hwang, J., Yu, T.-U.: Study of the pyrolysis of biomass using thermogravimetric analysis (TGA) and concentration measurements of the evolved species. J. Anal. Appl. Pyrol. 89(1), 66–73 (2010) 5. Várhegyi, G., Bobály, B., Jakab, E., Chen, H.: Thermogravimetric study of biomass pyrolysis kinetics. A distributed activation energy model with prediction tests. Energy Fuels 25(1), 24–32 (2011) 6. Sonobe, T., Worasuwannarak, N.: Kinetic analyses of biomass pyrolysis using the distributed activation energy model. Fuel 87(3), 414–421 (2008) 7. McGuinness, M.J., Donskoi, E., McElwain, D.L.S.: Asymptotic approximations to the distributed activation energy model. Appl. Math. Lett. 12(8), 27–34 (1999) 8. Kirtania, K., Bhattacharya, S.: Application of the distributed activation energy model to the kinetic study of pyrolysis of the fresh water algae Chlorococcum humicola. Biores. Technol. 107, 476–481 (2012) 9. Cai, J., Liu, R.: Application of Weibull 2-mixture model to describe biomass pyrolysis kinetics. Energy Fuels 22(1), 675–678 (2008) 10. Cai, J., Yang, S., Li, T.: Logistic distributed activation energy model—Part 2: Application to cellulose pyrolysis. Biores. Technol. 102(3), 3642–3644 (2011) 11. Sonoyama, N., Hayashi, J.-I.: Characterisation of coal and biomass based on kinetic parameter distributions for pyrolysis. Fuel 114, 206–215 (2013)

Pyrolysis Kinetics of Microalgae Residues—A Comparative Study …

91

12. Sfakiotakis, S., Vamvuka, D.: Development of a modified independent parallel reactions kinetic model and comparison with the distributed activation energy model for the pyrolysis of a wide variety of biomass fuels. Biores. Technol. 197, 434–442 (2015) 13. Várhegyi, G., Chen, H., Godoy, S.: Thermal decomposition of wheat, oat, barley, and brassica Carinata straws. A kinetic study. Energy and Fuels 23(2), 646–652 (2009) 14. Fiori, L., Valbusa, M., Lorenzi, D., Fambri, L.: Modeling of the devolatilization kinetics during pyrolysis of grape residues. Biores. Technol. 103(1), 389–397 (2012) 15. Bui, H.-H., Tran, K.-Q., Chen, W.-H.: Pyrolysis of microalgae residues—a kinetic study. Bioresour. Technol. (2015) 16. Kebelmann, K., Hornung, A., Karsten, U., Griffiths, G.: Intermediate pyrolysis and product identification by TGA and Py-GC/MS of green microalgae and their extracted protein and lipid components. Biomass Bioenerg. 49, 38–48 (2013) 17. Chen, W.H., Huang, M.Y., Chang, J.S., Chen, C.Y.: Thermal decomposition dynamics and severity of microalgae residues in torrefaction. Bioresour. Technol. 169, 258–264 (2014) 18. Cai, J., Wu, W., Liu, R.: An overview of distributed activation energy model and its application in the pyrolysis of lignocellulosic biomass. Renew. Sustain. Energy Rev. 36, 236–246 (2014) 19. Anthony, D.B., Howard, J.B.: Coal devolatilization and hydrogasification. AIChE J. 22(4), 625–656 (1976) 20. Cai, J., Jin, C., Yang, S., Chen, Y.: Logistic distributed activation energy model—Part 1: derivation and numerical parametric study. Biores. Technol. 102(2), 1556–1561 (2011) 21. Weibull, W.: J. Appl. Mech. 18, 293–296 (1951) 22. Jankovi´c, B.: A kinetic study of the isothermal degradation process of Lexan® using the conventional and Weibull kinetic analysis. J. Polym. Res. 16(3), 213–230 (2009) 23. Tran, K.-Q., Bach, Q.-V., Trinh, T.T., Seisenbaeva, G.: Non-isothermal pyrolysis of torrefied stump—a comparative kinetic evaluation. Appl. Energy 136, 759–766 (2014) 24. Grønli, M.G., Várhegyi, G., Di Blasi, C.: Thermogravimetric analysis and devolatilization kinetics of wood. Ind. Eng. Chem. Res. 41(17), 4201–4208 (2002) 25. Hu, S., Jess, A., Xu, M.: Kinetic study of Chinese biomass slow pyrolysis: comparison of different kinetic models. Fuel 86(17–18), 2778–2788 (2007) 26. Zhou, H., Long, Y., Meng, A., Chen, S., Li, Q., Zhang, Y.: A novel method for kinetics analysis of pyrolysis of hemicellulose, cellulose, and lignin in TGA and macro-TGA. RSC Adv. 5(34), 26509–26516 (2015) 27. Peng, W., Wu, Q., Tu, P.: Pyrolytic characteristics of heterotrophic Chlorella protothecoides for renewable bio-fuel production. J. Appl. Phycol. 13(1), 5–12 (2001) 28. Shuping, Z., Yulong, W., Mingde, Y., Chun, L., Junmao, T.: Pyrolysis characteristics and kinetics of the marine microalgae Dunaliella tertiolecta using thermogravimetric analyzer. Biores. Technol. 101(1), 359–365 (2010) 29. Manyà, J.J., Velo, E., Puigjaner, L.: Kinetics of biomass pyrolysis: a reformulated three-parallelreactions model. Ind. Eng. Chem. Res. 42(3), 434–441 (2003)

Low-Cost Anodic Material Made of Rice Husk Charcoal and Acrylic Paint for Soil-Based Microbial Fuel Cells Trang Nakamoto, Soichiro Hirose, Dung Nakamoto, Keisuke Nishida, and Kozo Taguchi

Abstract Microbial fuel cells (MFCs) have attracted attention as a technology that can simultaneously treat organic wastewater and generate electricity. The challenge facing MFC is to fabricate inexpensive electrodes. The performance of soil MFCs (SMFCs) is greatly affected by the material of the anode. In this study, a low-cost anodic material was fabricated based on charcoal made of rice husk (RHC) and commercially available acrylic paint (AP). Various RHC:AP ratios were tested to find out the optimum material composition. As a result, the SMFC with the anode made of a 75:25 wt% ratio generated a maximum power density of 6.7 µW/cm2 . RHC showed a significant impact on the performance of the anode electrode. Keywords Bioelectric · Bacteria electricity · Recycle

1 Introduction A microbial fuel cell (MFC) generates electricity based on electrogenic bacteria [1]. MFC technology can decompose organic wastewater and generate electric power simultaneously [2]. Soil microbial fuel cell (SMFC) is a type of MFC that operates in the soil. Microorganisms decompose organic matter in the soil to support the operation of SMFCs. In an SMFC, the anode supports biofilm formation and receives electrons from bacterial cells [3]. Accordingly, the anode performance depends significantly on microbial affinity and electrical conductivity characteristics [4]. Various low-cost carbon-based materials, such as carbon fiber, charcoal, and activated carbon, have a high microbial affinity. Therefore, they are usually used as anodic material [5, 6]. Rice husks are a byproduct of the rice harvesting process. Recently, the amount of rice harvested worldwide has reached about 800 million tons [7]. Rice husk disposal processes like open-pit mining or incineration generally dispose of rice husks. When T. Nakamoto (B) · S. Hirose · D. Nakamoto · K. Nishida · K. Taguchi Department of Electrical and Electronic Engineering, Ritsumeikan University, Kusatsu 525-8577, Japan e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 P.-C. Chiang (ed.), Environment and Renewable Energy, Springer Proceedings in Earth and Environmental Sciences, https://doi.org/10.1007/978-981-97-0056-1_8

93

94

T. Nakamoto et al.

rice husks are treated at high temperatures, rice husks will become rice husks-based charcoal (RHC). Since RHC has a high surface area, it is considered to have a high affinity with bacteria [8]. In addition, RHC is an electrically conductive material. Therefore, RHC is expected to be a low-cost carbon-based material for the anode electrode in SMFCs. Acrylic-based paints use acrylic as a binder, which are water-soluble paints that become water-resistant if dry. Acrylic paints (AP) could be used as a binder for conductive materials to make conductive paints [9]. Since AP is inexpensive and easy to find, it is expected to be a suitable binder for making the cost-effective and durable electrodes of SMFCs. This study utilized AP as a binder for RHC to fabricate a low-cost anodic material for SMFCs. RHC was mixed with AP in different ratios to make conductive carbonbased paints, which were coated on stainless steel mesh (SSM) to make the anodes of membrane-less single-chamber SMFCs. The performance of the fabricated anodes was investigated by measuring the discharging voltage and the maximum power density of the SMFCs.

2 Method and Experiment 2.1 Electrode Fabrication Method RHC (Tokorozawa Ueki Bachi Center, Ltd., Japan), carbonized by the contractor using the smoking method, was used for fabricating the conductive paint. RHC was treated in an alkaline NaOH solution (0.6 M) for 24 h at 100 ◦ C. After alkali treatment, RHC was washed with tap water five times, dried at 60 ◦ C for 48 h, and ground to powder with a pestle. The coating paint was created by blending RHC with AP (Daiso Industries Co., Ltd., Japan) using different ratios, as shown in Table 1. Then, the SSMs (#321 Hikari Co., Ltd., Japan) sized 2 × 2 cm were dip-coated with the as-prepared coating solutions. Then, coated SSMs were dried at 60 °C for 2 h to make the anodes 1 to 4. Photo images of Anodes 2, 3, and 4 are shown in Fig. 1. The cathode was fabricated by using activated carbon sheets (RCA-01L, Azumi filter paper co. ltd) [10]. RCA-01L with a thickness of 0.2 mm was cut into the size of 2 × 2 cm to make the cathode of the SMFCs. Table 1 Coating paint ratios for different anodes

RHC (wt%)

AP (wt%)

Anode 1

0

100

Anode 2

40

60

Anode 3

55

45

Anode 4

75

25

Low-Cost Anodic Material Made of Rice Husk Charcoal and Acrylic …

95

Fig. 1 Photo images of the anodes a Anode 2, b Anode 3, and c Anode 4

2.2 SMFC Setup and Measurement Figure 2 shows the schematic of the SMFC setup. The anode was plugged into the soil, and the cathode was floated on the surface. The soil was collected from rice paddies in Japan (at the location of 34°59' 42.7986” N, 135°57' 16.1892” E (34.9995222, 135.954497)). Tap water was added to the experiment box to keep 1 cm water layer on the soil. The SMFC was connected to a 10 kΩ external resistance. Thin SSM was used to connect the electrodes to the external circuit. SMFC operation and measurements were performed in an indoor environment (25 ± 1 °C). The voltage across the external resistance (10 kΩ) of the SMFC was monitored by a data acquisition system (NI USB-6210, National Instruments Corp., USA). The power density curves were calculated based on the steady-state discharge voltage measured by varying the external resistance in the range of 10–0.4 kΩ. The power density was measured several times during the experiment.

Fig. 2 Schematic of a MFC structure and b experiment setup

96

T. Nakamoto et al.

Fig. 3 SEM images of a SSM, b the Anode 4 surface before the experiment, and c the Anode 4 surface after the experiment

2.3 Observation of Electrode Surface A scanning electron microscope (SEM) (S-4300, Hitachi, Ltd., Japan) was used to observe the surface of the prepared electrodes. The surface was coated with gold by sputtering before observing by SEM. Also, the surface of the anode after the experiment was observed by SEM to check the bacterial biofilm. The anode was dehydrated with ethanol after the MFC operation for this measurement.

3 Result and Discussion 3.1 SEM Image of the Anode The surfaces of the SSM and anode were observed by SEM, as shown in Fig. 3. The SSM had a wire size of about 50 µm (Fig. 3a). After coating the SSM with RHC and AP, the surface showed a porous 3D structure (Fig. 3b). Moreover, bacterial cells can be observed on the anode surface after SMFC operation (Fig. 3c). This result suggests that the anode enables high bacteria affinity.

3.2 Anode Performance The time evolution of the voltages of SMFCs with the Anode 1–4 are shown in Fig. 4. The SMFC, which uses Anode 1 (without RHC), obtained a maximum voltage of 40 mV. The addition of RHC to Anodes 2, 3, and 4 increased the maximum voltage to 400 mV level. Furthermore, the time required for the voltage to reach at least 80% of its maximum value was about 70 h with Anode 2, 3, and 4, compared to 270 h with Anode 1. The reduction in start-up time is attributed to the improved affinity

Low-Cost Anodic Material Made of Rice Husk Charcoal and Acrylic …

97

Fig. 4 Evolution of the voltage output of the SMFCs using Anode 1–4 during 500 h of operation. Sharp drops on the graph happened after power density measurement events

Fig. 5 Comparison of the maximum power density of the SMFCs using Anode 1–4 was measured several times during the experiment

of the bacteria by the RHC and AP anodic material. Anode 4 showed the highest voltage and stability compared with other anodes. A comparison of the maximum power density of the SMFCs using Anodes 1–4 is shown in Fig. 5. When Anode 1 was prepared using only AP, the power density of the SMFC was almost zero. This result was attributed to the fact that AP was a non-conductive material, leading to the anode cannot receive electrons from bacterial cells. Meanwhile, by adding RHC to Anodes 2, 3, and 4, the power densities were much improved. Anode 4 generated the highest maximum power density of 6.7 µW/ cm2 at 160 h. These results suggest that the combination of RHC and AP as the anode material is effective in SMFCs.

4 Conclusion In this experiment, the use of RHC and AP to make the anodes of SMFCs was investigated. A maximum power density of 6.7 µW/cm2 was obtained with the anode with 75:25 wt% ratio of RHC and AP. Furthermore, the addition of RHC also improved

98

T. Nakamoto et al.

the start-up time of the SMFC. The experimental results suggest that the combination of RHC and AP has the potential for use as the anode in SMFCs. For future research, one thing to consider is optimizing the ratio of Bokuju to RHC because the amount of RHC in the anode was found to affect the performance of SMFCs.

References 1. Nguyen, D.T., Taguchi, K.: Enhancing the performance of E. coli-powered MFCs by using porous 3D anodes based on coconut activated carbon. Biochem. Eng. J. 151 (2019) 2. Carlos, M.C., Hu, Y., Chunbao, X., Amarjeet, B.: An overview of microbial fuel cell usage in wastewater treatment, resource recovery and energy production. Sci. Total Environ. 754 (2020) 3. Abbas, S.Z., Rafatullah, M.: Recent advances in soil microbial fuel cells for soil contaminants remediation. Chemosphere 272, 129691 (2021) 4. Nosek, D., Jachimowicz, P., Cydzik-Kwiatkowska, A.: Anode modification as an alternative approach to improve electricity generation in microbial fuel cells. Energies 13, 6596 (2020) 5. Tran, T.V., Lee, I.C., Kim, K.: Electricity production characterization of a sediment microbial fuel cell using different thermo-treated flat carbon cloth electrodes. Int. J. Hydrogen Energy 44, 32192–32200 (2019) 6. Nguyen, D.T., Taguchi, K.: A floating microbial fuel cell: Generating electricity from Japanese rice washing wastewater. Energy Rep. 6, 758 (2020) 7. Dizaji, H.B., Zeng, T., Hölzig, H., Bauer, J., Klöß, G., Enke, D.: Ash transformation mechanism during combustion of rice husk and rice straw. Fuel 307 (2022) 8. Singh, C., Tiwari, S., Gupta, V.K., Singh, J.S.: The effect of rice husk biochar on soil nutrient status, microbial biomass and paddy productivity of nutrient poor agriculture soils. CATENA 171, 485 (2018) 9. Mates, J.E., Bayer, I.S., Salerno, M., Carroll, P.J., Jiang, Z., Liu, L., Megaridis, C.M.: Durable and flexible graphene composites based on artists’ paint for conductive paper applications. Carbon 87 (2015) 10. Iwai, T., Nguyen, D.T., Taguchi, K.: Study of activated carbon sheets used for air-cathodes of portable quasi-solid aluminum-air batteries. IEEJ. Trans. Electr. Electron. Eng. 16, 653 (2021)

Water Pollution Control, Water Resource Management and Air Quality Assessment

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum, and Musa Paradisiaca Species as Natural Coagulants in the Treatment of the Cunas River—Peru Francis Eduardo Ambrosio Rosales, Ingrid Lucia Artica Cardenas, Leslie Alison Vargas Cordova, and Steve Dann Camargo Hinostroza

Abstract The research aimed to analyze the effectiveness of the skins of Solanum tuberosum (Potato), Musa paradisiaca (Banana), and Lepidium meyenni (Maca) as natural coagulants in the removal of physical and chemical parameters: Total suspended solids (TSS), turbidity, pH, and color, in the water sample obtained from the lower basin of the Cunas River. This study presents a level of explanatory research with experimental design, using the inductive method due to the analysis of samples, pre and posttest. In addition, for the elaboration of natural coagulants from skins of S. tuberosum, M. paradisiaca, and L. meyenni, these skins were obtained from restaurants and street commerce, to later perform the Jar-Test at different doses (25, 50, 75, and 100 ppm) using 4 water samples of 1000 mL for each coagulant. The results were: M. paradisiaca had a total average removal effectiveness of 69.39% with 75 ppm, S. tuberosum had a removal effectiveness of 55.86% with 75 ppm, and finally, L. meyenni had a lower removal effectiveness with a total average of 52.62% with 100 ppm. It is concluded that the natural coagulant of M. paradisiaca has a higher average effectiveness compared to S. tuberosum and L. meyenni. Keywords Natural coagulants · River Cunas · Chemical physical parameters · Effectiveness

1 Introduction The increase in demand for water has led to a considerable decrease in the availability of this resource [1]. In 2030 the land will have a water decrease of 40% [2], and by 2050 there will be an increase in water use worldwide of 20–30%. 46% of the F. E. Ambrosio Rosales (B) · I. L. Artica Cardenas · L. A. Vargas Cordova · S. D. Camargo Hinostroza Universidad Continental, Huancayo, Peru e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 P.-C. Chiang (ed.), Environment and Renewable Energy, Springer Proceedings in Earth and Environmental Sciences, https://doi.org/10.1007/978-981-97-0056-1_9

101

102

F. E. Ambrosio Rosales et al.

variation in the flow of a river is due to climatic effects, whereas 54% is due to the impacts caused by man; Of this percentage, 24.7% is given by wastewater [3]. These anthropogenic activities generate effluents, polluting water bodies such as lakes or rivers [4]. The pollution of rivers and the environment is reaching a critical level due to the lack of treatment plants [5]. In Latin America, 3/4 of wastewater is returned to rivers and other water sources [6]. This pollution will be driven by the increase in waste generation by 70%, due to population growth and urban expansion by 2050 [7]. In Peru, during the year 2020, a total of 92 822.84 tons of solid waste were recovered, of which 68 399.63 tons were municipal organic waste (remains of vegetables and/or fruits that come from markets, homes, and others) [8]. In many parts of the country, the proximity of urban areas to rivers is one of the main causes of water pollution. Pollution is not only produced by the dumping of solid waste into rivers, but also by washing clothes, vehicles, the dumping of fat, fuel, among others [9]. The waters of natural origin that do not have any treatment are called raw waters, these are found in rivers, streams, lakes [10]. Water treatment plants present equitable processes; the most important are the coagulation and flocculation process [11]. In the coagulation process, small particles are formed by the addition of a coagulant to the water or also by applying mixing energy. The suspended particles are destabilized by the neutralization of the charges of negative colloids, where they use mostly inorganic coagulants and these alter the physicochemical properties of water [12]. The use of plant resources as natural coagulants in the various water clarification processes turns out to be an appropriate technology in vulnerable sectors, in the face of scarce economic conditions. These plant resources are available to everyone, since the main source of production is nature [13]. When using natural coagulants, it can be observed that it minimizes and avoids the use of chemical coagulants, significantly reducing treatment costs; if the use of these natural coagulants is available [14], the delimitation of the dosage of natural coagulants is the most important part because it determines the efficiency of the measurable parameters in the sample [15]; therefore, the stock solutions obtained from plant residues are used as primary coagulants in treatment plants, which allows to have an effluent that meets the acceptable physical parameters of color and turbidity [16]. That is why the need arises to investigate about these natural coagulants, taking into account that the Junín region presents multiple tributary rivers to the Mantaro River. A main tributary river is the Cunas River; this represents a great economic importance because it is the main source for various activities such as agricultural, fishing, recreational, industrial, hydro electrical, and human consumption [17] Through the visits made to the body being studied, it was possible to verify the presence of various effluents of discharge of wastewater directly to the Cunas River, which cause the contamination of its waters [18, 19]. Numerous studies have evaluated the use of natural coagulants to replace chemical coagulants [20]. It is necessary to evaluate coagulants and flocculants more effectively, less harmful to the environment. In this sense, natural coagulants are a viable alternative due to their safety for health [21]. The coagulants of the skins of L. meyenii, S. tuberosum, and M. paradisiaca were collected from the waste of restaurants and street commerce. The present research focuses on analyzing the effectiveness of the skins of S. tuberosum, M. paradisiaca, and L. meyenni as natural coagulants in the

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum …

103

decrease of total suspended solids, turbidity, pH, and color in the lower basin of the Cunas River. The Cunas River was taken as an object of study because it is a main source of development of multiple economic activities for the Mantaro Valley, and in the lower basin is where there is a greater presence of urban area, 67% of this sector has poor quality water [22]. It should be noted that there are currently no studies on the use of the skin or pulp of L. meyenni as a natural coagulant.

2 Material and Methods The study presents a level of explanatory research with experimental design [23], and employs the inductive method due to the analysis of pre and posttest samples [24]. The water samples were extracted from 3 points of the lower basin of the Cunas River [25]; subsequently, the sample was separated into 13 beakers of 1000 mL each. Having 3 kinds of natural coagulants, it was decided to allocate 4 beakers, with a sample for each natural coagulant, and a control sample. To obtain the coagulant, skins were collected from restaurants and street commerce. Processing was carried out to obtain natural coagulants so that they could be used in the samples and thus know the optimal dose of the different coagulants from the skin of S. tuberosum, L. meyenii, and M. paradisiaca. Doses of 25, 50, 75, and 100 ppm were applied to each of the beakers [26]; At the same time, no alteration was made to the control group.

2.1 Water Sampling Following the National Protocol for monitoring the Quality of Surface Water Resources, it was considered to take 3 samples in the lower basin, this due to studies that mentioned that the quality of the Cunas River is affected by wastewater discharges [17]. For the collection of samples, 3 plastic containers (drums) with a capacity of 22.71 L were used. The analysis of the physicochemical parameters (turbidity, pH, color, and TSS), were performed in a laboratory certified by INACAL (National Institute of Quality).

2.2 Obtaining Natural Coagulants The procedure for obtaining natural coagulants was as follows: Initially, 2 kg of skin residues of S. tuberosum, L. meyenii, and M. paradisiaca were collected from the restaurants and markets located in the city of Huancayo-Peru [27]. Subsequently, the processes were carried out separately for each species where a first washing with drinking water was carried out, then in a container 5 L of water with 5 drops of NaClO (sodium hypochlorite) were placed, where the skins were immersed; after that, a third wash was carried out using distilled water for the elimination of impurities. As a fifth

104

F. E. Ambrosio Rosales et al.

step we proceeded with the drying in the laboratory stove at 100 °C for 24 h for S. tuberosum, L. meyenii, and 72 h for M. paradisiaca. Once the skins were dry, the grinding was carried out; finally, it was sifted with a mesh No. 40, where the fine flour obtained from the three coagulants were packed in polyethylene bags and preserved at room temperature [26].

2.3 Stock Solution For the preparation of the stock solution of M. paradisiaca, S. tuberosum, and L. meyenii, the same procedure was used for each one. The stock solution of 10,000 ppm was prepared, taking into account the methodologies of Aquino and Tovar [28], Salome and Salvatierra [29], Mallqui and Romero [30]; the three natural coagulants at 1%, using 2 g of the powders of each raw material in 200 ml of boiled distilled water (2 g/200 ml) [31], then the sample was homogenized for 30 min at 60 rpm. At the end of this time, it was left to rest for 10 min; finally, the stock solution of the natural coagulants M. paradisiaca, S. tuberosum, and L. meyenii was obtained. According to Chethana et al. [32] the solution has to be preserved for a maximum of one month.

2.4 Dosages of Natural Coagulants From the stock solution, the concentration volumes for each dose of 25, 50, 75, and 100 pm were obtained. The equation used to obtain the volumes [33]: C1 ∗ V 1 = C2 ∗ V 2 where C1 is the coagulant stock solution (2 g/200 mg), C2 is the dose with which 25, 50, 75, and 100 ppm were worked, V1 is the volume that was taken out of the stock solution and V2 is the volume of the water sampling with which the Jar-test was worked. In our case a sample volume of 1000 mL was taken. Table 1 shows the data obtained from the equation, with the volumes of the doses used being 2.5, 5, 7.5, and 10 mL respectively.

2.5 Jar-Test 4 treatments were performed where the doses of 25, 50, 75, and 100 ppm of each natural coagulant were applied. The Jar-test was performed on equipment (Model velp scientifica, JLT 4 series, 100–240 V, 50–60 Hz-F105A0109) that presented 4 pallets; where 4 beakers of 1000 mL were used, then each beaker was labeled with the

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum … Table 1 Doses used

105

Volumes found by dose Dosage—ppm

Volumes

25

2.5

50

5

75

7.5

100

10

corresponding dose and natural coagulant. Each beaker received the concentration volumes of 2.5, 5, 7.5, and 10 ml were applied per each coagulant. For the coagulation process, 100 rpm were considered for 10 min; after those 10 min the speed was decreased to 40 rpm for 20 min; this allowed the formation of flocs. At the end of the agitation, it was left to rest for 10 min, so that the colloids could settle [26]. After that, 300 mL were extracted from each beaker for the analysis of the physicochemical parameters; this procedure was performed in the same way for the three types of coagulants and the control treatment.

2.6 Parameter Analysis The analysis of the samples obtained from the lower basin of the Cunas River, before and after treatment, were carried out in a laboratory certified by INACAL, where physical-chemical parameters were evaluated: Color, turbidity, pH, and TSS.

3 Result 3.1 Pre-test Results To know the characteristics of the Cunas River, an initial measurement of the physicochemical parameters of turbidity, SST, pH, and color was made. Table 2 shows that turbidity had a value of 240 NTU. Regarding TSS, it was 235 mg/L; 6.93 for pH. According to previous research, it is shown that the Cunas River presented a pH of 7.21 in 2017 and 8.30 in 2019 [17]. As for the color, it was 900 PCU.

3.2 Post-test Results Turbidity. Table 3 shows that the control treatment (BK-00) results in 137 NTU (nephelometric turbidity unit) in turbidity after the Jar-test, whereas treatment with

106 Table 2 Characterization of the water of the Cunas River

F. E. Ambrosio Rosales et al.

Code

Parameter

Result

Unit

RC-0

Turbidity

240

NTUto

Sst

235

mg/L

pH

6.93

pH Unit

Color

900

PCUb

a b

NTU (nephelometric turbidity unit) PCU (platinum-cobalt unit)

natural coagulant of M. paradisiaca with doses of 75 ppm (CRT-PL03) gave the lowest turbidity value of 40.40 NTU than the other treatments. Also, it is observed that the highest value of 120 NTU was with the natural coagulant of S. tuberosum with a dose of 25 ppm (CRT-P01). Comparing the averages of the treatments with the control treatment, a notable reduction in turbidity could be observed in each treatment at different doses where the natural coagulant was applied; likewise, the treatment with CRT-PL shows better results of turbidity removal compared to CRT-P and CRT-M as it is shown in Fig. 1. Total suspended solids (TSS). Table 4 shows that the control treatment had a value of 204 mg/L after the Jar-test; similarly, it is observed that M. paradisiaca is the natural coagulant that presented the lowest values in all its applied doses, being the lowest value 42 mg/L with a dose of 100 ppm (CRT-PL04). Likewise, it was observed that L. meyenii is the natural coagulant that presents the highest values in all its applied doses, being the highest value 125 mg/L with a dose of 25 ppm (CRT-M01). Comparing the averages of all treatments with the control treatment, it Table 3 Turbidity results Natural coagulant

Code

Dosage—ppm

Result

Unit

Treatment control

BK a-00

0

137.00

NTU

Paradise muse

TRC-PL01

25

57.30

NTU

TRC-PL02

50

42.70

NTU

TRC-PL03

75

40.40

NTU

TRC-PL04

100

65.00

NTU

TRC-P01

25

120.00

NTU

TRC-P02

50

63.80

NTU

TRC-P03

75

60.80

NTU

TRC-P04

100

60.60

NTU

TRC-M01

25

100.00

NTU

TRC-M02

50

66.70

NTU

TRC-M03

75

66.20

NTU

TRC-M04

100

58.80

NTU

Solanum tuberosum

Lepidium meyenii

a

BK: Control treatment—target without application of natural coagulants

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum …

107

Fig. 1 Comparison of doses (PPM) with turbidity (NTU)

was observed that a reduction in TSS is shown in each treatment; likewise, treatment with PL-CRT shows better results than CRT-P and CRT-M as it is shown in Fig. 2. pH. Table 5 shows that the control treatment results in 7.1 pH; similarly, it is observed that the treatment with natural coagulant of M. paradisiaca with doses of 100 ppm (CRT-PL04) had the lowest pH of the other treatments, giving a result of 7.05 pH. Likewise, it is observed that the highest value was the treatment with the natural coagulant L. meyenii with 7.92 pH with doses of 100 ppm (CRT-M04). Figure 3 shows that in most treatments there is a slight increase in the pH parameter. This is due to DBO5 (Biochemical Oxygen Demand) caused by natural coagulants, which when applied in different measures, provide greater BOD5 in water, presenting higher protein content. Being biodegradable organic matter contained in the water Table 4 OSH results Natural coagulant

Code

Dosage—ppm

Result

Unit

Treatment control

BK-00

0

204

mg/L

Paradise muse

TRC-PL01

25

64

mg/L

TRC-PL02

50

58

mg/L

TRC-PL03

75

54

mg/L

TRC-PL04

100

42

mg/L

TRC-P01

25

119

mg/L

TRC-P02

50

106

mg/L

TRC-P03

75

86

mg/L

TRC-P04

100

84

mg/L

TRC-M01

25

125

mg/L

TRC-M02

50

110

mg/L

TRC-M03

75

94

mg/L

TRC-M04

100

92

mg/L

Solanum tuberosum

Lepidium meyenii

108

F. E. Ambrosio Rosales et al.

Table 5 pH results Natural coagulant

Code

Dosage—ppm

Result

Unit

Treatment control

BK-00

0

7.1

Unid.pH

Paradise muse

TRC-PL01

25

7.52

Unid.pH

TRC-PL02

50

7.51

Unid.pH

TRC-PL03

75

7.53

Unid.pH

TRC-PL04

100

7.05

Unid.pH

TRC-P01

25

7.06

Unid.pH

TRC-P02

50

7.38

Unid.pH

TRC-P03

75

7.45

Unid.pH

TRC-P04

100

7.43

Unid.pH

TRC-M01

25

7.13

Unid.pH

TRC-M02

50

7.46

Unid.pH

TRC-M03

75

7.66

Unid.pH

TRC-M04

100

7.92

Unid.pH

Solanum tuberosum

Lepidium meyenii

Fig. 2 Comparison of doses (PPM) with total suspended solids (Mg/L)

samples, it will be oxidized to CO2 and H2 O by microorganisms that use molecular oxygen [34]. Color. Table 6 shows that the control treatment has a value of 850 PCU (platinumcobalt unit) after the Jar-test; similarly, it is observed that M. paradisiaca is the natural coagulant that had lower values in all its applied doses, being the lowest value 290 PCU with a dose of 100 ppm (CRT-PL04). Likewise, it was observed that L. meyenii is the natural coagulant that presents the highest values in all its applied doses, being the highest value 460 PCU with a dose of 100 ppm (CRT-M04). Comparing the averages of all treatments with the control treatment, a notable reduction in color could be observed in each treatment of different doses where natural coagulants were applied;

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum …

109

Fig. 3 Comparison of doses (PPM) with pH (pH Unit)

Table 6 Color results Natural coagulant

Code

Dosage—ppm

Result

Unit

Treatment control

BK-00

0

850

PCU

Paradise muse

TRC-PL01

25

330

PCU

TRC-PL02

50

320

PCU

TRC-PL03

75

305

PCU

TRC-PL04

100

290

PCU

TRC-P01

25

370

PCU

TRC-P02

50

440

PCU

TRC-P03

75

390

PCU

TRC-P04

100

400

PCU

TRC-M01

25

405

PCU

TRC-M02

50

415

PCU

TRC-M03

75

450

PCU

TRC-M04

100

460

PCU

Solanum tuberosum

Lepidium meyenii

likewise, the TRC-PL treatment shows better results compared to CRT-P and CRT-M as it is shown in Fig. 4.

3.3 Effectiveness of Natural Coagulants Table 7 shows the effectiveness of natural coagulants; using M. paradisiaca, a turbidity removal of 70.51% (75 ppm) was obtained; for the TSS, a removal of 79.41% (100 ppm) was obtained; finally, for the color, a removal of 65.88% (100 ppm) was

110

F. E. Ambrosio Rosales et al.

Fig. 4 Comparison of doses (PPM) with color (PCU)

obtained. With S. tuberosum a turbidity removal of 55.77% (100 ppm) was obtained; for the TSS, a removal of 58.82% (100 ppm) was obtained; for the color, a removal of 56.7% (25 ppm) was obtained. Finally, using L. meyenni, it is observed that it reached a turbidity removal effectiveness of 57.08% (100 ppm). It can be observed that the TSS had a removal effectiveness of 54.90% (100 ppm); in the same way, it is observed that for the color, it was possible to remove 52.35% (25 ppm). Finally, for the pH, the variations after the Jar-test were considered. Figure 5 shows that the highest percentage of effectiveness of turbidity was achieved with the TRC-PL03 treatment, and the lowest effectiveness was that of the TRC-P01 treatment. For the TSS parameter, it is observed that the TRC-PL04 treatment presents the highest effectiveness unlike the TRC-M01 treatment that had the lowest result in the TSS parameter. In addition, it is observed that for the Color parameter, the TRC-PL04 treatment had the greatest effectiveness, and the TRC-M04 treatment had the lowest effectiveness.

3.4 Optimal Doses Table 8 shows the optimal doses selected by natural coagulant based on the average effectiveness. For M. paradisica, the optimal dose is 75 ppm, with an average effectiveness of 69.39%. For S. tuberosum, the optimal dose is 75 ppm, with an average effectiveness of 55.86%, and finally for L. meyenii, the optimal dose is 100 ppm, with an effectiveness of 52.62%. Figure 6 shows that the highest percentage of effectiveness was achieved in the CRT-PL03 treatment with an optimal dose of 75 ppm; in addition, it is appreciated that the TRC-P03 treatment had an intermediate effectiveness with an optimal dose of 75 ppm. Likewise, it can be observed that the least effective treatment is the CRT-M04 with an optimal dose of 100 ppm.

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum …

111

Table 7 Effectiveness of coagulants Natural coagulant

Paradise Muse

Solanum tuberosum

Lepidium meyenii

Code

Turbidity

Sst

Color

pH

BK-00 = 137 NTU

BK-00 = 204 mg/L

BK-00 = 850 PCU

BK.00 = 7.1 pH

Effectiveness (%)

Effectiveness (%)

Effectiveness (%)

Variation (±)

TRC-PL01

58.18

68.63

61.18

+0.42

TRC-PL02

68.83

71.57

62.35

+0.41

TRC-PL03

70.51

73.53

64.12

+0.43

TRC-PL04

52.55

79.41

65.88

−0.05

TRC-P01

12.41

41.67

56.47

−0.04

TRC-P02

53.43

48.04

48.24

+0.28

TRC-P03

55.62

57.84

54.12

+0.35

TRC-P04

55.77

58.82

52.94

+0.33

TRC-M01

27.01

38.73

52.35

+0.03

TRC-M02

51.31

46.08

51.18

+0.36

TRC-M03

51.68

53.92

47.06

+0.56

TRC-M04

57.08

54.90

45.88

+0.82

Fig. 5 Percentage of treatment effectiveness

Table 8 Optimal doses Natural coagulant

Code

Optimal dose (ppm)

Average effectiveness (%)

Paradise muse

TRC-PL03

75

69.39

Solanum tuberosum

TRC-P03

75

55.86

Lepidium meyenni

TRC-M04

100

52.62

112

F. E. Ambrosio Rosales et al.

Fig. 6 Average effectiveness of optimal treatments

4 Discussion The optimal dose used for the treatment of the lower basin of the Cunas River with the natural coagulant obtained from the skin of M. paradisiaca was 75 ppm, with a turbidity removal effectiveness of 70.51%, agreeing with Ahmad’s research [35], where it is used M. paradisiaca as a natural coagulant, obtaining a turbidity removal effectiveness of 64.67% with an optimal dose of 100 ppm. Also, it is related to Carrasquero [26] who used, in his research, M. paradisiaca as a natural coagulant and whose turbidity removal effectiveness was 94.5% when applying a dose of 25 ppm. Regarding the findings obtained for the color, with M. paradisiaca, the removal was 65.88% with an optimal dose of 100 ppm; this was the best result obtained at the level of the natural coagulants used. This finding coincides with Carrasquero [26] who, when using M. paradisiaca, obtained a color removal efficiency of 93.8% with an optimal dose of 25 ppm. On the other hand, the finding of TSS removal using M. Paradisiaca was 79.41 mg/L. Similar result to that of Carrasquero [26] that identified an average value of 253.33 mg/L of TSS. In addition, in the research of Moreno et al., [36] an optimal dose of 50 ppm of S. tuberosum was used, having a turbidity removal effectiveness of 93.31%. Also, in Carrasquero’s [26] research, an effectiveness of 97.8% was obtained using an optimal dose of 500 ppm when using the natural coagulant of the skin’s starch of S. tuberosum. Regarding the results of turbidity and color, the present research identified a removal of 55.77% and 56.47% respectively using S. tuberosum, keeping similarity with Carrasquero [26], who established that using doses greater than 500 ppm of S. tuberosum, the effectiveness of turbidity removal would be greater. For the parameter of TSS, Carrasquero [26] mentions that it was obtained as a result of the fact that the highest average value obtained was 240 mg/L, in a water with initial turbidity of 15 UNT applying an optimal dose of 250 mg/L. In the case of the results in the lower basin of the Cunas River, it was possible to obtain 58.82 mg/L of TSS, with an initial turbidity of 137 UNT applying an optimal dose of 100 ppm. The results found in this research would confirm that TSS removal increases when turbidity is greater. The natural coagulant obtained from the skin of L. meyenii presented a greater turbidity removal effectiveness of 57.08%

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum …

113

when applying the optimal dose of 100 ppm, demonstrating that at lower doses the effectiveness of turbidity removal is lower. With respect to the color parameter, an acceptable effectiveness percentage of 52.35% is shown with a dose of 25 ppm, reducing the control treatment to 405 PCU. For the TSS parameter, it is observed that the highest average value was 54.90 mg/L, in a sample with initial turbidity of 137 UNT, with an optimal dose of 100 ppm. With respect to the pH, the values obtained after applying the natural coagulant of the skin of M. paradisiaca vary from 7.52 to 7.05 pH units, the values obtained from the natural coagulant of the skin of S. tuberosum vary from 7.45 to 7.06 pH units, and the values obtained from the natural coagulant of the skin of L. meyenii vary from 7.92 to 7.13 pH units. The pH of the control sample was 7.1 pH units, demonstrating that when applying natural coagulants, no significant alterations of the pH parameter are observed. This would be in accordance with the research of Carrasquero [26], Vara [37], Camacho [10], and Moreno [36] who found that when using coagulants there are no extreme variations with respect to pH. At the same time, Carrasquero’s [26] research mentioned that when using the natural coagulant of the starch of the skin of S. tuberosum in higher doses, there is a decrease in pH, which is also reflected in our results. In Camacho’s [10] research it is mentioned that the pH must be in the range of 6.5–8.0 so that it can be subjected to the coagulation process without the need for a pH correction. As for the findings, there is no need to perform a pH correction because they are within the range mentioned by Camacho [10].

5 Conclusion The natural coagulant obtained from M. paradisiaca presented a higher average effectiveness of 69.39% when using an optimal dose of 75 ppm, higher than the other coagulants used; however, the greatest effectiveness of removal of TSS and color was given at a dose of 100 ppm. For pH it presents a variation of −0.05 with doses of 100 ppm. The natural coagulant obtained from the skin of S. tuberosum showed an average intermediate effectiveness of 55.86% with an optimal dose of 75 ppm; it has a greater effectiveness for turbidity removal of 55.77% with doses of 100 ppm; and also, it has a greater effectiveness of TSS removal of 58.82% with a dose of 100 ppm. For the color parameter, it has a greater removal effectiveness of 56.47% with doses of 25 ppm, and finally in the pH variation it presents − 0.04 with doses of 25 ppm compared to our control treatment. The natural coagulant obtained from the skin of L. meyenii showed a lower average effectiveness of 52.62% with an optimal dose of 100 ppm. This natural coagulant had less average effect compared to M. paradisiaca and S. tuberosum on the parameters evaluated. The highest effectiveness of turbidity removal is 57.08% with doses of 100 ppm; it is also obtained that the highest effectiveness of TSS removal is 54.90% with doses of 100 ppm. For the color parameter, the highest removal effectiveness is 52.35% with doses of 25 ppm. Finally, the pH presents a variation of +0.03 with doses of 25 ppm. Natural coagulants prepared from vegetable waste can be used in wastewater

114

F. E. Ambrosio Rosales et al.

treatment plants, in coagulation, flocculation, and sedimentation processes, being a more environmentally friendly alternative.

References 1. Immerzeel, W. et al.: Importance and vulnerability of the world’s water towers. 2(577) (2019) 2. Li, K., Liu, Q., Luo, R., Lu, Q., Zhou, W., Shuhao, H., Cheng, Q., Liu, J., Addy, M., Chen, P., Chen, D., Ruan, R.: Microalgae-based wastewater treatment for nutrients recovery: a review. 291(2) (2019) (Amsterdam, Elsevier) 3. Somorowska, U., Laszewski, M.: Quantifying streamflow response to climate variability, wastewater inflow, and sprawling urbanization in a heavily modified river basin. 656(4) (2019) (Amsterdam, Elsevier) 4. Salvador, L.: Water Management in Mucipalities in Latin America. 9(2) (2021) (El Ceprosimad) 5. Larramendi, R., Millán, G., Plana, A.: Scarcity and contamination of water, realities of the 21st century. Journal Estudiantil 60(279) (2021) 6. El País: Rivers of Latin America, among the most polluted in the world (Lima), p. 1. (2013) 7. De la Cruz, H.: Solid waste management and its impact on environmental education in an educational institution in Peru. 6(4) (2022) (Ciencia Latina) 8. Ministry of the Environment: Around 93 thousand tons of solid waste were valued in 2020 nationwide (Lima), p. 1. (2021) 9. El Comercio: They warn of dangers due to contamination in rivers of the jungle of San Martín (Lima), p. 1. (2021) 10. Camacho, H., Campos, D., Mercado, I., Cubillán, N., Castellar, G.: Use of potato peels (Solanum tuberosum L) in the clarification of water from the Ciénaga. Investig. Eng. Innov. 8(1) (2020) 11. Chee, T., Ta, W., Joon, J.: Optimization of agroindustrial wastewater treatment using unmodified rice starch as a natural coagulant, vol. 56. Elsevier, Amsterdam (2014) 12. Asharuddin, S., Othman, A., Mohd, N., Tajarudin, H., Din, H.: Flocculation and antibacterial performance of dual coagulant system of modified cassava peel starch and alum, vol. 31. Elsevier, Amsterdam (2016) 13. Fuentes, N., Rodriguez, E., Ariza, C.: Natural oagulants in continuous flow systems, as a substitute for Al2(SO4)3 for water clarification. 11(2) (2016) (Peru, Scielo) 14. Yin, C.: Emerging usage of plant-based coagulants for water and wastewater treatment. 45(9) (2010) (Amsterdam: Elsevier) 15. Zafar, M., Tausif, M., Mohsin, M., Waqas, S.: Potato starch as a coagulant for dye removal from textile wastewater. 226(244) (2015) (Berlin, Springer Link) 16. Saleem, M., Thomas, T.: A contemporary review on plant-based coagulants for applications in water treatment, vol. 72. Elsevier, Amsterdam (2019) 17. Custodio, M., Peñaloza, R., Chanamé, F., Hinostroza, J., De la Cruz, H.: Water quality dynamics of the Cunas River in rural and urban areas in the central region of Peru. 47(3) (2021) (Amsterdam, Elsevier) 18. Righetto, I., Juboori, R., Uzkurt, J., Mikola, A.: Wastewater Treatment with Starch-Based Coagulants for Nutrient Recovery Purposes: Testing on Lab and Pilot Scales, vol. 284. Elsevier, Amsterdam (2021) 19. Adedoja, A., Suzylawati, I.: Revolutionary Technique for Sustainable Plant-Based Green Coagulants in Industrial Wastewater Treatment—A Review, vol. 42. Elsevier, Amsterdam (2020) 20. Lugo, J., Lugo, E., Ovallos, D., Arango, J., De la Puente, M., Silva, J.: Effectiveness of the mixture of nopal and cassava starch as clarifying substances in water purification: a case study in Colombia. 6(6) (2020) (Netherlands, Heliyon)

Analysis of the Effectiveness of Lepidium Meyenii, Solanum Tuberosum …

115

21. Dearmas, D., Ramírez, L.: Nutrient removal by natural and chemical coagulants in a wastewater treatment plant, Valledupar Colombia. 6(2) (2015) (Spain, Dialnet) 22. Custodio, M., Chavez, E.: Quality of the aquatic environment of high Andean rivers evaluated through environmental indicators: a case of the Cunas River, Peru. 27(3) (2019) (Peru, Scielo) 23. Arroyo, J.: How Do I Run an Investigation? 1st ed., p. 14. Huancayo, Foundation for the development of science (2012) 24. Muñoz, C.: How to Develop and Advise Research Thesis, 2nd edn., p. 53. Prentice Hall, Mexico (2016) 25. National Water Authority: National Protocol for monitoring teh quality of surface water resources, p. 32. National Water Authority, Lima (2016) 26. Carrasquero, S., Montiel, S., Faría, E., Parra, P., Marin, J., Díaz, A.: Effectiveness of coagulants obtained from potato (Sonalum tuberosum) and plantain (Musa paradisiaca) residues in water clarification. 13(2), 12 (2017) (Peru, Neogranadina) 27. Reyes, B., Guevara, J.: Starch Obtaining from Plantain (Musa paradisiaca spp) Modified for the Coagulation-Flocculation Process Moyobamba, p. 45. (2017) 28. Aquino, K., Tovar, M.: Efficiency of Removal of Lead (II) from Mining Wastewater Using Potato Peel Starch (Solanum tuberosum) as a Natural Coagulant, p. 34. Peru, National University Of Central Peru (2021) 29. Salome, E., Salvaterria, J.: Evaluation of the Concentration of the Seed of Cassia Fistula as a Natural Coagulant and the Time of Depression, in the Treatment of Water for Purification in the Unidad Minera Poderosa–Huancavelica, p. 34. Peru, National University Of Central Peru (2019) 30. Mallqui, M., Romero, S.: Removal of Copper Ions from Wastwater Miners Using the Starch from the Shell of Banana. National University of Central Peru, Peru (2021) 31. Mas, M., Martínez, D., Carrasquero, M., Rincón, A., Vargas, L.: The efficiency of hymenaea courbaril seeds as a natural coagulant in the water clarification process. 2(2), 11 (2012) (Venezuela, Redieluz) 32. Chethana, M., Gayatri, L., Bhandari, V., Raja, V., Ranade, V.: Green approach to dye wastewater treatment using biocoagulants 4(5) (2016) (Spain, ACS Sustainable) 33. Huaringa, J., Vilcarano, D.: Effectiveness of the coagulant obtained from potato residues (Solanum tuberosum) in turbidity for purification (2019) (Peru, National University of Callao) 34. Choque, D., Ligarda, C., Ramos, B., Solano, A., Choque, Y., Peralta, D., Quispe, Y.: Optimization of the flocculating capacity of natural coagulants in water treatment. 87(212) (2020) (Peru: DINA) 35. Ahmad, A., et al.: Dosage-based application versus ratio-based approach for metal- and plantbased coagulants in wastewater treatment: merits, limitations, and applicability, vol. 334. Elsevier, Amsterdam (2022) 36. Moreno, G., Ortega, K., Valerio, L.: Application of Solanum tuberosum (potato) starch as a natural coagulant in the water treatment of the Punrún lagoon–Peru, p. 3 (2021) (Amsterdam, Elsevier) 37. Vara, S., Kumar, M., Kavitha, B.: Exploring Natural Coagulants as Impending Alternatives Towards Sustainable Water Clarification—A Comparative Studies of Natural Coagulants with Alum, vol. 32, p 7. Elsevier, Amsterdam (2019)

Geological Factors Influencing River Morphological Changes: Implications in the Agricultural Sector Akhmad Zamroni and Decibel V. Faustino-Eslava

Abstract Most studies often attributed changes in river morphology to natural and human influences, but they rarely explained geological factors. Furthermore, it is necessary to link morphological changes in rivers with social science perspectives. Therefore, this study aims to review geological factors influencing river morphological changes and their implications for the agricultural sector. To conduct this systematic literature review, the following steps include (1) formulating the research problem; (2) developing and validating the review protocol; (3) searching the literature; (4) screening for inclusion; (5) assessing quality; (6) extracting data; (7) analyzing and synthesizing data; and (8) reporting the findings. Geological factors that affect river morphological changes include slope, geological structure, lithology, and degree of weathering. While there are several effects of river morphological changes on the agricultural sectors, including increased or decreased agricultural land area, irrigation water quality pollution, and additional cost of adaptation strategy. Keywords Slope · Geological structure · Lithology · Degree of weathering · Agriculture

1 Introduction Numerous studies outlined the natural and human elements contributing to river morphological changes. Natural elements (e.g., climate change, volcanic eruptions, and major floods) or human elements (e.g., sediment mining, land-use change, and dams) can induce remarkable structural changes in river channels over short time intervals [1]. Some climate models predict a general trend toward decreased discharges and increasing precipitation. For the foreseeable future, rivers will likely continue to experience both these direct regulatory-related and climate-driven A. Zamroni (B) · D. V. Faustino-Eslava School of Environmental Science and Management, University of the Philippines, 4031 Los Baños, Laguna, Philippines e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 P.-C. Chiang (ed.), Environment and Renewable Energy, Springer Proceedings in Earth and Environmental Sciences, https://doi.org/10.1007/978-981-97-0056-1_10

117

118

A. Zamroni and D. V. Faustino-Eslava

changes, which could have profound and unanticipated effects on their morphodynamics and associated ecosystems. The hydrological regime is also starting to be impacted by human causes. People have modified river systems for millennia to meet their water demands and provide flood protection. Dams have been built to protect the water supply, control river flow for navigation, and produce electricity. As a result, many rivers’ flow regimes have been drastically changed from what they were naturally. Dams’ ability to alter flow can impact the size, timing, and duration of high and low flows [2]. Given its effects on basin-scale hydrology, water storage, irrigation systems, and drought and flood and disasters, studying past morphological changes in river reaches and their relationship to natural and man-made regulating factors is widely acknowledged as a powerful tool for defining evolutionary trends and sustainable river restoration [3]. This also has implications for the agricultural sector [4]. One of the most significant aspects of river geomorphology has been river morphology, concentrating on meander and braided rivers. Studies of river morphology also provide a geomorphologic foundation for protecting river landscapes and the environment and exploiting hydraulic resources. The geometric parameter of present rivers is a key to palaeohydrologic reconstruction and river evolution [5]. Most studies generally explained river morphological changes from natural and human factors but rarely explained river morphological changes due to geological factors. The origin and growth of drainage networks are influenced by the area’s endogenous and exogenous processes as well as the underlying geology [6]. The physical characteristics of the underlying rocks, tectonic deformation, and their structural elements are all included in geology. Geology interacts with climate to produce topography, including relief and the drainage system’s design. The geological characteristics of rocks, such as mineral composition, relative hardness, and the degree of weathering, which interact with climate, impact the grain size distribution and rate of sediment delivery to the stream system. The channel, valley floor, different hillslope processes, and landforms that affect the channel and valley bottom are all influenced by geology. For instance, debris flows or large landslides can quickly transport coarse material down rivers, causing constrictions, barriers, and natural dams. It can also displace large amounts of material or blocks into the channel [7]. Correlating river morphological changes with social science viewpoints is also crucial. River morphological changes harm the community’s socioeconomic well-being in several ways, including reduced navigability, loss of public and agricultural lands, the threat to fish production, and flooding. Climate change uncertainty, combined with other changes in land use, may result in changes in river shape, exacerbating the associated socioeconomic problems [8]. Therefore, this study aims to review geological factors influencing river morphological changes and their implications for the agricultural sector. This study is essential because stakeholders can input it to protect ecosystems, human life, and infrastructure.

Geological Factors Influencing River Morphological Changes …

119

2 Methodology The methodology of this study is based on guidance on conducting a systematic literature review [9], which includes “(1) formulating the research problem; (2) developing and validating the review protocol; (3) searching the literature; (4) screening for inclusion; (5) assessing quality; (6) extracting data; (7) analyzing and synthesizing data; and (8) reporting the findings” (Fig. 1). Step 1-Formulating the research problem, at this stage, the researcher creates research questions which are the main basis for researchers in developing a literature review. Research questions in this study include; (1) What are the geological factors influencing river morphological changes? and (2) How are the implications of river morphological changes to the agricultural sector? Step 2-Developing and validating the review protocol, in this stage, a research plan was designed. First, a list of keywords for the search that would be sufficient to find pertinent research topics was created. The inclusion and exclusion criteria were then established. Then, it was to develop search strategies for the related articles. Fourth, screening criteria and practices for papers that address the topics were established. Fifth, methods for gathering, combining, and summarizing information on the two research questions of this study were developed. Step 3-Searching the literature, keywords were merged into search strings for a thorough literature search. For instance, search strings (such as “river morphological changes”, “geological factors”, and “agricultural sector”) were used. In addition, using “AND”, and “OR” between some keywords were also used to obtain more specific literature results). Google Scholar is used as an openaccess database in this study. Step 4-Screening for inclusion, the inclusion criteria were: peer-reviewed, conference papers, empirical research papers, papers written

Fig. 1 Process of systematic literature review in this study

120

A. Zamroni and D. V. Faustino-Eslava

in English, year of publication not less than 2000, and focused on keywords used. The exclusion criteria were discussion paper, non-academic report, technical report, conceptual paper, non-English paper, year of publication less than 2000, and unavailable full text. Step 5-Assessing quality, an article met the quality assessment criteria if it was solely focused on river morphology, morphological changes, geological framework, and the agricultural sector. Reading the chosen references’ abstracts allowed researchers to understand the papers fully. Reading the entire text is necessary for clarification and a thorough understanding [10]. Step 6-Extracting data, researchers should carefully read the entire work and not only depend on the findings or the main conclusion. The only way to contextualize the results and avoid distorting the original paper is to do it in this manner. Step 7-Analyzing and synthesizing data, in order to propose a fresh contribution to knowledge, researchers must organize, contrast, collate, summarize, aggregate, or interpret the previously extracted information in this step [11]. Step 8-Reporting the findings, in this step, the researchers should organize the studies using a structure that ties them into significant themes, traits, or subgroups. Regardless of how stringent or accommodating the review procedures are, ensure the process is transparent and the data support the conclusions review [9].

3 Results and Discussion 3.1 Geological Factors Influencing River Morphological Changes Slope, geological structure, lithology, and degree of weathering are all geological elements that influence river morphological changes [7, 12, 13]. The slope significantly influences the pace and duration of water flow. Flatter surfaces are more susceptible to flooding because water moves more slowly, collects for a longer time, and accumulates there [14]. The sediment transport phenomena are closely related to the frequency of floods in the watershed. The process of transporting sediment could alter the morphology of the riverbed. During low water, erosion and deposition are the leading causes of many landforms. The river encourages the deposition of silt, which modifies the cross-section and is directly tied to the occurrence of floods. Due to the enormous flow observed during a flood, erosion is encouraged during flooding, changing the river’s morphology [15]. The drop in water level and stream flow will cause areas of silt deposition in the meander structures unless the basins are artificially changed. In addition, erosion and meander structure failures were brought on by the flood. After the flood, it was noticed that the flood-induced failures were repaired by artificial filling and natural sediment transfer [16]. So, steep or gentle slopes are not the keys to predicting narrowing or widening river morphology will be changed. On steep slopes, sediment might be more easily eroded. Still, on gentle slopes, if there is a flood, it will also be easy to erode sediment around the river bank

Geological Factors Influencing River Morphological Changes …

121

due to the significant accumulation of water on a flatter surface. In addition, at one point, the sediment eroded, but at another point, new sediment was deposited. The geological structure is one of the primary predisposing elements that regulate erosion occurrence at the regional scale [17]. Folds, faults, strata, brittle faults, ductile shears, contacts, foliations, lineations, structures with no preferred orientation (such as “miarolitic cavities”), joints, quartz veins, quartzo-feldspathic dykes, interlayer and intralayer dislocations, intrusives, deep fractures, and weak intercalations are examples of geological structures [18–21]. Geological setting, relief, drainage, geological boundaries, and trend of younger rocks have all been significantly influenced by the setting of geological structures. In general, the duration and degree of deformation determine the type and size of geological structures [18]. Narrowing and widening can result from the existence of geological formations, mainly if the faults follow or are cut across the river path. These geological formations favor down-cutting of the river bed rather than side-cutting, which causes the river channel to become deeper and narrower. Some lineaments/faults at specific spots cut across the river’s course, causing perturbations and convexities to form in the river bed. The incision, aggradation, and degradation of the river bank are geomorphic manifestations of the convexities along the river profile. Active and vigorous geologic processes are required to balance river convexities, which are typically unstable. Due to the migrating upstream and downstream convexities, the alluvial system may remain out of equilibrium long after the initial disturbance, even if the convexities can emerge on relatively rapid timescales [22]. Variations in grain size distribution, chemical content, and mineral composition are lithology principles [23]. Lithology consists of rock types, including alluvial, sedimentary, siltstone, limestone, shale, etc. Lithology is one of the key elements dictating the pace of surface erosion [24]. Changes in the drainage system are brought on by lithologically based variations in erosion resistance [25]. Due to the lithology’s low resistance to weathering, rivers and creeks are susceptible to erosion, which can result in side bank failures [26]. The erosion of river banks will lead to changes in river morphology. The transformation of rocks into sediments, soils, and/or the dissolution of minerals into ions through rock weathering is a crucial step in the formation of landforms and the evolution of the landscape [27]. In montane or upland areas where bedrock is likely to be exposed at the surface and where sufficiently steep slopes may carry rock straight to the streams, gravel, formed through physical rock weathering, frequently enters the stream system. Most gravel comes from linear sources along stream banks, like undercut slopes, or point sources in the terrain, like rock outcrops or landslides. The primary supply materials are unconsolidated colluvium, weathered rock, or glacial deposits. Recurrent debris flows may be the main mechanism for delivering coarse material from hillslopes into stream channels in steep mountains [28].

122

A. Zamroni and D. V. Faustino-Eslava

3.2 The Implications of the River Morphological Changes to the Agricultural Sector The essential component of relating decline and land-use changes to fluvial erosion is assessing bank erosion risks for the thorough study of fluvial erosion. Significant problems for human life, such as agricultural lands in floodplain zones, have been linked to meandering behavior. Several effects of river morphological changes on the agricultural sectors including decreasing the area of agricultural land, increasing the area of agricultural land, pollution of irrigation water quality, and additional cost of adaptation strategy. The erosion on the river banks is one indicator of the changes in river morphology [29]. In rivers with cohesive banks, erosion is primarily a result of the discharge, which accelerates the rate of change in river width as the downstream distance grows. However, at one point, the development of sand bars and the central island accelerates the erosion of the external bank, raising the width change rate. A strong correlation exists between event peak discharge and river bank erosion. Bank erosion results from several physical processes, including weathering, river erosion, and geotechnical instability [30]. The loss of houses, agricultural land, and subsequent unemployment among displaced individuals after forced human migration due to river bank erosion has also been noted [29, 31]. The development of new silt deposition on the river banks is one indicator of morphological changes in the river. Reduced channel width increases flow rates, increasing sediment movement, especially fine and small silt [32]. Distance from the main river, microtopography, and flood height were the key factors influencing sediment deposition rates. Deeply mobilized sediments were among the many sources from which the deposited sediments came [33]. Farmers can use the fresh sediment deposits on the riverbanks as new cropland. Most agricultural lands benefited from sediment deposition, particularly those near river banks where alluvial deposits (silt, loamy, and clayey soils) predominated and had a high humus content. Farmers in those places can use the river for intense irrigation. Conversely, the detrimental effects of sediment deposits include decreasing their productivity since the arable land is more likely to become unproductive due to the new deposit of infertile soils (coarse sandy soils) on fertile clayey soils [34]. There is typically a significant amount of suspended substrate in irrigation water. This comes from various sources common in natural water habitats, including continental weathering, coastal erosion, atmospheric deposition, in-situ biogeochemical processes, and industrial discharges [35–38]. Due to the abundance of eroded material in the area near the river banks, irrigation water quality is susceptible to contamination. It will result in irrigation water contamination, including raised turbidity, lowered pH, and raised levels of heavy metals in water bodies. The erosion process is associated with the degradation of surrounding river morphology and the decline in farmland production. Rainfall is one of the main causes of soil erosion because it erodes the soil, moves it from its location, and then washes it away as runoff. The growth of plants, water quality, and agricultural yields are all negatively impacted by soil erosion. Heavy metals, pesticides, and nonpoint nutrient pollutants are also

Geological Factors Influencing River Morphological Changes …

123

carried by soil particles, increasing the sediment levels that eventually eutrophic the water and upset fragile aquatic ecosystems. Extreme soil erosion causes excessive silt to be exported to waters, which disrupts aquatic life and lowers environmental quality [39, 40]. Maintaining sustainable development in irrigated agriculture is essential to social stability and economic development since it plays a significant role in the agricultural sector [41]. Adaptation methods include workable steps to help a region develop a resilient and adaptable system for future changes [42]. When calamities like floods or erosion affect agricultural land, especially along riverbanks, there is a substantial danger of crop loss, agricultural land loss, food insecurity, and economic losses. Policymakers and scholars have committed themselves to lower the detrimental impact of flood disasters on agriculture after becoming aware of the harmful consequences of flood events on agriculture [43]. By putting adaptation measures in place, such as building stronger disaster-resistant structures and establishing protection infrastructure, the expected increases in risk from natural disasters can be reduced [44]. However, implementing those adaptation measures comes at an additional expense, paid for out of the government budget or the farmers’ own pockets.

4 Conclusion Slope, geological structure, lithology, and weathering intensity are all geological elements that influence variations in river morphology. Sediment may be more easily eroded on steep slopes. However, on gentle slopes, in the event of a flood, it will also be simple to erode the sediment near the river bank because of the substantial water buildup on a flatter surface. Geological formations may cause a river to narrow or widen, especially if faults follow or cut across the river’s course. Rivers and creeks are vulnerable to erosion because of the lithology’s low susceptibility to weathering, which can cause side bank failures. River shape will alter as a result of bank erosion. A critical stage in landforms’ development and the landscape’s evolution is the weathering of rocks, which involves the dissolving of minerals into ions, sediments, and/or soils. River morphological changes have a variety of repercussions on the agricultural sectors, including increased or decreased agricultural land area, irrigation water quality pollution, and additional cost of adaptation strategy. Acknowledgements We thank Deutscher Akademischer Austauschdienst (DAAD) and Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA) for the research funding.

124

A. Zamroni and D. V. Faustino-Eslava

References 1. Yousefi, S., Mirzaee, S., Keesstra, S., Surian, N., Pourghasemi, H.R., Zakizadeh, H.R., Tabibian, S.: Effects of an extreme flood on river morphology (case study: Karoon River, Iran). Geomorphology 304, 30–39 (2018) 2. Van Oorschot, M., Kleinhans, M., Buijse, T., Geerling, G., Middelkoop, H.: Combined effects of climate change and dam construction on riverine ecosystems. Ecol. Eng. 120, 329–344 (2018) 3. Clerici, A., Perego, S., Chelli, A., Tellini, C.: Morphological changes of the floodplain reach of the Taro River (Northern Italy) in the last two centuries. J. Hydrol. 527, 1106–1122 (2015) 4. Langat, P.K., Kumar, L., Koech, R.: Understanding water and land use within Tana and Athi River Basins in Kenya: opportunities for improvement. Sustain. Water Resourc. Manag. 5(3), 977–987 (2019) 5. Zhang, B., Ai, N., Huang, Z., Yi, C., Qin, F.: Meanders of the Jialing River in China: morphology and formation. Chin. Sci. Bull. 53(2), 267–281 (2008) 6. Mahala, A.: The significance of morphometric analysis to understand the hydrological and morphological characteristics in two different morphoclimatic settings. Appl. Water Sci. 10(1), 1–6 (2020) 7. Grant, G.E., Schmidt, J.C., Lewis, S.L.: A geological framework for interpreting downstream effects of dams on rivers. Water Sci. Appl. 7, 209–225 (2003) 8. Shrestha, S., Imbulana, N., Piman, T., Chonwattana, S., Ninsawat, S., Babur, M.: Multimodelling approach to the assessment of climate change impacts on hydrology and river morphology in the Chindwin River Basin. Myanmar. Catena. 188, 104464 (2020) 9. Xiao, Y., Watson, M.: Guidance on conducting a systematic literature review. J. Plan. Educ. Res. 39(1), 93–112 (2019) 10. Suprapto, N., Zamroni, A., Yudianto, E.A.: One decade of the “LUSI” Mud Volcano: physical, chemical, and geological dimensions. Chemistry 26(4), 615–629 (2017) 11. Templier, M., Paré, G.: A framework for guiding and evaluating literature reviews. Commun. Assoc. Inf. Syst. 37(1), 6 (2015) 12. Aswathy, M.V., Vijith, H., Satheesh, R.: Factors influencing the sinuosity of Pannagon River, Kottayam, Kerala, India: an assessment using remote sensing and GIS. Environ. Monit. Assess. 138(1), 173–180 (2008) 13. Zamroni, A., Sugarbo, O., Prastowo, R., Widiatmoko, F. R., Safii, Y., Wijaya, R. A.: The relationship between Indonesian coal qualities and their geologic histories. In: AIP Conference Proceedings 2020, vol. 2245(1), pp. 070005. AIP Publishing LLC. 14. Hagos, Y.G., Andualem, T.G., Yibeltal, M., Mengie, M.A.: Flood hazard assessment and mapping using GIS integrated with multi-criteria decision analysis in upper Awash River basin, Ethiopia. Appl. Water Sci. 12(7), 1–8 (2022) 15. Gharbi, M., Soualmia, A., Dartus, D., Masbernat, L.: Floods effects on rivers morphological changes application to the Medjerda River in Tunisia. J. Hydrol. Hydromech. 64(1), 56 (2016) 16. Akay, S.S., Özcan, O., Sanlı, ¸ F.B.: Quantification and visualization of flood-induced morphological changes in meander structures by UAV-based monitoring. Eng. Sci. Technol. Int. J. 27, 101016 (2022) 17. Ilinca, V., Sandric, ¸ I., Jurchescu, M., Chi¸tu, Z.: Identifying the role of structural and lithological control of landslides using TOBIA and weight of evidence: case studies from Romania. Landslides 19(9), 2117–2134 (2022) 18. Mallo, S. J., Yarekes, P.: Preliminary study of rocks and soils types in mangun. Plateau State, North Central Nigeria (2012) 19. Simons, B. A.: The GeoSciML logical data model of geological concepts. (2014) 20. Yang, Z. X., Zhou, J. J.: Mine 3D geological interface modeling by NURBS technology. In: Advanced Materials Research, vol. 1065, pp. 44–47. Trans Tech Publications Ltd. (2015) 21. He, H., Liu, Y., Cui, J., Hu, D.: Scenario expression method for regional geological structures. ISPRS Int. J. Geo Inf. 11(1), 47 (2022)

Geological Factors Influencing River Morphological Changes …

125

22. Dar, R.A., Mir, S.A., Romshoo, S.A.: Influence of geomorphic and anthropogenic activities on channel morphology of River Jhelum in Kashmir Valley, NW Himalayas. Quat. Int. 507, 333–341 (2019) 23. Murdmaa, I., Borisov, D., Dorokhova, E., Dara, O.: Lithology. In: The Ioffe drift, pp. 53–97. Springer, Cham (2021) 24. Jamwal, A., Sharma, V.: Vulnerability assessment of landslide with the help of geospatial approach in Western Himalayas, Upper Basin of River Sutlej, India. In: Geospatial technology for environmental hazards, pp. 415–433. Springer, Cham (2022) 25. Moussi, A., Rebaï, N., Chaieb, A., Saâdi, A.: GIS-based analysis of the stream length-gradient index for evaluating effects of active tectonics: a case study of Enfidha (North-East of Tunisia). Arab. J. Geosci. 11(6), 1–5 (2018) 26. Tsai, Y.J., Chen, Y.J., Weng, C.H., Syu, F.T., Hsu, K.A., Lee, W.L.: Application of stream conductivity to activity of potential large-scale landslide. Front. Earth Sci. 10, 759556 (2022) 27. Krklec, K., Domínguez-Villar, D., Perica, D.: Use of rock tablet method to measure rock weathering and landscape denudation. Earth Sci. Rev. 212, 103449 (2021) 28. Church, M.: Gravel-bed rivers. Sediment cascades: An integrated approach. pp. 241–69 (2010) 29. Pal, S., Khatun, S., Mahato, S.: Charland mapping and analyzing suitability for settlement and agriculture in ganga river within the stretch of West Bengal, India. Geocarto Int. 37(25), 1–24 (2021) 30. Ashraf, M., Shakir, A.S.: Prediction of river bank erosion and protection works in a reach of Chenab River, Pakistan. Arab. J. Geosci. 11(7), 1–1 (2018) 31. Das, T.K., Haldar, S.K., Gupta, I.D., Sen, S.: River bank erosion induced human displacement and its consequences. Living Rev. Landscape Res. 8(3), 1–35 (2014) 32. Yousefi, S., Moradi, H.R., Keesstra, S., Pourghasemi, H.R., Navratil, O., Hooke, J.: Effects of urbanization on river morphology of the Talar River, Mazandaran Province, Iran. Geocarto Int. 34(3), 276–292 (2019) 33. Omengo, F.O., Alleman, T., Geeraert, N., Bouillon, S., Govers, G.: Sediment deposition patterns in a tropical floodplain, Tana River, Kenya. Catena. 143, 57–69 (2016) 34. Sadiq, A. A., Faruk, B. A.: Impact evaluation of sediment deposition on arable lands of Yola and environs of Adamawa State, Nigeria. Afr. J. Agric. Food Sci 3, 22–37. ISSN: 2689–5331 (2020) 35. Li, B., Peng, L., Wei, D., Lei, M., Liu, B., Lin, Y., Li, Z., Gu, J.: Enhanced flocculation and sedimentation of trace cadmium from irrigation water using phosphoric fertilizer. Sci. Total Environ. 601, 485–492 (2017) 36. Asih, A. S., Zamroni, A., Alwi, W., Sagala, S. T., Putra, A. S.: Assessment of heavy metal concentrations in seawater in the coastal areas around Daerah Istimewa Yogyakarta Province, Indonesia. Iraqi Geol. J. 14–22 (2022) 37. Zamroni, A., Sugarbo, O., Trisnaning, P. T., Sagala, S. T., Putra, A. S.: Geochemical approach for seawater intrusion assessment in the area around Yogyakarta International Airport, Indonesia. Iraqi Geol. J. 1–1 (2021) 38. Zamroni, A., Trisnaning, P. T., Prasetya, H. N., Sagala, S. T., Putra, A. S.: geochemical characteristics and evaluation of the groundwater and surface water in limestone mining area around Gunungkidul Regency, Indonesia. Iraqi Geol. J. 189–198 (2022) 39. Issaka, S., Ashraf, M.A.: Impact of soil erosion and degradation on water quality: a review. Geol. Ecol. Landscapes 1(1), 1–1 (2017) 40. Trisnaning, P. T., Zamroni, A., Sugarbo, O., Prasetya, H. N., Sagala, S. T., Hardiansyah, M. Y.: Quality of surface water due to sand mining activity: a case study from the Progo River, Daerah Istimewa Yogyakarta Province, Indonesia. In: IOP conference series: earth and environmental science, vol. 1098(1), pp. 012031. IOP Publishing (2022) 41. Fa-Hu, L.I., Zhang, L.J.: Combined effects of water quality and furrow gradient on runoff and soil erosion in North China. Pedosphere 20(1), 35–42 (2010) 42. Eikelboom, T., Janssen, R.: Interactive spatial tools for the design of regional adaptation strategies. J. Environ. Manage. 127, S6-14 (2013)

126

A. Zamroni and D. V. Faustino-Eslava

43. Alhassan, H.: Farm households’ flood adaptation practices, resilience and food security in the Upper East region, Ghana. Heliyon 6(6), e04167 (2020) 44. Haer, T., Botzen, W.W., Zavala-Hidalgo, J., Cusell, C., Ward, P.J.: Economic evaluation of climate risk adaptation strategies: cost-benefit analysis of flood protection in Tabasco, Mexico. Atmósfera. 30(2), 101–120 (2017)

Assessment of Particle Filter Technique for Data Assimilation in the Forecasting of Streamflows for the Tocantins River Basin in Brazil Karena Quiroz Jiménez

Abstract The Particle Filter (PF) technique is applied to forecasting of streamflows in the Tocantins River located in Brazil in this paper. This technique used as a data assimilation method is coupled to a semi-distributed hydrological model named MGB at hourly time intervals. The states variables are generated by computing rainfall forcing, considering time and spatial correlated errors. Sensibility tests were performed to highlight the importance of the precipitation error value and the particles number, as well as the low dependence on time and spatially correlated errors. The PF performance has been compared with streamflows predicting without assimilation, together with an empirical method. The resulting forecasts agreed well with the observations and maintained meaningful in terms of Nash–Sutcliffe at all stations analyzed even for long lead times. Also, PF technique performed well in greater lead time of forecasting when compared with empirical method. Keywords Particle filter · Forecasting · Data assimilation

1 Introduction Hydrological models are indispensable tools commonly used to predict, forecast and understand the terrestrial phase of the water cycle. Internal process associated with hydrological models for forecasting applications are plagued by uncertainties, especially in the model structure, chosen parameters and meteorological forcing data [1]. To improve the forecasting of hydrologic events, Data Assimilation (DA) methods are optimal options that incorporate measured data into mathematical models with the goal of optimizing the estimates of the system states [2]. Also, these methods are better advantage in spatially semi-distributed models when spatial input data with high-resolution is available. The Ensemble Kalman Filter (EnKF) method is one of the most common methods advocated to data assimilation, which consists in an K. Q. Jiménez (B) Peruvian University of Applied Sciences, Lima, Peru e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024 P.-C. Chiang (ed.), Environment and Renewable Energy, Springer Proceedings in Earth and Environmental Sciences, https://doi.org/10.1007/978-981-97-0056-1_11

127

128

K. Q. Jiménez

approach of simulations with sequential updates of model error covariance routing to an ensemble of model states, which are individual realizations perturbed by an assumed model error [3]. Another data assimilation method is Particle Filter, which utilizes the Bayesian filtering theory for non-linear and non-gaussian systems. This method seeks to represent the posterior probability density function of an ensemble of randomly chosen samples (particles), where each particle is associated with a given weight [4]. The overall procedure consists of three main steps: particle generation, weight computation (or updating) and particle regeneration (or resampling). For the second step, the updated weights are computed by means of the particle number and not from ensemble of state variables in order to reduce numerical instabilities, especially in semi-distributed physical models. However, in some cases resampling techniques make use of a significant amount of particles. Lately, several variants have been proposed in the literature to manage filter degeneration and particles size [5]. In this work, we aim to identify the limitations of the PF method in the computation of streamflow forecasts using a semi-distributed hydrological model applied to a large basin located in Brazil.

2 Methodology A large-scale semi-distributed hydrological model named MGB is described in the sequel, where the Particle Filter technique is used to assimilate MGB state variables and forcing precipitation method to generate the desired rainfall field.

2.1 Semi-Distributed MGB Hydrological Model The MGB model is a large-scale semi-distributed hydrological model widely applied to the study of several hydrological processes, including streamflow and river level [6]. This model uses input data derived from geographical information systems of the studied basin such as land use, topography, vegetation cover and soil types. The drainage area is divided into catchments for which hydrologic and hydraulic computations are made, where the Grouped Response Unit (GRU) approach is used for hydrological classification of combination of soil and land cover. That is, soil water budget is computed independently for each GRU by describing canopy interception, evapotranspiration, infiltration, surface runoff, subsurface flow, baseflow and soil water storage. In this manner, runoff generated from different GRUs in the cell is then summed up and propagated through the river network.

Assessment of Particle Filter Technique for Data Assimilation …

129

2.2 Particle Filter (PF) The PF method is based on a Bayesian filtering approach that consists in constructing the posterior probability density function of a system state by an ensemble of randomly drawn samples, named particles, with associated weights. The state-space and discrete-time dynamic systems for hydrologic models we shall consider here follow the general formulation as stated in the next equations. xt+1 = f (xt , u t ) + wt+1

(1)

yt+1 = h(xt+1 ) + νt+1

(2)

In Eq. 1, the function f is non-linear model operator, expresses the system transition in response to the forcing data u t expressing the precipitation, as well as the forecasted state variables. On the other hand, Eq. 2 prescribes the function h necessary to transfer the current states to the measurement states. The independent random vectors wt+1 and vt+1 represent the model and the measurement error. The principal idea of Particle Filtering is to approximate the posterior distribution of model state variables at time t by means of the following equation. p(xt+1 /yt+1 ) ≈

Np ∑

i i ωt+1 δ(xt+1 − xt+1 )

(3)

i=1 i where N p is the number of particles, ωt+1 denotes the weight of the ith particle and δ ∑N p i denotes the Dirac delta function. The weights are normalized so that i=1 ωt+1 = 1 and are obtained through importance sampling. The first step to estimate the posterior weights is to compute the likelihood based on the following term.

]2 yt+1 − h(xt+1 ) L(yt+1 \xt+1 ) = √ ) exp(− 2σ 2 2π σ 1

[

(4)

where yt+t denotes perturbed observations and it was computed from the available observation yt with an unidimensional error of state expressing the streamflow. The likelihood is applied on updating the particle weights based upon the available observations, i.e., the posterior weights for the ith particle are updated in a discrete manner with the following expression. ωi L(yt+1 \xt+1 ) i ωt+1 = ∑N p t k k=1 ωt L(yt+1 \x t+1 )

(5)

where ωti is the prior weight. In this paper, the resampling procedure applied is the Sampling Importance Resampling.

130

K. Q. Jiménez

2.3 Uncertainty in Rainfall Forcing Here, model state variables were perturbed by adding a noise in rainfall forcing. To apply the proposed methodology, the basin is divided into Nc ordered catchments, where each catchment center is associate to a local precipitation value. We shall denote by Pi,t ∈ R Nc the ith term of rainfall at time t, while Pi,t∗ denotes the perturbed precipitation at time t, which is computed in the following manner. / ( ) 1+β exp( ln E 2 + 1 ∈)Pi,t Pi,t∗ = √ E2 + 1

(6)

where E denotes a certain relative error expressed in percentage, β is a relative bias and ∈ is a random normal deviate with zero mean and unit variance. Spatially correlated pseudo random fields w were generated by means of the algorithm based on the two dimensional Fourier transform, having zero mean, unit variance and isotropic covariance function decreasing to the e−1 value at the distance τx called spatial decorrelation length. At each spatial location, temporal correlation was also considered using the following equation for simulating the time evolution of errors: ∈t = α∈t−1 +



1 − α 2 wt−1

(7)

where α is a temporal decorrelation parameter computed from Eq. (9), with τt being a temporal decorrelation length at time t. α =1−

Δt τt

(8)

For more information about this method, the reader can be referred to the work by [3].

3 Experimental Model 3.1 Area of Study and Available Data The Tocantins River basin is located in the central region of Brazil and presents a drainage area of 310,000 km2 up to the confluence with the Araguaia River as shown in Fig. 1. The basin topography elevations range from 83 to 1640 m. The annual rainfall mean is 1480 mm.year−1 and the streamflow is 3300 m3 .s−1 according to the computations performed for the period 2008–2014 at the Estreito station. The streamflow data were extracted from six stations from the Electric System National Operator, where the naturalized streamflow data is available at daily time intervals, while the meteorological data were obtained from fifteen climate stations

Assessment of Particle Filter Technique for Data Assimilation …

131

Fig. 1 Location of the Tocantins River basin

located around the basin according to the National Water and Sanitation Agency of Brazil. To obtain hourly data, all this data was linearly interpolated in this work. Because the precipitation data was obtained from only fifty rain-gauging stations, which is deemed a low density value for the studied area (1 station at every 6,200 km2 ), it was decided to combine the rainfall gauges with satellite precipitation product. For such scope, the results of the MGB model presented in [7] were employed. Here, the Tocantins River basin was discretized into 410 catchments and 45 sub-basins, while the integration of the use and soil type maps generated six different types of hydrological response units. Also, the outcomes of the prediction model reproduced reasonably well the observed streamflow data based on maximum values and for the recession period as well. Indeed, a Nash–Sutcliffe efficiency greater than 0.60 for locals over a drainage area greater than 20,000 km2 was obtained.

132

K. Q. Jiménez

3.2 Integration of Particle Filter with the MGB Hydrological Model The used state variables of the MGB model for the integration of the PF technique were the water storage in the soil layer, the volume from reservoirs (surface, subsurface and groundwater) and routing streamflow. The state variables at the beginning of the data assimilation process were estimated based on the initial conditions of the hydrological model and repeated for all particles. The rainfall was altered by means of synthetic generation according to the model described in the item 2.3. State variables were obtained for each particle through the simulation process. Sensitivity analysis was also performed to determine the parameters that connect the rainfall synthetic generation model with the particle number of the PF assimilation method. From the results of the simulation with the PF technique, the mean of the streamflows of all particles at each time interval was considered. The parameters of MGB model were considered invariant in time, i.e., the model parameters after the calibration process are kept constant throughout the assimilation process and streamflow forecasting. On the other hand, the error in observations was defined based on other studies, in which a value of 10% for daily streamflow was suggested [8, 9], however in this work a percentage error of 20% was defined for all observations. This value is greater than that considered with hourly data, which is 10%.

3.3 Parameters of the PF Scheme The parameters related with the PF scheme and rainfall forcing such as particles number (N P ), relative error precipitation (E), spatial decorrelation (τx ) and temporal decorrelation (τt ) were calculated by a sensitivity analysis. The main point of such analysis is to identify the most important parameters. Results were evaluated in terms of mean changes in root-mean-squared error (Δr ms) between observed and simulated streamflows computed for two samples. The first sample includes the streamflow used for data assimilation and the second one does without the assimilation (simulation model). Results were evaluated for six months from January to June in 2012 by means of the root-mean-square error that compares simulation results with observed ones: Δr ms = 100(r ms 2 − r ms 1 )/r ms 1 , , ranging from -100% (optimum) to +∞, where r ms 1 and r ms 2 are the root-mean-square errors from simulation model and PF simulations, respectively.

Assessment of Particle Filter Technique for Data Assimilation …

133

4 Discussion of Results First, the sensitivity analysis for the PF technique and forcing precipitation was computed and then the forecasting application on Tocantins River basin was performed.

4.1 Sensitivity Test First, Lajeado station was selected to compose the station temporal series for the assimilation process due to its central location in the basin, whereas the Serra da Mesa, Cana Brava, São Salvador, Peixe Angical and Estreito stations were used for the validation process. According to the analysis showed in Fig. 2, it may be observed that the performance of the assimilation process measured by the Δr ms values (resulting in more negative values) is better than those of the verification stations for all selected parameters. It is noted that the data assimilation scheme performed by the PF technique has a stronger dependency on the particle number and precipitation relative error, but with a minor dependency on τx and τt parameters. The election of parameters has been performed based on the Δr ms values for the validation stations, where a value of N p equal to 300 was elected because larger values will lead to a high computational cost. In this context, it was decided to consider the following values: N p = 300; E = 60%, τx = 2◦ and τt = 10 hours based on a sensitivity analysis.

4.2 Streamflow Forecasting Streamflow forecasting was realized for the period from January to May in 2013, obtaining a total of 3575 hourly time intervals. Here, the observed rainfall was taken as the forecast rainfall yielded from a scenario of real-time forecasting, in which the rainfall forecasts do not present errors. For instance, this consideration has already been made in [8]. The results of the streamflow forecasting are shown at the gauges located in the main network of the Tocantins River for a period of 144 h (6 days) with a frequency of one hour. These results were compared with those obtained from an Empirical method embedded in the MGB model, which it is based on the use of a correction factor obtained since the observed and computed streamflows and accumulated drainage areas. This condition makes the statistical terms exact, presenting null error and perfect Nash–Sutcliffe indexes with values equal to 1 at sites with measured data. Figure 3 shows the efficiency of the Nash–Sutcliffe index based on the lead time for four representative stations. Here, the gray vertical arrows highlight the superiority in performance of the PF technique over the Empirical method, in terms of NS,

134

K. Q. Jiménez

Fig. 2 Particle Number of PF and parameters of the rainfall synthetic model. Changes of the rootmean-square errors for gauges with PF (star dots, red color) and gauges with verification (circle dot, black color)

primarily for lead times larger than 40 h, 60 h and 120 h for Peixe Angical, Lajeado and Estreito stations, respectively. Also, it is sketched that the assimilation with both methods was better when compared to the efficiency of simulations without assimilation. Indeed, the assimilation used in both methods proved to mitigate the errors since the start of the forecast. A visual analysis of the forecasts using the DA methods is shown in Fig. 4. This analysis was performed for maximum events during the first period of analysis for the stations of Sao Salvador, Peixe Angical, Lajeado and Estreito. It was observed that the forecasting hydrographs computed with both DA methods showed noticeable differences when compared to the maximum observed streamflows, e.g., the maximum streamflow obtained with the Empirical method is overestimated for the Peixe Angical station, while this is underestimated with the PF technique in relation to the observed maximum streamflow from January 16, 2013. Conversely, the streamflow forecasting with both data assimilation methods was better adjusted when compared to the observed maximum streamflow in January 26, 2013. Otherwise, the maximum streamflow obtained with the PF technique was anticipated in January

Assessment of Particle Filter Technique for Data Assimilation …

135

Fig. 3 Nash–Sutcliffe efficient coefficient (NS) as a function of lead time with a forecasting period from December 15, 2003 to March 15, 2014

18, 2013 when compared to the maximum observed streamflow for Lajeado station. Furthermore, in São Salvador station the maximum streamflow was better adjusted with the observed streamflow by the PF technique in January 26, 2013. Thus, the forecasting analysis carried out has shown the usefulness of the data assimilation technique by Particle Filter technique for several forecast horizons.

136

K. Q. Jiménez

Fig. 4 Streamflow forecasting hydrographs with the beginning of forecasting at two instants in time corresponding to the period of analysis

5 Conclusion A data assimilation method namely Particle Filter (PF) was tested to evaluate its usefulness for real-time streamflow forecasting in this paper. The results were presented in terms of statistical indexes using a semi-distributed hydrological model called MGB applied to the Tocantins River basin in Brazil. Also, a comparison of this technique with an Empirical method for data assimilation was studied. As such, the state variables were obtained from a synthetic generation model of rainfall by means of the hydrological model using PF. The parameters of the rainfall generation model such as the relative error of rainfall and the ensemble size of PF were then calculated using a sensitivity analysis. According to the results presented herein, the data assimilation scheme with PF proved to be partially reliable. Although the empirical method behaves better in terms of computational time for the streamflow forecasting, the simulations using PF proved to be promising because the computed Nash–Sutcliffe index indicated a better performance of it in relation to the Empirical method for greater lead time of forecasting, e.g., 45 h for the Peixe Angical and 120 h for the Estreito stations. Another

Assessment of Particle Filter Technique for Data Assimilation …

137

limitation of the Empirical method is related to the management of uncertainties from various sources, which can be better treated with more advanced assimilation methods such as PF. Finally, it is highly recommended that the results incorporate forecasting rainfall models over the Tocantins River basin.

References 1. Vrugt, J.A., Diks, C.G.H., Grupta, H.V., Bouten, W., Verstraten, J.M.: Improved treatment of uncertainty in hydrologic modeling: combining the strengths of global optimization and data assimilation. Water Resour. Res. 41, W01017 (2005) 2. Camporese, M., Girotto, M.: Recent advances and opportunities in data assimilation for physicsbased hydrological modelling. Front. Water 11 (2022). https://doi.org/10.3389/frwa.2022.94883 3. Evensen, G.: The ensemble Kalman filter: theoretical formulation and practical implementation. Ocean Dyn. 53, 343–367 (2003) 4. Arulampalam, M., Maskell, S., Gordon, N., Clapp, T.: A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking, signal processing. IEEE Trans. 50(2), 174–188 5. Albarakati, A., Budiši´c, M., Crocker, R., Glass-Klaiber, J., Iams, S., Maclean, J., Marshall, N., Roberts, C., Van Vleck, E.S.: Model and data reduction for data assimilation: particle filters employing projected forecasts and data with application to a shallow water model. Comput. Math. Appl. 116, 194–211 (2022) 6. Collischonn, W., Allasia, D., Silva, B.C., Tucci, C.M.: The MGB-IPH model for large-scale rainfall–runoff modeling. Hydrol. Sci. J. 52, 878–895 (2007) 7. Quiroz, K., Collischonn, W., de Paiva, R. C. D.: Data assimilation using the ensemble Kalman filter in a distributed hydrological model on the Tocantins River. RBRH 24:e14 8. Paiva, R.C.D., Collischonn, W., Bonnet, M.P., Gonçalves, L.G.G., Calmant, S., Getirana, A., Silva, J.S.: Assimilating in situ and radar altimetry data into a largescale hydrologichydrodynamic model for streamflow forecast in the Amazon. Hydrol Earth Syst. Sci. 17, 2929–2946 (2013) 9. Clark, M.P., Rupp, D.E., Woods, R.A., Zheng, X., Ibbitt, R.P., Slater, A.G., Schmidt, J., Uddstrom, M.J.: Hydrological data assimilation with the ensemble Kalman filter: use of streamflow observations to update states in a distributed hydrological model. Adv. Water Resour. 31, 1309–1324 (2008)

Development of a Biodegradable Detergent Based on Quinoa as an Alternative to Minimize Eutrophication Karina Yupanqui Pacheco, José Vladimir Cornejo Tueros, and Fiorella Milagros Pacheco Dawson

Abstract Synthetic detergents were created more than a century ago and their use has been essential, however, during the SARS-CoV-2 pandemic, the Ministry of Health issued recommendations for hygiene and disinfection in which the use of these detergents was further enhanced. Due to this problem, it was proposed to develop a biodegradable detergent with low concentrations of compounds containing phosphorus and nitrogen, in order to minimize eutrophication in surface bodies that cause ecosystemic imbalances. The experiment consisted of washing two garments in the same conditions (dirty) and each one with each detergent (conventional and biodegradable), as a result it was obtained that both garments were clean after washing. For the laboratory analysis, wastewater samples were taken from each container. The results obtained from the laboratory were as follows: the conventional detergent had the following concentrations: pH (11.25), T° (12.3 °C), Conductivity (11,350 uS/cm), Phosphorus (4.11 mg/L) and total Nitrogen (20.16 mg/L); while for the biodegradable detergent the resulting concentrations were: pH (7.02), T° (12.6 °C), Conductivity (445 uS/cm), Phosphorus (0.318 mg/L) and total Nitrogen (