Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama

Meless Djedjro Franck-Renaud; Gbagbo Tchape Aubin; Kpaibe Sawa Andre Philippe; Yapo Toussaint Wolfgang; Kouassi-Agbessi Therese Brah; Amin N’cho Christophe

doi:doi:10.11648/j.ajbes.20220804.12

Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama

Meless Djedjro Franck-Renaud, Gbagbo Tchape Aubin, Kpaibe Sawa Andre Philippe, Yapo Toussaint Wolfgang, Kouassi-Agbessi Therese Brah, Amin N’cho Christophe

Published in American Journal of Biological and Environmental Statistics (Volume 8, Issue 4)

Received: 22 September 2022 Accepted: 7 November 2022 Published: 16 November 2022

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Context: In the village of M'pody in the Anyama district, located about 60 kilometers from the town of Anyama, a diarrhea epidemic was detected in January 2020 and affected 69 people, mostly children aged 0 to 5 years. According to the affected population, these cases of diarrhea were related to the consumption of water from the improved village water system, which had not been maintained for nearly three years. The objective of this work was to develop a bacteriological characterization model of the water table in the village of M'pody (Ivory coast) based on physicochemical parameters and meteorology in order to estimate the concentration of indicator germs of fecal pollution (Escherichia coli) by well. Methods: The methodology consisted of four water sampling campaigns per well during the year's four seasons on all 72 wells in this region, for a total of 288 visits. Conventional physico-chemical parameters were determined using electrochemical and spectrophotometric methods. Bacteriological parameters were determined by the membrane filtration technique. A sanitary inspection was also carried out. The development of the prediction model for the Escherichia coli indicator was performed using a linear mixed model. The performance of our model was evaluated by bootstrap and k-fold cross-validation techniques. Results: The mixed linear model with random intercept (log transformation) chosen following the spaghetti plot and likelihood ratio test gave the following results: The predictive model explained 30,24% of the variance in Escherichia coli concentrations (log transformation). It is based on 9 variables. Validation of the model performance by bootstrap gave us a very low relative bias < 5%, average prediction errors (RMSE) and absolute prediction errors per K-fold lower than 2,5. Conclusion: The development of the statistical model for predicting concentrations of fecal pollution indicator bacteria in wells was made possible by the existence of reliable databases. These databases made it possible to use 9 explanatory variables in a scientific approach to explaining the variable explained Escherichia coli. The validation of the predictive performances by K-fold and bootstrap showed that the model predictions are accurate and the bootstrap estimates of the parameters are unbiased. This implemented model could be used in the event of a declaration of waterborne diseases in this locality before the results of the microbiological analysis are returned.

Published in	American Journal of Biological and Environmental Statistics (Volume 8, Issue 4)
DOI	10.11648/j.ajbes.20220804.12
Page(s)	102-111
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2022. Published by Science Publishing Group

Keywords

Mixed Linear Model, Bootstrap, Principal Component Analysis

References

[1]	Otchoumou KFE, Bachir SM, Etienne AG, Issiaka S. Contribution of remote sensing and GIS in the identification of groundwater resources in the Daoukro region (Central-Eastern Ivory Coast). Int J Innov Appl Stud 2012; 1: 35–53.
[2]	Festy B, Hartemann P, Ledrans M, Levallois P, Payment P, Tricard D. Water Quality. Environ Santé Publique-Fond Prat 2003: 333–68.
[3]	Guergazi S, Achour S. Physico-chemical characteristics of the town of Biskra’s feedwater. Practice of chlorination. Larhyss J 2005; 4: 119–27.
[4]	Adelaïde OYMJ, Yaya OL, Bernard YO, Veronique M. Impacts of Environmental Management on the Quality of Traditional Well Water in the Soubré Region (South-West of Côte d’Ivoire). J Water Resour Prot. 27 déc 2017; 9 (13): 1634-44Ernest AK, Nagnin S, Gbombélé S, Théophile L, Solange OM, Pacôme ZS. Groundwater pollution in Africans biggest towns: case of the town of Abidjan (Côte d’Ivoire). Eur J Sci Res 2008; 20: 302–16.
[5]	Yapo O, Mambo V, Seka A, Ohou MJA, Konan F, Gouzile V, et al. Evaluation of the quality of domestic well water in the deprived areas of four municipalities of Abidjan (Ivory Coast): Koumassi, Marcory, Port-Bouet and Treichville. Int J Biol Chem Sci 2010; 4.
[6]	Lacina C, Dramane D, Adama C, Germain G. Use of water resources, sanitation and health risks in the precarious neighbourhoods of the commune of Port-Bouët (Abidjan; Ivory Coast). VertigO- Electronic Rev En Sci Environ 2004; 5.
[7]	Fofana F. Assessment and mapping of the vulnerability of the Abidjan groundwater to pollution using Drastic and God methods. Memo DEA Univ Abobo-Adjamé Côte D'Ivoire 2005: 72.
[8]	Ahoussi KE, Koffi YB, Kouassi AM, Soro G, Biémi J. 2013. Étude hydrochimique et microbiologique des eaux de source de l’ouest montagneux de la Côte d’Ivoire : Cas du village de Mangouin-Yrongouin (sous-préfecture de Biankouman). Journal of Applied Biosciences, 63: 4703– 471 n.d.
[9]	Jourda JP, Kouamé KJ, Saley MB, Kouadio BH, Oga YS, Deh S. Contamination of the Abidjan aquifer by sewage: An assessment of extent and strategies for protection. Groundw. Pollut. Afr., CRC Press; 2006, pp. 305–14.
[10]	KOACI. Ivory Coast: 69 cases of “detected” diarrhea in Anyama, no loss of life. KOACI n.d. https://www.koaci.com/article/2020/01/24/cote-divoire/sante/cote-divoire-69-cas-de-diarrhee-detectes-a-anyama-aucune-perte-en-vie-humaine_138802.html (accessed November 12, 2021).
[11]	Rompré A, Servais P, Baudart J, De-roubin M-R, Laurent P. Detection and enumeration of coliforms in drinking water: current methods and emerging approaches. J Microbiol Methods. 2002; 49 (1): 31 54. n.d.
[12]	DGPR (GENERAL CENSUS OF POPULATION AND HABITAT). (2014): Socio-demographic data. Permanent Technical Secretariat of the RGPH Technical Committee, p 26 n.d.
[13]	Nomades DC. 2020 Anyama weather. Hist Météo n.d. https://www.historique-meteo.net/afrique/cote-d-ivoire/anyama/2020/ (accessed May 19, 2022).
[14]	Palm R. Use of bootstrap for statistical issues related to parameter estimation. Biotechnol Agron Society Environ 2002; 6: 143–53.
[15]	Modelling the bacteriological quality of a potential bathing site in Beauport Bay, QUEBEC, ETE, 2003 - Google Search n.d. accessed May 16, 2022.
[16]	N’Diaye AD, Sid MO, Kankou AO, Namr KI. Assessment of faecal coliform contents by coupling with physic o-chemical parameters in ACP: case of the WWTP effluents of the Sebkha, Nouakchott market perimeter. Cameroon J Exp Biol 2011; 7: 35–40.
[17]	Iron and infectious diseases François Bricaire, Paris - Google Search n.d. https://www.google.com/search?q=Fer+et+maladies+infectieux+Fran%C3%A7ois+Bricaire%2C+Paris&oq=Fer+et+maladies+infe
[18]	Escherichia coli (E. coli) s.d. https://www.who.int/fr/news-room/fact-sheets/detail/e-coli (consulté le 16 mai 2022).
[19]	Snee RD. Validation of regression models: methods and examples. Technometrics 1977; 19: 415-28.
[20]	Yang Y, Huang S. Suitability of five cross validation methods for performance evaluation of nonlinear mixed-effects forest models – a case study. For Int J For Res 2014; 87: 654–62.
[21]	Wheeler W. Evaluating linear regression models using RMSE and R2. wwblog 2021. https://medium.com/wwblog/evaluating-regression-models-using-rmse-and-r%C2%B2-42f77400efee (consulté le 17 mai 2022).
[22]	Thai H-T, Mentré F, Holford NH, Veyrat Follet C, Comets E. A comparison of bootstrap approaches for estimating uncertainty of parameters in linear mixed effects models. Pharm Stat 2013; 12: 129-40.

Cite This Article

Plain Text BibTeX RIS

APA Style

Meless Djedjro Franck-Renaud, Gbagbo Tchape Aubin, Kpaibe Sawa Andre Philippe, Yapo Toussaint Wolfgang, Kouassi-Agbessi Therese Brah, et al. (2022). Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama. American Journal of Biological and Environmental Statistics, 8(4), 102-111. https://doi.org/10.11648/j.ajbes.20220804.12

Copy | Download

ACS Style

Meless Djedjro Franck-Renaud; Gbagbo Tchape Aubin; Kpaibe Sawa Andre Philippe; Yapo Toussaint Wolfgang; Kouassi-Agbessi Therese Brah, et al. Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama. Am. J. Biol. Environ. Stat. 2022, 8(4), 102-111. doi: 10.11648/j.ajbes.20220804.12

Copy | Download

AMA Style

Meless Djedjro Franck-Renaud, Gbagbo Tchape Aubin, Kpaibe Sawa Andre Philippe, Yapo Toussaint Wolfgang, Kouassi-Agbessi Therese Brah, et al. Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama. Am J Biol Environ Stat. 2022;8(4):102-111. doi: 10.11648/j.ajbes.20220804.12

Copy | Download

@article{10.11648/j.ajbes.20220804.12,
  author = {Meless Djedjro Franck-Renaud and Gbagbo Tchape Aubin and Kpaibe Sawa Andre Philippe and Yapo Toussaint Wolfgang and Kouassi-Agbessi Therese Brah and Amin N’cho Christophe},
  title = {Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama},
  journal = {American Journal of Biological and Environmental Statistics},
  volume = {8},
  number = {4},
  pages = {102-111},
  doi = {10.11648/j.ajbes.20220804.12},
  url = {https://doi.org/10.11648/j.ajbes.20220804.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajbes.20220804.12},
  abstract = {Context: In the village of M'pody in the Anyama district, located about 60 kilometers from the town of Anyama, a diarrhea epidemic was detected in January 2020 and affected 69 people, mostly children aged 0 to 5 years. According to the affected population, these cases of diarrhea were related to the consumption of water from the improved village water system, which had not been maintained for nearly three years. The objective of this work was to develop a bacteriological characterization model of the water table in the village of M'pody (Ivory coast) based on physicochemical parameters and meteorology in order to estimate the concentration of indicator germs of fecal pollution (Escherichia coli) by well. Methods: The methodology consisted of four water sampling campaigns per well during the year's four seasons on all 72 wells in this region, for a total of 288 visits. Conventional physico-chemical parameters were determined using electrochemical and spectrophotometric methods. Bacteriological parameters were determined by the membrane filtration technique. A sanitary inspection was also carried out. The development of the prediction model for the Escherichia coli indicator was performed using a linear mixed model. The performance of our model was evaluated by bootstrap and k-fold cross-validation techniques. Results: The mixed linear model with random intercept (log transformation) chosen following the spaghetti plot and likelihood ratio test gave the following results: The predictive model explained 30,24% of the variance in Escherichia coli concentrations (log transformation). It is based on 9 variables. Validation of the model performance by bootstrap gave us a very low relative bias Conclusion: The development of the statistical model for predicting concentrations of fecal pollution indicator bacteria in wells was made possible by the existence of reliable databases. These databases made it possible to use 9 explanatory variables in a scientific approach to explaining the variable explained Escherichia coli. The validation of the predictive performances by K-fold and bootstrap showed that the model predictions are accurate and the bootstrap estimates of the parameters are unbiased. This implemented model could be used in the event of a declaration of waterborne diseases in this locality before the results of the microbiological analysis are returned.},
 year = {2022}
}

Copy | Download

TY  - JOUR
T1  - Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama
AU  - Meless Djedjro Franck-Renaud
AU  - Gbagbo Tchape Aubin
AU  - Kpaibe Sawa Andre Philippe
AU  - Yapo Toussaint Wolfgang
AU  - Kouassi-Agbessi Therese Brah
AU  - Amin N’cho Christophe
Y1  - 2022/11/16
PY  - 2022
N1  - https://doi.org/10.11648/j.ajbes.20220804.12
DO  - 10.11648/j.ajbes.20220804.12
T2  - American Journal of Biological and Environmental Statistics
JF  - American Journal of Biological and Environmental Statistics
JO  - American Journal of Biological and Environmental Statistics
SP  - 102
EP  - 111
PB  - Science Publishing Group
SN  - 2471-979X
UR  - https://doi.org/10.11648/j.ajbes.20220804.12
AB  - Context: In the village of M'pody in the Anyama district, located about 60 kilometers from the town of Anyama, a diarrhea epidemic was detected in January 2020 and affected 69 people, mostly children aged 0 to 5 years. According to the affected population, these cases of diarrhea were related to the consumption of water from the improved village water system, which had not been maintained for nearly three years. The objective of this work was to develop a bacteriological characterization model of the water table in the village of M'pody (Ivory coast) based on physicochemical parameters and meteorology in order to estimate the concentration of indicator germs of fecal pollution (Escherichia coli) by well. Methods: The methodology consisted of four water sampling campaigns per well during the year's four seasons on all 72 wells in this region, for a total of 288 visits. Conventional physico-chemical parameters were determined using electrochemical and spectrophotometric methods. Bacteriological parameters were determined by the membrane filtration technique. A sanitary inspection was also carried out. The development of the prediction model for the Escherichia coli indicator was performed using a linear mixed model. The performance of our model was evaluated by bootstrap and k-fold cross-validation techniques. Results: The mixed linear model with random intercept (log transformation) chosen following the spaghetti plot and likelihood ratio test gave the following results: The predictive model explained 30,24% of the variance in Escherichia coli concentrations (log transformation). It is based on 9 variables. Validation of the model performance by bootstrap gave us a very low relative bias Conclusion: The development of the statistical model for predicting concentrations of fecal pollution indicator bacteria in wells was made possible by the existence of reliable databases. These databases made it possible to use 9 explanatory variables in a scientific approach to explaining the variable explained Escherichia coli. The validation of the predictive performances by K-fold and bootstrap showed that the model predictions are accurate and the bootstrap estimates of the parameters are unbiased. This implemented model could be used in the event of a declaration of waterborne diseases in this locality before the results of the microbiological analysis are returned.
VL  - 8
IS  - 4
ER  -

Copy | Download

Author Information

Meless Djedjro Franck-Renaud

National Institute of Public Hygiene, Abidjan, Ivory Coast
Gbagbo Tchape Aubin

National Institute of Public Hygiene, Abidjan, Ivory Coast
Kpaibe Sawa Andre Philippe

National Institute of Public Hygiene, Abidjan, Ivory Coast
Yapo Toussaint Wolfgang

National Institute of Public Hygiene, Abidjan, Ivory Coast
Kouassi-Agbessi Therese Brah

National Institute of Public Hygiene, Abidjan, Ivory Coast
Amin N’cho Christophe

National Institute of Public Hygiene, Abidjan, Ivory Coast

Download PDF

Sections

Plain Text BibTeX RIS

APA Style

Meless Djedjro Franck-Renaud, Gbagbo Tchape Aubin, Kpaibe Sawa Andre Philippe, Yapo Toussaint Wolfgang, Kouassi-Agbessi Therese Brah, et al. (2022). Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama. American Journal of Biological and Environmental Statistics, 8(4), 102-111. https://doi.org/10.11648/j.ajbes.20220804.12

Copy | Download

ACS Style

Meless Djedjro Franck-Renaud; Gbagbo Tchape Aubin; Kpaibe Sawa Andre Philippe; Yapo Toussaint Wolfgang; Kouassi-Agbessi Therese Brah, et al. Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama. Am. J. Biol. Environ. Stat. 2022, 8(4), 102-111. doi: 10.11648/j.ajbes.20220804.12

Copy | Download

AMA Style

Meless Djedjro Franck-Renaud, Gbagbo Tchape Aubin, Kpaibe Sawa Andre Philippe, Yapo Toussaint Wolfgang, Kouassi-Agbessi Therese Brah, et al. Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama. Am J Biol Environ Stat. 2022;8(4):102-111. doi: 10.11648/j.ajbes.20220804.12

Copy | Download

@article{10.11648/j.ajbes.20220804.12,
  author = {Meless Djedjro Franck-Renaud and Gbagbo Tchape Aubin and Kpaibe Sawa Andre Philippe and Yapo Toussaint Wolfgang and Kouassi-Agbessi Therese Brah and Amin N’cho Christophe},
  title = {Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama},
  journal = {American Journal of Biological and Environmental Statistics},
  volume = {8},
  number = {4},
  pages = {102-111},
  doi = {10.11648/j.ajbes.20220804.12},
  url = {https://doi.org/10.11648/j.ajbes.20220804.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajbes.20220804.12},
  abstract = {Context: In the village of M'pody in the Anyama district, located about 60 kilometers from the town of Anyama, a diarrhea epidemic was detected in January 2020 and affected 69 people, mostly children aged 0 to 5 years. According to the affected population, these cases of diarrhea were related to the consumption of water from the improved village water system, which had not been maintained for nearly three years. The objective of this work was to develop a bacteriological characterization model of the water table in the village of M'pody (Ivory coast) based on physicochemical parameters and meteorology in order to estimate the concentration of indicator germs of fecal pollution (Escherichia coli) by well. Methods: The methodology consisted of four water sampling campaigns per well during the year's four seasons on all 72 wells in this region, for a total of 288 visits. Conventional physico-chemical parameters were determined using electrochemical and spectrophotometric methods. Bacteriological parameters were determined by the membrane filtration technique. A sanitary inspection was also carried out. The development of the prediction model for the Escherichia coli indicator was performed using a linear mixed model. The performance of our model was evaluated by bootstrap and k-fold cross-validation techniques. Results: The mixed linear model with random intercept (log transformation) chosen following the spaghetti plot and likelihood ratio test gave the following results: The predictive model explained 30,24% of the variance in Escherichia coli concentrations (log transformation). It is based on 9 variables. Validation of the model performance by bootstrap gave us a very low relative bias Conclusion: The development of the statistical model for predicting concentrations of fecal pollution indicator bacteria in wells was made possible by the existence of reliable databases. These databases made it possible to use 9 explanatory variables in a scientific approach to explaining the variable explained Escherichia coli. The validation of the predictive performances by K-fold and bootstrap showed that the model predictions are accurate and the bootstrap estimates of the parameters are unbiased. This implemented model could be used in the event of a declaration of waterborne diseases in this locality before the results of the microbiological analysis are returned.},
 year = {2022}
}

Copy | Download

TY  - JOUR
T1  - Development of a Groundwater Quality Prediction Model for the M'pody Village of Anyama
AU  - Meless Djedjro Franck-Renaud
AU  - Gbagbo Tchape Aubin
AU  - Kpaibe Sawa Andre Philippe
AU  - Yapo Toussaint Wolfgang
AU  - Kouassi-Agbessi Therese Brah
AU  - Amin N’cho Christophe
Y1  - 2022/11/16
PY  - 2022
N1  - https://doi.org/10.11648/j.ajbes.20220804.12
DO  - 10.11648/j.ajbes.20220804.12
T2  - American Journal of Biological and Environmental Statistics
JF  - American Journal of Biological and Environmental Statistics
JO  - American Journal of Biological and Environmental Statistics
SP  - 102
EP  - 111
PB  - Science Publishing Group
SN  - 2471-979X
UR  - https://doi.org/10.11648/j.ajbes.20220804.12
AB  - Context: In the village of M'pody in the Anyama district, located about 60 kilometers from the town of Anyama, a diarrhea epidemic was detected in January 2020 and affected 69 people, mostly children aged 0 to 5 years. According to the affected population, these cases of diarrhea were related to the consumption of water from the improved village water system, which had not been maintained for nearly three years. The objective of this work was to develop a bacteriological characterization model of the water table in the village of M'pody (Ivory coast) based on physicochemical parameters and meteorology in order to estimate the concentration of indicator germs of fecal pollution (Escherichia coli) by well. Methods: The methodology consisted of four water sampling campaigns per well during the year's four seasons on all 72 wells in this region, for a total of 288 visits. Conventional physico-chemical parameters were determined using electrochemical and spectrophotometric methods. Bacteriological parameters were determined by the membrane filtration technique. A sanitary inspection was also carried out. The development of the prediction model for the Escherichia coli indicator was performed using a linear mixed model. The performance of our model was evaluated by bootstrap and k-fold cross-validation techniques. Results: The mixed linear model with random intercept (log transformation) chosen following the spaghetti plot and likelihood ratio test gave the following results: The predictive model explained 30,24% of the variance in Escherichia coli concentrations (log transformation). It is based on 9 variables. Validation of the model performance by bootstrap gave us a very low relative bias Conclusion: The development of the statistical model for predicting concentrations of fecal pollution indicator bacteria in wells was made possible by the existence of reliable databases. These databases made it possible to use 9 explanatory variables in a scientific approach to explaining the variable explained Escherichia coli. The validation of the predictive performances by K-fold and bootstrap showed that the model predictions are accurate and the bootstrap estimates of the parameters are unbiased. This implemented model could be used in the event of a declaration of waterborne diseases in this locality before the results of the microbiological analysis are returned.
VL  - 8
IS  - 4
ER  -

Copy | Download