Spatial Assessment of Water River Pollution Using the Stochastic Block Model: Application in Different Station in the Litani River, Lebanon

  • Alya ATOUI Univ Paris Est Creteil, France
  • Abir El Haj Universite de Poitiers, France
  • Yousri Slaoui Universite de Poitiers, France
  • Ali Fadel CNRS - Universite de Tours, France
  • Kamal Slim National Council for Scientific Research (CNRS), Beirut, Lebanon
  • Samir Abbad Andaloussi Univ Paris Est Creteil, France
  • Régis Moilleron Univ Paris Est Creteil,France
  • Zaher KHRAIBANI Lebanese University, Nabatieh, Lebanon
Keywords: Litani River, Water quality, Physicochemical parameters, Stochastic Block Model, Clustering, Lebanon.

Abstract

Water pollution is a major global environmental problem. In Lebanon, water pollution threatens public health and biological diversity. In this work, a non-classical classification method was used to assess water pollution in a Mediterranean River. A clustering proposal method based on the stochastic block model (SBM) was used as an application on physicochemical parameters in three stations of the Litani River to regroup these parameters in different clusters and identify the evolution of the physicochemical parameters between the stations. Results showed that the used method gave advanced findings on the distribution of parameters between inter and intra stations. This was achieved by calculating the estimated connection matrices between the obtained clusters and the probability vector of belonging of the physicochemical parameters to each cluster in the different stations. In each of the three stations, the same two clusters were obtained, the difference between them was in the estimated connection matrices and the estimated cluster membership vectors. The power of SBM proposed methods is demonstrated in simulation studies and a new real application to the sampling physicochemical parameters in Litani River. First, we compare the proposed method to the classical principal component analysis (PCA) method then to the Hierarchical and the K-means clustering methods. Results showed that these classical methods gave the same two clusters as the proposed method. However, unlike the proposed SBM method, classical approaches are not able to show the blocks structure of the three stations.

References

Hayek, A. , Tabaja, N. , Andaloussi, S. , Toufaily, J. , Garnie-Zarli, E. , Toufaili, A. and Hamieh, T. Evaluation of the Physicochemical Properties of the Waters on the Litani River Station Qaraoun, American Journal of Analytical Chemistry, 11 (2020) 90-103. doi: 10.4236/ajac.2020.112007.

Airoldi, E., Blei, D., Fienberg, S., and Xing, E. Mixed membership stochastic block models. The Journal of Machine Learning Research, 9 (2008) 1981-2014.

Anderson, C. J., Wasserman, S., and Faust, K. Building stochastic block models, Social Networks, 14 (1992) 137-161.

Barbillon, P., Donnet, S., Lazega, E., and BarHen, A. Stochastic block models for multiplex networks: an application to a multilevel network of researchers, Journal of the Royal Statistical Society: Series A (Statistics in Society), 180 (2017) 295-314.

Blei, D. M., Ng, A. Y., and Jordan, M. I. Latent Dirichlet allocation, The Journal of Machine Learning Research, 3 (2003) 993-1022.

Bradley, P.S., Fayyad, U.M., Reina, C. Scaling clustering algorithms to large databases. In: KDD’98 proceedings of the fourth international conference on knowledge discovery and data mining, (1998) 9–15.

Celisse, A., Daudin, J. J., and Pierre, L. Consistency of maximum-likelihood and variational estimators in the stochastic block model, Electronic Journal of Statistics, 6 (2012) 1847-1899.

Chaden M.H., Nada N, Sadek A, Bachar K., Mohamad F., Ali Y., Joumana T., Frederic V.,Tayssir H. Water Quality of the upper Litani River Basin, Lebanon, Physics Procedia, 55 (2014) 279-284.

Daudin, J., Picard, F., and Robin, S. A mixture model for random graph, Statistics and Computing, 18 (2008) 1-36.

Dudgeon, Angela HA, Gessner MO, Kawabata ZI, Knowler DJ, Levêque C, Naiman RJ, Prieur RAH, Soto D, Stiassny MLJ, Sullivan CA. Freshwater biodiversity: importance, threats, status, and conservation challenges. Biological Reviews of the Cambridge Philosophical Society, 81(2) (2006) 163-182. doi: 10.1017/s1464793105006950.

El Haj, A., Slaoui, Y., Louis, P-Y., and Khraibani, Z. Estimation in a Binomial Stochastic Block model for a Weighted Graph by a Variational Expectation Maximization Algorithm, Communication in Statistics Simulation and Computation, (2020) doi:10.1080/03610918.2020.1743858.

Fadel, A., Lemaire, B. J., Atoui, A., Vinçon‐Leite, B., Amacha, N., Slim, K., & Tassin, B. First assessment of the ecological status of Karaoun reservoir, Lebanon. Lakes & Reservoirs, Research & Management, 19(2) (2014) 142-157.

Fadel, A., Sharaf, N., Siblini, M., Slim, K., & Kobaissi, A. A simple modeling approach to simulate the effect of different climate scenarios on toxic cyanobacterial bloom in a eutrophic reservoir, Ecohydrology & Hydrobiology, 19(3) (2019) 359-369.

Fadel, A., Atoui, A., Lemaire, B. J., Vinçon-Leite, B., & Slim, K. Environmental factors associated with phytoplankton succession in a Mediterranean reservoir with a highly fluctuating water level. Environmental Monitoring and Assessment, 187(10) (2015) 633.

Fortunato, S. Community detection in graphs, Physics Reports, 486 (2010) 75-174.

Groppo, JD., De Moraes JM., Beduschi, CE., Genovez, AM. and Martinelli, LA. Trend analysis of water quality in some rivers with different degrees of development within the São Paulo State, Brazil. River Res. Applic., 24(8) (2008) doi: 1056-1067. doi:10.1002/rra.1091.

Esterby, S.R. Trend analysis methods for environmental data, Environmetrics, 4 (1993) 459-481, doi:10.1002/env.3170040407.

Holland, P. W., Laskey, K. B., and Leinhardt S. Stochastic blockmodels: First steps, Social Networks, 5 (1983) 109-137.

I. Jolliffe. Principal Component Analysis. Springer-Verlag, New York, 1986.

Hubert, L. and Arabie, P. Comparing partitions, Journal of Classification, 2 (1985) 193-218.

Latouche, P., Birmel'e, E., and Ambroise, C. Overlapping stochastic block models with application to the French political blogosphere, Annals of Applied Statistics, 5 (2011) 309-336.

Latouche, P., Birmel'e, E., and Ambroise, C. Variational Bayesian inference and complexity control for stochastic block models, Statistical Modelling, 12, (2012) 93-115.

Mariadassou, M., Robin, S., and Vacher, C. Uncovering latent structure in valued graphs: a variational approach. Annals of Applied Statistics, 4 (2010) 715-742.

Mark S., Lucy S., Nabil A. Physicochemical Evaluation of Upper Litani River Watershed, Lebanon, Scientific World Journal, (2012) 1-8.

Matias, C., and Miele, V. Statistical clustering of temporal networks through a dynamic stochastic block model, Journal of the Royal Statistical Society: Series B (Statistical Methodology), 79 (2017) 1119-1141.

Nowicki, K., and Snijders, T. A. B. Estimation and prediction for stochastic block structure, Journal of the American Statistical Association, 96 (2001) 1077-1087.

Simpi B., Hiremath SM., Murthy KNS. Analysis of Water Quality Using physico-Chemical Parameters Hosahalli Tank in Shimoga District, Karnataka, India, Global Journal of Science Frontier Research, 3 (2011) 31-34.

Snijders, T. A., and Nowicki, K. Estimation and prediction for stochastic blockmodels for graphs with latent block structure, Journal of Classification, 14 (1997) 75-100.

Zreik, R., Latouche, P., and Bouveyron, C. The dynamic random subgraph model for the clustering of evolving networks, Computational Statistics, 32 (2017) 501-533.

Published
2022-07-10
How to Cite
ATOUI, A., El Haj, A., Slaoui, Y., Fadel, A., Slim, K., Andaloussi, S. A., Moilleron, R., & KHRAIBANI, Z. (2022). Spatial Assessment of Water River Pollution Using the Stochastic Block Model: Application in Different Station in the Litani River, Lebanon. Statistics, Optimization & Information Computing, 10(4), 1204-1221. https://doi.org/10.19139/soic-2310-5070-1547
Section
Research Articles