Risk-based water quality decision-making under small data using Bayesian network

Abstract

A knowledge-based network for Section Yidong Bridge, Dongyang River, one tributary of Qiantang River, Zhejiang Province, China, is established in order to model water quality in areas under small data. Then, based on normal transformation of variables with routine monitoring data and normal assumption of variables without routine monitoring data, a conditional linear Gaussian Bayesian network is constructed. A “two-constraint selection” procedure is proposed to estimate potential parameter values under small data. Among all potential parameter values, the ones that are most probable are selected as the “representatives”. Finally, the risks of pollutant concentration exceeding national water quality standards are calculated and pollution reduction decisions for decision-making reference are proposed. The final results show that conditional linear Gaussian Bayesian network and “two-constraint selection” procedure are very useful in evaluating risks when there is limited data and can help managers to make sound decisions under small data.

References

[1] GanoulisJ. G.Engineering risk analysis of water pollution [M], 2005BeijingTsinghua University Press1-10

[2] WottonC. L., LenceB. J.. Water quality management of ammonia under uncertainty [C]. DOMENICA M F. Integrated Water Resources Planning for the 21st Century, 1995CambridgeAmerican Society of Civil Engineers456-459

[3] VernulaV. R. S., MujumdarP. P., GhoshS.. Risk evaluation in water quality management of a river system [J]. J. Water Resources Planning & Management, 2004, 130: 411-423

[4] ChenC. F., MaH.-w., ReckhowK. H.. Assessment of water quality management with a systematic qualitative uncertainty analysis [J]. Science of the Total Environment, 2007, 374: 13-15

[5] NieX. H., HuangG. H., WangD., LiH. L.. Robust optimisation for inexact water quality management under uncertainty [J]. Civil Engineering and Environmental Systems, 2008, 25(2): 167-184

[6] ReckhowK.. Water quality prediction and probability network models [J]. Canadian Journal of Fisheries and Aquatic Sciences, 1999, 56(7): 1150-1158

[7] PortieljeR., Hvited-JacobsenT., Schaarup-JensenK.. Risk analysis using stochastic reliability methods applied to two cases of deterministic water quality models [J]. Water Resource, 2000, 341: 153-170

[8] McintyreN. R., WagenerT., WheaterH. S., ChapraS. C.. Risk-based modelling of surface water quality: A case study of the Charles River, Massachusetts [J]. Journal of Hydrology, 2003, 27(1/2/3/4): 225-247

[9] XueL.-q., WangJ., HuangS., WuC.-lin.. Stochastic modelling of river water quality and risk analysis [C]. The 2nd Conference on Environmental Science and Information Application Technology, 2010WuhanInstitute of Electrical and Electronics Engineers186-190

[10] HU Guo-hua, XIA J. Grey-stochastic risk method for risk analysis [J]. Journal of Hydraulic Engineering, 2001 (4): 1–6. (in Chinese)

[11] HeL., ZengG.-ming.. Study on Grey water quality in water environment [J]. Chongqing Environmental Science, 2002, 24(1): 55-57

[12] DarwicheA.Modelling and reasoning with Bayesian networks [M], 2009New YorkCambridge UP20-32

[13] CharniakE.. Bayesian networks without tears [J]. AI Magazine, 1991, 124: 50-63

[14] HeckermanD., MamdaniA., WellmanM. P.. Real world applications of Bayesian networks [J]. Communications of the ACM, 1995, 38(3): 24-26

[15] JensenF. V.An introduction to Bayesian networks [M], 1996LondonUCL Press12-15

[16] NiedermayerD.HolmesD. E., JainL. C.. An introduction to Bayesian networks and their contemporary applications [C]. Innovations in Bayesian Networks, 1998BerlinSpringer1-12

[17] ZengY.-h., EJ.-q., ZhuH., GongJ.-ke.. Fault diagnosis on cooling system of ship diesel engine based on Bayes network classifier [J]. Journal of Central South University of Technology: Science and Technology, 2010, 41(4): 1379-1384

[18] BorsukM., StowC. A., ReckhowC. H.. A Bayesian network of eutrophication models for synthesis, prediction, and uncertainty analysis [J]. Ecological Modelling, 2004, 173: 219-239

[19] SaidA.. The implementation of a Bayesian network for watershed management decision [J]. Water Resources Management, 2006, 20: 591-605

[20] HenriksenH. J., BarlebH. C.. Reflections on the use of Bayesian belief networks for adaptive management [J]. Journal of Environmental Management, 2007, 88(4): 1025-1036

[21] PollinoC., WoodberryO., NicholsonA., KorbK., HartB. T.. Parameterization and evaluation of a Bayesian network for use in an ecological risk assessment [J]. Environmental Modelling & Software, 2007, 22: 1140-1152

[22] FarmaniR., HenriksonH. C., SavicD.. An evolutionary Bayesian belief network methodology for optimum management of groundwater contamination [J]. Environmental Modelling & Software, 2009, 24: 203-310

[23] CowellR. G., LauritzenP., SpiegelhalterS. L.Probabilistic networks and expert systems [M], 1999New YorkSpringer Press25-37

[24] Water Resources Department of Zhejiang Province. Partition plan of water function regions and water environmental function regions in Zhejiang Province [R]. 2005. (in Chinese)

[25] XiaQ., ZhongY.-q., LiuX.-b.Water quality benchmarks and quality standards [M], 2004BeijingChina Standards Publishing House1-13

[26] XuY.-ping., ZhangQ.-q., NiuS.-f., LouZ.-hua.. Water quality analysis and assessment of the main stream of Dongyang river [J]. Shanghai Environmental Science, 2009, 28(5): 190-195

[27] SakiaR. M.. The box-cox transformation technique: A review [J]. The Statistician, 1992, 41: 169-178

[28] ShachterR. D., KenleyC. R.. Gaussian influence diagrams [J]. Management Science, 1989, 35(5): 527-550