Prediction of Dissolved Oxygen in Bang Pakong River by Regression Analysis

IJEP 42(10): 1268-1271 : Vol. 42 Issue. 10 (October 2022)

Jatupat Mekparyup1,2 and Kidakan Saithanu1,2*

1. Burapha University, Department of Mathematics, Faculty of Science, Chonburi, Thailand
2. Centre of Excellence in Mathematics, Commission on Higher Education, Ratchathewi, Bangkok, Thailand

Abstract

Dissolved oxygen (DO) values in Bang Pakong river located in the east of Thailand were determined by multiple regression analysis. The collected data was split into 2 sets to cogitate which were training and validation sets. The stations of water quality monitoring for each set were breakdown into 4 periods: January-March, April-June, July-September and October-December. The prediction of DO in each period using data in the first set were remarkably achieved as 4 equations with the subsequent regression standard errors of 0.7342, 0.4349, 0.6436 and 0.4319 as well as the adjusted coefficients of determination of 0.3760, 0.8670, 0.2480 and 0.7320, respectively. The prediction accuracy was definitely measured by 4 performance indexes: mean absolute percentage error, mean absolute error, root mean squared error and mean squared error in each period (1st period: 0.1416, 11.0695, 1.5781, 2.4905; 2nd period: 0.2882, 9.7404, 1.3987, 1.9562; 3rd period: 0.2933, 8.6648, 1.3806, 1.9060; 4th period: 0.2454, 6.3330, 0.8890, 0.7903), accordingly applying data in the second set. 

Keywords

Water quality index, Dissolved oxygen

References

  1. Water Quality Management Division, Pollution Control Department, Ministry of Natural Resources and Environment. 2021. Available at : http://iwis.pod. go.th/index.php?method=water quality & etc= 1550340427183.
  2. Altinkaynak, A., M. Ozger and M. Cakmakci. 2005. Fuzzy logic modelling of the dissolved oxygen fluctuations in Golden Horn. Ecol. Modelling. 189(3-4):436-446.
  3. Costa, M. and A. M. Goncalves. 2011. Clustering and forecasting of dissolved oxygen concentrations on a river basin. Stochastic Env. Res. Risk Assess., 25(2):151-153.
  4. Antanasijevic, D., et al. 2013. Modelling of dissol-ved oxygen content using artificial neural networks: Danube river, North Serbia, a case study. Env. Sci. Poll. Res., 20(12):9006-9013.
  5. Abyanesh, H.Z. 2014. Evaluation of multivariate linear regression and artificial neural networks in prediction of water quality parameters. J. Env. Health Sci. Eng., 12(1):40.
  6. Nemati, S., et al. 2015. Estimation of dissolved oxygen using data-driven techniques in the Tai Po river, Hong Kong. Env. Earth Sci., 74(5):4065-4073.
  7. Ross, A.C. and C.A. Stock. 2019. An assessment of the predictability of column minimum dissolved oxygen concentrations in Chesapeake Bay using a machine learning model. Estuarine Coastal Shelf Sci., 221:53-65.
  8. Saber, A., D.E. James and D.F. Hayes. 2020. Long-term forecast of water temperature and dissolved oxygen profiles in deep lakes using artificial neural networks conjugated with wavelet transform. Limol. Oceanogr., 65(6):1297-1317.
  9. Anderson, T.W. and D.A. Darling. 1952. Asymptotic theory of certain ‘goodness of fit’ criteria based on stochastic processes. Annals Math. Statistics. 23(2):193-212.
  10. Durbin, J. and G.S. Watson. 1950. Testing for serial correlation in least squares regression. Biome-trika. 37(3-4):409-428.
  11. Breusch, T.S. and A.R. Pagan. 1979. A simple test for heteroskedasticity and randsom coefficient variation. Economatrica. 47(5):1287-1294.