Modeling the monthly flows of AL-SIN Spring in the Syrian Coast by using the time series analysis


Abstract in English

The study and design of water-intakes on springs is based on the analysis of time series of historical measurements to achieve prediction of incoming water volumes or future expected. The research aims to model the monthly water flows of AL-SIN Spring in Syrian Coast and future expectations of these flows, by adopting the Box-Jenkins models to analyze the time series data, due to its reliable accuracy. Monthly water flows, thus, monthly volumes, for 101 month (from June 2008 to October 2016) were processed. Performing the stability of the time series on variance and median and non-seasonality and making the wanted tests on model residuals, we found that the best model to represent the data is SARIMA(2,0,1) (2,1,0)12 , and after dividing the data into 81 month to build the model and 20 month to test it. Depending on the smallest of weighted mean of criteria RMSE, MAP, MAE,. The best predicted model was SARIMA (3,1,0) (1,1,0)12 and the model gave the nearest predicted values to actually measured data in spring.

References used

Springflow Simulator Manual Vers. 1.0.2, 2007. ACSAD-BGR TECHNICALCOOPERATION PROJECT - NO.: 2004.2032.3, Management, Protection and Sustainable Use of Groundwater and Soil Resources
NELSON, G. R., Applied Time Series Analysis For Managerial Forecasting, Holden- Day, Inc. 1973, 78-91
Box, G. M. P. and PIERCE, D. A., Distribution of Residual Autocorrelation in Autoregressive Integrated Moving Average Time Series Models, John Wiley & Sons. 1970, 115-132

Download