STOCHASTIC BASED ARIMA MODEL FOR FORECASTING SEDIMENT YIELD FOR KAL RIVER IN KONKAN REGION OF INDIA

K.D. GHARDE¹*, MAHESH KOTHARI², HEMANT MITTAL³, D.M. MAHALE^4
1Department of Soil and Water Conservation Engineering, CAET, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli, 415712, India
²Department of Soil and Water Engineering, CTAE, Maharana Pratap University of Agriculture and Technology, Udaipur, Rajasthan,313001, India
³Department of Soil and Water Engineering, CTAE, Maharana Pratap University of Agriculture and Technology, Udaipur, Rajasthan,313001, India
⁴Department of Soil and Water Conservation Engineering, CAET, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli, 415712, India
* Corresponding Author : ghardek@rediffmail.com

Received : 24-04-2016     Accepted : 21-06-2016     Published : 09-10-2016
Volume : 8     Issue : 45       Pages : 1916 - 1922
Int J Agr Sci 8.45 (2016):1916-1922

Keywords : Runoff, sediments¸ Auto correlation Function, Partial Correlation Function, Akai information criteria
Academic Editor : Dr Pravin Ashok Dahiphale, Srivastava Parul, Harshalkumar Namdevrao Bhange
Conflict of Interest : None declared
Acknowledgements/Funding : None declared
Author Contribution : None declared

Forecasting of sediment yield and runoff is very useful for designing of soil and water conservation structures and planning of watershed activities, design of hydraulic structures, etc. Stochastic based model are found to be very reliable in estimating and forecasting sediment yield. In present study, sediment yield on daily basis of Kal river is tributary of Savitri basin were forecasted using seasonal ARIMA model developed in SPSS software following initial analysis of time series, identification model, apply diagnostic checks, adopting a model and forecast the runoff by adopted model with checking the statistical indices such R, RMSE, CE, EV, MAD, MAPE. The sediment yield data for duration of 2003 to 2009 (7 years) measured at Birwadi station were used to calibrated, validate and forecast the sediment yield by the ARIMA model. It is observed that, the data used in ARIMA model is stationary, no trends and no seasonality found in data sets on daily basis. The two models were identified as ARIMA (111,111)31 and ARIMA(101,000)31 as best model for forecasting the sediment yield with R value more than 0.95 during calibration and forecasting period. The other statistical indices RMSE, CE, EV, MAD, MAPE were also found in appropriate range to forecast the sediment yield with goodness of fit. Hence, developed seasonal ARIMA model could be adopt for forecast the sediment yield on daily basis for sub tropical coastal region of Maharashtra by adopting the estimated parameters.