STOCHASTIC MODELING OF KODAR RESERVOIR INFLOWS FOR OPTIMAL WATER RELEASE AND CROP PLANNING IN RIGHT BANK CANAL COMMAND
Loading...
Files
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Indira Gandhi Krishi Vishwavidhyalaya, Raipur
Abstract
The present study entitled “Stochastic Modeling of Kodar Reservoir Inflows for Optimal water Release and Crop Planning in Right Bank Canal Command” was undertaken considering the stochastic nature of inflow to Kodar river in Mahasamund district of Chhattisgarh.
One of the objectives of the present study is to develop and validate appropriate stochastic model for Kodar river inflows. Daily inflow data of Kodar river for a period of 21 years (1995-2015) were collected and then the monthly and annual inflows were estimated. Various stochastic models were evaluated for predicting river inflow. Statistical analysis of the data showed that, the Kodar river is a monsoonal river having most of its inflow in wet season and less than 5% inflow in the dry season. The turning point test, Kendall’s correlation test and regression test was performed on data for testing of randomness and trend.
The autocorrelation function, partial autocorrelation function, of annual and monthly series were analyzed to identify the class and order of stochastic model to represent the Kodar river inflows. The autocorrelation function and partial autocorrelation function were plotted on the basis of annual and monthly inflow basis. For annual inflows AR model of lower order were found to be suitable and they were further analyzed. For selecting an appropriate model for annual inflows, AR models were tested by various statistical tests and they were compared by various statistical error functions. It was found that the performance of AR (1) model is slightly better than AR(2). However, both the models can be used for predicting annual river inflows of Kodar reservoir. Apart from this, 12 linear nonstationary stochastic models of Autoregressive Integrated Moving Average i.e. ARIMA (p,d,q), having non seasonal parameters, (AR and MA), p=0,1,2, 3,4,5, and 6, q= 1 and 2 respectively, and non seasonal differencing operation d=0, were selected for the present study. The parameters of the identified models were estimated by using SPSS 16.0 software, out of 12 models, all 12 models passed the t- test and out of 12 models, 4 ARIMA models passed the ACF & PACF residual test, the 12 models also passed the L-Jung box statistics tests. Finally, 4 ARIMA models namely ARIMA (4,0,1), ARIMA (5,0,2), ARIMA (6,0,1), ARIMA (6,0,2) were used for validation of model.
The root mean squared error (RMSE), mean relative error (MRE) and maximum absolute error criterion were used for selecting the most appropriate ARIMA model. It is revealed that, ARIMA (5,0,2) model was proved to be an appropriate model among 12 selected models, having lowest RMSE, MRE and MAE, values. Therefore, considering the overall process of model building and comparing the results obtained from forecasting by all the 12 ARIMA models, it can be stated that the Kodar river monthly inflows can at best be forecasted by ARIMA(5,0,2)model which may be used for water resources planning.
Quantification of amount of water delivered in the command area were also estimated by utilizing the canal release data of 12 years of Shahid Veer Narayan Singh Dam (also called Kodar reservoir) and amount of water utilized in the command area estimated by estimating the crop water requirement of the crops grown in the command area with the help of CROPWAT 8.0 software. The area covered under crops other than paddy is very less. Hence, only crop water requirement of paddy was estimated for assessing the amount of water utilized in the command area.It was found that there is a significant difference between the amount of water delivered and utilized in the command area.
The linear programming (LP) model was formulated for arriving at an optimal cropping pattern in Rabi season and reservoir release for the command area. The crop water requirements for the major Kharif and Rabi crops were computed by Penman- Monteith method using CROPWAT 8.0 software and these were used in the formulation of LP model. The model has been developed considering the stochastic nature of inflows in to the storage dam. The linear programming optimization model was solved for two objective functions namely benefit maximization and production maximization under the constraints of reservoir capacity, water requirement of the crops, land availability, etc. The computeroptimization software called TORA 2.0 was used to solve the LP model. The solution of LP model revealed that, wheat is the most profitable crop in the command area. The acreage under wheat crop with objective of benefit maximization was 778ha, which is almost 38% of the net Rabi command area.
In view of above, it can be concluded that linear programming technique is a very useful tool for determining the optimal cropping pattern. Therefore, the present study can be considered as guideline for knowing optimum crop plans and deciding the operational schedule of canal releases as per the results obtained by CROPWAT 8.0 irrigation schedule in the RBC Command area of Kodar reservoir .
Description
STOCHASTIC MODELING OF KODAR RESERVOIR INFLOWS FOR OPTIMAL WATER RELEASE AND CROP PLANNING IN RIGHT BANK CANAL COMMAND
Keywords
null