Fluctuations in groundwater level prediction with Bayesian network (case study)

Geohydrology issues ranging from the water table is very important . Therefore, research is necessary to estimate the missing data .In this study nourabad plain the 11 wells was observed that all the annual statistics of the statistical interpolation method ( Krigng and co ) places no statistics are calculated. This study aims to predict the groundwater level fluctuations , using intelligent Bayesian networks and artificial neural networks has been modeled .For this purpose, the latitude and longitude on a monthly time scale as the input and output parameters were selected as the groundwater level fluctuations .Criteria of correlation coefficient , root mean square error and coefficient of Nash suttclif and performance models were used to assess. The results showed that the neural network model has a correlation coefficient ( 0.880 ) , root mean square error ( 0.024m) and the standard Nash suttclif ( 0.900 ) in step verification succeeded with considerable accuracy to estimate water level fluctuations would .