NEURAL NETWORK MODELING FOR WATER TABLE FLUCTUATIONS: A CASE STUDY ON HOSHANGABAD DISTRICT OF MADHYA PRADESH
Journal Title: International Journal of Agriculture Sciences - Year 2016, Vol 8, Issue 60
Abstract
The accurate prediction of groundwater level is quite essential for ecological and sustainable development and management of groundwater resources. In this study artificial neural networks, namely multilayer perceptron (MLP) was used for predicting water tables in a selected aquifer system. The inputs for the ANN models consisted of rainfall, temperature and river stage and water table data. The ANN model was trained using gradient descent with momentum (GDM) algorithm. The predictive ability of ANN model was developed for each of the seven sites was evaluated using four statistical indicators (bias, RMSE, NSE and MSE) as well as visual examinations. Based on the results of this study, the neural network model was found to be efficient in predicting monthly water tables at almost all the sites. The study concluded that the neural network techniques can be efficiently used for predicting water table fluctuations, particularly in data-scarce conditions
Authors and Affiliations
SOURABH NEMA
Keywords
Related Articles
- EP ID EP171533
- DOI -
- Views 89
- Downloads 0
MANAGERIAL EFFICIENCY OF COCONUT PLANTATION GROWERS IN COASTAL AREA OF SAURASHTRA REGION
The study was conducted in the coastal areas of Saurashtra where coconut is a major crop. Total five talukas i.e. 3 talukas of Junagadh, 1 taluka of Porbandar and 1 taluka of Bhavnagar district were selected purposively,...
ASSESSMENT OF KNOWLEDGE LEVEL AND GAP IN POTATO CULTIVATION PRACTICES AT HILLY AREAS OF CHHATTISGARH
The present study was conducted in Mainpath block of Surguja District of Chhattisgarh. The study revealed that, majority of respondents (89.16%) were having full level of knowledge about land preparation, 70.83 per cent...
DETERMINANT OF RESISTANCE IN BREAD WHEAT GENOTYPES AT DIFFERENT DATE OF SOWING TO SPOT BLOTCH DISEASE
Present investigation was carried out to evaluate the eight varieties of wheat against spot blotch disease and their performance under different date of sowing. Among the three different sowing date the plant height of a...
STOCHASTIC BASED ARIMA MODEL FOR FORECASTING SEDIMENT YIELD FOR KAL RIVER IN KONKAN REGION OF INDIA
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...
EFFECTS OF PARENTS ON PER SEPERFORMANCE OF SINGLE CROSS HYBRIDS FOR GRAIN YIELD IN MAIZE (Zea mays L.)
An investigation was carried out to assess the combining ability and nature of gene action in respect of grain yield and its attributing traits in 87 single cross hybrids of maize developed by crossing three testers with...