Assessment Of Mine Spoil Genesis Influencing Restoration In Chronosequence Iron Mine Overburden Spoil Using Artificial Neural Network

Research Article
Pasayat, M and Patel, A.K
DOI: 
http://dx.doi.org/10.24327/ijrsr.2017.0807.0455
Subject: 
science
KeyWords: 
Artificial neural network, iron mine spoil, mine spoil genesis, QSAR, restoration
Abstract: 

Comparative assessment of soil variables in different iron mine overburden spoil and NF soil endow with valuable information about the pace and progress of mine spoil genesis, which can be implemented for improving mine spoil restoration through sustainable use of resources. About 14 mine spoil variables were selected to develop the QSAR equation based on brute-force approach and genetic function approximation for prediction of mine spoil restoration required for fresh iron mine overburden spoil to reach the soil features of the nearby NF soil. The training and test sets with statistically best fitted with r 2 = 1.0 and r2 LOO = 0.996. The predictive ANN model with 14-11-1 structure was recognized as the best model illustrating the time period required for mine spoil restoration across the sites. The standard error for the proposed model was estimated to be 0.276, which can be used as indicator of the robustness of the fit and suggested that the predicted years for mine spoil restoration based on the model is reliable. The validity of the developed model was confirmed with higher calculated value of squared correlation coefficient determination (r2 = 0.999) and lower root mean square error (RMSE = 0.194 kPa), which revealed good predictability. Hence, IB0 shall take  38.319 years to reach the soil features of nearby NF soil depending on the variability in physico-chemical properties, enzyme activities and fungal PLFA biomarkers as sensitive and reliable indicator influencing mine spoil genesis in different age series iron mine spoil over time.