28.01.2015 Views

Kouli_etal_2008_Groundwater modelling_BOOK.pdf - Pantelis ...

Kouli_etal_2008_Groundwater modelling_BOOK.pdf - Pantelis ...

Kouli_etal_2008_Groundwater modelling_BOOK.pdf - Pantelis ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

PREFACE<br />

<strong>Groundwater</strong> (GW) is one of the most valuable natural resources and for that reason, the<br />

GW protection and management is vital for human evolution, socio-economic development<br />

and ecological diversity. During the last decades, the continuously increasing need of water<br />

has led to a rapidly growing awareness in the field of GW management. At the same time<br />

over exploitation and pollution of water resources are threatening the ecosystems. The<br />

combination of these two problems which have acquired worldwide dimensions has forced<br />

many scientists working in relative fields to search new, multidisciplinary approaches to<br />

address them. Effective management and protection of groundwater resources require detail<br />

knowledge and quantitative/qualitative characterization of aquifers. Thus, modeling and<br />

planning of the GW through the use of modern technologies and approaches have become of<br />

high priority towards this direction. This book provides leading-edge research on this field.<br />

Artificial neural networks are empirical mathematical tools proven to represent complex<br />

relationships of hydrological systems. Neural networks are increasingly being applied in<br />

subsurface modeling where intricate physical processes and lack of detailed field data prevail.<br />

Two types of ANN models: Back propagation neural network (BPNN) and radial basis<br />

function neural network (RBFN) are examined to predict the pesticide contamination of<br />

domestic wells. Because sample collection, analysis, and re-sampling are expensive, a large<br />

dataset is not available for ANN use in this study. This Short Communication presents<br />

analyzes of raw data for preparation of input subsets for ANN use. Thus, a clustering<br />

technique is used to divide the whole dataset into three subsets: training, validating, and<br />

testing. The sensitivity analysis was carried out by deleting one or more input variables from<br />

the input data set to measure the importance of one variable over the other in terms of ANN<br />

prediction performance. It provides a sense of the effect of each parameter on pesticide<br />

occurrence in a well. The well depth, depth to aquifer material from land surface, and on-site<br />

pesticide storage are found to be important parameters in pesticide occurrence in well.<br />

<strong>Groundwater</strong> (GW) is one of the most valuable natural resources and for that reason, the<br />

GW protection and management is vital for human evolution, socio-economic development<br />

and ecological diversity. During the last decades, the continuously increasing need of water<br />

has led to a rapidly growing awareness in the field of GW management. At the same time<br />

over exploitation and pollution of water resources are threatening the ecosystems. The<br />

combination of these two problems which have acquired worldwide dimensions has forced<br />

many scientists working in relative fields to search new, multidisciplinary approaches to<br />

address them. Effective management and protection of groundwater resources require detail

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!