Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 78
PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON THE APPLICATION OF ARTIFICIAL INTELLIGENCE TO CIVIL AND STRUCTURAL ENGINEERING
Edited by: B.H.V. Topping
Paper 56

Prediction of Phytotoxicity of Metal Uptake in Plants using Artificial Neural Networks

M.A. Gharaibeh+ and K.A. Bani-Hani*

+Department of Natural Resources and The Environment
*Department of Civil Engineering
Jordan University of Science and Technology, Irbid, Jordan

Full Bibliographic Reference for this paper
M.A. Gharaibeh, K.A. Bani-Hani, "Prediction of Phytotoxicity of Metal Uptake in Plants using Artificial Neural Networks", in B.H.V. Topping, (Editor), "Proceedings of the Seventh International Conference on the Application of Artificial Intelligence to Civil and Structural Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 56, 2003. doi:10.4203/ccp.78.56
Keywords: phytotoxicity, plant, neural network, wastewater, selenium, irrigation, heavy metals.

Summary
With the advances in neural networks technology, increased interest have been shown and applied to solve different problems in agriculture. Neural networks have shown a superior performance over the statistical models when the models are complex and highly nonlinear. Neural network enjoy some exclusive humanlike capabilities in information processing. Probably, learning from examples is the most important capability in information processing. Neural networks are capable of learning complex phenomena and predicting the behaviour of the phenomena. Therefore, they are characterized by their adaptability, noise immunity, and generalization ability. All of these advantages are exactly what is needed for solving some of the practical agricultural problems.

Wastewater is a preferred unconventional water source, since the supply is increasing because of population growth, there is enhanced awareness of environmental quality, and its costs are relatively low. It can serve as a source of both water and nutrients for agricultural uses, thus also reducing fertilization costs. Benefits of agricultural reuse of wastewater are expressed when agricultural production is maintained while water sources and environmental qualities are preserved. At the same time, wastewater irrigation may be hazardous to the environment, since the influent contains pollutants such as macro- and micro- organic and inorganic matter. Moreover, recycled wastewater effluent is potentially an important resource of irrigation water in arid and semi-arid regions. However, irrigation water quality is one of the main factors limiting plant growth in these areas. Effluent is mainly comprised of water (99.9%) together with relatively small concentration of suspended and dissolved solid, both organic and inorganic (e.g. Cl, Na, B, and selected heavy metals). It follows then that the major concern in application of wastewater to agriculture land is the effect of these metals on soil properties and plant growth especially after several years of application.

In fact, the limiting factor in application of wastewater to agricultural lands is the excessive accumulation of heavy and trace elements such as Zn, Cu, Ni, Cd, and others in soil, and the resultant phytotoxicity due to metal uptake by plants [1,2]. Therefore, addition of wastewater may increase the accumulation of many elements in soil.

Green house experiment was conducted to test the ability of three different crops to accumulate significant tissue concentration (uptake) of selected heavy metals. Simulated wastewater was added in different levels in addition to different nitrate levels. The experiment was conducted with sandy soil to test the ability of these plants without any competitions with the growing medium. Plant tissues accumulated high concentration of added heavy metals. All these experiments were employed to develop and train an artificial neural networks model that is capable to predict the resultant phytotoxicity, accumulations and dry masses due to wastewater irrigation. The neural network models were intelligent to capture the effect of such uptakes to the resultant phytotoxicity. Moreover, the neural networks showed great ability to generalize and to predict the phytotoxicity to different solvents percentage. The neural networks are tested and verified experimentally and results showed robustness and accuracy even when the data are trimmed or noisy.

References
1
Chaney, Rufus L. 1973. Recycling Municipal Sludges and Effluents. P: 129-141.
2
Page, et.al., Methods of soil analysis, Part 2: (2nd Edition). Am. Soc. Agron., Madison, Wisc., USA. Madison.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £82 +P&P)