\documentclass[english,german]{scrartcl}
\usepackage[latin1]{inputenc}
\usepackage{geometry}
\geometry{verbose,a4paper,tmargin=1.75cm,bmargin=2cm,lmargin=2cm,rmargin=2cm,headheight=0.8cm,headsep=0.5cm,footskip=0.5cm}
\usepackage{babel}
\usepackage{url}
\urlstyle{same}
\newcommand\email{\begingroup \urlstyle{same}\Url}\usepackage{graphicx}
\usepackage{units}
\usepackage{tabularx}
\usepackage{wasysym} % enable use of \permil instead of \textperthousand
\usepackage{marvosym}
\usepackage{arabtex}
\usepackage{hyphenat}
\usepackage{booktabs}
\usepackage{mdwlist}
\renewcommand{\thefootnote}{\relax}
\usepackage{index}
\newindex{key}{kdx}{knd}{Keywords}
\newindex{author}{idx}{ind}{Authors}
\usepackage{fancyhdr}
\pagestyle{fancy}
\renewcommand{\headrulewidth}{0pt}
\fancyfoot{}
\usepackage{ifpdf}
\ifpdf
%\usepackage[activate]{pdfcprot}
%\usepackage[owner=Tapioka, print=true, copy=true, edit=false, annotate=false,set]{pdfcrypt}
%encryption: not supported any more (since pdftex 1.10)
\usepackage[pdftex]{thumbpdf}
\usepackage[pdftex]{hyperref}
\hypersetup{pdftitle={Identifying the Most Important Factors Affecting Zinc Concentration of Wheat Grain Using Artificial Neural Network},pdfauthor={Mojtaba Norouzi, Shamsollah Ayoubi, Amir Hossein Khoshgoftarmanesh, Majid Afyuni},pdfsubject={Tropentag 2010: Abstract},pdfkeywords={Artificial neural network modelling, terrain attributes, wheat grain, zinc concentration},pdfpagemode=None,colorlinks=true}
\else
\usepackage{ae}
\usepackage{textcomp}
\fi
\begin{document}
\ifpdf
\parbox[b]{2.5cm}{\resizebox*{2.5cm}{!}{\includegraphics{dtt2010.jpg}}}
\else
\parbox[b]{2.5cm}{\resizebox*{2.5cm}{!}{\includegraphics{dtt2010.eps}}}
\fi
\hfill
\parbox[b]{13.4cm}{\centering \large{\textbf{Tropentag, September 14-16, 2010, Zurich}}\\[1ex] \Large{``World Food System  ---\\A Contribution from Europe''\\[2ex]}}
\hspace{0.1cm}
\begin{center}
\Large{\textbf{Identifying the Most Important Factors Affecting Zinc Concentration of Wheat Grain Using Artificial Neural Network\footnote{\textbf{Contact Address:} Mojtaba Norouzi, Isfahan University of Technology, Department of Soil Science, 8415683111~Isfahan, Iran, \mbox{e-mail}: \email{m_norouzi@ag.iut.ac.ir}}\\[0.8ex]}}
\normalsize{\textsc{Mojtaba Norouzi, Shamsollah Ayoubi, Amir Hossein Khoshgoftarmanesh, Majid Afyuni}}
\end{center}
\begin{itemize*}
\item[]{\small{\textit{Isfahan University of Technology, Department of Soil Science, Iran}}}
\end{itemize*}
\index[author]{Norouzi, Mojtaba}
\index[author]{Ayoubi, Shamsollah}
\index[author]{Khoshgoftarmanesh, Amir Hossein}
\index[author]{Afyuni, Majid}
\begin{center}
\vspace{1.1cm}
\textbf{Abstract}
\begin{abstract}
\normalsize{
In Iran, zinc (Zn) malnutrition is considered to be a serious health problem. Zinc is an essential micronutrient required in small but critical amounts by humans, plants and animals. The Zn concentration of wheat grain is one of the most important nutritional elements in many developing countries. The objectives of this study were to predict the zinc concentration of wheat grain in hilly regions using artificial neural network and to identify the most important topographic attributes affecting the variability of the zinc concentration of wheat grain under rainfed condition in the semiarid regions of western Iran. Wheat yield data were collected from 1m$^{2}$ plots at 100 selected points. The sampling points were chosen in a stratified random manner on the given geomorphic surfaces including summit, shoulder, backslope, footslope, and toeslope at the site. Primary and secondary terrain attributes were calculated using the digital elevation model. The artificial neural network model for grain Zn concentration of wheat in the study area resulted in R$^{2}$ and root mean square error of 0.78 and 0.043, respectively. The plan curvature was identified as the most important topographic attribute influencing the grain Zn concentration. Other important factors for predicting grain Zn concentration of wheat were included mean curvature, wetness index, slope and sediment transport index. Zinc concentration of wheat grain showed less sensitivity to other terrain attributes such as profile curvature, stream power index, elevation, specific catchment area and aspect. Overall, our results indicated that the artificial neural network models could explain \mbox{78\,\%} of the total variability in grain Zn concentration of wheat at the study site. Also, the predictability grain Zn concentration of wheat could be further improved by considering soil properties and management practices followed during the growing season.   



                    



}

\vspace{0.5cm}
\end{abstract}
\end{center}
\noindent \textbf{Keywords:} Artificial neural network modelling, terrain attributes, wheat grain, zinc concentration
\index[key]{Artificial neural network modeling}
\index[key]{Terrain attributes}
\index[key]{Wheat grain}
\index[key]{Zinc concentration}
\end{document}