International Journal of Agricultural Policy and Research
Vol.2 (10), pp. 352-361, October 2014
Article 13/ID/ JPR135, 11 pages
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 3.0 International License.
Original Research Article
Artificial Neural Networks (ANNs) application to predict occurrence of phenological stages in wheat using climatic data
B. Safa1*, A.Khalili2, M. Teshnehlab3 and A. Liaghat4
1Ph.D Candidate in Agricultural Meteorology Department of Agronomy and Horticulture University of Nebraska-Lincoln, USA.
2Department of Irrigation and Reclamation Engineering College of Agriculture, University of Tehran, Karaj, Iran.
3Department of ElectricalEngineering K.N. Toosi Technology University, Tehran, Iran.
4Department of Irrigation and Reclamation Engineering College of Agriculture, University of Tehran, Karaj, Iran.
* Corresponding Author Email: babaksafa(at)hotmail.com
The main purpose of the study was to estimate the occurrence of phenological stages in dry farming wheat at the time interval of short duration before their occurrence using meteorological data. The study was accomplished using Sararood station data in Kermanshah Province (Iran) having the most complete homogeneous statistics. The data of climatology for four meteorological factors during the period between 1990 to 1999 including degree days (heat units), total daily rainfall, sum of sun hours and sum of water requirement for each of the eleven phenological stages in wheat (sowing, germination, emergence, third leaves, tillering, stem formation, heading, flowering, milk maturity, wax maturity and full maturity) were collected separately for each farming year and arranged in two matrices in order to apply Artificial Neural Network for numerical analyses to the best approximation thus: A matrix whose rows were repetitions of the statistical years (i) at each phenological stage of wheat (j) and the columns were meteorological factors (k) were the basic elements of 3-D matrix (M ijk). The obtained model had the capabilities to: 1. Predict the date of occurrence of phenological stages (from stem formation till full maturity) with a maximum error of 3 to 6 days with at least five days before the occurrence of each stage and; 2. Determine the sensitivity of each phenological stage with respect to meteorological factors.
Key words: Artificial neural network, climatic data, wheat phenological stage, crop model