Artificial Neural Networks (ANNs) Application for the Occurrence Date Prediction of Each Phenological Stage in Wheat Using Climatic Data

Prediction of the time of crop phenological stages occurrence mainly strategic plants such as, wheat, corn and rice help us to achieve to the exact time in order to control pests, weeds and pathogens and also, to the best time for operating such as, fertilizing and irrigation. The main purpose of such study is to estimate the occurrence of phenological stages in dry farming wheat at the time interval of short duration before their occurrence using meteorological data. Recently, the application of Artificial Neural Networks (ANNs) has developed into a powerful tool that can compute most complicated equations and numerical analyses to the best approximation. According to the available data and information from different areas in Iran, this research was accomplished using Sararood station data in Kermanshah Province which has the most complete homogeneous statistics. In this study, the results of climatology for four meteorological factors in period (1990-99) including degree days (heat units), total daily rainfall, sum of sun hours and sum of water requirement for each of eleven phenological stages in wheat including 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: A matrix whose rows are repetitions of the statistical years (i) at each phenological stages of wheat (j) and the columns are meteorological factors (k). A matrix whose rows are each of the statistical years (i) and the columns are meteorological factors (k) at each phenological stage (j). In fact, statistical years (i), phenological stages (j) and meteorological factors (k) are the basic elements of 3-D matrix (M ijk) arranged as above. Finally, different networks were made for each stage and the optimum values of network parameters were obtained by trial and error. It should be reminded that two of the eight-year farming were randomly excluded network training computations and the comparison of the estimated data with the real data for these two years were used to test the accuracy of the models. The model which obtained has the following capabilities: 1. Prediction of the date of phenological stages occurrence (From stem formation till full maturity) with maximum error 3 to 6 days at least five days before the occurrence of each stage. 2. Determination of the sensitivity of each phenological stage with respect to meteorological factors.