Dr. P. R. VaishnavSATVINDER KOUR2018-03-212018-03-212017http://krishikosh.egranth.ac.in/handle/1/5810042299A timely and reliable forecast of yield of crop needs little emphasis for monsoon dependent country like India where, the economy is mainly based on agricultural production. Weather is a major factor affecting crop production in advanced agricultural systems. The large variation in yield from year to year and place to place is dominated by the weather parameters. In view of fluctuating weather effects, a timely and reliable forecast of crop productivity could help in deciding the policies. The present study has been taken up to, 1) Explore the possibility of suggesting suitable statistical model for pre-harvest forecasting of productivity of pearl millet for Kheda district of Gujarat, using weather parameters. 2) Compare the efficiency of different statistical models. 3) Assess the impact of weather parameters on pearl millet productivity. To estimate the effect of weather variables and technological advances, productivity data from 1982 to 2013 were collected. The weekly averages of weather variables viz., bright sunshine hours (BSS), maximum temperature (MAXT), minimum temperature (MINT), morning relative humidity (RH1), afternoon relative humidity (RH2) and weekly total rainfall (RF) from 26th to 38th standard meteorological week for kharif and from 5th to 16th standard meteorological week for summer pearl millet of the respective year were considered in the study. The week-wise approach was used with original weather variables, whereas generated variables with correlation coefficient and week number as weight were also tried. Time trend was also included as independent variable in all the approaches. To provide early forecasts different week intervals (12 to 13) for kharif and (11 to 12) for summer pearl millet were considered. Three sets of multiple linear regression equations consisting of 27, 28 and 29 years data for each model were fitted. The effects of all the weather variables in relation to their quantum and direction differed in models over the approaches. However, they were found important for prediction point of view in the pearl millet productivity. Week-wise approach using original weather variables were found suitable for kharif and summer pearl millet models, while correlation coefficient as weight with generated weather variables was found suitable for summer pearl millet model only. These approaches provided pre-harvest forecasting models of Kheda district, one week before expected harvest for kharif pearl millet and 2 weeks before expected harvest for summer pearl millet. The selected models for pre-harvest forecast explained more than 50 per cent of total variation in pearl millet productivity with simulated forecast error below 15 per cent. The best fitted pre-harvest forecast model based on week-wise approach using original weather variables are Y = 5048.35 - 98.60 X3W29 + 46.81 X3W30 – 26.53 X5W35 – 39.60 X1W34 (R2 = 50.40%) for kharif and Y = -495.03+278.60 X1W7 + 19.18 X5W8 – 13.90 X5W15 – 40.41 X3W12 (R2 = 70.00%) for summer pearl millet. Model based on correlation coefficient as weight using generated weather variables for summer pearl millet is Y = -264.02 + 3.77 Z131 + 49.03 Z61 + 8.61 T– 1.60 Z361 (R2 = 70.00 %) The comparison between selected regression models and ANN model on the basis of RMSE and R2 value showed that the ANN model was superior as compared to regression models for kharif as well as summer season pearl millet crop. Based on over all comparison ANN model is recommended for pre-harvest forecasting of pearl millet productivity in kheda district of middle Gujarat. The perusal of the results indicated that effect of the time trend was not observed in week-wise approach (MLR model) on kharif pearl millet productivity. The effect of trend was positive and significant in summer pearl millet productivity. Effect of bright sunshine hours was found unfavorable on kharif pearl millet productivity during 9th week which corresponded to dough stage. Whereas, bright sunshine hours had favorable effect on summer pearl millet productivity during 3rd week which corresponded to fifth leaf stage. The minimum temperature was found to have detrimental effect on kharif pearl millet productivity at 4th week which corresponded to panicle initiation stage and favorable effect at 5th week which corresponded to flag leaf stage. Morning relative humidity had negative influence on kharif pearl millet productivity during 10th week which corresponded to dough stage. Whereas, the morning relative humidity had negative influence on summer pearl millet productivity during 11th week corresponded to physiological maturity stage and afternoon relative humidity had positive influence during 2nd week which corresponded to fifth leaf stage. Further, the results indicated that effect of the afternoon relative humidity was not observed in week-wise approach on kharif pearl millet productivity.enagriculture, statisticsEvaluationThesis