Browsing by Author "Mallikarjuna, G.B."
Now showing 1 - 2 of 2
Results Per Page
Sort Options
ThesisItem Open Access EVALUATION OF STATISTICAL CORRECTIVE METHODS TO MINIMIZE BIAS WITH RESPECT TO OBSERVATORY DATA IN MODELLED CLIMATIC PARAMETERS(UNIVERSITY OF AGRICULTURAL SCIENCES GKVK, BENGALURU, 2017-07-24) SRUTHI, U.; Mallikarjuna, G.B.Climate change refers to any systematic change in the long-term weather parameters. The study aims at reducing the bias in the climatic parameter values obtained from the satellite models by comparing with the data obtained from the observatories. Nine years day wise data for the climatic parameters such as Rainfall, Maximum and Minimum temperature are collected from the AICRP on Agro meteorology UAS GKVK Bengaluru, for the study. The day wise data was converted to, standard meteorological weekly and monthly data for the period wise analysis. Climatic parameter data obtain from the observatories are always more accurate than those from the satellite model. Bias correction methods such as difference method (DM) and modified difference method (MDM) were attempted to minimize bias of the satellite model data compared to observatories data. Best bias correction method is identified based on the coefficient of variation. Results revealed that, among them MDM was better for all the three periods of rainfall. Similarly, MDM is an ideal correction measure for daily maximum and minimum temperature, for weekly and monthly maximum and minimum temperature data DM was found to be an ideal measure. Probability distribution functions were attempted for the climatic factors considered and best fit of them were identified using chi square test. The best fitted probability distribution for the different periods were identified as Gamma and Weibull distribution as most suitable for the rainfall data. For Maximum and minimum temperature, no generalize single model was found as best fit.ThesisItem Open Access TIME SERIES ANALYSIS OF AREA, PRODUCTION AND PRODUCTIVITY OF SELECTED PLANTATION CROPS IN DAKSHINA KANNADA DISTRICT OF KARNATAKA(UNIVERSITY OF AGRICULTURAL SCIENCES GKVK, BENGALURU, 2017-08-01) CHAITHRA, M.; Mallikarjuna, G.B.Plantation crops are high valued commercial crops which are export oriented. In addition to commercial importance these also generate huge employment opportunities. In the present study a prudent attempt was made using data of thirty five years to understand the trend in area, production and productivity of major plantation crops such as arecanut, cashewnut and coconut of Dakshina Kannada. Further, an attempt was also made to forecast the area and production of these crops. The polynomial regression models were fitted to assess the trend in area and production. Based on the model adequacy linear model was the best fit for area and production of arecanut. For the productivity of arecanut none of the fitted models were significant indicating that there was non-significant change. Further, cubic and quartic models were found to be best fit for production and productivity of coconut and cashewnut respectively. Due to the presence of autocorrelation in the data, ARIMA and exponential smoothing methods were used for forecasting. The appropriate ARIMA models were identified after removing the outliers. Using 10 per cent of data as testing set, ARIMA (1,1,1) and ARIMA (0,1,0) were found to be the best fitted model based on RMSE and MAPE values to forecast area and production of arecanut. Whereas, ARIMA (0,1,1) found to be the most suitable model to forecast area and production of coconut. However, Damped trend model and Brown’s linear trend model were the best fitted model for predicting the area and production of cashewnut.