MSc

ABSTRACT
The research work entitled ‘Statistical modelling of climate change in Southern Kerala’ was carried out at College of Agriculture Vellayani during 2018-2020. The objective was to develop statistical models to analyze climate change overtime across different regions in Southern Kerala and to determine its impact on the yield of paddy and cardamom. The weather data corresponds to maximum temperature, minimum temperature and rainfall was collected from four agrometeorological stations under KAU in Southern Kerala viz., College of agriculture, Vellayani, RARS, Kumarakom, CRS, Pampadumpara for a period of 29 years (1991-2019) and from RRS, Moncompu for a period of 22 years (1997-2018). Secondary data on production of Paddy and cardamom was collected from the Report of Agricultural Statistics, GOK for a period of 24 years (1995-2018) and from the Spices Board for a period of 22 years (1997-2018) respectively. The entire analysis was done with the help of Microsoft excel, SPSS, Gretl and R programming software.
Descriptive statistics includes mean, range, standard deviation and coefficient of variation and box plot was used to describe the features of monthly weather parameters over the years. The results from the box plot and descriptive statistics indicated that heavy rainfall was received in July, August 2018 at Kumarakom and Moncompu followed by Pampadmpara and Vellayani. Rainfall patterns in Kumarakom, Moncompu and Idukki showed more deviations from normal rainfall as compared to Vellayani over the study period. Among the four regions the highest annual maximum temperature was observed at Kumarakom and the lowest at Pampadumpara and highest precipitation was observed in Kumarakom followed by Moncompu, Pampadumpara and lowest in Vellayani.
Trend analysis was performed to understand the climate change on the basis of rainfall, maximum and minimum temperature with the help of Mann-Kendall test and Sen’s slope estimator for annual and different seasons in four regions. No significant trend was noticed in annual and different seasons rainfall while a decline in rainfall was noticed for south west and north east monsoon at Vellayani. At the same time a significant increasing trend was observed for maximum temperature in all the seasons at Vellayani. Whereas a significant and decreasing trend in annual and north east monsoon rainfall with a significant increase in north east and south west monsoon maximum temperature and a decrease in winter maximum temperature at Kumarakom is an indication of the adverse change in climate change overtime. Even though no significant trend was noticed in rainfall except for annual and north east monsoon, the slope estimator was negative indicating a decline in rainfall with a non-significant increase in maximum temperature at Moncompu. Pampadumpara in the high range zone also recorded with decrease in rainfall during all the seasons.
The magnitude of trends in rainfall and maximum and minimum temperature of different stations provides a more precise picture about the change in weather variables. The magnitude of the negative trend for the annual rainfall was highest at Moncompu (-43.5 mmyear-1) and lowest at Vellayani (-0.871 mm year-1). But for maximum temperature a positive trend was obtained in majority of the stations and a significant negative trend was seen at Pampadumpara. The magnitude of the positive slopes of the annual maximum temperature was highest at Vellayani (0.038 0 Cyear-1) and lowest at Moncompu (0.004 0 C year-1). Pampadumpara had recorded with a highest positive slope of 0.121 0 C year-1 for minimum temperature followed by 0.024 0 Cyear-1 at Moncompu. In general a decrease in annual rainfall and an increase in temperature were observed in Southern Kerala over the years.
Seasonal ARIMA models were used to model rainfall data of the three stations. Rainfall data in the level form was stationary and a prominent seasonality was found out which indicated that the order of integrating factor was 0 for non-seasonal component and 1 for seasonal component. Using trial and error method the best model among the randomly chosen models was selected on the basis of least AIC, BIC and Hannan- Quinn criteria. The best identified SARIMA models for rainfall were respectively ARIMA (1, 0, 0) (0, 1, 1)12 for Vellayani, ARIMA
(0, 0, 0) (0, 1, 1)12 for Kumarakom and ARIMA (0, 0, 0) (0, 1, 1)12 for Pampadumpara to forecast monthly rainfall in these regions.
Multiple regression analysis was performed to analyse the impact of weather parameters on the production of paddy in Kottayam indicated that rainfall during July had a negative significant effect, maximum temperature during August and minimum temperature during June had a positive significant influence on the production of paddy in Viruppu season without considering the influence of other factors of production. While maximum and minimum temperature during October had a positive significant effect and minimum temperature during December and excess rainfall during October and November had a negative significant effect on Puncha paddy yield.
The estimated regression coefficients of rainfall during July to September, maximum temperature during April to June and minimum temperature during January to March and July to September were found to be significant and the negative sign of the parameters indicating the adverse influence weather parameters on the production of cardamom. However, maximum temperature during July to September, minimum temperature during April to June had a positive impact on yield once again suggesting the negative impact of excess rain during this period. Moreover, the coefficient of rainfall in all the quarters had a negative sign indicating the negative effect of excess rainfall on cardamom productivity particularly during the harvesting period.
The results of the study based on trend analysis indicated clear evidence about climate change in terms of weather parameters occurred across different regions in Southern Kerala during the study period. SARIMA model was found to be the best model for prediction of rainfall in the selected regions. The impact analysis of weather parameters on the production concluded that the weather parameters during different months had either a positive or a negative influence on the production of paddy and cardamom.