| Popis: |
This paper presents trend analysis and change point detection of time-series data of climatic rainfall and temperature data for more than one century (Year 1901–2017) for Jodhpur, Rajasthan, India. For detailed analysis, non-parametric test Mann-Kendall trend test (MK-test), Modified Mann-Kendall trend test (Modified MK test), and Modified Mann-Kendall trend test with pre-whitening (Modified MK-test PWMK), have been carried out for trend detection. In addition, Sen's Slope test has been performed to analyzed the magnitude of trend, whereas, Pettitt's test has been carried out to assess change point detection. Results indicated that Z-test of Mann-Kendall has shown overall positive trend in annual and all seasonal except winter season with respect to both the data i.e. Indian Meteorological Department (IMD) and the Climate Research Unit (CRU) data. Temperature analysis trends have shown a positive value for Z in yearly and all seasons, which indicated that there has been a general tendency toward rising temperatures. For annual and seasonal rainfall time series analysis, no shift has been observed; however, climatic variables have shifted for the past few years as a result of global changes in temperature and rainfall. Results revealed that trend assessment and shift point detection (annual and seasonal) rainfall was not more significantly changed at the 10% significant level. Three non-parametric statistical tests have shown that air temperature has increased at 10% significant level annually and on seasonal basis. Analysis of this study may be helpful to show the relationship among different parameters and determine the prediction of climatic data variables. |
