Spatio-temporal estimation of rainfall patterns in north and northwestern states of India between 1901 and 2015: change point detections and trend assessments

Change point detection and trend analysis are the adopted techniques of time series analysis. We have applied non-parametric methods on the temporal and spatial-scale data of 115 years between 1901 and 2015 from ten different federal states in the north and the northwestern India to examine the change points as well as to estimate the future scenarios by examining the past trends. The change points were examined by Pettitt's test, Standard Normal Homogeneity (SNH) test, and Buishand's test, whereas the trend analyses of monthly, annual, and seasonal rainfall data were carried out using Sen's slope estimator after assessing their statistical significance by Mann-Kendall (M-K) test. The trend analyses showed non-zero slope values and a few among them were of statistical significance. The results of our statistical experiment concluded that the trends of reduction in winter, pre-monsoon, and post-monsoon rainfalls would have notable effects on the rain-fed agricultural production in the near future, particularly in the areas without proper irrigation facility (e.g., parts of Uttar Pradesh). More extreme events of higher rainfall in some states (e.g., Goa, Maharashtra, and Jammu and Kashmir), however might cause disasters like landslide and flooding.