Detection of recognizing events in lower atmospheric temperature time series (1941-2010) of Midnapore Weather Observatory, West Bengal, India

Abstract

In this study, trends of mean monthly surface temperature time series of each month were examined for Midnapore station, West Bengal, India for the period 1941-2010. The data-set has the observations of two variables: the mean monthly maximum and mean monthly minimum temperature. Major changes in the mean monthly temperature trends over the period considered, have been detected using the Sequential version of Mann-Kendall rank statistic. The plots of the values of sequential statistic and backward sequential statistic intersect each other at points each of which indicates a change in the trend of considered data-set. The analyses of the mean monthly temperature signify warming of the winter temperature. Statistically significant (p?0.05) abrupt changes has been detected in 1985, 1992 and 1995 in the temperature time series for February, October and November, respectively.

Keywords: sequential Mann-Kendall test statistic, change point detection, climate change.