Observed trends in temperature and rainfall in Bangladesh using pre-whitening approach

Understanding the trend of temperature and rainfall is of prime importance for climate change related research. This research aims to identify the trends of temperature and rainfall for the last 50 years (1966-2015) for Bangladesh. To detect the trends, the non-parametric Mann-Kendall test in combination with a trend-free prewhitening approach for correcting the time series to eliminate the influence of serial correlation has been used whereas the Sen's slope method is used for identification of the trend value. Maximum, minimum, and mean monthly temperatures have been analyzed to identify the monthly, seasonal and annual trend. Four seasons are distinct in Bangladesh, namely, pre-monsoon, monsoon, post-monsoon and winter. In case of rainfall, mean monthly rainfall, 1-day maximum rainfall and consecutive 3-day maximum rainfall are analyzed in a similar fashion. For maximum, minimum and mean monthly temperature, the majority of the stations show an increasing trend for all three indices. Monsoon season observes highest increasing trend of 0.015 degrees C/yr for mean monthly temperature whereas the winter season shows the negative trend of - 0.002 degrees C/yr. On an annual basis, the trend is increasing with a value of 0.008 degrees C/yr or 0.4 degrees C in 50 years having significance level, p < .05. On the other hand, mean rainfall shows an increasing trend for the three seasons except for winter, however, significance level in this case is p > .10. The trend of mean rainfall for annual basis shows an increasing trend of 5.675 mm/yr. The trend in one day and consecutive three-day maximum rainfall depicts that the trend is increasing in few places whereas in other places the trend is decreasing with no significant change. The research concludes that the climate of Bangladesh is getting cooler and drier for winter, on the other hand, warmer and wetter for the rest of the year.