Trend analysis of rainfall data Rangpur, Bangladesh

The study area lies in Bangladesh that is in the tropical area. The analysis of rainfall data reveals that the average annual rainfall in Rangpur was 2099.25 mm that varies from 427 mm to 3748 mm within the investigated period of time. The highest amount of annual rainfall was recorded in 1984. Heavy rainfall occurs in the month of July of the year. The highest amount of total monthly rainfall was recorded in July, 1987 and measured as 1314 mm. The rainfall trend can be expressed as monsoon rainfall > pre-monsoon rainfall> post monsoon rainfall. The total amount of annual rainfall is strongly significantly positively correlated with the total monthly rainfall of the months of June, July August, September and October and significantly positively correlated with the total monthly rainfall of the month April. The analysis shows that it rained in every year in the month of June as the minimum monthly value for June is greater than zero. Pearson correlation index shows that the mean daily temperature is significantly negatively correlated with daily rainfall which implies that fall of daily mean temperature with the daily rainfall. The increasing trend of annual rainfall suggests the climate change in Rangpur within the investigated period of time. In Rangpur, the value of skewness for all rainfall data are positive that indicate the data are skewed to the right. The positive values of kurtosis indicate that the distribution is not normal.


Introduction
Rainfall is the major form of precipitation in Bangladesh. The rainfall is one of the principal factors that control the agriculture, health condition and life style of people of any region (Islam et al. 2019). Sometimes excess rainfall becomes curse for the people of Bangladesh. The study area Rangpur lies in the northern part of Bangladesh ( Figure 1) and is characterized by heavy rainfall during the summer and low rainfall during pre-monsoon and post monsoon periods (Rahman et al. 2016). Rainfall is the main input of the hydrogeological cycle of the study area. The climate change is measured by the rainfall patterns where Bangladesh is considered as one of the most vulnerable countries to climate change (IPCC 2007: Richard 2012. The study area is drained by numerous rivers like Tista, Ghagat, Jamuneswari and Karatoa. It is the extension of the Himalayan piedmont plain that slope southward (Saha et al. 2020). The surface geology is mainly composed of gravel, sand, silt and clay of Recent age and in some areas by the clay deposits of Pleistocene age. The present research work deciphers the total monthly rainfall and trend of annual rainfall. The annual rainfall is mainly influence by the monsoon wind. The daily rainfall also helps to lower the daily temperature especially in monsoon summer. The amount of annual rainfall shows positive correlation with time.

Methods
Daily rainfall data of Rangpur, Bangladesh was downloaded from website. From the daily rainfall data, total monthly rainfall was calculated using necessary computer software. The total annual rainfall was calculated by adding the total monthly rainfalls of twelve months of that specific year. The mean monthly rainfall and mean annual rainfall were analyzed. The required graphs were drawn. The Pearson correlation coefficients were calculated. The other statistical parameters such as mean, median standard deviation, skewness and kurtosis were calculated to analyze the yearly and monthly data (Islam et al. 2019). The daily rainfall data of the wettest month of July, 1987 were correlated with the mean daily temperature to show the influence of daily rainfall on daily mean temperature.   Table 2 depict that the amount of total annual rainfall is positively correlated with time. The analyses of daily rainfall data reveal the maximum amount of daily rainfall was 294 mm in September 25, 2002. The heaviest rainfall in any given month during the study period was recorded in July, 1987.  Table 1 shows the distribution of mean monthly rainfall of Rangpur station during 1954-2013. Figure 3 shows that 80.21% annual rainfall occurs during the monsoon months of June-October, while 18.67% annual rainfall occurs in the pre-monsoon months of February-May and 1.11% annual rainfall was recorded in post monsoon periods November-January.  January: The mean monthly rainfall for the month of January is 8.45 mm in the study area, while the total rainfall varies from 0 mm to 49 mm ( Figure 4). It did not rain 22 years in the month of January in Rangpur. The mean monthly rainfall has a decreasing trend with time (years).  February: The total amount of rainfall varies from 0 mm to 57 mm for the month of February ( Figure 5). The mean monthly rainfall for the month of February is 9.145 mm in the study area, while it did not rain 19 years in the month of February in Rangpur. The mean monthly rainfall of this month is increasing with time (years). March: The total amount of rainfall during the month of March was highest in 1959 that was recorded as 257 mm ( Figure 6). There was no rainfall during the month of March in seven years like 1954,1957,1971,1977,1992,1996 and 2010. The total monthly rainfall for the month of March is negatively correlated with time. The mean monthly rainfall of March is 27.36 mm.   June: The amount of June rainfall varies from 104 mm to 989 mm with the mean rainfall value of 440.55 mm. The median value for the June rainfall is 426.5 mm. The distribution of total amount of rainfall remained even for the investigated time period in Rangpur ( Figure  9). The maximum amount of June rainfall was reported in the year of 1984. July: July is the wettest month of the year. The total amount of July rainfall remained more or less uniform over time ( Figure 10). It did not rain only two years (1958 and 1981) in the month of July. The median July rainfall value is 460 mm ( Table 1). The average magnitude is 447.10 mm while the highest amount of July rainfall was 1314 mm in the year of 1987.     1954  1957  1960  1963  1966  1970  1973  1977  1980  1983  1986  1989  1992  1995  1998  2001  2004  2007  2010  2013 Monthly

Rainfall
Year October: The frequency of October rainfall is significantly marked by the decrease of mean value. The total amount of October rainfall varies from 6 mm (2011) to maximum 586 mm in 1959. The average value of October rainfall was 158.84 mm and the median value was 121 mm. The amount of October rainfall is positively correlated with time ( Figure 13). November: The mean rainfall value of November was 7.64 mm. The median was 1 mm. The maximum amount of rainfall in November was 109 mm in 1995 while there was no rainfall in 27 years ( Figure 14). The rainfall pattern was almost steady over the period of investigation in Rangpur.  December: The median value of December rainfall was 0 mm whereas the mean rainfall was 7.21mm. There was no rainfall in 36 years in the month of December. The maximum amount of rainfall was recorded in the year of 1959 which was 103 mm. The pattern of December rainfall is decreasing with time ( Figure 15). Statistical Analysis of Rainfall Table 1 shows the results of statistical analyses of rainfall data in millimeters. The amount of rainfall starts to increase from the month of March, reaches to the peak in July and then begins to fall. The highest total monthly rainfall was recorded in July 1987 that was 1314 mm. Most of the rainfall occur in the months of May, June, July, August, September and October. The analysis shows that it rained in every year in the month of June as the minimum monthly value for June is greater than zero. The standard deviation and range describe the variability of annual rainfall and signify how reliable the rainfall in terms of its persistence as constant and stable replenishing source (Islam 2019). A high standard deviation value means that the rainfall is spread wide on the either sides of the mean. Skewness and kurtosis were computed to test the pattern of annual rainfall data. The skewness for a normal distribution is zero, and any symmetrical data should have skewness value close to zero. Negative skewness values indicate that the data are skewed to the left while the positive skewness values suggest that the data are skewed to the right. In Rangpur, the value of skewness for all rainfall data are positive that indicate the data are skewed to the right. Kurtosis is a measure of peakedness or flatness to normal distribution. The positive value of kurtosis means a peaked distribution and a negative value reveals the flat distribution with the same mean and standard deviation. The annual rainfall distribution under this circumstance, did not follow the normal distribution.

Discussion
The rainfall varies in different months of the year. The amount of rainfall influences the agricultural activities of a region. The analysis of rainfall data is important for proper crop management. The rainfall also controls the hydrogeological characters of an area. In Rangpur region, the water from the rainfall recharge the groundwater aquifer, surface water bodies and the quality of groundwater and surface water bodies depends on the recharge and discharge of them.   The total amount of annual rainfall is strongly significantly positively correlated with the total monthly rainfall of the months of June, July August, September and October and significantly positively correlated with the total monthly rainfall of the month April (Table 3). This study reveals that the total annual rainfall is mainly influenced by the total amount of monsoon rainfall. The heavy rainfall in May-September is suggestive that the study area is located in humid tropical area and the rainfall is influenced by the South West Monsoon. The amount of pre-monsoon rainfall also influences the total annual rainfall of Rangpur. On an average 18.67% rainfall of the year occurs in Pre-monsoon periods. The importance of pre-monsoon period is increasing day by day with the increased cultivation of rice in this time span. The Pearson correlation matrix index of daily rainfall and mean daily temperature (of July, 1987) shows that they are significantly negatively correlated. This result suggests that the daily temperature in Rangpur during the summer monsoon is controlled by the amount of rainfall.

Conclusion
The trend of annual rainfall is increasing and the mean annual rainfall is 2099.25 mm. The maximum amount of rainfall is recorded during the summer monsoon. The total annual rainfall is significantly positively correlated with the monsoon rainfall. The analysis shows that it rained in every year in the month of June as the minimum monthly value for June is greater than zero. The climate change can be revealed by the increment of pre-monsoon rainfall especially in the month of May. May is the harvesting time of paddy in Rangpur. The heavy rainfall in May might have impact on the production of rice. The daily temperature in the summer months lowered by the amount of daily rainfall in those months. In Rangpur, the value of skewness for all rainfall data are positive that indicate the data are skewed to the right. The positive values of kurtosis indicate that the distribution is not normal. Further research works can be taken on the seasonal variability of chemical composition of ground and surface waters and the change in groundwater levels.

Conflict of interest statement
Author does not have any conflicts of interest.