Trend Analysis of Temperature Data for Narayani River Basin, Nepal

Abstract

Study of spatiotemporal dynamics of temperature is vital to assess changes in climate, especially in the Himalayan region where livelihoods of billions of people living downstream depends on water coming from the melting of snow and glacier. To this end, temperature trend analysis is carried out in Narayani river basin, a major river basin of Nepal characterized by three climatic regions: tropical, subtropical and alpine. Temperature data from six stations located within the basin are analyzed. The elevation of these stations ranges from 460 to 3800 m asl. and the time period of available temperature data ranges from 1960–2015. Multiple regression and empirical mode decomposition (EMD) methods are applied to fill in the missing data. Annual as well as seasonal trends are analyzed and Mann-Kendall test is employed for testing the statistical significance of detected trend. Results indicate significant cooling trends before 1970s, and warming trends after 1970s in the majority of the stations. The warming trends range from 0.028 ∘ C per year to 0.035 ∘ C per year with a mean increasing trend of 0.03 ∘ C per year after 1971. Seasonal trends show highest warming trends in monsoon season followed by winter, pre-monsoon, and post-monsoon season. However, difference in warming rates between different seasons isn’t sufficiently large. An average temperature lapse rate of −0.006 ∘ C per m with the steepest value (−0.0064 ∘ C per m) in pre-monsoon season and least negative (−0.0052 ∘ C per m) in winter season is observed for this basin. A comparative analysis of the gap-filled data with freely available global climate data sets shows reasonable correlation thus confirming the suitability of the gap filling methods.