DefinePK

Abstract

Long-term seasonal and annual precipitation trends are among the effective tools to assess the climate change of a region. In this study, long-term seasonal and annual trends were assessed for six meteorological stations of the Gilgit Baltistan (GB) (a mountainous belt of Pakistan) during 1980−2015. Non parametric Mann-Kendall (MK) and Sen's slope (SS) techniques were applied to find strength and magnitude of trends at
10% significance level. Serial correlation in data was removed by using Trend Free Pre-Whitening (TFPW) technique. Results revealed increasing trends in mean annual precipitation at five stations (significant only at Gupis with a SS of 7.34 mm/year) and decreasing trend at one station (Astore) with a SS of -3.05 mm/year. The evidence of increasing trends in mean annual precipitation in five out of six hydro-climatic stations indicated perennial flow in the Indus River and which ensure sufficient ground water recharge for the downstream plain areas. All stations showed increasing trends in precipitation during monsoon season while in other seasons both positive and negative trends found were non-significant. Precipitation trend in post- monsoon season is decreasing in most parts of the GB that might cause temporary dry condition in the Indus River downstream. Based on the overall mean precipitation analysis for the entire GB, there were non- significant increasing seasonal and annual trends except the post-monsoon season. These seasonal variations in precipitation might affect water supply to the Indus River resulting in less availability of water for agriculture sector and in turn affecting crop production in the downstream plains. Seasonal trends in precipitation could directly affect agricultural activities which is a vital source of livelihood in Pakistan. To be on safe side, construction of water storage structures (dams) is highly needed to avoid possible floods during monsoon season and store most of the precipitation to meet off season demands especially in dry season.