Spatio-statistical analysis of rainfall fluctuation, anomaly and trend in the Hindu Kush region using ARIMA approach

This paper focuses on the spatio-statistical analysis of rainfall fluctuation, anomaly and trend in the Hindu Kush region using auto-regressive integrated moving averages (ARIMA) approach. In the study area, trend in rainfall has significant impact on fluctuations in river discharge, which ultimately led to floods and hydrological drought. In this study, rainfall has been used as a climatic parameter. For this study, average annual and mean monthly rainfall data for Dir, Timergara, Saidu, Chitral, Drosh, Malam Jabba and Kalam meteorological stations located in the study region were gathered from Regional Meteorological Center Peshawar. In the study area, the rainfall is mostly received during two prominent periods, i.e., summer rainfall from monsoon, whereas winter and spring rainfall from western depressions. In the study area, Malam Jabba has recorded the heavy mean annual rainfall (1647 mm) and is considered as the humid station followed by met station Dir with a 1362 mm mean annual rainfall. Similarly, Saidu met station received 1050 mm mean annual rainfall and Kalam 1038 mm, whereas Timergara, Drosh and Chitral recorded 796 mm, 568 mm and 458 mm, respectively. The temporal data regarding rainfall were calculated and simulated in Addinsoft Excel state 2014 by applying ARIMA statistical model for trend prediction, fluctuations and anomaly. The analysis indicates that in terms of rainfall, an increasing trend has been detected at Dir, Chitral, Saidu and Kalam meteorological stations, whereas a declining trend has been recorded at Timergara, Drosh and Malam Jabba meteorological stations. In terms of rainfall anomaly, the met station Dir has indicated comparatively high positive anomaly. Contrary to this, the met stations of Saidu and Drosh have experienced negative rainfall anomaly.