Mid-century change analysis of temperature and precipitation maxima in the Swat River Basin, Pakistan

Abstract

This study analyzes trends in historical (1989–2018) and projected (2041–2060) temperature and precipitation maxima in the Swat River Basin, Pakistan. This basin has a history of climate-related disasters that directly affected livelihood and personal safety in local communities and are becoming more intense and more frequent due to changing climate. Major economic sources of this basin are agriculture and tourism, both highly sensitive to extreme climate events. Therefore, it is very important to assess future trends in extremes of temperature and precipitation. Non-parametric tests were employed for currently acquired data, while future projections were assessed using the statistical downscaling model (SDSM) with CanESM2 GCM under three scenarios: representative concentration pathways (RCPs) of 2.6, 4.5, and 8.5. The R2 value between monthly observed and simulated temperatures varied from 0.82 to 0.91 and 0.92 to 0.96 for training and confirmation periods, respectively. For areal precipitation, an R2 value of 0.49 was noted for calibration and 0.35 for validation. Observed temperatures showed a decreasing trend at all stations except Saidu Sharif, but the differences were not significant. Precipitation showed an increasing trend at two stations, Kalam and Malam Jabba, and a decreasing trend at two other stations, Dir and Saidu Sharif. A >2°C rise was noted for the annual projected maximum temperature (2041–2060) at areal and Dir, while Kalam, Malam Jabba, and Saidu Sharif showed a 1°C rise. For precipitation, an approximately 12% increase in annual maximum (areal) and seasonal precipitation (summer and autumn) was seen under all scenarios except RCP 4.5 in which there was a 20% and 32% increase in summer and autumn, respectively. The performance of SDSM in simulating maximum temperature and precipitation was satisfactory.