Tree-ring based climate reconstructions are significant for understanding past climate variability, in the context of increasing population, rising water demands, and changing climate. With trees being widespread, precipitation patterns can be reconstructed in various regions, including areas with limited instrumental data. This study presents a tree-ring based multi-century precipitation reconstruction for Skardu, located in Upper Indus Basin (UIB) in northern Pakistan, using a stepwise linear regression model. The reconstruction, spanning the period 1472-1993, revealed significant variability in precipitation patterns. The results showed a strong correlation between tree-ring data and precipitation, with a calibration R² value of 0.52 and a predicted R² value of 0.35. The reconstruction highlighted a rich history of wet and dry periods. The presence of notable wet periods in the late 15th and late 16th/early 17th centuries suggested that the Skardu region has experienced significant hydrological variability over the centuries. In contrast, the late 18th century and the early 20th century, and the mid-16th and mid-17th centuries recorded some of the lowest precipitation values, pointing to dry conditions that could have impacted agriculture and water resources. Future work directions for this study include enhancing the reconstruction with the addition of new tree-ring chronologies, combining tree-ring data with other climate proxies to enhance the reconstruction methodology, and applying this methodology to additional regions within the UIB for a comprehensive understanding of spatial climate variability.

Keywords- tree ring chronology, Upper Indus Basin, streamflow reconstruction, Skardu, Northern Pakistan