Based on the Mann-Kendall (MK) test and Budyko hypothesis, this study utilized the China Natural Runoff Dataset (CNRD v1.0) and the China Meteorological Forcing Dataset (CMFD) to analyze the characteristics of annual runoff depth (R) in the Pearl River Basin from 1979 to 2018 and attributed the observed changes. The results are as follows: (1) The average R in Pearl River basin is 887.7 mm, exhibiting a spatial distribution pattern of higher values in the east and lower in the west. The basin has high runoff production efficiency, with an average runoff coefficient of 57.8 %. (2) In the eastern and western regions of the basin, R generally shows a decreasing trend, with 5.7% of grids exhibiting significant decreases. In contrast, the central part of the basin predominantly shows increasing trends, with 2.5% of grids exhibiting significant increases. Overall, the basin shows an insignificant increasing trend in R, with a change rate of 0.81 mm/a. (3) Sensitivity analysis shows that, R in the Pearl River basin is positively correlated with precipitation (P) changes, negatively correlated with potential evapotranspiration (ET0) and underlying surface parameter (ω) changes. Among these factors, R is most sensitive to changes in P. For every 1.0% change in P, R increases by 1.2% to 2.2%. (4) P and ω are the primary driving factors of changes in R in the Pearl River basin, contributing 55.4 % and 47.1 %, respectively. In contrast, ET0 has a smaller impact, with a contribution of only -4.3 %. The findings of this study provide a valuable scientific reference for the rational development and utilization of water resources in the Pearl River Basin under the context of climate change.