The water level of reservoirs， especially large reservoirs， is a crucial parameter for flood forecasting and reservoir operation. Taking the Danjiangkou Reservoir as an example， this study utilizes the ATL13 data from the spaceborne photon-counting LiDAR （ICESat-2/ATLAS） from 2018 to 2022. By extracting parameters such as longitude， latitude， geodetic height， and geoid undulation， the water level information is derived and validated against ground-based hydrological station data. This approach enables the acquisition of high-precision water level data even in the absence of direct measurements. Furthermore， considering factors such as water surface area and temperature， stepwise regression and multiple linear regression are applied to construct an area-water level relationship model， enabling the estimation of the Danjiangkou Reservoir’s water levels over multiple years. The results show that： （1） The water level of the Danjiangkou Reservoir exhibits significant seasonal variations； （2） The method's accuracy is validated by comparing the derived water levels with observed data， with an average error of 0.03 m and an R2 of 0.999； （3） The area-water level relationship curve established using a cubic polynomial model provides the best fit， with an R2 of 0.956 and a Mean Absolute Error （MAE） of 0.009 m. These findings demonstrate that spaceborne photon-counting LiDAR technology offers valuable data support for water level estimation and dynamic change studies of large reservoirs.