In precision agriculture, methods for analysing 3D point clouds of plants have been introduced, particularly pointing to the high accuracy of light detection and ranging (LiDAR) laser scanning under field conditions. In the present work, LiDAR-based 3D point clouds of cherry trees (n = 255) were analysed for estimating the leaf area as the main factor for water interception. Canopies were scanned for segmenting leaf area pointing to a high variability of canopy surface. The derived tree-specific data of leaf area index (LAI) were implemented into the Community Land Model (CLM), which takes into account canopy interception processes during rainfall events. During canopy development of perennial trees the LAI increased resulting in increased water interception. Events with low rain fall the interception reached 38–100 % capturing LAI of 0.76 – 2.11 m2/m2, respectively. In high rainfall events, interception varied 10–14 % capturing the same LAI range. An equation for describing the varying effects of rainfall intensity and LAI is proposed. The evapotranspiration and water interception data point to a substantial decrease of effective water supply that varies tree-individually during the season. In commercial fruit production, the proposed method can support precise irrigation management.