LiDAR-based assessment of canopy size variability and its impact on irrigation efficiency and yield in almond orchards

Water use in almond orchards is strongly conditioned by tree canopy size. However, conventional irrigation management does not take into account the structural variability in tree size present within management units, which can lead to inefficient use of water. This study quantifies the impact of tree structural heterogeneity, estimated from Ground Cover (GC) measurements obtained through LiDAR technology, on water–yield relationships in a commercial almond orchard (cv. Lauranne). A theoretical and uniform model based on the weighted average GC calculated with LiDAR (Scenario A) was compared against a water–yield model that incorporates the ground cover variability of individual trees in the orchard response (Scenario B). Results show that the model accounting for canopy heterogeneity enables quantification of yield losses associated with treating all trees uniformly. At the maximum yield point estimated in Scenario A, Scenario B predicts a reduction in production due to the variations in size which is equivalent to the potential gain achievable through the implementation of precision irrigation. Moreover, the analysis of Marginal Water Productivity (MWP) revealed that approaches based on orchard average size (Scenario A) fail to capture the true marginal returns of water application. Under conditions of structural variability (Scenario B), MWP is greater than in Scenario A over a wider range of applied water. This implies that decisions based on Scenario A lead to selecting irrigation depths below the optimum, forgoing profitable production, or failing to recognize when differential water application would generate greater returns per unit of water. These findings demonstrate that ignoring intra-plot variability leads to inaccurate characterization of water-yield relationships and masks spatial inefficiencies. The integration of LiDAR-derived structural data increases the accuracy of water-yield models and identifies relevant opportunities to optimize irrigation scheduling at the sector scale, thereby improving water use efficiency even before the adoption of variable-rate irrigation technologies.