Measurement and modelling of Moringa transpiration for improved irrigation management

A greater understanding of Moringa ( Moringa oleifera Lam.) transpiration (T) can assist in the development of accurate irrigation management tools. This study aimed at quantifying Moringa T by measuring and modelling the sap flow (SF) of intact stems using an improved heat balance technique. The study was conducted during two consecutive seasons (2021–2022 (Season 1) and 2022–2023 (Season 2)) at the Roodeplaat Experimental Farm of the Agricultural Research Council in South Africa. EXO-Skin sap flow sensors were used. Transpiration-related drivers such as weather and plant physiological parameters were measured simultaneously. The measured SF data in Seasons 1 and 2 were used to respectively parameterize and validate a canopy conductance T model. There was a positive correlation between the measured SF and its drivers, evidenced through coefficients of determination (R 2 ) of 0.82, 0.99 and 0.92 for the relationships between SF and short-grass reference evapotranspiration (ET o ), stem area and stomatal conductance, respectively. The measured and simulated SF varied from 0.82–1.29 and 0.71–1.19 mm tree −1 day −1 (model parameterization), as well as from 0.77–3.54 and 1.10–3.10 mm tree −1 day −1 (model validation). Despite the slight discrepancies between measured and predicted SF values during model performance evaluation, an acceptable agreement was achieved through root mean square errors (RMSEs) of 0.32 and 0.37 mm day −1 and model efficiencies (Efs) of 0.93 and 0.88, for model parameterization and validation, respectively. The current study showed that the canopy conductance T model has the potential to accurately predict Moringa T and contribute to optimizing irrigation water management. • Moringa tree transpiration was highly influenced by tree size. • Sap flow was affected by the study morpho-physiological and microclimatic variables. • Stomatal conductance transpiration model for Moringa irrigation management was successfully parameterized and validated.