Answers are grounded in this paper and cite their sources.
This study investigated water stress in eggplant (Solanum melongena L.) grown under field conditions in Bafra, Türkiye, from 2015 to 2017. Five irrigation treatments were evaluated with three replicates: full irrigation (I1), 75% (I2), 50% (I3), 25% (I4), and rainfed conditions (I5). The crop water stress index (CWSI) was calculated for each treatment. Five artificial intelligence algorithms—adaptive neuro-fuzzy inference system (ANFIS), fuzzy logic (FL), gene expression programming (GEP), support vector regression (SVR), and random forest (RF)—were used to estimate the calculated CWSI. Models were trained with combinations of canopy-to-air temperature difference (Tc − Ta), vapor pressure deficit (VPD), and solar radiation (Rs), producing 35 models across seven input combinations. Self-organizing maps (SOMs) were used separately to explore the topological relationships among the inputs and CWSI. Among the evaluated models, ANFIS4, based on Tc − Ta and VPD, produced estimates closest to the calculated CWSI values. Because the reference CWSI itself was derived from Tc − Ta and VPD, this result should be interpreted as accurate reproduction of the empirical CWSI relationship under the experimental conditions. The approach may provide a basis for sensor-assisted irrigation decision support, although independent validation is required before operational application.
No takes yet. Share an insight, caveat, or question.
Cantürk et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: