Predicting Growth and Dry Matter Production of Tomato in Response to Exogenous Proline Application under Drought Stress
Zin Shwe Thar Nu *
Department of Horticulture, YAU, Yezin, Myanmar.
San Shwe Myint
Department of Horticulture, YAU, Yezin, Myanmar.
Wai Wai Lwin
Department of Horticulture, YAU, Yezin, Myanmar.
Pan Ei Ei Kyaw
Department of Soil and Water Science, YAU, Yezin, Myanmar.
*Author to whom correspondence should be addressed.
Abstract
Aims: To evaluate the drought response of tomato under varying levels of proline application, to build a growth model for leaf area and shoot dry weight under drought and proline sprays, and to evaluate the model performance.
Study Design: Split-plot design: main factor (water supply: well-watered, drought-stressed), sub-factor (proline: 0, 20, 40, 60, 80 ppm) with four replications.
Place and Duration of Study: Polyhouse at the Department of Horticulture, Yezin Agricultural University (YAU), between October 2024 and May 2025.
Methodology: There were two pot experiments (Expt.): Expt. 1 for parameterization, Expt. 2 for model evaluation. Withholding of water and proline sprays was employed at five-leaf stage. Drought response of relative leaf expansion rate (RLER) to fraction of transpirable soil water (FTSW) was fitted by a linear–plateau regression (LPR) at each level of proline sprays and related parameters were generated for model simulation. Daily leaf expansion rate under well-watered condition (LERw) was calculated from plant leaf area (PLA) estimated from measurements of individual leaf length and width. In model, LER under drought (LERd) was described as the product of LERw and RLER and PLA was the integral of LER. For each water regime, shoot dry weight (SDW) was the product of specific shoot mass and PLA.
Results: Moderate proline levels (20-40 ppm) delayed the drought response of RLER, exhibited by lower FTSW thresholds (0.49-0.5) compared to control (0.75). Predicted PLA under well-watered condition showed higher goodness of fit than droughted condition (R²=0.60 vs 0.55). Simulated SDW represented more under drought than the well-watered condition (R²=0.57 vs 0.42). Model performance across proline levels revealed RMSD ranges of 112-286 for PLA and 0.85-2.26 for SDW, with accuracies of 0.88-0.96.
Conclusion: Current growth model showed varying drought responses to proline sprays with some magnitudes of errors, which needed further calibration and validation.
Keywords: Model, tomato, drought, proline, leaf area, dry matter production