Growth dynamics and logarithmic modelling of multi-leaf lettuce cultivars ‘Ribbon’ and ‘Lime’ in greenhouse hydroponics
1School of Life Sciences & Chemical Technology, Ngee Ann Polytechnic, Singapore. 2Urban Food Solutions Division, Singapore Food Agency, Singapore. Conservation Division, National Parks Board, Singapore. 4Yong Loo Lin School of Medicine, National University of Singapore, Singapore. E-mail: ckk@np.edu.sg
DOI: https://doi.org/10.37855/jah.2026.v28i01.38
This study examined the growth dynamics of two multi-leaf lettuce (Lactuca sativa L.) cultivars, ‘Ribbon’ and ‘Lime’, grown hydroponically using the deep flow technique (DFT). The objective was to characterize cultivar-specific vegetative growth patterns and evaluate the suitability of logarithmic regression models for predicting harvest timing in hydroponic systems. Growth parameters including shoot fresh weight (FW), stem FW, plant height, stem height, and leaf number were recorded at 1-2 day intervals from 20 to 35 days after transplanting (DAT). ‘Ribbon’ exhibited rapid vegetative growth between approximately 26 and 29 DAT, with shoot FW increasing from 80 g to 100 g and stem height reaching about 5 cm within the 3-day period. Logarithmic regression models showed strong fits for shoot FW (R² = 0.9449), plant height (R² = 0.8817), and stem height (R² = 0.9430), while leaf number increased from 7 to 19 (R² = 0.8478). These characteristics indicated that ‘Ribbon’ was suitable for short-cycle production emphasizing rapid leaf biomass accumulation. In contrast, ‘Lime’ showed a slower but more gradual growth pattern. After 29 DAT, shoot FW reached 93 g with stem height approaching 5 cm, requiring an additional day to exceed the 100 g harvest threshold. Logarithmic models showed excellent fits for shoot FW and leaf number (R² = 0.9837 and 0.9917, respectively), with leaf numbers increasing from 6 to 24 during the observation period. The results demonstrated that logarithmic regression could be applied as an empirical approximation over an observed harvest window rather than full-cycle growth predictions in hydroponic lettuce. Such models may assist growers in predicting crop development and optimizing harvest scheduling in controlled hydroponic production systems.
