عنوان مقاله [English]
Background and objectives: The complete (C), rapid (R) and uniform (U) germination of vigorous seedlings (V) are traits that all together (CRUV) can ensure early canopy closure and ultimately increase plant yield. The probable no strong positive relation (correlation greater than +0.95) between one or some components of CRUV causes that one treatment level appears to be the best treatment level for increasing of one or some components of CRUV, but the worst or medium for other components. In such situation it is hard to select one treatment level (Judgment predicament). Genetic algorithm-based optimization as a complementary analysis is a solution (overcoming the predicament) to such problems by which the treatment levels are interpolated for possible simultaneous maximum increase in CRUV. This experiment was firstly aimed to find whether there is Judgment predicament or not? Secondly, in case of positive answer, it was aimed at finding the best components of ultrasonication (as increasing treatment for CRUV) of mung bean seeds. The components of ultrasonication were seed pre-soaking, ultrasonication temperature, and duration of ultrasonication.
Materials and methods: The mean values of CRUV of 10 published papers were subjected to correlation analysis. Then a germination experiment based on completely randomized design with 3 replications was conducted for mung bean. Treatments were factorial arrangement of seed pre-soaking (2, 4, 6, 8, 10, 12 hours), ultrasonication temperature (17, 22, 27, 32 oC), and ultrasonication duration (0, 3, 6, 9, 12 minutes). Due to no strong relation between all components of CRUV, the genetic algorithm was used to optimize them. For this reason, firstly the desirability function was calculated; then the value of general desirability was determined. For predicting the response variables, different linear and non-linear functions were examined among which, the quadratic multiple regression function was found to be more appropriate. Finally, the best combination of experimental factors for possible simultaneous increase in CRUV was interpolated, using MATLAB software.
Results: The results indicated that like those plants mentioned in published papers which their CRUVs were reviewed here, there was no strong positive relation between all components of CRUV in mung bean. Therefore, the judgment predicament tends to be true in mung bean too. The result of analysis of variance revealed that in addition to main effects, the triple interactive effects of factors were significant on each components of CRUV. The interpolated value of factors was ultrasonication temperature of 24.89 oC, ultrasonication duration of 4.125 minutes, and pre-soaking of seeds for 6.013 hours. This combination of factors could result in the highest possible increase in CRUV simultaneously.
Conclusion: Based on the results of this study, it seems that the judgment predicament tends to be widespread over all plants. In such situation, as it was seen in mung bean, the value of treatment combinations (or treatment level) is not the same for having simultaneous increase in all components of CRUV. The optimization as complementary analysis beside the mean comparison analysis, can estimate the best treatment level (mono-factor experiments) or treatment combination (multi-factor experiments); due to such simultaneous increase in CRUV, the canopy closure will be accelerated.