عنوان مقاله [English]
Introduction: Seed germination is an important stage in the life history of plant affecting seedling development, survival, and population dynamics. Germination begins with seed, water uptake and terminates with the elongation of the embryonic axis from the seed coat. Germination and seedling emergence are the most important phonological development stages in pigweed and have a vital role in its establishment. Accordingly, predicting the phenological stages would be resulted in improvement of crop management as the number and time of pigweed emergence could be quantified. Sigmoidal curves, also known as growth models, have wide application in agricultural research and can be evaluated by means of nonlinear models, which operates through data modelling by a nonlinear combination of parameters depending on one or more independent variables. This study was conducted to evaluate various regression models to describe the response of germination rate to temperature range in two pigweed ecotypes (Alborz and Fars).
Materials and Methods: A glasshouse experiment was carried out as a completely randomized design with four replicates. The seeds were sterilized by soaking in a 3% solution of hypochlorite sodium for 30 seconds. After the treatment, the seeds were washed several times with distilled water. 25 seeds were put in each Petri dish (with 9 cm diameter). The petri dish is monitored on a daily basis and afterwards germinated seeds (according to exit of radicles to the size of 2 mm) were measured and recorded daily in each Petri dish. Six regression models were applied to quantify the germination patterns of two pigweed ecotypes (Alborz and Fars) under a range of temperature between 5 to 35 °C. For both regions, during spring and summer, the range of temperatures was selected in order to simulate the temperature changes. The models were included Weibull, Lognormal, Logistic, Gompertz, Sigmoidal and Chapman. Some criteria were used to describe the goodness of fits of the models, including coefficient of determination (R2), root mean square of error (RMSE) and Akaike index (AIC). Moreover, a simple program called Germin was used to calculate D10, D50 and D90 (the time interval to maximum 10, 50 and 90% of germination, respectively).
Results and Discussion: Results showed that Weibull four-parameter and logistic models were the best for describing the germination rate in Alborz and Fars ecotypes, respectively compared to the other models. The difference between ecotypes could be attributed to their base temperatures and thermal time requirements at each developmental stage. Therefore, it can be concluded weed germination during different seasons is not a random phenomenon. However, the germination and emergence of a clear pattern over time follows the pattern of different environmental conditions is subject to change. Results also indicated that the time to D90 was only 4 and 5 days in Alborz and Fars ecotypes, respectively meaning that with increasing temperature during early spring, this weed would germinate much sooner than spring crops and consequently resulted in crop damage. The results showed that the in Alborz population, with increases temperature from 10 to 30 °C, germination percentage linearly increased and with increasing temperature to the desired temperature, it decreased. However, germination in Fars ecotype showed that, in temperatures 10, 15 and 20 °C respectively, the germination was 36, 56 and 84 percent, while, with an increase in temperature from 25 to 35 °C, this component was always a constant process. Germination rate increased with increasing temperature from 10 to 35 °C which was higher in Alborz ecotype compared to Fars. At lower temperatures, the main reason for less germination rate could be decrement of water imbibition and enzyme activity in biochemical processes of germination.
Conclusion: Increasing public awareness and concern about the impacts of herbicides on the environment, development of herbicide-resistant weeds, and the high economic cost of herbicides have increased the need to reduce the amount of herbicides used in agriculture. Prediction of weed emergence timing would help reduce herbicide use through the optimization of the timing of weed control. It was concluded that Weibull four-parameter and logistic models could be used as decision making tools in Alborz and Fars, respectively, to predict seed emergence of pigweed which in turn resulted in efficient management as well as reduction of herbicides usage. Future research should be addressed to determine a wider validation of the models, which could be valuable tools for farmers and practitioners for adequate timing of control in pigweed weed.