عنوان مقاله [English]
Introduction: Wild melon (Cucumismelo) is a monoecious, annual, trailing-vine plant of the Cucurbitaceae family. The reproduction of wild melon is only by seeds. Wild melon is native to Africa but is aggressively invading crop lands in the northern Iran. Because of rapid growth and limited control measures of wild melon, it is fast becoming a dominant weed in soybean fields. As wild melon and soybean share similar ecological niches, information pertaining to wild melon germination, and emergence would be beneficial.
The ability to predict the time of seedling emergence is an important step towards increasing the timeliness and efficiency of chemical and cultural weed control measures. Controlling weed seedlings that emerge early may help to reduce competition during the critical phase of crop seedling establishment, but later emerging cohorts may escape control and contribute to the soil seed bank. Therefore, understanding the factors that regulate the time of weed emergence is crucial for both short- and long-term crop production goals.
The time of weed germination and emergence in the field is influenced by environmental factors such as light, and soil moistureand temperature. Soil temperature is often considered as the primary environmental factor regulating emergence. The cumulative effect of temperature, expressed as thermal time, is correlated with the processes of after-ripening and seedling elongation.
The aim of this study was to determine the effects of different management operations on the seedlings emergence of wild melon and to develop a soil thermal time (STT) model to predict the emergence of this species in soybeans.
Material and Methods: Experiment was arranged as split-split plot in a completely randomized block design with three replications. Conventional tillage (CT) and no-till (NT) systems were main plots, three seed rates of soybean (200,000; 300,000 and 400,000 seed ha-1) were sub-plots and sub-sub plots comprised five different doses of imazethapyear (0, 50, 75 100 and 125 g ai ha-1). The soybean cultivar used in the experiment was Telar (BP).
Before performing the experiment, field had also been in a continuous soybean production and contained natural infestations of annual weed, especially wild melon. The experimental design was then carried out during the growing seasons in 2016 and 2017. The experimental farm was divided into two uniform areas each year to adopt the CT and NT treatments. After seedbed preparation, planting in the first year was carried out on 17/6/2016 and in the second year on 25/6/2017 in both tillage systems. The spacing of the planting rows was 19 cm and the dimensions of each plot were 6×6 m. Each year after planting until flowering, irrigation was held once by week starting at 20 days after planting in each year, the indoxacarb (Avaunt) was used to control of heliotis. To evaluate the effect of different doses of imazethapyear (Pursuit) on the emergence pattern of wild melon during each year, spraying was carried out as PRI based on the mentioned treatments by using Marina-type 20 liter sprayer equipped with a flood jet nozzle and 2.5 bar pressure (calibrated based on applying 200 to 300 lit ha-1 water).
To predict the wild melon emergence pattern in each plot, a fixed 50×50 cm quadrate was placed at the center of each plot, and from the beginning of the season after the first irrigation, the newly emerged seedling was counted. The counts were repeated every week, and then the seedlings counted at each stage were removed until new emerged seedling was not observed.
Statistical Analysis: Thermal time models were developed and evaluated, to describe observed wild melon seedling emergence pattern. Therefore, to estimate the cumulative seedlings emergence during the growing season, number of purslane seedlings was calculated based on the number of seedlings m-2.STT was obtained with the soil temperature by the following:
whereSTmean is the mean daily soil temperature, Tbase is the minimum temperature at which purslane seed germinates, and n is the number of days after sowing. Base temperature used in the calculations of STT for wild melon was 10 °C (Baker and Reddy 2001). Temperature Model software (STM2) (Spokas and Forcella 2009) was used to predict daily soil temperature at the 3cm depth. Daily precipitation, minimum and maximum air temperature were taken from the nearest meteorological station located near Sari, Mazandaran province. Soil texture properties and organic matter (%), along with the latitude, longitude, and elevation (1 m) of the research site were also used to predict daily soil temperature.
To describe the pattern of seedling emergence, cumulative emergence data were fitted using a three-parameters logistic model (Equation 2) (Brown and Mayer 1988; Eizenberg et al. 2005):
where a represents the upper asymptote (final CE), T50 represents the STT when the CE is 50% of the final CE (median), and b represents the slope of the curve at approximately T50. The parameter estimates were compared by using a two-tailed t-test (p < 0.05).
The goodness of model fit was evaluated with root mean square error (RMSE) and adjusted coefficient of determination (R2adj) (Izquierdo et al. 2009). RMSE was calculated as follows:
wherePi and Oi are the predicted and observed emergence values, respectively, andn is the number of observations. Preliminary analysis of variance was conducted using PROC GLM in SAS to determine treatment effects and their interactions. Nonlinear regression analyses were applied using Sigma Plot ver. 12.5.
Results: The statistical measures of the model's performance showed that STT model is a good predictor for wild melon emergence, with adjusted coefficient of determination (R2adj) values ranging from 0.92 to 0.99 and root-mean-square error (RMSE) values ranging from 2.59 to 12.69. The time of wild melon emergence was significantly affected by management practices such as tillage systems, soybean seeding rates, and imazethapyr doses. In the CT system, a seeding density of 400,000 seeds ha-1 and a 125 g ai ha-1imazethapyr dose showed a reduced wild melon emergence rate and had the highest T50, thus resulting in a long lag phase of wild melon emergence at the beginning of season. This provides more time for crops to establish and improves soybean competitive ability. The current study provides valuable information on prediction of the time of wild melon emergence to develop effective management strategies for this species. The proposed emergence models precisely predict the emergence pattern of wild melon seedlings as a function of STT, and STT models developed on the basis of field observations offer relatively strong predictions. With the continued development of such models, IWM will be more practical, enabling us to predict how farm management changes affect weed populations, the biological outcomes of the various management options, and the economic and environmental impact of crop-weed competition.