Introduction
Among all weeds, cogongrass (Imperata cylindrica L. Beauv) has been identified as a problematic weed in the sugarcane (Saccharum officinarum L.) fields of Iran for over 20 years since 1996. Failure to control of cogongrass regrowth during spring develops it rapidly and, in turn, reduces the sugarcane yield in subtropical Iran. In effect, the identification of weak points in cogongrass phenology and frequency is deemed to help decision makers adopt effective controlling strategies in optimum time. Storage of carbohydrate reserves by plants is, in turn, the best mechanism for optimizing the timing of management strategies when reserves are at the lowest levels in storage organs. Glyphosate has been also widely used in Iran to control of weeds in the sugarcane fields. Currently, despite double spraying, the spring control of cogongrass is not satisfactory in the sugarcane fields in subtropical Iran.
 
 
Materials and Methods
Two studies were carried out in Salman Farsi Agro-Industrial Company in about 40 km south-west of the center of Khuzestan Province, Ahvaz, Iran, in 2017. In the first experiment, monthly (February–November) total nonstructural carbohydrate allocation patterns of cogongrass were evaluated to determine carbohydrate low points in the seasonal phenological cycle. To that end, glyphosate (00, 0.92, 1.84, 2.76 and 3.68 L a.i ha^-1) was applied in April and June, 2017, in a 10×2 factorial design under randomized complete block designs with four replications. Data were analyzed using standard ANOVA procedures with the subsequent use of Least Significant Difference test (Little and Hills 1978( to separate means.
 
Results and Discussion
The occurrence of carbohydrate low points (CLPs) (88.69±13.15 mg g^-1) was considered from May to August, CLP was found to have two important points with the potential control of cogongrass reported. The first and second points were in May (579 GDD) and August (2683GDD), respectively. The suppression of cogongrass (58-100%) was achieved with the three highest rates (1.84, 2.76 and 3.68 L a.i ha^-1) in both application times, although substantial regrowth occurred in all plots by 5 months after the first treatment. Double spraying with a 60-day interval did not result in the effective control of such a troublesome weed in the sugarcane fields. Applying glyphosate two times in the spring and mid-summer (the end of CLP, 2683 GDD) was determined effective for the long-term control (>6 months) of cogongrass. The results suggest that glyphosate 1.84 L a.i ha^-1 should be performed on cogongrass in the spring for improved control.
 
Conclusion
We concluded that frequent spring spraying may be excluded to control of cogongrass in the sugarcane fields and only spraying should be made on twice the growth seasons. The first use of glyphosate in its minimum optimum dose (1.84 L a.i ha^-1) is made on early spring. The second is sprayed in its maximum optimum dose (3.68 L a.i ha^-1) at the midsummer, once the single-sprayed cogongrass gets appropriate leaf surface for transferring the herbicide.