Growth, Germination and Dormancy of Herbicide-Resistant Echinochloa phyllopogon, and Implications for Management

Echinochloa phyllopogon is an important weed of paddy rice in California, where it has evolved multiple resistance to most relevant herbicides, creating a need for alternative control measures. To enhance the functioning of such measures, a series of experiments were conducted to determine the growth, competitiveness, germination, emergence and dormancy of two herbicide-resistant (R) and two -susceptible (S) E. phyllopogon populations. Through a series of growth and competition studies we confirmed that interference with rice by R and S plants was similar and mostly driven by root interactions, even though R plants were smaller than S plants. Evaluations of fecundity and timing of seed shatter found R plants produce less seed, suggesting R populations might decline in the absence of herbicide selection pressure; while a shorter time to anthesis in R plants implies a greater chance of avoiding seed removal from fields during harvest and higher rates of soil seed bank regeneration. We applied population-based threshold models (PBTM) that could quantify and test the temperature, moisture and oxygen conditions for achieving the highest rates of germination and emergence in non-dormant seed, and found both R and S seed germinated under a wide range of water potentials and oxygen concentrations. By predicting hydrothermal time accrual in 13 distinct fractions of each population; we used PBTM to define the relative size and quantities of the emergence flushes that constitute final recruitment in field soils, finding emergence in R populations to be more sensitive to moisture and oxygen stress. PBTM also showed dormancy increased base water potential and hydrotime to germination, while lower final germination of non-stratified S seeds could imply dormancy is weaker among R populations. We then subjected seeds to fluctuating temperatures and a variety of stratification temperatures, moisture levels and durations, and measured subsequent changes in germination rates to show that decreases in water potential, and prolonged stratification time both reduced the extent of dormancy release, while differences in temperature below the base level for germination had little impact on dormancy release. We applied these models toward optimization of the stale seedbed weed control technique for controlling R plants and recommend winter flooding for 3–4 weeks to remove dormancy, and an initial flood to promote germination in spring followed by successive flushes to foster early growth