Contributed by Fernando Oreja and Judit Barroso, Oregon State University
Soil seedbanks are the reservoirs of weed seeds present in the soil. These seeds can be either freshly shed by mature weed plants or seeds that have accumulated over time. The soil seedbank plays a crucial role in the persistence of weed populations. Despite efforts to reduce troublesome weed populations, seedbanks are important buffers that can mitigate the effects of weed management programs. Even in cases of successful herbicide programs that achieve weed-free conditions, seedbanks can retain enough viable seeds to cause reinfestations in the field in the following season/s.
Seedbanks are classified into three types: ‘transient’ for species with seeds that persist in the soil for less than one year, ‘short-term’ for species with seeds that persist for more than one year but less than five years, and ‘long-term’ for species with seeds that persist for more than five years. An understanding of the type of soil weed seedbank should help determine the best weed management strategies to reduce it.
In addition, understanding when a problematic weed species is likely to emerge and the duration of its emergence period is important to plan effective weed control programs. Knowledge of weed emergence dynamics can inform planting date decisions, allowing for the delay of planting to target early-emerging species, or an early planting date to enhance competitiveness against late-emerging species. This knowledge can also be valuable to optimize the control with residual and/or post-emergence herbicides. Improving herbicide efficacy can reduce herbicide applications, the risk of new cases of resistance, and control costs.
Recently, at the Columbia Basin Agricultural Research Center (CBARC-OSU) (Adams, OR), experiments have been conducted to determine Russian thistle (RT) seedbank longevity and seedling emergence dynamics. Two experiments were started in 2020 (referred to as sites A and B in Figure 1) and another two experiments were started in 2021 (referred as sites C and D in Figure 1). Around 14,400 RT seeds were spread per site at the beginning of each experiment and since then, RT germination has been recorded continuously.
The highest germination percentage was observed in the first year after the experiment establishment at all sites (Figure 1). Years two, three, and four had much lower germination, particularly at sites A and B where the germination in the first year had been very high (52% on average). The first year at sites C and D (2021) was warmer and drier than 2020, which may explain the lower germination registered that year. All fields were in no-till management, however, we observed significantly more RT germination when fields were seeded with spring wheat than when left in fallow. In 2020, site B (in spring wheat) had 72% RT germination compared to 32% RT germination at site A (in fallow), and in 2021, site D (in spring wheat) had 46% RT germination compared to 7% at site C (in fallow).
Except for site C, that had 5% germination in the second year, seedling emergence was less than 2% in the years following the initial year. Based on these findings, the seedbank for RT can be classified as short-term persistence (lasting more than one year but less than five). Implementing measures to prevent new RT seeds from entering the seedbank will be essential to quickly reduce the population levels of this species.
In the first year, two peaks of emergence were registered in April at sites A and B, achieving 60% of relative emergence in that year (Figure 2a). Meanwhile, sites C and D exhibited a single peak between March 31 and April 14, contributing to 85% relative emergence for that year (Figure 2b). Following these initial peaks, seedling emergence continued in all sites until early June. In the second year, sites A and B experienced a single peak in April, while sites C and D had several small peaks, extending the RT emergence until mid-June. The amount of moisture received late in spring 2022 could have played a role in those multiple peaks. In the third year, the emergence period at all sites was delayed compared to the first two years. This agrees with previous work that found old RT seeds germinating later than newer RT seeds.
Figure 1. Seedling emergence registered for four years, at sites A and B, and for three years at sites C and B. Bars are the mean and whiskers the standard error of the mean (SEM). The SEM is a measure of how far the sample mean (average) is likely to be from the true population mean. The inset is the last three years at sites A and B and the last two years at sites C and D.
Figure 2. Seedlings emergence (%), a) of sites A and B in 2020, 2021 and 2022, and b) of sites C and B in 2021, 2022 and 2023. Symbols (square, circle, triangle) are the mean and whiskers the standard error of the mean (SEM).
In conclusion, we observed that most RT seeds germinated during the first season, although germination continued for four years at very low percentages. For new seeds, most germination happened at the end of March or during April, although a small percentage germinated until early June depending on precipitation. Spring wheat seems to enhance RT germination compared to chemical fallow or winter wheat. So, if a field has a large RT seedbank and needs to be seeded with spring wheat, an adequate integrated weed management strategy should be implemented.
We would like to thank USDA-NIFA for funding this research, Jennifer Gourlie for helping in the establishment and development of the experiments, and Kyle Harrison for conducting some of the farming in the experiments. We also thank John Rietmann and Keith Morter (two growers of Morrow County, OR) for allowing us to collect RT plants from their fields to conduct this study.