Abstract
Weeds remain a major constraint to cost effective crop production by interfering with yield, harvesting and quality. However, many wild plants, which may be considered weeds in cropping situations, are also considered to have beneficial roles in the environment in their contribution towards biodiversity; both directly and also by supporting other organisms as a source of food and shelter. The nature of vegetation management (to remove or to encourage) therefore changes with the perceived role of the species concerned. However, in any situation a good understanding of the species life cycle and its ability to establish seedlings in different environments is essential to be able to manage the species. This understanding does not currently exist, and in particular our knowledge of the annual dormancy cycle of seeds in the soil seedbank is limited. Seed dormancy determines whether the seed can germinate or not, and during dormancy cycling the level of dormancy is continually changing in response to the environment. In this way, germination of the seed is controlled so that it occurs in the correct season and conditions for growth of that species. Not all seeds in the population respond uniformly to the environment and so individual seeds will germinate at different times thereby reducing the risk of failure of seedling establishment in the whole population. This behaviour contributes to species survival and is ecologically sound, but frustrates attempts to make predictions of the timing of seedling emergence or seedling abundance, and therefore management of the plant population. A number of seedbank management techniques act directly on the germinability of the weed seeds by either stimulating or suppressing germination. Many tactics used to deplete the seedbank by stimulating emergence, such as the stale seedbed technique, require the viable seedbank to be in a non-dormant state for them to be most effective. Getting the timing of cultivation wrong can lead to poor emergence, or even dormancy induction which is potentially more problematic. For example, as in the case of volunteer oilseed rape, poor timing of post-harvest incorporation can lead to the induction of secondary dormancy and long-term persistence problems with this species in the seedbank. In practice, the way we manage our wild plants is increasingly constrained by the loss of herbicide products available to remove unwanted plants (weeds) following changes in EU regulations (EU 91/414) coupled with environmental pressures to reduce herbicide inputs. A major challenge for managers of vegetation is therefore how to improve the reliability of non-chemical and integrated methods. For the reasons above, the main obstacle facing this challenge is arguably our limited understanding of seed dormancy cycling in the soil. This limits the introduction, accurate targeting and success of many novel low-input methods of weed control and proactive methods to enhance plant biodiversity such as sowing plant mixtures for in-crop and margin flora. In line with Defras’ requirements, the proposed project aims to improve understanding of seed dormancy cycling in broad-leaved species to facilitate more informed vegetation management decisions. A general problem with and limitation to the study of dormancy in a variable environment during dormancy cycling, is that it can only be negatively defined by the absence of germination or post-facto by the germination event. Recent progress made on our Defra projects has provided us with the potential to minimise this limitation using modelling techniques and molecular markers in addition to the more conventional approach of germination potential. Armed with these techniques, the overall plan for the project is to monitor germination and emergence (in situ and from recovered seeds) over two years following seed shedding (e.g. from September 2007) in field and/or controlled environment experiments. We will use two selected species of the Brassicacae which have practical significance and the genetically related model species Arabidopsis thaliana (Cape Verdi ecotype). In the final full year it is intended to test the understanding gained in the first two years by manipulation of the germination environment in field experiments. Manipulation will be carried out using techniques that have potential for use in the control (positively or negatively) of seedling establishment in farm practice. We therefore aim to: 1. Generate comprehensive data sets concerning the annual dormancy cycle of broad-leaved weed species through controlled environment and field experimentation. 2. Improve understanding of annual dormancy cycling in the soil seedbank. 3. Establish the link between applied practices and the annual dormancy cycle of seeds to support informed decision making in weed management and seedling establishment practice.
United Kingdom