Modelling weed emergence patterns

Anticipating weed pressure may be important in selecting and timing weed control measures in order to optimize their effectiveness, and thus reduce herbicide use. Therefore, a predictive model of the time of emergence and the numbers of seedling emerging (the weed emergence pattern) after soil cultivation may be a useful tool in integrated weed management. In this study, a simulation model was developed in order to increase the quantitative understanding of weed emergence in the field in relation to weather, soil and cultivation measures. In the model, three phases were distinguished in the process of weed emergence in the field, and modelled in separate modules: annual changes in dormancy, germination, and pre-emergence growth. The model was parameterized and tested for three arable weed species: Polygonum persicaria, Chenopodium album and Spergula arvensis.Simulation of annual cycles in dormancy and germination is based on a physiological model concerning the action of phytochrome in the seed. Dormancy is related to the amount of an hypothetical phytochrome receptor, that fluctuates in an annual pattern. The simulation model gave a reasonably accurate description of cyclic changes in germinability of seeds exhumed in a three years' burial experiment. The timing of germination was simulated by means of the thermal time concept.A physiologically based model describes the effects of temperature, soil penetration resistance, burial depth and seed weight on pre-emergence growth of seedlings. The model provided a good description of seedling emergence observed in a laboratory experiment.The separate modules simulating the consecutive processes of dormancy release, germination and pre-emergence growth were linked to form a model simulating seasonal weed emergence patterns in the field. Input variables of the model were the date and method of soil cultivation, soil temperature and soil penetration resistance. Output of the model was seedling density and the timing of seedling emergence. The model was evaluated with data from a field experiment. When using the germination results of the exhumed seed lots to estimate the degree of dormancy at the time of soil cultivation, the extent of the emergence flushes following soil cultivation could be described well. Although the dormancy model gave a good description of annual cycles in dormancy, the quantitative prediction of seasonal changes in dormancy and germination was not accurate enough for predicting field emergence, and appeared to be the weak point in predicting weed emergence patterns. When there was substantial emergence as a result of soil cultivation, the timing of emergence could be predicted accurately