Abstract. The performance of a mechanistic simulation model of crop-weed competition was tested with data on the critical period of weed competition in sugarbeets and both seeded and transplanted tomatoes. In general, there was good agreement between simulated and observed yields for different periods of weed interference in each crop. The model was then used to evaluate the influence of weed density, weed height, and weather conditions on timing of the critical period. Simulations suggested that the greater the weed density, the shorter the period of time that the crop could tolerate early-season competition, and the longer the period of time that the crop must be kept weed free to prevent yield losses. Simulations also suggested that the length of time that a crop can tolerate early-season weed competition is related more to the availability of soil moisture, or possibly essential nutrients, than to light limitations.

Destructive measurements to collect input data for models that predict yield loss from relative leaf area of weeds can be laborious. Alternative methods were tested in seven field experiments with sugar beet or spring wheat. Weeds with different morphologies showed the same linear relationships between relative leaf area, measured destructively, and cover, assessed by means of a frame, until 3 or 4 weeks after crop emergence. At later growth stages, differences in weed morphology resulted in different relationships. Visual estimates of weed cover corresponded only roughly with cover assessments with a frame. The possibility of estimating relative leaf area of weeds with a reflectance technique was tested, assuming that for early growth stages the leaf area index of weeds can be considered as additional to that of the crop. In spring wheat, relative leaf areas of Sinapis alba L., sown at different times and densities, correlated well with characteristics based on infra-red reflectance. In sugar beet, these relationships were not as distinct. Techniques d'estimation de Ia surface folia ire relative et de Ia couverture des mauvaises herbes dans les cultures, en vue de predictions de pertes en rendementDes prelevements destructifs peuvent alimenter en donnees les modeles qui predisent les pertes de rendement a partir de la surface foliaire relative des mauvaises herbes, mais ils sont exigeants en temps. Des methodes alternatives ont ete testees lors de sept experiences au champ dans de la betterave ou du ble de printemps. Jusqu'a trois ou quatre semaines a pres la levee de la culture, des mauvaises herbes possedant differentes morphologies montraient les memes relations lineaires entre d'une part Ia surface foliaire relative mesuree de maniere destructive et d'autre part Ia couverture mesuree grace a une grille.

Summary The effects of three different weed management strategies on the required input of hand weeding in an arable organic farming system, the weed seedbank in the soil and the emerging weed seedling emergence were studied from 1996 to 2003. Strategies were based on population dynamic models and aimed for (1) control of weeds as carried out in standard organic farming practice, (2) control of all residual weeds that grow above the crop and (3) prevention of all weed seed return to the soil. Under all strategies, the size of the seedbank increased during the conversion from conventional to organic farming systems. The increase under strategy 3 was significantly smaller than the increase under the other strategies. From 1999 onwards, the weed densities in plots treated with strategy 3 became significantly lower than the weed densities in plots treated with the other strategies. The time needed for hand‐weeding required to prevent weed seed return, in addition to the time needed in standard organic farming practices, reduced during the course of the study. A management strategy aimed at the prevention of seed return (strategy 3) can reduce the size of the increase of the seedbank, which is usually observed after transition from conventional to organic farming. This study provides unique real‐world data that are essential for evaluating population dynamic models. The results may contribute to the development of weed management systems based on ‘no seed’ threshold strategies and to a further decrease in the dependence on herbicides.

No abstract

Most studies on weed population dynamics in farming systems have focused on the effects of different weed control strategies. Those studies usually assume that farmers, operating within a particular system, have a uniform management style. However, it is likely that weed management decision making also varies between farmers that operate within a system. In this study, the relationship between weed management behavior and the outcome of that behavior within an organic farming system is studied. It is hypothesized that differences in weed pressure between organic farms can be related to differences in farmers' weed management behavior. We explore which weed and general management factors are of main influence on the weed pressure, and investigate the influence of farmer's beliefs and knowledge on weed control techniques and the observed weed pressure. Preventive measures and timing of main soil tillage operation were identified as the weed management factors most influential for weed pressure. With the increasing number of preventive measures applied, weed pressure decreased, with a stale seedbed being the most important preventive measure. The weed pressure increased with the number of days after September 1st on which the main tillage operation was carried out. Because of this postponement of the tillage treatments, the growing season of weeds was extended and more species were able to reproduce before winter, thereby enhancing weed pressure. Field size, rather than weed pressure, determined the number of hand-weeding hours per ha; with increasing field size the amount of hand weeding per surface area was reduced. On farms with lower weed pressures a higher percentage of competitive crops were grown than on farms with higher weed pressures. The farmer's beliefs and knowledge on weed control techniques differed between farmers with different weed pressures. Market-oriented growers had a higher on-farm weed pressure than crop-growth–oriented growers. It was concluded that studies on weed management behavior and the effect of that behavior can lead to a better understanding of farming systems and to more effective weed management in those systems.