Weeds, insect fauna, and the lack of food chains in conventional cereal fields

Summary in the original language of the document

Weeds are important as food for both herbivorous insects and farmland birds. Especially monophagous, herbivorous insects depend on the presence of one or a few weed species, and consequently a shortage of such weed species may hamper parts of the food web of the agricultural ecosystem. In any ecosystem on the earth food webs and food chains are expected, i.e. that there are species from one trophic level feeding on the lower trophic levels. These levels may be organised as organic matter – degrader, plant – herbivore and predator – prey, where preys may be degraders, herbivores and other predators. In describing ecosystems the trophic connections are often described as nodes linking species from one trophic level to another, and in most ecosystems it is possible to find a number of nodes connecting herbivores to their food plants/host plants, different kinds of organic matter to degraders, herbivores and degraders to their predators, of which some may feed on both degraders and herbivores (Dunne et al., 2002, Albrechts et al., 2007). Therefore, nodes are also expected in modern agricultural ecosystems even though these systems are highly disturbed ecosystems.
n agricultural ecosystems farmers do large efforts to control weeds in the fields and these efforts have caused a dramatic decrease in weed frequency in conventional Danish agricultural fields between the late 1960’ies and the late 1980’ies (Andreassen et al., 1996). However, this decline has to some extent been reversed during the 1990’ies and the first years after the turn of the century (Andreasen and Stryhn, 2008, Andreasen and Streibig, 2011). The decline since the l960’ies is also very likely to have been followed by a decline in arthropod density in the fields and since many arthropods are herbivores this ecosystem component theoretically should follow the weed abundance. This connection has been demonstrated by Hald and Reddersen (1990) who found a substantially higher weed biomass in organic cereal fields than in conventional ones (59 g/m2 versus 11 g/m2 in June). Equivalently, total arthropod density, diversity and biomass were higher in organic cereal fields (Hald & Reddersen 1990, Reddersen 1997), which suggests a correlation between biomass and diversity of weeds and arthropods.
Only few studies have dealt with the general relationship between the weed, crop and arthropod fauna in terms of both biomass and species composition in arable fields. In a large-scale field study of effects of reduced pesticide use on field flora and fauna, Esbjerg & Petersen (2002) found no positive correlations between total arthropod biomass and weed density, which will often be correlated to weed biomass. However, they did find positive correlations between several groups of herbivorous arthropods and weed density in both barley and wheat fields. In the extensive British study of the response of weeds and invertebrates to changes in herbicide regimes (the Farm Scale Evaluations of genetically modified herbicide-tolerant crops), herbivore densities co-varied with weed biomass (Hawes et al. 2003, Haughton et al. 2003). Strandberg & Bruus Pedersen (2002) found densities of both herbivorous and non-herbivorous arthropod groups in beet fields to correlate with weed biomass, with quadratic equations fitting better for most arthropod groups.
This apparent connection between weed density and arthropod fauna (or groups of arthropods) in arable fields suggests that it must be possible to identify the food web nodes in the field by investigating the weed flora and the arthropod fauna at the same time. Therefore, we here we present the results of an investigation of the weed flora and the arthropod fauna in terms of both biomass and species composition in conventionally managed fields in order to: 1) explore the relationship between crop, weed and arthropod biomass, and 2) investigate the food chain relationships between plants (weeds and crop) and herbivores, and 3) assess the amount of possible food items available for birds in conventional arable fields.