Bekæmpelsesmiddelforskning fra Miljøstyrelsen, 87 – Adskillelse af effekter af herbicider og kvælstof på vegetation og leddyr i hegn og græslandsvegetation

Separate and Combined Effects of Herbicides and Nitrogen on Vegetation and Arthropods in Hedges and Grassland

Summary

This report contains the results of three years of investigation on the effects of Glyphosate drift and simulated ammonia deposition on experimental vegetation and associated anthropods. Further, the publication reports two years of investigation of vegetation and arthropods in hedges on organic and conventional farms. The aim of the work is to document and quantify separate and combined effects of glyphosate and nitrogen on vegetation and arthropods in small biotopes along fields in the agricultural landscape.

The investigation in hedges was carried out in 2001 in the southwestern part of Jutland on sandy Saalian moraines or Weichelian outwash plains and in 2002 on Weichelian moraine loamy soils. A total of 56 hedges were selected, using narrow criteria with an ideal age of 10-15 years and with rotational fields on both sides. Half of the hedges were located on organic farms (minimum 10 years of organic farming period), and the rest on conventional farms. On sandy soils a comparison was made of plots in the centre of the hedges with plots in the open strip of vegetation between the trees and the field (hedge margin). On the loamy soils, the comparison focused on the open strip of vegetation. Frequency of plants was registered and arthropods were collected by vacuum-sucking in the same plots. Soil analysis of physical and chemical characteristics were conducted.

The experimental vegetation was established in 2001 on sandy soil after deep ploughing and consisted of 30 species with different ecological attributes. Species abundance in hedges, germination ability and seed availability were additional criteria. Ten replica of 12 unique combinations of 0. 25 and 100 kg N ha^-1 yr^-1 and 0, 1, 5 and 25 % of field dose of glyphosate in plots of 49m² was established. The plots were fertilised with phosphorus, potassium, sulphur and copper. The vegetation was registered and arthropods were collected yearly with methods resembling the hedge study. Physical soil characteristics were analysed once and chemical characteristics in 2001 and 2003.

The hedge studies showed very clear differences in diversity of plants between the two farming practices. Organic hedges had 34 % more plant species than corresponding conventional hedges on sandy soils and 50 % more species on loamy soils. The differences occurred in the hedge margins. More bryophyte species, more species related to semi-natural areas (fens, meadows), more species from fields and ruderal areas were found in the organic hedges. Fewer plant species with a competitive strategy were found in the organic hedges. It was not possible to test for differences between obligatory forest species (hedges are too young) or differences in number of herbicide tolerant species. Many of the indicator species of organic hedges are declining in the agricultural landscape and organic hedges may serve as refuge for certain species.

Herbivorous arthropods generally follow plant distribution variables, and species composition differed significantly between organic and conventional hedgerows. Numbers of species per sample, number of animals per sample and Shannon-Wiener diversity index were not affected by farming practice on sandy soils, whereas significant differences were found between internal hedge plots and hedge margin plots, with significantly more arthropods in the hedge margins. On loamy soils, however, more individuals and more species of arthropods were found in organic hedges compared to conventional (average of 240 individuals and 31 species per sample in organic versus 218 individuals and 17 species per sample in conventional hedges). Mould beetle larvae and thrips were the most frequent groups on loamy soils, but also butterflies, cicadae, certain bugs, leaf beetles and adult mould beetles were common. For weevils in particular the crop of associated fields (grass ley versus other crops) had significant impact on abundance. Field crop was an important variable, which was not totally independent of farming practice, as grass ley was more frequent on organic farms.

The experiment of separating the effects of glyphosate and nitrogen revealed significant interactions on the biomass of vegetation, litter biomass, frequency of bryophytes and diversity of species. The effects of glyphosate increased by increased nitrogen level, and in general species with a competitive strategy (C) and with ruderal strategy (R) decreased strongly with increased glyphosate level, whereas plants with a Stress-tolerant (S) strategy increased as C and R decreased. Sheep fescue (Festuca ovina) showed remarkable tolerance towards glyphosate and dominated the biomass of the S strategy. Also bryophytes (colonising species) were more frequent at 25 % glyphosate and high nitrogen level compared to lower herbicide doses. With low herbicide doses, the bryophytes had higher frequencies at low N doses. Unfortunately, several common hedge species (C and R strategies) such as couch grass (Agropyron repens), Canada thistle (Circium arvense), mugwort (Artimisia vulgaris) and Common nettle (Urtica dioeca) did not establish very well in the experiment. The plant species react very differently to glyphosate, and therefore the results cannot be compared directly to the conditions in the hedges.

The number of arthropods increased from 2001 to 2002, reflecting the establishment of the vegetation. Only in a few groups significant interactions of glyphosate and nitrogen were found. Glyphosate had significant effects on numbers of several groups (several kinds of bugs, large carabid beetles, small rove beetles and mould beetles), numbers of animals and numbers of species in 2003. Significant effects of nitrogen were found for numerous groups (cicadae, several kinds of bugs, small rove beetles, rove beetle larvae and mould beetles) and for numbers of individuals and numbers of species. Numbers of species and individuals per sample were reduced at 25 % glyphosate, whereas plots with 100 kg N had more individuals and species without herbicide than other N levels. Glyphosate and Nitrogen had a significant impact on species composition of arthropods, and the species were distributed more clearly along the herbicide gradient than along the nitrogen gradient.

Glyphosate alone could reduce the number of plant species by approximately ¼, and the same reduction could be achieved by nitrogen alone. Both factors influenced significantly and with equal strength on quantitative biodiversity. Glyphosate had a stronger impact on plant community composition as compared to nitrogen. Glyphosate had a very strong impact on plant biomass at high nitrogen level, but hardly any effect on biomass at low nitrogen level. Diversity and quantity of arthropods was significantly reduced at high glyphosate level (47 and 26 %, respectively) – especially at high nitrogen level. This reflected the large change of the vegetation under these circumstances and illustrated the arthropod dependence on the vegetation.

When the result of the hedgerow study and the experiment are seen together the following main conclusions are reached. Organic hedges contain more plant species, extending the basis for herbivorous arthropods and adding more diversity to food-web of the agricultural landscape. The interactive effect of herbicide and nitrogen found in the experiment shows that also nitrogen drift into hedgerows affects hedgerow flora, and that the effect will differ between organic and conventional farms, even at similar nutrient levels.

The results of the experiment on vegetation and anthropods are comparable to the expected development in planted hedges the first three years after establishment. The hedge studies could reflect the `result' after 10-15 years of influence from conventional agriculture with pesticide drift and organic farming without pesticide drift. The study as a whole showed that absence of pesticides on organic farms gave a richer nature in the hedges. If nutrient levels could additionally be reduced, we expect even higher diversity. Field edge zones free of pesticides and fertilizer may improve the conditions for nature in the small farmland biotopes. Other technological methods reducing herbicide and fertiliser drift into hedgerows will have similar positive effects.