| Front page | | Contents | | Previous | | Next |

Effekter af sprøjtemiddelafdrift på bærbærende buske og træer som indikator for biodiversitetsforandringer

Summary and conclusions

It is known from studies in arable fields that the density and biodiversity of weeds, insects, and birds are affected by the use of pesticides. However, pesticides do not only affect species within the fields, but also organisms in adjacent habitats like hedgerows, field edges, and forest lots. The effect of spray drift on trees and shrubs is not known.

This project uses hawthorn as a representative of hedgerow trees as it is commonly found in Danish hedgerows, and supports a wide range of insect species. It is known from English studies that hawthorn berries are an important food item for birds.

Aim of the project

We want to study and quantify the effects of herbicide spray drift onto perennial berry-producing plants. In order to design such an experiment optimally it is necessary to know the variability of the selected response variables. This measure is used in a statistical power analysis in order to calculate the necessary sample size.

Therefore, the primary aim of the project is to assess the necessary sample size in order to document statistically a specified effect level on the production of leaf biomass and reproductive endpoints (i.e. flowers and berries). These two endpoints represent the availability of food for herbivorous insects and birds.

In addition the pilot experiments should establish the effects of herbicide treatment on hawthorn at the point of berry production. A controlled spray trial with the herbicide metsulfuron assessed the effects on the berries.

Finally insects associated to the hawthorn hedgerows were collected throughout the season.

All results were used in the planning of an experiment to establish the possible consequences of herbicide spray drift on biodiversity.

Results

The statistical power analyses showed that samples needed were highly variable for different endpoints (leaves, buds, flowers, green berries and mature berries). For leaves, buds and flowers it was possible to obtain a high statistical power with relatively few trees, even in a single hedgerow. For juvenile berries it was also possible to limit the set-up to only one hedgerow. Mature berries were the most variable endpoint and therefore the most resource demanding. A summary of the results of statistical power analyses is presented in Table 1.

Table 1
Summary of the minimum number of samples needed for a statistical power of at least 80% when the number of hedgerows, developmental stage of the hawthorn and the response variable is given. The figures are the result of the statistical power analyses performed in Chapter 2.

   
Economy

We wanted to have the highest confidence in the conclusions drawn from a new experiment. Therefore we aimed to sample as many hedgerows as practically and economically feasible. Consequently, we calculated the necessary resources needed to measure a 20% reduction with a significance level of 0.05 and a statistical power of 80% for both types of response variables.

The calculation showed that sampling a single hedgerow would be much more time consuming than collecting samples from several hedgerows because sampling then should be made from a high number of trees. The resulting design is presented in Table 1 and the cost (hours) associated with this design is presented in Table 2.

Table 2.
Cost (man-hours) associated with the design presented in Table 1.

  
Spraying experiment

It was found that the herbicide metsulfuron caused a clear negative effect on hawthorn when the plant was exposed to dosages between 10 and 40% of the recommended field rate. The dose – response relationship can be described by means of the following equation:

Proportion of berries present = 0.287 – 0.0067 x Dose

where ‘Dose’ is the percentage of the recommended field rate, and ‘Proportion of berries present’ is the relative amount of berries remaining on the tree 22 days after spraying. The present dose – response relationship was established on the basis of counting both berries attacked by insects and undamaged berries. Figure 1 (below) shows that the results are highly dependent on whether the berries were attacked.

Figure 1
Relationship between the spray dose and the proportion of berries remaining on hawthorn 22 days after spraying with metsulfuron.

This experiment shows us that:

	Effects on berry production can be expected at dosages of 10 percent of the recommended field rate and higher
	The herbivore impact is highly important for the assessment of the results

Insects

Twenty-six species of insects were collected. In particular three of these are interesting due to their effect on hawthorn and because there were large numbers of them. Both Anthonomus pedicularius and A. sorbi are weevils that attack flowers and early berries in May and June. It is not known to what extent these insects provoke abortion of berries; however, our data show that there is a high loss of flowers. Another weevil, Rhynchites aequatus, attacks the green berries and as a result the berries are lost.

The pilot experiment shows that there is a number of insect species, which damage early flowers and berries. The different species attack at different times of the season. Therefore it is necessary to sample buds, flowers and berries throughout the season. Furthermore lost units should be sampled in order to assess the proportion damaged by insects.

| Front page | | Contents | | Previous | | Next | | Top |