Summary

Udvaskning af glyphosat ved juletræsproduktion på lerjord

The aim of the project was to investigate the possible leaching of glyphosate and the degradation product AMPA when a glyphosate based herbicide was used for growing Christmas trees on clayey soil. In the project, two experimental plots were established at a field site at Slæggerup, Sealand, Denmark. This area had previously been use for traditional farming. Nordmann fir (Abies nordmanniana, also known as Caucasian fir) was planted on this farmland and two growth strategies were compared:

Conventional treatment where weed was controlled by spraying the area with a glyphosate formulation ("P" plot)
Reduced pesticide application and use of mechanical weed control ('mulching', "M" plot)

In addition, a control plot was established, where the trees were left to grow without treatment of any kind after planting. The leaching of the pesticide glyphosat and its transformation product AMPA was monitored for three growth seasons (May 2004 to October 2006).

Glyphosate and AMPA residue analysis was made on soil samples. Further, water samples from horizontal wells as well as from drainage systems were used for analysis of leaching. The quality of the trees was monitored throughout the study to evaluate the effect of management practise on the trees, and to assure that the treatments use would produce trees usable for selling as Christmas trees. The quality of the trees from each treatment was compared and related to the quality of the trees from the untreated control. Also, a soil variability study at field scale was made, including measurements of soil intrinsic parameters and microbial diversity in three depths below the soil surface. Laboratory experiments were integrated in the project for characterisation of processes such as sorption and residue transport in soil columns. The main conclusion of the project can be summarised as follows:

In a few water samples from drainage content of glyphosate and AMPA was detected at levels below the residue limit of 0.1 µg/l (in 3 of 11samples from P and 1of 15 from M)
Residuals of glyphosate and AMPA was found in the upper soil layers below the Christmas trees on the P as well as the M area
Glyphosate or AMPA was not detected in water samples from the horizontal wells 3m below the surface
The Christmas tree quality was affected by the treatment as conventional pesticide treatment (P) resulted in a better quality when compared to the quality of the trees grown under the alternative treatment conditions (M)
The alternative primarily mechanical based treatment (M) could be used for production of Christmas trees on clay soil but it was anticipated that another growth season would be required for the trees to be fully developed

Glyphosate and AMPA

Overall it was concluded that the use of glyphosate for the production of Christmas trees on clay soil was associated with low risk for leaching. In this context it should be noted that the application doses used were within the approved range but above what is normally applied in the Christmas tree production. The glyphosate and AMPA contents detected in 4 out of 26 samples from drain water were at a low concentration level, and the drain water volume corresponding to these samples was small in comparison to the annual drainage volume (less than 1 %). Thus, the amount that is leaching through the drainage system is limited.

In water samples from horizontal wells no content of glyphosate of AMPA was detected. Content of glyphosate and AMPA was observed when analysing water samples from the drainage systems below plot P as well as plot M. In all samples, the content was below the residue limit of (0.1 µg/l):

In 2005 the findings were: Glyphosate (0,074 µg/l) and AMPA (0.025 µg/l) from drainage systems relating to plot P and glyphosat (0,049 µg/l) and AMPA (0.015 µg/l) from M
In 2006 content was observed in drain water samples from the P plot only: Glyphosate was detected at two occasions (0.018 and 0.011 µg/l), and a single observation of AMPA was made at 0.016 µg/l

In soil samples from both experimental plots (P and M) residuals of glyphosate and AMPA were detected and an overall tendency to higher AMPA than glyphosate content was observed. After two growth seasons with two applications on the M plot and five pesticide applications on the M P plot 1,5 and 5,4 kg active compound was applied pr. hectare respectively. Throughout the experiment the average level of residue content present in the soil below the P plot varied between 15 % and 65% of the pesticide amount applied. In soil from the M plot a higher but also more uncertain fraction of the applied pesticide could be recovered from the soil samples. The uncertainty related to the measurements made on the M soil samples was related to both the lower amount applied under this treatment and field variability. At the end of the field experiment the pesticide recovered from soil samples corresponded to 34 % and 69 % of the active compound applied to the P and M plot respectively.

Overall, the findings from the field experiment demonstrated that no significant leaching of pesticide or AMPA could be found in the drainage system or water from horizontal wells whereas glyphosate as well as AMPA could be detected in soil below both P and M plots, and findings were made throughout a longer period. Due to a strong binding to soil matrix components the contents of glyphosate and AMPA observed in the soil may be less assessable for microbial degradation. Consequently, there is a possibility that the soil content of glyphosate and AMPA may be present in the soil for a prolonged period of time. The dissipation time and residue levels identified in the field experiment may be considered high, but they are at the same level as measured in other projects in Flakkebjerg and Hvidovre (both are Danish locations). The residue levels measured was higher for AMPA than glyphosate and this is likely to be related to higher sorption of AMPA in the soil matrix and to degradation of glyphosate.

Summing up on the residue analysis and comparing to the use of glyphosate pesticide in more conventional farming the results of the present study did not indicates that glyphosate use for weed control in cultures of Abies nordmanniana on loamy sites should cause increased leaching of glyphosate or AMPA. In this context it is noteworthy that rather high residue levels of both glyphosate and AMPA was found in soil samples, and that these findings were made over a prolonged time period. However, due to the variations in the contents detected the quantitative aspects are to be interpreted with caution. Comparing the results of the present project with studies made elsewhere it is clear that differences do exist, and care must be taken when using data from other regions for prediction of the pesticide fate in Danish soils. Overall, the results of the study indicate that the residuals detected in the soil are related to the combination of soil and pesticide properties rather than the crop and the growth strategy.

Christmas tree quality and management

In relation to tree quality and managing procedures it was observed that during the three years the conventional, herbicide based treatment (P) resulted in increasingly better height, clearly improved needle colour when compared to the primarily mechanical treated plants (M). The differences in needle colour between trees from the two treatments were related to pronounced differences in concentration and content of nitrogen in the needles. This indicates a difference in nitrogen uptake under the two growth conditions investigated. The two treatments did not cause differences for any of the other nutrients measured. In general, the content of nutrients in the needles was within the optimal range for all trees on the experimental site.

Significant differences in growth of the trees were observed when comparing trees from the P and M plots. The results indicates that the reduced pesticide treatment implemented at plot M may cause a need for at least one additional growth season to produce trees at a suitable height, and possibly the trees would be of reduced quality causing a increase in production cost.