scope:

This guideline should be used to monitor butterfly and moth fauna when monitoring the environmental effects of genetically modified organisms in accordance with the specifications of the European Directive 2001/18/EC and Council Decision 2002/811/EC.

The main objective is to identify potentially significant changes in the inventory or abundance of butterfly and moth communities over an extended period of time. This VDI guideline defines methodological standards for monitoring the species inventory and the individual abundances of butterfly and moth imagines and larvae. The need to monitor butterfly and moth fauna will depend on the GMO being monitored. The standardisation of monitoring methods ensures a high level of reproducibility and comparability of the recorded data. The methods must be practical and efficient, require a justifiable level of effort and costs, and generate data suitable for statistical analysis. If possible, the chosen approach to sampling and site selection should enable conclusions to be drawn about the potential cause-effect relationship of a GMO effect, or failing that, should enable the deduction of specific hypotheses about the effects of GMOs on the lepidopteran fauna, which can then be tested subsequently. The sampling design described here can also be transferred to other groups of organisms.

This guideline deals with butterflies and moths (Lepidoptera) as a whole. It describes the required monitoring periods and methods to be applied to the respective areas, whilst taking into account the different monitoring requirements of diurnal and nocturnal Lepidoptera and their imaginal and larval stages. It makes recommendations about site selection and design, including statistical power analysis. The use of this guideline should ensure that regional and transregional effects in different, relevant habitats are recorded by using representative sampling procedure.

A number of fundamental difficulties are associated with farming practices and the possible distribution of GM and GM-free areas. In contrast to a planned scientific field trial, site selection for a monitoring programme cannot be made on the basis of the comparability of variants alone; the fields chosen by farmers for the cultivation of GM crops are the starting points. On this basis, laboratory-like conditions are very difficult to achieve and unfeasible for realistic monitoring strategies.

A key component of this guideline is to address this fundamental problem. When using this guideline, (environmental) variability can be reduced and the explanatory power improved by applying the following recommendations, with due consideration to the effort involved:

• Paired comparisons between similar GM and GMfree areas are conducted to minimise interference from heterogeneous environmental factors.

• The general setting of the paired areas must be checked including a documentation of this assessment.

• A variety of monitoring methods (light trap, transect method, visual larvae search, beating method) are applied to obtain a comprehensive illustration of the various relevant aspects of the data recorded.

• A sufficiently accurate and efficient monitoring strategy is aimed for (number of visits/light trap days, e. g. two-week monitoring cycle compared with a weekly cycle as used by Butterfly Monitoring Germany (TMD), for example).

• Crambid snout moths will also be recorded when performing butterfly transects. This increases the recorded species spectrum specifically in agricultural areas without significant additional effort and expense. Monitoring of species communities and their abundances as a broad-based method is combined with larval surveys of selected focus species for a more detailed approach.

• The explanatory power of the number of replications (number of paired comparisons) is estimated by means of a prospective power analysis to provide transparent information on the statistical confidence generated by the invested effort.

• Various different statistical methods to improve the explanatory power are put forward to ensure that the best possible information is obtained from the data collected, depending on the data structure.

Compliance with these recommendations will result in the best possible treatment of the subject currently available by employing a reasonable level of effort and expenditure to optimise the explanatory power of the results whilst taking into account the practical constraints of field monitoring.