Frieder Hofmann

Maize pollen deposition in relation to distance from the nearest pollen source under common cultivation - results of 10 years of monitoring (2001 to 2010)

BackgroundInformation on pollen dispersal is essential for the risk assessment and management of genetically modified organisms (GMOs) such as Bt maize. We analyzed data on maize pollen deposition at 216 sites in Germany, Switzerland, and Belgium from 2001 to 2010. All data were collected using the same standardized sampling method. The distances between sampling site and the nearest maize field ranged from within the field to 4.45 km.ResultsMaize pollen deposition was negatively correlated with distance from the nearest pollen source. The highest pollen deposition was within the field, but depositions of several thousand pollen grains per square meter were recorded over the kilometer range. A power function model most accurately described the relationship between deposition and distance from the nearest pollen source, rather than the exponential model currently used in EU risk assessment and management, which underestimates exposure for distances greater than 10 m. Regression analysis confirmed the high significance of the power relationship. The large variation in pollen deposition at a given distance reflected the influences of wind direction and other meteorological and site conditions. Plausible variations of single values and the predicted mean pollen count at a given distance were expressed by confidence intervals.ConclusionsThe model described here allows estimations of pollen deposition in relation to distance from the nearest field; therefore, it will be valuable for the risk assessment and management of GMOs. Our results indicate that buffer zones in the kilometer range are required to prevent harmful exposure of non-target organisms to GMOs.

A New Method for in Situ Measurement of Bt-Maize Pollen Deposition on Host-Plant Leaves.

Maize is wind pollinated and produces huge amounts of pollen. In consequence, the Cry toxins expressed in the pollen of Bt maize will be dispersed by wind in the surrounding vegetation leading to exposure of non-target organisms (NTO). NTO like lepidopteran larvae may be affected by the uptake of Bt-pollen deposited on their host plants. Although some information is available to estimate pollen deposition on host plants, recorded data are based on indirect measurements such as shaking or washing off pollen, or removing pollen with adhesive tapes. These methods often lack precision and they do not include the necessary information such as the spatial and temporal variation of pollen deposition on the leaves. Here, we present a new method for recording in situ the amount and the distribution of Bt-maize pollen deposited on host plant leaves. The method is based on the use of a mobile digital microscope (Dino-Lite Pro, including DinoCapture software), which can be used in combination with a notebook in the field. The method was evaluated during experiments in 2008 to 2010. Maize pollen could be correctly identified and pollen deposition as well as the spatial heterogeneity of maize pollen deposition was recorded on maize and different lepidopteran host plants (Centaurea scabiosa, Chenopodium album, Rumex spp., Succina pratensis and Urtica dioica) growing adjacent to maize fields.

Prioritizing stream types according to their potential risk to receive crop plant material — A GIS-based procedure to assist in the risk assessment of genetically modified crops and systemic insecticide residues

Crop plant residues may enter aquatic ecosystems via wind deposition or surface runoff. In the case of genetically modified crops or crops treated with systemic pesticides, these materials may contain insecticidal Bt toxins or pesticides that potentially affect aquatic life. However, the particular exposure pattern of aquatic ecosystems (i.e., via plant material) is not properly reflected in current risk assessment schemes, which primarily focus on waterborne toxicity and not on plant material as the route of uptake. To assist in risk assessment, the present study proposes a prioritization procedure of stream types based on the freshwater network and crop-specific cultivation data using maize in Germany as a model system. To identify stream types with a high probability of receiving crop materials, we developed a formalized, criteria-based and thus transparent procedure that considers the exposure-related parameters, ecological status--an estimate of the diversity and potential vulnerability of local communities towards anthropogenic stress--and availability of uncontaminated reference sections. By applying the procedure to maize, ten stream types out of 38 are expected to be the most relevant if the ecological effects from plant-incorporated pesticides need to be evaluated. This information is an important first step to identifying habitats within these stream types with a high probability of receiving crop plant material at a more local scale, including accumulation areas. Moreover, the prioritization procedure developed in the present study may support the selection of aquatic species for ecotoxicological testing based on their probability of occurrence in stream types having a higher chance of exposure. Finally, this procedure can be adapted to any geographical region or crop of interest and is, therefore, a valuable tool for a site-specific risk assessment of crop plants carrying systemic pesticides or novel proteins, such as insecticidal Bt toxins, expressed in genetically modified crops.

Procedure to select test organisms for environmental risk assessment of genetically modified crops in aquatic systems

For a long time, the environmental risk assessment (ERA) of genetically modified (GM) crops focused mainly on terrestrial ecosystems. This changed when it was scientifically established that aquatic ecosystems are exposed to GM crop residues that may negatively affect aquatic species. To assist the risk assessment process, we present a tool to identify ecologically relevant species usable in tiered testing prior to authorization or for biological monitoring in the field. The tool is derived from a selection procedure for terrestrial ecosystems with substantial but necessary changes to adequately consider the differences in the type of ecosystems. By using available information from the Water Framework Directive (2000/60/EC), the procedure can draw upon existing biological data on aquatic systems. The proposed procedure for aquatic ecosystems was tested for the first time during an expert workshop in 2013, using the cultivation of Bacillus thuringiensis (Bt) maize as the GM crop and 1 stream type as the receiving environment in the model system. During this workshop, species executing important ecological functions in aquatic environments were identified in a stepwise procedure according to predefined ecological criteria. By doing so, we demonstrated that the procedure is practicable with regard to its goal: From the initial long list of 141 potentially exposed aquatic species, 7 species and 1 genus were identified as the most suitable candidates for nontarget testing programs. Integr Environ Assess Manag 2017;13:974-979.