Pesticides are the dominant stressors for vulnerable insects in lowland streams.

Despite elaborate regulation of agricultural pesticides, their occurrence in non-target areas has been linked to adverse ecological effects on insects in several field investigations. Their quantitative role in contributing to the biodiversity crisis is, however, still not known. In a large-scale study across 101 sites of small lowland streams in Central Europe, Germany we revealed that 83% of agricultural streams did not meet the pesticide-related ecological targets. For the first time we identified that agricultural nonpoint-source pesticide pollution was the major driver in reducing vulnerable insect populations in aquatic invertebrate communities, exceeding the relevance of other anthropogenic stressors such as poor hydro-morphological structure and nutrients. We identified that the current authorisation of pesticides, which aims to prevent unacceptable adverse effects, underestimates the actual ecological risk as (i) measured pesticide concentrations exceeded current regulatory acceptable concentrations in 81% of the agricultural streams investigated, (ii) for several pesticides the inertia of the authorisation process impedes the incorporation of new scientific knowledge and (iii) existing thresholds of invertebrate toxicity drivers are not protective by a factor of 5.3 to 40. To provide adequate environmental quality objectives, the authorisation process needs to include monitoring-derived information on pesticide effects at the ecosystem level. Here, we derive such thresholds that ensure a protection of the invertebrate stream community.

Surrogate for Electron Beam Inactivation of Salmonella on Pumpkin Seeds and Flax Seeds.

ABSTRACT: This study aimed to identify a suitable nonpathogenic surrogate for industrial validation of irradiation process by high-energy electron beam (5 MeV) of dried seeds. Pumpkin seeds (Cucurbita pepo var. styriaca) and golden flax seeds (Linum usitatissimum) were contaminated with a five-strain Salmonella cocktail comprising five serovars or a two-strain Escherichia coli cocktail comprising pathogenic strains, including E. coli O157:H7. Comparison of log survival fractions of the E. coli and Salmonella cocktails revealed that on both types of seeds, the Salmonella cocktail exhibited higher tolerance against high-energy electron beam at doses of 4 kGy than the E. coli cocktail, with a log survival fraction of -4.1 ± 0.7 compared with -6.0 ± 0.2 on pumpkin seeds and -4.7 ± 0.7 compared with reduction from 1.8 × 108 CFU/g to below the limit of detection (1 × 102 CFU/g) on flax seeds. For surrogate selection, the Salmonella cocktail and the strains E. coli DSM 18039 (strain MG1655) and Enterococcus faecium NCCB 86023 (strain NRRL B-2354) were subjected to electron beam processing at doses of 2 to 6 kGy. The calculated D10-values of the Salmonella cocktail were not significantly different (P > 0.05) from those of E. coli DSM 18039, i.e., 1.07 ± 0.10 kGy compared with 1.20 ± 0.07 kGy on pumpkin seeds and 0.88 ± 0.04 kGy compared with 1.07 ± 0.03 kGy on flax seeds. E. faecium NCCB 86023 exhibited significantly higher tolerance on pumpkin seeds (3.07 ± 0.18 kGy) and on flax seeds (2.22 ± 0.29 kGy), ∼3 log and 2 log higher than the Salmonella cocktail, respectively. Hence, the nonpathogenic E. coli DSM 18039 is suggested to serve as a surrogate for Salmonella in industrial validation trials. Because on both types of seeds E. faecium NCCB 86023 showed significantly higher tolerance against electron beam than the Salmonella cocktail, this nonpathogenic strain could serve as a process control indicator for the decontamination of dried seeds by electron beam.

Predicting low-concentration effects of pesticides.

We present a model to identify the effects of low toxicant concentrations. Due to inadequate models, such effects have so far often been misinterpreted as random variability. Instead, a tri-phasic relationship describes the effects of a toxicant when a broad range of concentrations is assessed: i) at high concentrations where substantial mortality occurs (LC50), we confirmed the traditional sigmoidal response curve (ii) at low concentrations about 10 times below the LC50, we identified higher survival than previously modelled, and (iii) at ultra-low concentrations starting at around 100 times below the LC50, higher mortality than previously modelled. This suggests that individuals benefit from low toxicant stress. Accordingly, we postulate that in the absence of external toxicant stress individuals are affected by an internal "System Stress" (SyS) and that SyS is reduced with increasing strength of toxicant stress. We show that the observed tri-phasic concentration-effect relationship can be modelled on the basis of this approach. Here we revealed that toxicant-related effects (LC5) occurred at remarkably low concentrations, 3 to 4 orders of magnitude below those concentrations inducing strong effects (LC50). Thus, the ECx-SyS model presented allows us to attribute ultra-low toxicant concentrations to their effects on individuals. This information will contribute to performing a more realistic environmental and human risk assessment.