GLYPHOSATE, OTHER HERBICIDES, AND TRANSFORMATION PRODUCTS IN MIDWESTERN STREAMS, 20021

ABSTRACT: The use of glyphosate has increased rapidly, and there is limited understanding of its environmental fate. The objective of this study was to document the occurrence of glyphosate and the transformation product aminomethylphosphonic acid (AMPA) in Midwestern streams and to compare their occurrence with that of more commonly measured herbicides such as acetochlor, atrazine, and metolachlor. Water samples were collected at sites on 51 streams in nine Midwestern states in 2002 during three runoff events: after the application of pre-emergence herbicides, after the application of post-emergence herbicides, and during harvest season. All samples were analyzed for glyphosate and 20 other herbicides using gas chromatography/mass spectrometry or high performance liquid chromatography/mass spectrometry. The frequency of glyphosate and AMPA detection, range of concentrations in runoff samples, and ratios of AMPA to glyphosate concentrations did not vary throughout the growing season as substantially as for other herbicides like atrazine, probably because of different seasonal use patterns. Glyphosate was detected at or above 0.1 μg/1 in 35 percent of pre-emergence, 40 percent of post-emergence, and 31 percent of harvest season samples, with a maximum concentration of 8.7 μg/1. AMPA was detected at or above 0.1 μg/1 in 53 percent of pre-emergence, 83 percent of post-emergence, and 73 percent of harvest season samples, with a maximum concentration of 3.6 μg/1. Glyphosate was not detected at a concentration at or above the U.S. Environmental Protection Agency's maximum contamination level (MCL) of 700 μg/1 in any sample. Atrazine was detected at or above 0.1 μg/1 in 94 percent of pre-emergence, 96 percent of post-emergence, and 57 percent of harvest season samples, with a maximum concentration of 55 μg/1. Atrazine was detected at or above its MCL (3 μg/1) in 57 percent of pre-emergence and 33 percent of post-emergence samples.     

The impact of disturbance and ensuing forestry practices on Collembola in monitored stands of windthrown forest in the Tatra National Park (Slovakia)

Soil Collembola communities were investigated in spruce forest stands of the High Tatra Mts that had been heavily damaged by a windstorm in November 2004 and subsequently by a wildfire in July 2005. The study focused on the impact of these disturbances and forestry practices on collembolan community distribution and structure 4 years after the disturbance. Four different treatments were selected for this study: intact forest stands (REF), non-extracted windthrown stands (NEX), clear-cut windthrown stands (EXT) and burnt windthrown stands (FIR). From a total of 7,820 individuals, 72 species were identified. The highest total abundance mean was recorded in FIR stands followed by NEX and EXT stands and, surprisingly, the lowest in REF stands. The highest total species richness was observed in REF stands, followed by NEX stands and FIR stands and the lowest in EXT stands. In REF and NEX stands, the most abundant species were Folsomia penicula and Tetracanthella fjellbergi, while in heavily damaged stands, the most abundant was Anurophorus laricis. The ordination method used demonstrated a significant influence of treatment on the abundance of Collembola. ANOVA used confirmed significant differences for all dominant species between treatments. The present study shows the negative impact of windthrow on Collembola communities as reflected in decreased species richness and abundance. However, disturbance by fire caused a considerable increase in collembolan abundance 3 years after the event. Moreover, we show that clearing of windthrown spruce forests after a windstorm is less favourable for communities of soil collembolans and slows down the recovery process.     

Do national advisories serve local consumers: an assessment of mercury in economically important North Carolina fish

Consumption of marine fish provides both benefits (lean protein, omega-3 fatty acids and essential nutrients) and risks (main source of mercury (Hg) exposure for humans). Mercury is a potent neurotoxin and the source of more fish advisories nationwide than any other toxicant. Despite the widespread nature of Hg, it is unknown whether local Hg contamination reflects national and regional levels often used as bases to inform consumers of potential fish consumption risk. Thus, the objectives of our study were to examine Hg levels of six commonly consumed marine species harvested locally off the North Carolina coast and to compare our results to published regional (Monterey Bay Aquarium's Seafood Watch List) and national (Environmental Protection Agency, EPA, and Food and Drug Administration, FDA) Hg averages, action levels, and guidelines. We found significant differences in Hg concentrations among collected species, and we identified correlations between Hg concentration and fish length and trophic levels. Collected mahi mahi and triggerfish were below the EPA fish tissue action level (0.3ppm). Wahoo and grouper exceeded the EPA action level but were below the FDA action level (1.0ppm). King mackerel had the highest Hg concentration among targeted species, exceeding both EPA and FDA action levels. Further, our local results were not always consistent with calculated averages from EPA and FDA databases for the same species, and although many of our findings were consistent with Monterey Bay Aquarium's Seafood Watch List (southeast region), recommendations based on Hg levels would conflict with recommendations they provide based on sustainability. We find regional and national averages are not always reflective of local Hg contamination and suggest local data may be needed to accurately assess consumer risk.     

Phytoestrogens and mycotoxins in Iowa streams: an examination of underinvestigated compounds in agricultural basins

This study provides the first broad-scale investigation on the spatial and temporal occurrence of phytoestrogens and mycotoxins in streams in the United States. Fifteen stream sites across Iowa were sampled five times throughout the 2008 growing season to capture a range of climatic and crop-growth conditions. Basin size upstream from sampling sites ranged from 7 km2 to > 836,000 km2. Atrazine (herbicide) also was measured in all samples as a frame-of-reference agriculturally derived contaminant. Target compounds were frequently detected in stream samples: atrazine (100%), formononetin (80%), equol (45%), deoxynivalenol (43%), daidzein (32%), biochanin A (23%), zearalenone (13%), and genistein (11%). The nearly ubiquitous detection of formononetin (isoflavone) suggests a widespread agricultural source, as one would expect with the intense row crop and livestock production present across Iowa. Conversely, the less spatially widespread detections of deoxynivalenol (mycotoxin) suggest a more variable source due to the required combination of proper host and proper temperature and moisture conditions necessary to promote Fusarium spp. infections. Although atrazine concentrations commonly exceeded 100 ng L(-1) (42/75 measurements), only deoxynivalenol (6/56 measurements) had concentrations that occasionally exceeded this level. Temporal patterns in concentrations varied substantially between atrazine, formononetin, and deoxynivalenol, as one would expect for contaminants with different source inputs and processes of formation and degradation. The greatest phytoestrogen and mycotoxin concentrations were observed during spring snowmelt conditions. Phytoestrogens and mycotoxins were detected at all sampling sites regardless of basin size. The ecotoxicological effects from long-term, low-level exposures to phytoestrogens and mycotoxins or complex chemicals mixtures including these compounds that commonly rake place in surface water are poorly understood and have yet to be systematically investigated in environmental studies.