No residential use of OPs was reported by participants. Our assumption that 100% of absorbed OP dose was excreted as urinary diethyl and dimethyl metabolites may underestimate dose, as approximately 20% of the OPs used in the study area do not metabolize to any of the DAP metabolites . Furthermore, the OPs that do devolve into DAP metabolites are not excreted entirely as DAP metabolites within 24 h, as they may be excreted in other biological media and as non-DAP urinary metabolites . Factors such as the route of exposure may also impact the proportion of parent OPs excreted as DAPs, with previous studies finding a higher recovery for oral versus dermal exposures . Another limitation is that we did not administer a comprehensive dietary assessment. We asked mothers to state whether their child had consumed any fruits or vegetables in the previous day. Compared to a more rigorous Food Frequency Questionnaires , our assessment may have underestimated dietary exposures. Moreover, the USDA PDP program publishes food residue data from food samples acquired from across the country without regard to region of origin. Employing these data inherently assumes participants consumed fruits and vegetables with similar exposure profiles of produce sold throughout the U.S. It is possible that participants from an agriculture region are more likely to consume locally grown produce, resulting in exposure profiles that may or may not reflect those sold nationwide. For example, we observed that dose estimates based solely on nearby agricultural pesticide use were significantly higher than dietary dose estimates,growing raspberries in container in part due to the higher proportion of exposure from more toxic pesticides such as oxydemeton methyl and disulfoton in PUR dose estimates.
If specific OPs that were sprayed locally in this time frame were also present to a higher degree on locally consumed produce, our use of national food residue data may have underestimated dietary dose estimates. When determining the proportion of exposure to attribute to diet, we chose to incorporate data from an organic diet intervention study in a similar population of children living in Salinas and Oakland, CA in 2006 . Various studies, including other intervention studies that have observed decreases in DAP concentrations from 70 to 89% among children and adults following an organic diet intervention , suggest that diet is the primary source of OP exposure among children in non-agricultural areas . However, longitudinal studies of children living in agricultural and suburban areas in Washington State suggest that DAP concentrations may vary temporally and that diet may not necessarily be the primary source of OP exposure among agricultural children during spray seasons . Furthermore, the overall interpretation regarding the predictive power of FMV and non-FMV spots remained consistent between main analyses and sensitivity analyses in which we varied the proportion of exposure from diet. Additional studies are needed to disentangle the proportion of exposure from diet, nearby agricultural pesticide use, and other sources among children living both in agricultural and non-agricultural regions. Regardless of the proportion of exposure assigned to each source, our overall conclusions that non-FMV spots may underestimate exposure remain the same. Two primary goals of restoring natural habitat are to conserve biodiversity and restore ecosystem functions and services . Agriculture is the world’s largest land use and constitutes a principle driver of biodiversity loss, increased homogenization and decreased ecosystem services .
Agricultural lands also constitute much of the matrix that surrounds protected patches of natural habitat. Managing this matrix both to provide resources for species in these patches and to improve connectivity among patches is perhaps the most important current task for biodiversity conservation . While restoring habitat within agricultural areas might enhance species abundances in the matrix or promote movement through the matrix, such schemes are thought to primarily promote common and resilient species and thus provide few conservation benefits for species of concern . Such species, it is thought, are likely to have specific functional traits like high mobility and generalist habits that permit them to survive even in intensive agricultural landscapes . Thus, trait composition could be used to assess whether restoration simply bolsters populations of such species or, alternatively, promotes species that are sensitive to habitat loss, fragmentation and degradation . Here, we examine how restoration of native plant hedgerows in an intensive agricultural setting influences the response trait composition of flower visitor communities, as an indicator of conservation effectiveness of this technique. For bees and syrphid flies, two dominant groups in many flower visitor communities , abundance, body size, specialization in diet or microhabitat, and sociality are response traits that are sensitive to land-use changes and might, therefore, differentiate flower visitor communities in response to restoration . Abundance was the single most important trait influencing persistence in a study of bees and flies , while population size, but not habitat area, was related to persistence in a solitary bee species . Diet specialization was associated with sensitivity to land-use changes for both bees and syrphid flies .
Microhabitat specialization also influenced flower visitor response to land-use changes. In flies, Schweiger et al. found that larval habitat specialists were most sensitive to land-use changes. In bees, several studies found that cavity nesters were more affected by land-use changes , as are above-ground nesters that either used existing cavities or excavated their own nests . Body size, sociality and parasitism displayed conflicting responses to land-use changes in different studies. Body size is a proxy for mobility in bees and flies . Larger-sized individuals may be more resilient to land-use changes because they can disperse further through inhospitable landscapes in search of resources. However, large-bodied species may also have larger resource needs and smaller population sizes, reducing their resilience to land-use changes. These opposing tendencies may explain the wide variation found in the responses of body size to land-use changes which include non-significant for bees , significant positive for bees and significant negative for bees and flies . Social bees responded more strongly to land-use changes than solitary bees in several studies , but others found no effect of sociality or effects that varied by bee family . Cleptoparasitic bees, which are generally specialized on their hosts and are considered to occur at a higher trophic level because they feed on the nest provisions and/or larvae of other bees , were found to be more sensitive to land-use changes than non-parasitic bees in one study , but less sensitive in another . These discrepancies among studies may reflect not only true differences among study systems, but also methodological differences, such as coding of qualitative traits. While many studies have examined how traits of flower visitor communities change as communities disassemble in response to land-use changes , only a few studies have used a trait-based approach to examine how restoration influences the reassembly of flower visitor communities . If restoration in intensive agricultural landscapes merely promotes common generalist species, then we would expect to see increases in mean occurrence of species between restored and unrestored sites,raspberry container size but no relative increases in the occurrence of species that are more sensitive to land-use changes. Here, we present results from a long-term restoration study. Specifically, we examine how restoration of native plant hedgerows within an intensive agricultural landscape in California’s Central Valley influences species occurrences of bees and flies and how these effects on species occurrences are modulated by response traits. We predict that hedgerows promote species more sensitive to land-use changes and thus will disproportionately increase occurrence of species that have some or all of the following response traits: less abundant, narrow larval and/ or adult diet breadths, cavity-nesting bees, large body size for bees , small body size for flies and parasitic bees. We predict no difference in sociality for bees, however, since in our study system, some social bees are least responsive to agricultural intensification , whereas others are most sensitive to agricultural intensi- fication . If hedgerows promote species with these response traits disproportionately relative to controls, then hedgerows may be partially reversing the community disassembly that has occurred in response to agricultural intensification in this region .Our study landscape, located in the Central Valley of California , is an intensively managed agricultural landscape comprised principally of conventional row crops, vineyards and orchards .
The 1-km buffers around our sites contained on average <0 6 0 2% natural habitat cover; thus, these areas are examples of ‘cleared landscapes’ . We utilized a before-after control-impact design to assess the impact of hedgerows on pollinator communities, as recommended for evidence-based assessment of conservation and agri-environment management schemes . We selected five farm edges to be restored and paired these with 10 control sites that would not be restored. As recommended, we selected a larger number of controls than restoration sites . Monitoring began in 2006 prior to restoration and continued through 2013. Hedgerows were planted in 2007 or 2008 with native perennial shrubs and trees . Hedgerows are approximately 350 m long and 3–6 m wide, bordering an irrigation ditch or slough and adjacent to large crop fields. After initial planting, hedgerows were irrigated and weeded for 3 years until well-established . Control sites were selected to roughly match conditions surrounding paired restoration sites, including adjacency to an irrigation ditch or slough and similar crop system , within the same landscape context . Controls reflect the variety of potential conditions on edges of crop fields that could be restored . Such edges may at times be tilled, treated with pesticides or left alone; plants on these edges include predominantly non-native forbs and grasses, with occasional shrubs and trees. The most common flowering plants at these sites are the non-native weeds: Convolvulus arvensis, Brassica spp., Lepidium latifolium, Picris echioides and Centaurea solstitialis. Many of these weeds also occurred at restoration sites. We sampled flower visitor communities at each site a minimum of three times between April and August each year, except for two sites which were sampled only twice in the first year . For logistical reasons, no sampling was conducted in 2010. In each sample round, sites were sampled in random order during allowed weather conditions, which were bright overcast to clear skies, wind speed <2 5ms 1 , temperature >21 °C. Beginning in the morning, all flower visitors that contacted the reproductive parts of the flower were netted along a 350-m transect for 1 h, pausing the timer while handling specimens and recording the plant species on which each specimen was collected. Honeybees were not collected because their abundance is determined largely by the placement of hives throughout the region by bee-keepers. Here, we focus our analyses on the two most abundant and effective wild pollinator groups in the data set: bees and syrphid flies . Bee specimens were identified to species or morpho-species by expert taxonomist Dr. Robbin Thorp , and syrphid specimens were identified to species by expert taxonomist Dr. Martin Hauser .Qualitative traits for bees included sociality, nesting location and nesting habit. Following Burkle, Marlin & Knight , we classified bees as social , solitary or cleptoparasitic, based on Michener . Following Williams et al. , we classified nesting location as above- or below-ground or mixed and nesting habit as constructing a nest or using a pre-existing cavity. Nesting location was based on Krombein et al. , Michener , Cane, Griswold & Parker , Sheffield et al. , and nesting habit was based on Michener . Cleptoparasitic bees were not scored for nesting habit since they do not collect pollen or construct nests. For flies, we assessed the type of larval diet , but dropped the latter two classes because they were utilized by only one species each. Fly traits were provided by taxonomists Dr. Martin Hauser and Dr. Francis Gilbert . Quantitative traits for bees and flies included mean body size, abundance and floral resource specialization. We used intertegular distance for bees and wing length for flies as proxies for mobility , measuring from one to five specimens under a dissecting microscope. We calculated floral resource specialization and abundance, using not only the data from this study, but also data collected in the same study area on an additional 56 hedgerow and control sites using identical sampling methods during the same sample years .