Nonparametric functional data estimation applied to ozone data: prediction and extreme value analysis

The study of extreme values and prediction of ozone data is an important topic of research when dealing with environmental problems. Classical extreme value theory is usually used in air-pollution studies. It consists in fitting a parametric generalised extreme value (GEV) distribution to a data set of extreme values, and using the estimated distribution to compute return levels and other quantities of interest. Here, we propose to estimate these values using nonparametric functional data methods. Functional data analysis is a relatively new statistical methodology that generally deals with data consisting of curves or multi-dimensional variables. In this paper, we use this technique, jointly with nonparametric curve estimation, to provide alternatives to the usual parametric statistical tools. The nonparametric estimators are applied to real samples of maximum ozone values obtained from several monitoring stations belonging to the Automatic Urban and Rural Network (AURN) in the UK. The results show that nonparametric estimators work satisfactorily, outperforming the behaviour of classical parametric estimators. Functional data analysis is also used to predict stratospheric ozone concentrations. We show an application, using the data set of mean monthly ozone concentrations in Arosa, Switzerland, and the results are compared with those obtained by classical time series (ARIMA) analysis.     

Immunosuppression in the infaunal bivalve Scrobicularia plana environmentally exposed to mercury and association with its accumulation

This study aimed to test the hypothesis whether mercury (Hg) activates or suppresses inappropriately the immunity of the bivalve Scrobicularia plana inhabiting a Hg contaminated area (Laranjo basin, Ria de Aveiro, Portugal). Immunity endpoints, as well as lipid peroxidation (LPO) as a sign of damage, were evaluated in parallel with total Hg burden. Bivalves from both moderately (MO) and highly (HI) contaminated sites displayed higher haemolymph Hg load and reduced plasma agglutination. Increased haemocytes density and decreased phagocytosis were observed at HI, whereas increased oxidative burst activity (OBA) was observed at MO, pointing out that the immunotoxicity is a result of Hg direct contact involving no ROS intervention. OBA observed at MO was concomitantly associated to peroxidative damage as depicted by LPO increase in haemocytes and haemolymph plasma. Thus, S. plana can be suggested as a suitable bioindicator of metal pollution in coastal areas on the basis of Hg bioaccumulation and immunotoxicity responses.     

Comparing the sensitivity of soil invertebrates to pesticides with that of Eisenia fetida

The sole routine testing of the standard earthworm Eisenia fetida for the terrestrial risk assessment of pesticides has been under much debate since other soil invertebrates may be more sensitive than this standard test species. However, the very low availability of laboratory toxicity data for taxa other than E. fetida has greatly hampered sensitivity comparisons. In the present study, the relative tolerance (T_rel) approach was used to enable comparing toxicity thresholds obtained from the US-EPA ECOTOX database, for main terrestrial taxonomic groups and pesticidal types of action (insecticides, fungicides, herbicides, and other) separately. Analyses confirmed previously reported lower and higher sensitivity of collembolans to fungicides and insecticides, respectively. However, various other discrepancies in susceptibility relative to E. fetida were encountered as indicated by species sensitivity distributions and/or calculated 95% confidence intervals of T_rel values. Arachnids and isopods were found to be more sensitive to insecticides, and nematodes to fungicides, as compared to E. fetida. Implications of study findings for the terrestrial risk assessment of pesticides are discussed.     

Plant communities in relation to flooding and soil contamination in a lowland Rhine River floodplain

Using canonical correspondence analysis (CCA), relationships were investigated between plant species composition and flooding characteristics, heavy metal contamination and soil properties in a lowland floodplain of the Rhine River. Floodplain elevation and yearly average flooding duration turned out to be more important for explaining variation in plant species composition than soil heavy metal contamination. Nevertheless, plant species richness and diversity showed a significant decrease with the level of contamination. As single heavy metal concentrations seemed mostly too low for causing phytotoxic effects in plants, this trend is possibly explained by additive effects of multiple contaminants or by the concomitant influences of contamination and non-chemical stressors like flooding. These results suggest that impacts of soil contamination on plants in floodplains could be larger than expected from mere soil concentrations. In general, these findings emphasize the relevance of analyzing effects of toxic substances in concert with the effects of other relevant stressors.     

Integrated analysis of the effects of agricultural management on nitrogen fluxes at landscape scale

The integrated modelling system INITIATOR was applied to a landscape in the northern part of the Netherlands to assess current nitrogen fluxes to air and water and the impact of various agricultural measures on these fluxes, using spatially explicit input data on animal numbers, land use, agricultural management, meteorology and soil. Average model results on NH₃ deposition and N concentrations in surface water appear to be comparable to observations, but the deviation can be large at local scale, despite the use of high resolution data. Evaluated measures include: air scrubbers reducing NH₃ emissions from poultry and pig housing systems, low protein feeding, reduced fertilizer amounts and low-emission stables for cattle. Low protein feeding and restrictive fertilizer application had the largest effect on both N inputs and N losses, resulting in N deposition reductions on Natura 2000 sites of 10% and 12%, respectively.     

Farm, land, and soil nitrogen budgets for agriculture in Europe calculated with CAPRI

We calculated farm, land, and soil N-budgets for countries in Europe and the EU27 as a whole using the agro-economic model CAPRI. For EU27, N-surplus is 55 kg N ha⁻¹ yr⁻¹ in a soil budget and 65 kg N₂O–N ha⁻¹ yr⁻¹ and 67 kg N ha⁻¹ yr⁻¹ in land and farm budgets, respectively. NUE is 31% for the farm budget, 60% for the land budget and 63% for the soil budget. NS values are mainly related to the excretion (farm budget) and application (soil and land budget) of manure per hectare of total agricultural land. On the other hand, NUE is best explained by the specialization of the agricultural system toward animal production (farm NUE) or the share of imported feedstuff (soil NUE). Total N input, intensive farming, and the specialization to animal production are found to be the main drivers for a high NS and low NUE.     

Cocaine and other illicit drugs in airborne particulates in urban environments: a reflection of social conduct and population size

Levels of cocaine and other psychoactive substances in atmospheric particulate matter (PM) were determined in urban environments representing distinct social behaviours with regard to drug abuse: night-life, university and residential areas. Three cities (with population >1 million and <0.3 million inhabitants) were selected. Mean daily levels of drugs in PM were 11-336 pg/m³ for cocaine, 23-34 pg/m³ for cannabinoids, and 5-90 pg/m³ for heroin. The highest levels were recorded on weekends, with factors with respect to weekdays of 1-3 for cocaine, 1-2 for cannabinoids and 1.1-1.7 for heroin. Higher levels were detected in the night-life areas, pointing towards consumption and trafficking as major emission sources, and possibly ruling out drug manufacture. The similarities in temporal trends at all sites suggested a city-scale transport of psychoactive substances. Correlations were detected between cocaine and amphetamine consumption (r² = 0.98), and between heroin and cannabinoids (r²>0.82).     

Percolation and batch leaching tests to assess release of inorganic pollutants from municipal solid waste incinerator residues

In this study, percolation and batch leaching tests were considered in order to characterize the behaviour of air pollution control (APC) residues produced in a municipal solid waste incinerator (MSWI) as a function of the liquid to solid ratio (L/S). This waste is hazardous, and taking into account their physical and chemical properties, leaching of contaminants into the environment is the main concern. In our work the leaching behaviour of toxic heavy metals (Pb, Zn, Cr, Ni and Cu) and inorganics associated with soluble salts (Na, K, Ca and Cl) was addressed. Although pH of the leaching solution is the most important variable, L/S may also play an important role in leaching processes. In our work, results from column and batch tests were compared in terms of concentration (mg/L) and releasing (mg/kg). The APC residues revealed to be hazardous according to both tests, and both Pb and Cl⁻ far exceeded the regulatory thresholds. The material exhibits high solubility, and when the liquid to solid ratio was high, more than 50% can be solubilised. The patterns of release may be in some cases availability or solubility controlled, and the former was easier to identify. When the results from column and batch experiments were compared by representing the cumulative released amounts (in mg/kg) as a function of L/S, both curves match for Zn, Ni, Cu, K, Na, Cl and Ca, but for Cr and Pb a significant difference was observed. In fact, the column experiments revealed that under percolation conditions it should be expected slow releasing of Pb along time. From this study, it can be concluded that the released amounts obtained in batch experiments for a certain L/S should be considered as the worst case for medium term. Some simple models proposed on the literature and based on local equilibrium assumption showed good fitting to experimental data for soluble species (non-reactive solutes).     

Assessment of organochlorine pesticide residues in Atlantic Rain Forest fragments, Rio de Janeiro, Brazil

A superficial water quality survey in a watershed of the Paraíba do Sul River, the main water supply for the most populated cities of southeastern Brazil, was held in order to assess the impact of the expansion of agricultural activity in the near border of the Atlantic Rain Forest. The aim of this study was to investigate the presence of priority organochlorine pollutants in soils and superficial waters of Atlantic rainforest fragments in Teresópolis, Rio de Janeiro State. Soil sample preparations were compared by using ultrasound, microwave assisted extraction and Soxhlet extraction. Recoveries of matrix spiked samples ranged from 70 to 130%. Analysis of a certified soil material showed recoveries ranging from 71 to 234%. Although low concentrations of organochlorine residues were found in water and soil samples, this area is of environmental importance and concern, thus demanding a monitoring program of its compartments.     

Application of QuEChERS method and gas chromatography-mass spectrometry for the analysis of cypermethrin residue in milk

Analytical methods for the isolation and determination of cypermethrin in milk, based on the solid-phase microextraction (SPME) and QuEChERS methods (Quick, Easy, Cheap, Effective, Rugged, and Safe) are presented. The SPME technique was not appropriate to analyse cypermethrin in milk, even establishing the best extraction conditions, polydimethylsiloxane fiber, 60 min time extraction, 60 °C temperature extraction, addition of salt (NaCl) and stirring rate. The extraction efficiency was low probably because of the matrix constituents. The QuEChERS method involves the extraction of the analyte with acetonitrile and simultaneous liquid-liquid partitioning formed by adding anhydrous MgSO(4) plus NaCl, followed by the removal of residual water and cleanup using a procedure called dispersive solid-phase extraction, in which anhydrous MgSO(4) plus PSA and C18 are mixed with 1 mL of acetonitrile extract. The detection and quantification limits were 0.01 and 0.04 mg kg(-1), respectively, and the percentage recovery obtained ranged from 92 to 105% with relative standard deviations below 7%.