Critical evaluation of soil contamination assessment methods for trace metals.

Correctly distinguishing between natural and anthropogenic trace metal contents in soils is crucial for assessing soil contamination. A series of assessment methods is critically outlined. All methods rely on assumptions of reference values for natural content. According to the adopted reference values, which are based on various statistical and geochemical procedures, there is a considerable range and discrepancy in the assessed soil contamination results as shown by the five methods applied to three weakly contaminated sites. This is a serious indication of their high methodological specificity and bias. No method with off-site reference values could identify any soil contamination in the investigated trace metals (Pb, Cu, Zn, Cd, Ni), while the specific and sensitive on-site reference methods did so for some sites. Soil profile balances are considered to produce the most plausible site-specific results, provided the numerous assumptions are realistic and the required data reliable. This highlights the dilemma between model and data uncertainty. Data uncertainty, however, is a neglected issue in soil contamination assessment so far. And the model uncertainty depends much on the site-specific realistic assumptions of pristine natural trace metal contents. Hence, the appropriate assessment of soil contamination is a subtle optimization exercise of model versus data uncertainty and specification versus generalization. There is no general and accurate reference method and soil contamination assessment is still rather fuzzy, with negative implications for the reliability of subsequent risk assessments.

Measurement instability and temporal bias in chemical soil monitoring: sources and control measures.

Concern that human impacts on the environment may be harmful to natural resources such as soils as well as to living conditions is the major motivation for long-term environmental monitoring. However, the evidence that measurement bias is not constant through time affects time series as an artifact; this also holds true for chemical soil monitoring. Measurement instabilities occur along the whole measurement chain, from soil sampling to the expression of results. The first step in controlling measurement instability is to identify its relevant sources, and the second is to control it by stabilizing, minimizing, or quantifying measurement instability. For all five steps in the measurement process, from soil sampling to the expression of the analytical results, sources of measurement instability are identified and measures of control discussed, leading to the main conclusion concerning the requirement to continuously control the relevant environmental and measurement boundary conditions that may affect measurement instability. The innovative aspect of this paper consists in explicitly addressing measurement instability in chemical soil monitoring and tracking it along the whole measurement chain. The paper is also a plea for a change of paradigm in long-term environmental monitoring, namely to consider temporal measurements as unstable unless their degree of stability is traceably demonstrated, adequately quantified, and included in interpretation.

The role of metadata and strategies to detect and control temporal data bias in environmental monitoring of soil contamination.

It is crucial for environmental monitoring to fully control temporal bias, which is the distortion of real data evolution by varying bias through time. Temporal bias cannot be fully controlled by statistics alone but requires appropriate and sufficient metadata, which should be under rigorous and continuous quality assurance and control (QA/QC) to reliably document the degree of consistency of the monitoring system. All presented strategies to detect and control temporal data bias (QA/QC, harmonisation/homogenisation/standardisation, mass balance approach, use of tracers and analogues and control of changing boundary conditions) rely on metadata. The Will Rogers phenomenon, due to subsequent reclassification, is a particular source of temporal data bias introduced to environmental monitoring here. Sources and effects of temporal data bias are illustrated by examples from the Swiss soil monitoring network. The attempt to make a comprehensive compilation and assessment of required metadata for soil contamination monitoring reveals that most metadata are still far from being reliable. This leads to the conclusion that progress in environmental monitoring means further development of the concept of environmental metadata for the sake of temporal data bias control as a prerequisite for reliable interpretations and decisions.

PAH and PCB in soils of Switzerland--status and critical review.

The surface soil concentrations (0-20 cm) of the Swiss soil monitoring network (NABO) with 105 observation sites representing all major land use types ranged for the sum of 16 EPA PAH (PAH(16)) from 32 to 8465 microg kg(-1) (median 163 microg kg(-1)), for benzo[a]pyrene (BaP) from 0.5 to 1129 microg kg(-1) (median 13 microg kg(-1)) and for the sum of seven IRMM PCB (PCB(7)) from 0.5 to 12 microg kg(-1) (median 1.6 microg kg(-1)). The legal guide values of Switzerland were exceeded for PAH(16) at only three and for BaP at two sites. The PCB(7) concentrations were clearly below any assessment value. The concentration ranges were overlapping between all land use types. Tendencies for higher concentrations were observed at urban and viticulture sites. The overall measurement precision at repeatability conditions ranged from 1 to 37% RSD for PAH(16), BaP and PCB(7). The median bias for the chemical analysis was around zero for PAH(16), +5% for BaP and -5% for PCB(7) with spreads ranging from less than -20% up to more than +30%. The PAH profiles were clearly dominated by phenanthrene. Stratification by land use revealed a prevalence of benzo[a]pyrene at urban and naphthalene at conservation sites. For PCB, the general congener rank order was PCB no. 153 > 138 > 101 > 180. From a broad correlation screening only PAH(16)/BaP (r = 0.88**) were relevant for practical soil protection. The extensive comparison with other studies was severely biased by the lack of harmonisation, especially concerning sampling depth, sampling support, analytical method and the sum calculation procedure.

Critical evaluation of PAH source apportionment tools using data from the Swiss soil monitoring network.

In this study a large dataset on the polycyclic aromatic hydrocarbon (PAH) content of Swiss soils was analysed to evaluate two source apportionment tools, i.e., characteristic PAH ratios/molecular markers and a linear mixing model. Population density and total organic carbon (TOC) content were identified by a multiple regression model as independently and positively influencing the PAH concentrations in Swiss background soil. Specifically, TOC was more strongly positively correlated with the sum of light PAH (naphthalene to phenanthrene) than with the sum of heavy PAH (anthracene to benzo[ghj]perylene), whereas population density was more strongly positively correlated with the sum of heavy PAH than with light PAH. In addition, the sum of the heavy PAH as well as the total sum correlated negatively with sample site altitude. It is therefore hypothesised that heavy PAH are less mobile, whereas light PAH were closer to equilibrium with TOC in the soil. Similar results were found for polychlorinated biphenyls (PCB). The characteristic ratios and molecular markers pointed to pyrogenic origin of PAH in Swiss background soil but did not allow for further differentiation of individual fuel contributions, even though attempts to take environmental fractionation processes into account were made. The comparison of three soil profiles identified with a linear mixing model from the pattern of 16 PAH with >300 PAH emission profiles from the literature suggested urban dust, wood combustion and binders from asphalt as PAH sources. However, also here, environmental fractionation processes probably obscured source characteristic PAH ratios and fingerprints, which thus need to be interpreted with caution.

Correlation of PCDD/F and PCB concentrations in soil samples from the Swiss soil monitoring network (NABO) to specific parameters of the observation sites.

Concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/F) and polychlorinated biphenyls (PCB) were determined in 23 soil samples collected at reference sites of the Swiss national soil monitoring network (NABO). Total PCDD/F and PCB concentrations are well correlated and proportional (R2=0.720). The total PCDD/F levels were between 72 and 703 ng/kg corresponding to 1.1-11 ng I-TEQ/kg, total PCB concentrations (sum of seven congeners IUPAC no. 28, 52, 101, 118, 138, 153, and 180) were from 1.1 to 12 microg/kg. In all samples, the PCDD/F and PCB concentrations were in the range of background levels for these contaminants in Central Europe. Ninteen samples revealed PCDD/F levels below the guide value of 5 ng I-TEQ/kg set by the Swiss ordinance relating to impacts on the soil (OIS); concentrations for the remaining four samples were below the trigger value (20 ng I-TEQ/kg). All PCB concentrations were below the guide value of 100 microg/kg. The elevated PCDD/F and PCB levels in some of the samples originating from forested sites could be attributed to the scavenging effect of forest canopies for semivolatile organic compounds. This interpretation could be substantiated by the observed distortion of the congener and homolog patterns of PCDD/F and PCB which is characteristic for forested sites.

Polycyclic aromatic hydrocarbons, black carbon, and molecular markers in soils of Switzerland.

Polycyclic aromatic hydrocarbons (PAH) were analysed in 23 soil samples (0-10 cm layer) from the Swiss soil monitoring network (NABO) together with total organic carbon (TOC) and black carbon (BC) concentration, as well as some PAH source diagnostic ratios and molecular markers. The concentrations of the sum of 16 EPA priority PAHs ranged from 50 to 619 microg/kg dw. Concentrations increased from arable, permanent and pasture grassland, forest, to urban soils and were 21-89% lower than median numbers reported in the literature for similar Swiss and European soils. NABO soils contained BC in concentrations from 0.4 to 1.8 mg/g dw, except for two sites with markedly higher levels. These numbers corresponded to 1-6% of TOC and were comparable to the limited published BC data in soil and sediments obtained with comparable analytical methods. The various PAH ratios and molecular markers pointed to a domination of pyrogenically formed PAHs in Swiss soils. In concert, the gathered data suggest the following major findings: (1) gas phase PAHs (naphthalene to fluorene) were long-range transported, cold-condensated at higher altitudes, and approaching equilibrium with soil organic matter (OM); (2) (partially) particle-bound PAHs (phenanthrene to benzo[ghi]perylene) were mostly deposited regionally in urban areas, and not equilibrated with soil OM; (3) Diesel combustion appeared to be a major emission source of PAH and BC in urban areas; and (4) wood combustion might have contributed significantly to PAH burdens in some soils of remote/alpine (forest) sites.