Toward estimation of map accuracy without a probability test sample

The time and effort required of probability sampling for accuracy assessment of large-scale land cover maps often means that probability test samples are not collected. Yet, map usefulness is substantially reduced without reliable accuracy estimates. In this article, we introduce a method of estimating the accuracy of a classified map that does not utilize a test sample in the usual sense, but instead estimates the probability of correct classification for each map unit using only the classification rule and the map unit covariates. We argue that the method is an improvement over conventional estimators, though it does not eliminate the need for probability sampling. The method also provides a new and simple method of constructing accuracy maps. We illustrate some of problems associated with accuracy assessment of broad-scale land cover maps, and our method, with a set of nine Landsat Thematic Mapper satellite image-based land cover maps from Montana and Wyoming, USA.     

Expositionsabschätzung: Human-Biomonitoring vs. Modellrechnungen

Exposure assessment represents an important and integral part of the assessment of health risks associated with the human exposure to toxic substances in soil and other environmental media. In recent times model calculations are widely used to assess the doses of toxic substances which have been incorporated into the human body via different routes of exposure. The present paper discusses the possibilities and limitations of such model calculations with regard to exposure and risk assessment. A critical point is that the results largely depend on the basic assumptions introduced into the model calculations. Moreover, the results refer to a more or less typical scenario of exposure, although they do not allow any conclusion with regard to individual exposure levels. In contrast, the use of biomarkers of exposure (human biological monitoring) represents a methodological approach to estimate individual exposure levels of subjects exposed to toxic substances in a given situation. In practice, the identification of persons at risk and the exclusion of individual health risks can only be based on human biological monitoring studies. Exposure assessment based on theoretical calculations and exposure assessment based on biomarkers of exposure and effects should be regarded as complementary approaches in exposure and risk assessment.     

Correlation of PCDD/PCDF and CO values in a MSW incinerator--indication of memory effects in the high temperature/cooling section

The correlation of PCDD/PCDF levels with the CO emissions in a full-scale municipal waste incinerator was assessed during a four-week measurement effort. PCDD/PCDF concentrations in fly ashes-containing more than 99% of the total PCDD/PCDF burden of the fluidized bed incinerator (FBI)-were measured and compared with the emitted CO concentrations. The CO concentration during the sampling time showed no significant correlation to the PCDD/ PCDF amount in fly ash (R2 = 0.078). However, a comparison of the time integrated CO concentration several hours before sampling lead to a correlation with the PCDD/PCDF burden. Maximum correlation was found for the time integrated CO values of 3 and 4 h before sampling (R2 = 0.467 and R2 = 0.457 respectively). This indicates a memory effect in the high temperature cooling section of several hours. Possible mechanisms leading to the memory effect are discussed. The correlation of PCDD/PCDF with CO concentration demonstrate that the combustion conditions play an important role for PCDD/PCDF formation in FBIs. However the variability in the correlation of CO to PCDD/PCDF levels show that other factors have a significant influence on PCDD/PCDF formation.     

Biomonitoring and environmental specimen banking

The environmental specimen banking program (ESB) of the Federal Republic of Germany implies a collection of specimens from representative areas of Germany, stored under stable conditions for deferred analysis. It is an important part of the ecological assessment program. The ecological concept is presented as the frame of the entire ESB program. It is based on the selection of representative sampling areas and specimens. Standard operating procedures (SOP’s) for sampling and characterization of specimens, as well as sampling designs specific to the areas, are presented as further important parts of the quality assurance system in relation to the correct sampling of specimens.     

Clearance of atrazine in soil describing xenobiotic behavior

The primary aim of this study was to evaluate the “clearance concept” as a tool for describing the behavior of xenobiotic movement into and through soils. As an example, degradation of 2-chloro-4-ethylamino-6-isopropylamino-s-triazine (atrazine) with the formation of metabolites 2-chloro-6-isopropylamino-s-triazine (desethylatrazine) and 2-chloro-4-ethylamino-s-triazine (desisopropylatrazine) was investigated. Atrazine was sprayed post-emergently in doses of 0.125 or 0.5 g active ingredient/m² each on four test plots. Soil type was a sandy-loam, on which corn (Zea mays L.) was cultivated. Soil samples were taken as cores of 0.2 m depth 0, 1, 2, 4, 8, 12, 16 and 20 weeks after application of atrazine, and analyzed by HPLC. Soil concentrations of atrazine were highly correlated (r=0.993, p< 0.001) between the two applications of 0.125 g/m² and 0.5 g/m². Up to 50% of the atrazine was measured as metabolites during the whole vegetation period. Clearance of atrazine from soil was calculated as the total load of atrazine divided by the area under the soil atrazine concentration time curve. Soil atrazine clearance was calculated as 5.13 +/? SD 1.10 and 5.17 +/? SD 1.02 liter of soil per day for doses of 0.125 g/m² and 0.5 g/m², respectively (from a “soil unit” of 1 × 1 × 0.2 meter). The clearance concept might be a tool for risk assessment of xenobiotics.     

The ACS division of analytical chemistry establishes home page on world wide web

Environmental Science and Pollution Research -     

Determination of selected pharmaceuticals and caffeine in sewage and seawater from Tromsø/Norway with emphasis on ibuprofen and its metabolites

Selected pharmaceuticals, among them analgesics, ss-blockers and anti-depressants as well as caffeine, the anti-bacterial triclosan and the insect repellent N,N-diethyl-3-toluamide (DEET) were determined in different sewage samples (sewage treatment plants, hospital effluents) from Troms?/Norway and in seawater from Troms?-Sound, into which the sewage is discharged. While caffeine, triclosan, ibuprofen and its major metabolites hydroxy- and carboxy-ibuprofen were present in all sewage samples, additional pharmaceuticals were observed in sewage containing hospital effluents. Concentrations were in the range of 20-293 microg/l (caffeine), 0.2-2.4 microg/l (triclosan) and 0.1-20 microg/l (sum ibuprofen + metabolites). In seawater, only caffeine (7-87 ng/l), DEET (0.4-13 ng/l) and ibuprofen + metabolites (sum concentration < LOQ-7.7 ng/l) were detected. Ibuprofen and its metabolites hydroxy- and carboxy-ibuprofen were quantified individually by use of the respective reference compounds. Relative amounts of the three compounds were determined in different types of water showing characteristic patterns, with hydroxy-ibuprofen being the major component in sewage whereas carboxy-ibuprofen was dominant in seawater samples. The patterns which were compared to those observed in similar samples from Germany indicated different transformation behaviour under limnic and marine conditions.     

Fate of 14C-bisphenol A in soils

Bisphenol A (BPA; 2,2-(4,4(')-dihydroxydiphenyl)propane) is predominantly used as an intermediate in the production of polycarbonate plastics and epoxy resins. Traces of BPA released into the environment can reach the soil via application of sewage sludge from wastewater treatment systems that receive wastewaters containing BPA, or from leachate from uncontrolled landfills. The biodegradability of BPA has been previously investigated in several studies designed to simulate surface waters and biological wastewater treatment systems. However, there is little information available about the fate of BPA in soil. Therefore, laboratory soil degradation and batch adsorption studies were conducted with 14C-BPA and four soils according to international guidelines. The soils represented a broad range of physico-chemical properties. An important result of the degradation study was that, independent of the soil type, 14C-BPA was rapidly dissipated and not detectable in soil extracts following 3 days of incubation. Based on this result, a dissipation half-life of less than 3 days was estimated. The major route of dissipation of 14C-BPA in soil was the formation of bound residues that could not be recovered by exhaustive Soxhlet extraction. 14C-BPA was also shown to be transiently converted to up to five metabolites, but within 3 days, neither 14C-BPA nor 14C-metabolites were detectable in the soils. After 120 days incubation, significant amounts (up to 20% of the radioactivity applied) of the parent compound were recovered as 14CO(2). Soil adsorption experiments indicated that the distribution coefficients (K(oc)) were between 636 and 931, classifying BPA as having low mobility for all tested soils. From the results of this study, it was concluded that if BPA reaches the soil compartment, it is not expected to be stable, mobile, or bioavailable.     

Comparison of three extraction methods for 17beta-estradiol in sand, bentonite, and organic-rich silt loam

Extraction is an important procedure for samples that contain soil because other compounds in soil may affect analysis of estrogens. This study was conducted to evaluate three different extraction methods for 17beta-estradiol in soil. Sand, bentonite, and organic-rich silt loam were spiked with 1 mg kg(-1) of 17beta-estradiol as a model compound of estrogens. 17beta-estradiol and its metabolites, estrone and estriol, were extracted using (i) a modified Bligh and Dyer extraction, (ii) a pressurized fluid extraction, and (iii) a diethyl ether extraction, and measured by liquid chromatography tandem mass spectrometry. There were significant differences in the extraction efficiency for 17beta-estradiol among the extraction methods and the soils: the efficiencies ranged from 10% to 97%. Overall, the diethyl ether extraction method had the largest efficiency of 17beta-estradiol with 45% and 57% for bentonite and silt loam, respectively. Transformation of 17beta-estradiol to estrone and estriol in the different extraction methods was less than 3.6% during the extraction procedures. This study underlined the importance of sample preparation for estrogen analysis in soil samples.     

Ecotoxicological Characterisation and Classification of Existing Chemicals

GOAL, SCOPE AND BACKGROUND: In 1998, the International Council of Chemical Associations (ICCA) launched a global initiative to investigate more than 1,000 HPV chemicals (High Production Volume, > or = 1,000 t/a) within the refocused OECD HPV Chemicals Programme. Up to the OECD SIDS Initial Assessment Meeting in April 2004 (SIAM 18) 147 ICCA dossiers (ca. 230 CAS-No) have been assessed based on a harmonised data set. The environmental profile and an ecotoxicological characterisation of these chemicals are presented here. Data for acute aquatic toxicity were correlated among each other, as well as data for fish (LC50, LD50) and rodents (LD50). The data for acute aquatic toxicity are compared with other existing chemicals. METHODS: Data of the ICCA HPV chemicals from the OECD SIAM 11-18 are presented for: log Kow (as an indicator for bioaccumulation potential), biodegradation, acute aquatic toxicity and availability of long-term toxicity data. Correlation analysis was performed with log transformed data and a linear regression model was fitted to the data, if a significant correlation was found. Acute toxicity for fish and acute oral toxicity for rodents were correlated on a molar basis. Acute aquatic toxicity of the chemicals is compared with data from BUA reports 1-234 and a random EINECS sample (Knacker et al. 1995). RESULTS AND DISCUSSION: According to the dossier information, 71 of the 147 ICCA chemicals are not 'readily biodegradable', 21 have a log Kow > or = 3, and 44 are 'toxic' (LC/EC50 < or = 10 mg/L) or 'very toxic' (LC/EC50 < or = 1 mg/L) to aquatic organisms. For 77, only the base set (acute fish, Daphnia and algae) is available, for the rest at least one long-term test (fish or Daphnia) is available and three tests for a mere 14 others. Based on the data presented, the SIAM gives recommendations for Environment and Human Health. 22 chemicals have been identified as a 'candidate for further work' for Environment and 16 for Human Health. The highest correlation coefficient was obtained correlating fish and Daphnia (r2 = 0.79). LC50 (fish) is significantly correlated with LD50 (rodent), but data are widely scattered. The correlation is not improved after transforming LC50 (fish) to LD50 (fish), using BCF QSAR. Based on acute aquatic toxicity, 25.1% of the chemicals from the BUA reports 1-234 are classified as 'very toxic' (LC/EC50 < or = 1 mg/L). This proportion is 2.5-fold higher than the ICCA HPV chemicals and 1.4-fold higher than the random EINECS sample. CONCLUSIONS: Correlation coefficients for aquatic toxicity data are rather uniform (0.57-0.79) compared with literature data, but also the best correlation was observed between fish and Daphnia. Because the scatter around the regression lines is still considerable, simple predictions of ecotoxicity between species are not possible. Correlation of LC50 (fish) and LD50 (rodent) indicates that toxicity is different. Surprisingly, the correlation of fish and rodent toxicity is not improved by transforming LC50 values to internal LD50s. The selection of ICCA chemicals by market significance (production volume) leads to a classification of toxicity, which is more comparable to a random sample of EINECS chemicals than to German BUA chemicals. The latter were chosen for concern (for Environment or Human Health). RECOMMENDATIONS AND OUTLOOK: Of 147 dossiers assessed between SIAM 11-18, ca. 75% were sponsored by the three following countries: Germany (42), USA (37) and Japan (33). The current output is about 50 dossiers per year (70-100 CAS-No), but a trend for an increase of output is noticeable. Industry, national authorities, and OECD work on a further development to speed up the output. The number of chemicals with 'low priority for further work' and the work recommended for the 'candidates' (mainly exposure assessment) indicate that the data presented were adequate for an initial hazard assessment according to OECD requirements. From the ICCA HPV list (n = 880, state of 1999) 44% of the chemicals have data available to cover all SIDS endpoints for Environment and only 33% for Human Health (Allanou et al. 1999). This indicates the importance of the Initiative to provide information on existing chemicals. The authors agree with the expectation "...that the scientific information provided by this global initiative will be considered as an internationally accepted and harmonised basis for further steps of chemicals management." (ICCA 2002 b).