Analysis of two-way layout of count data with negative binomial variation

A number of methods has been proposed for dealing with single-factor or factorial experiments when the requirements for performing the normal theory analysis of variance procedure are not satisfied. This paper suggests the use of the likelihood ratio statistic for testing the main effects and the interaction between the factors in two-way layout of count data following negative binomial distributions with a common dispersion parameter. The likelihood ratio statistic for testing the equality of the dispersion parameters of several groups of count data is also derived. The methods is illustrated by an example concerning the study of spatial and temporal variation of bacterial counts.     

Spatial analysis and mapping the annual mean concentrations of acidity and major ions in precipitation over the United Kingdom in 1986

The composition of precipitation is being monitored in collectors sited throughout the United Kingdom. In the first year, 1986, hydrogen ion, nitrate, ammonium and non-marine sulphate were measured in samples from 47 sites and annual mean concentrations determined. Spatial analysis based on Regionalized Variable Theory revealed autocorrelation in all four ions and reasonable bounded models were fitted to the sample variogram of all four. These were then used to estimate values of a fine mesh grid by simple kriging, and the estimates were then contoured. The estimation errors were similarly mapped, and both sets of maps are presented. The maps show that concentrations of these ions are greatest in eastern parts of the country, as expected from consideration of European emissions and weather patterns. For the first time estimation errors were also mapped and show how the network could be improved.     

Organochlorine contaminants in seabird eggs from the Pacific coast of Canada, 1971–1986

Eggs were collected from seven seabird species at colonies on the British Columbia coast from 1983 to 1986 and analyzed for organochlorine contaminants. Total PCB levels (wet weight) were highest in double-crested cormorants (Phalacrocorax auritus) from the Fraser estuary (2.91 mg kg^-1) and the Strait of Georgia (3.79 mg kg^-1). Highest DDE levels were in fork-tailed storm-petrels (Oceanodroma furcata) from the Queen Charlotte Islands (1.68 mg kg^-1). Organochlorine levels were generally lower in eggs from the mid 1980s than in those collected in the early 1970s. Organochlorine levels in Pacific alcids and hydrobatids foraging in offshore locations were compared to those in the same or ecologically similar species from the Canadian Atlantic coast. DDT- and HCH-related compounds were higher in Pacific populations while levels of dieldrin, oxychlordane, and HCB were generally lower. With the exception of -HCH, levels of all measured organochlorines were lower in cormorants breeding in the Fraser River estuary than in cormorants from the St. Lawrence River estuary on the Atlantic coast.     

Robust graphical methods for diagnosing trend in irregularly spaced water quality time series

Graphical methods can play an important role in the reliable assessment of trends in typically ill behaved river quality data series both as diagnostic tools and as visual corroborative evidence when assumptions required for formal statistical tests are not met. Robust, graphically-oriented trend diagnosis procedures are presented for data series characterized by nonnormal populations, uneven time spacing, nonmonotonic trend and other factors which can create serious problems for standard parametric time series methods. Cleveland's robust locally weighted regression (RLWR) developed for investigating nonlinearity in x-y scatterplots is adapted as a robust/resistant smoothing filter for the analysis of irregular time series comprising quantitative observations. Low powered RLWR trend lines reveal temporally local phenomena, e.g. abrupt jumps (often associated with point source impacts) and periodicities, while higher powered RLWR yields smooth lines characterizing medium and longer term trends. Simple variants of Tukey smoothing concepts are developed for series with censored observations. Applications to Ontario river quality series reveal that graphical evidence is frequently sufficient to obviate the need for formal trend testing. The methods are generally applicable to most time series.     

Organochlorine and mercury residues in Swainson's and ferruginous hawk eggs collected in North and South Dakota, 1974–79

Residues of organochlorine pesticides, polychlorinated biphenyls (PCBs), and mercury were measured in eggs of Swainson's hawks (Buteo swainsoni) and ferruginous hawks (B. regalis) collected in North and South Dakota during 1974–79. DDE was the most common compound detected in the eggs, but residues were below levels known to have adverse effects on reproduction. Other organochlorine compounds and mercury were found at low levels. Eggs of ferruginous hawks tended to contain more compounds with higher residues than eggs of Swainson's hawks.     

The characteristics of size distribution of suspended particulates in the air for sulphates and selected metals in different areas and seasons

The characteristics of the distribution of particles by size were investigated for sulphates and selected metals (Pb, Fe, Mn, and Cu) in the air in two urban and an industrial area during the winter and summer periods using an Andersen cascade impactor.In the case of metals, but not sulphates, the character of the area affected the value of mass median diameter. Lowest values were obtained in a sparsely populated urban area, whereas in a densely populated urban area and in the industrial area the values were higher.In the industrial area a high correlation coefficient was found between sulphate and manganese, and between sulphate and lead, as well as a high value of the equivalent Mn²⁺/SO _inf4 ^sup2- ratio for the total sample. It may therefore be assumed that in the industrial area manganese produces a catalytic effect on SO₂ conversion to sulphate. Among the investigated metals lead stands out as a dominant cation which binds to the sulphate ion in the industrial and densely populated areas.     

An application of circular-linear correlation analysis to the relationship between Freon concentration and wind direction in Woods Hole,Massachusetts

Two statistical tests for correlation between a circular variate and a linear variate are presented. The tests are applied to a small data set concerning Freon-12 concentration and wind direction in Woods Hole, Massachusetts during the summer of 1987. A significant correlation is found. Further analysis suggests that this directional effect is related to onshore Freon-12 release.Woods Hole Oceanographic Institution Contribution Number 6750     

Exploring data quality for some inorganic and organic constituents of inhalable particulate matter (PM10)

Quality assurance (QA) assessments of air pollution data sets provide a basis for evaluating the significance of various substances in the atmosphere. For non-criteria pollutants, QA results are seldom reported in the technical literature and are often difficult to estimate. The present report provides a summary of QA results such as recovery, precision and accuracy data. Of the six trace elements and nine organic constituents compared in detail, recoveries, laboratory precision and laboratory accuracy values were 100%, ±3% to ±19% and-2 to-13% for the former group and 69% to 98%, ±7% to 23% and-15 to-34% for the later group. System precision varied from ±22% to ±47% for the trace elements and ±42% to ±83% for the organic constituents. Limitations in the interpretation of non-criteria particulate-phase pollutant data bases are discussed with some emphasis placed on receptor-modelling and risk assessment applications. Finally, the relevance of NBS certified materials for QA estimates in non-criteria air pollutant studies is also reviewed.     

Population monitoring for genetic damage induced by environmental physical and chemical agents

The short-term tests performed in vitro on different systems, from phage to human cells, or in vivo, on laboratory animals, allow only a qualitative estimate of the action of mutagenic agents, and the extrapolation of such experimental results to man may encounter many difficulties.Direct biomonitoring of populations exposed to chemicals could represent a more realistic approach for an evaluation of the hazards to man. Certain methods are still under development. Nevertheless, other ones can already by used routinely, e.g. the cytogenetic observations on peripheral blood lymphocytes and the Ames test for mutagenic substances in urine, and can provide useful suggestions how to set threshold limits for chemical substances encountered in the working environment.     

Changes in levels of organochlorines in woodcock wings from 1971 to 1975

Wings from woodcock (Philohela minor) were first monitored for organochlorine pesticides and polychlorinated biphenyls (PCBs) in the eastern U.S. in 1971. Regional differences in these compounds were clearly demonstrated and baseline residue levels were obtained for later comparisons. An expanded sampling of wings in 1972 revealed that residues in wings of adult woodcock may differ significantly from those in immatures, and that residues of several important agricultural insecticides and PCBs had declined significantly. More extensive sampling was undertaken in 1975 to determine if changes in residue levels had taken place in the intervening years. PCBs, mirex, and heptachlor epoxide increased significantly between 1972 and 1975 in adult woodcock wings. In immature woodcock wings mirex, dieldrin, and PCBs increased significantly between 1972 and 1975, but DDT residues decreased significantly.