Graduation date: 1978

Short-term changes in the distribution, surface to 10 meters, of phytoplankton-associated trace metals (Fe, Mn, Cr, Zn, Co, Ni, and Cu) were studied in the near-shore waters off Humboldt Bay, California, through the summer of 1971. The depth distribution of phytoplankton-associated trace metals was related to local hydrography. During periods of upwelling higher concentrations of phytoplankton-associated Fe, Mn, Cr, and Zn were found at depth (10 meters) than at the surface. During periods of non-upwelling there was less stratification in their distribution. Phytoplankton-associated Cu, Go, and Ni showed no significant increase with depth during either period. The rate of change in trace metal distribution patterns of Fe, Mn, Cr, and Zn were regulated by surface wind patterns and hydrographic regimes. Again, Cu, Co, and Ni did not fluctuate in a significant fashion. The elements studied tended to fall into three ordered groups based on their behavior with respect to 1) depth distribution, 2) correlation with other metals studied, and 3) correlation with biomass indicators. The groupings were Fe, Mn, and Cr; Zn, Co, and Ni; and finally Cu. This ordering is similar to the Irving- Williams series and to the elements relative 'hardness" as a Lewis acid where Fe(III), Cr(III), and Mn(II) are considered hard Lewis acids while Zn(II), Co(II), Ni(II) and finally Cu(II) are considered borderline hard to soft Lewis acids. These properties are related to the relative affinities and selectivities of the elements for organic ligands. There also appears to be some relation between the physical (dissolved versus particulate form) and the chemical (oxidation state and ionic potential) behavior of the elements in sea water and their behavior in the phytoplankton. There was a negative correlation between most phytoplankton-associated trace metal concentrations (Fe, Mn, Cr, Zn, Co, and Ni) and biomass during non-upwelling periods and a positive correlation during upwelling periods. Copper, on the other hand, showed a positive correlation during both periods. This aberrant behavior of copper may be related to its high affinity for organic ligands and its relative softness as a Lewis acid. Since no simple negative correlation was found between phytoplankton biomass and phytoplankton-associated trace metal concentrations, biological dilution does not appear to be as important as the hydrographic regime and the chemical activities of the element in determining the concentrations of trace metals in near shore phytoplankton populations. / Graduation date: 1975

Graduation date: 1971

Thesis (M.S.)--Oregon Graduate Center, 1980.

Thesis (Ph. D.)--Oregon Graduate Center, 1979.

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references. Also available in electronic version. Access restricted to campus users.

Thesis (M. Phil.)--University of Hong Kong, 1987.

Thesis (M. Phil.)--University of Hong Kong, 1977.

Trace element compositions, expressed as ratios relative to Ca (Mg/Ca, Sr/Ca and Ba/Ca), in drip water and modern speleothem calcite were measured at multiple sites within a single cave system (Natural Bridge Caverns) in central Texas. These measurements are used to investigate how water and calcite compositions respond to changes in climate. Drip water trace element ratios respond to changes in climate and in soil, vadose zone and/or in-cave processes, which are in turn influenced by climate. It is commonly assumed that speleothem calcite directly reflects these changes in the drip water composition. To test this assumption, this study quantifies the partitioning of trace elements into speleothem calcite in a natural cave setting. To determine the controls on calcite trace element ratios, empirical partition coefficients (K [subscript D]) for Mg, Sr, and Ba are measured using a unique time series of water and modern calcite geochemistry at two drip sites. One drip site, dominantly supplied by conduit flow, has relatively invariant calcite trace element compositions that reflect correspondingly small variations in drip water chemistry. A second drip site, supplied by a combination of conduit and diffuse flow, exhibits a seasonal change in drip water composition due to changes in cave-air CO₂ concentrations. The drip water seasonality at this site is recorded in the calcite trace element compositions; however the partitioning of Mg/Ca from drip water to calcite is not controlled by the same mechanism(s) that control Sr/Ca and Ba/Ca partitioning. Results of this study indicate that the partitioning of Mg changes with drip water Mg concentration, temperature and location of the calcite relative to the point of drip impact. Calcite Sr/Ca and Ba/Ca ratios are more strongly influenced by changes in cave-air CO₂ that cause changes in CO₂ degassing, affecting calcite precipitation. This element specific partitioning complicates the interpretation of speleothem trace element records, since calcite Mg/Ca and Sr/Ca may not always covary even at drip sites that experience PCP. Although there is a strong correlation between Mg/Ca and Sr/Ca ratios in the drip water at these sites, there is considerable scatter between these two values in calcite. Average K [subscript D] values at both sites fall within the range of previous theoretical and empirical studies and are 0.025, 0.12 and 0.15 for Mg, Sr and Ba respectively. It is important to note that not all trace element partitioning is controlled by the same mechanism, since this has implications for interpretations of hydrologic processes from speleothems. / text

The trace element content of clinkers (and possibly of cements) can be used to identify the manufacturing factory. The Mg, Sr, Ba, Mn, Ti, Zr, Zn and V content of clinkers give detailed information for the determination of the origin of clinkers produced in different factories. However, for the analysis of such complex data there is a need for algorithmic tools for the visualization and clustering of the samples. This paper proposes a new approach for this purpose. The analytical data are transformed into a twodimensional latent space by factor analysis (probabilistic principal component analysis) and dendograms are constructed for cluster formation. The classification of South African clinkers is used as an illustrative example for the approach.