Date of Award

1990

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Oceanography and Coastal Sciences

First Advisor

William H. Patrick, Jr

Abstract

Palynological and carbon-13 analyses were studied to determine their practicality as tools for paleoecological reconstruction of marsh vegetation communities of the Mississippi delta plain. Both were calibrated on marsh surface sediment samples from four vegetation zones within Barataria Basin, Louisiana: fresh, intermediate, brackish and salt marsh. Samples were taken concurrent to vegetation sampling to allow comparison of sediment data to plant biomass and clip plot data from each of the four zones. The $\partial\sp{13}$C of sedimentary carbon reflects the proportion of C-3 and C-4 species at a site (which corresponds to a salinity gradient in the basin). The average from all sites within each wetland type is $-$27.8, $-$22.1, $-$16.9 and $-$16.2$\perthous$, for fresh, intermediate, brackish and salt marshes. This method can distinguish among the fresh, intermediate and brackish vegetation zones. A seasonal study of pollen carried by Mississippi River was conducted to determine the impact of river floodwaters on a marsh pollen assemblages. River water assemblages vary seasonally and reflect pollen sources from without the drainage basin. Types abundant during high river stage, indeterminate grains (assumed to present reworked material), Pinus, Quercus, Taxodium, TCT, Ambrosia and Chenopodiaceae-Amaranthaceae type, are assumed to be over-represented in marshes subject to river flooding. Each of the four vegetation zones has a characteristic pollen assemblage. Assemblages are not affected by over-representation of local sources, but appear to reflect the composition of the entire vegetation zone (extra-local sources). Classification of modern pollen assemblages by discriminant functions is highly successful (94% correct classification rate) even when common river-borne taxa are excluded from the analysis. When applied to buried marsh deposits (ca. 2,000 yr B.P.) classification results from both techniques generally agree indicating that $\partial\sp{13}$C values are not significantly shifted by diagenetic processes in the sediments. Comparison of information provided by both techniques, however, also demonstrates that important analogues (i.e., progradational marshes) are missing from the suite of modern samples.

Pages

147

DOI

10.31390/gradschool_disstheses.4903

Download

Share

COinS