Date of Award

May 2019

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Earth Sciences

Advisor(s)

Christopher A. Scholz

Keywords

deconvolution of grain size distribution, image processing, Lake Bosumtwi, lake turbidites, skew normal distribution, tropical climate change

Subject Categories

Physical Sciences and Mathematics

Abstract

Sedimentary archives contain rich climate signals from the Earth’s past. These signals are preserved and encrypted in different properties of sedimentary deposits, and can be proxies for local and regional climate characteristics. The objectives of this dissertation research are to explore new ways and methods of extracting and interpreting these climatic signals that are preserved in sedimentary archives. Sample sets analyzed in these studies were collected from scientific drill cores recovered from Lake Bosumtwi, Ghana, as part of a drilling campaign supported by the U.S. National Science Foundation and the International Continental Scientific Drilling Program.

Laminations in sedimentary profiles can represent not only seasonal variability and longer rhythms, but their properties such as thickness and color can also be proxies for environmental and depositional processes. However, identification and measurement of abundant laminae in sedimentary records have been manual processes that were extremely labor intensive, subjective, and error-prone. The first part of this dissertation project is to develop a new method that can automatically identify and measure laminae and their properties using sedimentary core images. A software system named LA-2D was developed for this objective and is comprised of four major components: 1) image enhancement that includes noise reduction and contrast enhancement to improve the signal-to-noise ratio; 2) identification of 1-D laminae; 3) lamina connectivity analyses of the 1-D laminae to obtain a lamina stratigraphy; and 4) extraction and retrieval of the primary and derived lamination stratigraphic data. Sediment core images from Lake Hitchcock and Lake Bosumtwi were used for algorithm development and testing. Our experiments show a complete match between laminae produced by the software and manual process for images from Lake Hitchcock, and an insignificant discrepancy in the number of laminae identified by the software and determined manually. Results also revealed that over 90% of mismatches were less than one pixel, between the software and the manual method for the experimental images from Lake Bosumtwi. A partial result of this work was published in Computers and Geoscience (Gan and Scholz, 2013).

The second chapter of my research is to extract climatic signals from clastic grain size distribution data. This research included a complete cycle of grain size analyses starting from sample collection from sedimentary core archives, sample preparation, sample analysis, and data processing and interpretation. A partial result of this research was published in the Journal of Sedimentary Research (Gan and Scholz, 2017). This study presents a new nonparametric method called the skew normal distribution deconvolution (SNDD) that parses a sample GSD into one or more component subpopulations, each of which can be fully quantified by a parameter set (C_i,ξ_i,ω_i^2 〖,α 〗_i) of a skewed normal distribution (SND). A key advance of this method is that each SND has a set of rich statistical parameters (different mean and median, characteristic skewness and kurtosis) that can be used to characterize sediments and depositional environments. Application of this method to 530 samples from Lake Bosumtwi reveals seven disjointed subpopulation groups in the mean grain size distribution (MGSD) field, and each of their volume frequencies can be fit to a normal distribution. There is one dominant subpopulation with the median grain size of 11.8 µm, and six minor subpopulations with median grain sizes of 0.86 µm, 7.61 µm, 24.4 µm, 60.1 µm, 129.8 µm, and 439.6 µm. They provide indications that seven discrete depositional processes and/or environments may be responsible for depositing these clastic grains over 94,500 years of deposition in Lake Bosumtwi.

The third chapter of this dissertation measures, analyzes and interprets temporal and thickness data of lake turbidites observed in scientific drill cores from Lake Bosumtwi, in conjunction with carbon and nitrogen data from organic matter of bulk sediments to assess their climatic implications. A total of 315 layers of flood-induced turbidites (hyperpycnites) were identified in drill cores covering the past 70,000 years. Their temporal distribution and thickness reveal eleven major millennial-scale alternations of humid and arid climate. Rhythmic analysis of unevenly spaced time series indicates that distribution of turbidites has a cyclicity of ~13,500 years with Lomb-Scargle significance above 0.5 for their thickness distribution and 0.05 for their occurrence frequencies in 100 year bins. This periodicity is not similar to any previously identified cyclic modulation of global climate variation. Accordingly, it could be a unique aspect of low-latitude West African climate that was impacted by both hemispheres. Total organic carbon (TOC) and total nitrogen (TN) composition of bulk sediments, and δ13C composition of organic matter were measured and examined at different scales and resolutions. Anomalously low TOC and TN were observed over two intervals: a mega-drought event between 74.4 ka and 68.4 ka (TOC ranged 0.39% to 0.46%), and an arid period from 51.9 ka to 47.6 ka (TOC = 2.3% to 2.78%). TOC, δ13C, and C/N values consistently suggested a long, arid climate from 100.0 ka to around 79.2 ka, during which o.m. were clearly of lacustrine origin. TOC, TN, mean grain size and C/N values showed distinct signatures of mega-drought events from 74.4 ka to 68.4 ka. This is also consistent with the dust index of the Sahara Desert observed in the Ionian Sea. These proxies also showed that the sediment o.m. may have mainly been a mix of C3 and C4 plants of terrestrial origin, with some lacustrine o.m. deposited between mega-drought and Younger Dryas intervals, which implied a frequent alternation of arid and humid climate in this period.

Access

Open Access

Download

Share

COinS