The Use of Chemical Element Fingerprints to Investigate the Transformation of Lake Sediments to Land Soils in Drying Lakes: A Case Study at Lake Mead, Nevada
Abstract
Drying lakes are a worldwide catastrophe. The goals of this dissertation research were (1) to investigate whether chemical fingerprints of lake sediments, and nearby land soils, could be established by the analysis of 44 chemical elements, moisture content, Eh, pH, and leachable anions and if so, (2) whether the fingerprints could help explain the process of physical and chemical changes in the lake sediments as they became soils when the lake's water level fell and the anoxic sediments were exposed to the air and weather.
Lake Mead was chosen to model drying lakes because its water level has fallen over 150 feet from full pool, stranding acres of former lake sediments to be transformed into land soils. Since the lake was formed, it has experienced three cycles of drying and refilling. This history complicates the chemical fingerprint, but has yielded a new sediment dating method. The research was conducted at Crawdad Cove in Lake Mead, Nevada, from 2012 to 2018. Six soil or sediment cores were collected at three locations along each of two transects on the east and west sides of Crawdad Cove Road at Lake Mead. One core was at the water’s edge and two spaced equidistant going inland from the water’s edge towards the historical full lake water elevation line. Each core was sliced vertically in up to 50 samples. Data were analyzed using ICPAES, ICPMS, XRF, alpha spectrometry, and ion chromatography to examine element concentrations, percent moisture, and anion concentrations and Principal Component Analysis (PCA) to process ten thousands of the experimental results.
Most chemical elements, including Mg, Al, P, S, Ca, V, Cr, Ni, Cu, Zn, Rb, Sr, Y, Zr, Ba, La, Ce, Nd, Pb, Th, and U, had their highest concentrations in the submerged sediment core and lower concentrations in the inland soil cores. Higher concentrations of F-, Cl-, and SO42- are found in the water’s edge than the farthest inland and they are strongly correlated to each other. HCO2- and NO3 are higher in the farthest inland core than at the water’s edge. Sulfur by ICP and sulfate by ion chromatography have a strong correlation, as they should.
The results of PCA showed that: (1) every cores had an unique fingerprint; and (2) rather than showing a consistent pattern of change in the cores along each transect based on distance from shoreline, the patterns of chemical fingerprint in the cores reflected changes in elevation. The change in chemical fingerprints in the cores is most closely associated with time the sediments have been out of the water. The elevation is the marker of lake sediments’ wetting/drying history due to the changes in lake level.
The results of this analysis indicate that there is a clear change in chemical fingerprint as drying lake sediments become soils. The most likely explanation for the chemical fingerprint change is that it reflects changes in the redox conditions from reducing to oxidizing which creates soluble element species. Other explanations are possible if the sediment composition changes over time. This could occur due to factors such as wind erosion or deposition that would change surface material composition, water erosion or deposition of the surface material and vegetation growth in the mudflats after lake water recedes that affects uptake and cycling of chemical elements. Investigation of such factors would be an interesting future extension of this analysis.
Core dating or calculating sedimentation rates was done by two methods: one by an established Pb-210 dating and a new method using the historical records of lake water levels and the ratios of the natural abundances of the elements divided by the concentrations of the elements in mid-inland cores from the two transects. Results of these two dating methods agree closely, 0.6 cm/year for the Pb-210 method and 0.7 cm/year for the natural abundance ratio method. This suggests that the historical records and ratio method can substitute well for the Pb-210 dating method. It also suggests that the historical record of sediment input to the lake is somewhat preserved since Lake Mead was created in 1935.