A new procedure for the lithic fraction characterization in marine sediments from high productivity areas: Optimization of analytical and statistical procedures

Abstract

The lithogenic fraction in marine sediments can be a key proxy to reconstruct past changes in atmospheric circulation, river runoff among other sedimentary processes. But, accurate isolation of the lithic component within biogenic-rich sediments can be particularly challenging in areas of high marine productivity. We present an experiment to isolate and characterize this lithogenic fraction that ensures the full elimination of biogenic and authigenic components. This experiment demonstrates that standard procedures used in most of the studies of terrigenous fraction in marine sediments, are inadequate to remove efficiently most of the biogenic components in sediments from high productivity regions such as the Eastern Equatorial Pacific. The experiment reviews each of the procedure steps designed for the removal of organic matter, carbonate, Fe-Mn oxides, and biogenic silica. The most critical steps were the removal of organic matter incorporated in fecal pellets, and the elimination of the biogenic silica. The continuous monitoring of the grain-size distribution of the samples along the experiment steps allows evaluating the impact that an inadequate procedure can induce on the grain size. Finally, lithic grain-size distributions were characterized using k-means statistical technique providing objective information for climate interpretation. The proposed approach ensures the quality of the results and allows magnifying the information provided by the lithic fraction in marine sediments.

Atmosphere-ocean linkages in the eastern equatorial Pacific over the early Pleistocene

“…According to the obtained modal distribution, k-means clustering analysis characterizes the grain size data set into six clusters (cluster A to cluster F), previously presented in Povea et al [2015]. Clusters A, B, and C are the most dominant patterns, reaching almost 69% of the frequency (Figure 3c).…”

“…Biogenic silica concentrations were measured in 200 mg samples of bulk sediment. The extraction method is modified from Povea et al [] and includes a two‐step leaching with a 1.5 M NaOH solution. Si concentration in both leachates was determined through inductively coupled plasma optic spectrometer (ICP‐OES), using a PerkinElmer model Optima 3200 RL at the CCiT‐UB.…”

“…The difference between these weights gives the biogenic fraction removed. Thus, the lithic percentage was obtained after subtracting the biogenic percentage from the total percentage (lithic percentage = 100% − % biogenic fraction).To aid in the interpretation of the large grain size data set, a statistical method using a k‐means clustering algorithm was applied [ Povea et al , ]. This method divides the data set into n clusters, which are represented by its centroid and formed by the data for which the centroid is the nearest [ Hastie et al , ].…”

“…Grain size analysis and mineralogical description have been conducted to characterize the lithic fraction that reached the eastern equatorial Pacific. Prior to the grain size analysis, samples were processed with a protocol specifically designed to ensure the complete removal of all the biogenic (organic matter, carbonate, and biogenic silica) and authigenic components (diagenetic oxides), leaving only the lithogenic components [Povea et al, 2015]. After this procedure, smear slides were prepared for each sample using a UV curing adhesive Paleoceanography 10.1002/2015PA002883 and dried under an ultraviolet light [Backman and Shackleton, 1983;Rothwell, 1989].…”

“…After this procedure, smear slides were prepared for each sample using a UV curing adhesive Paleoceanography 10.1002/2015PA002883 and dried under an ultraviolet light [Backman and Shackleton, 1983;Rothwell, 1989]. They were then observed with a microscope to ensure the quality of the procedure as well as to determine the percentages of the main mineralogical components of the lithic fraction [Povea et al, 2015]. Mineral abundance was estimated using the comparison chart for visual percentage estimation [Terry and Chilingar, 1955].…”