Particulate Trace and Minor Element Behavior in a Dust-dominated Basin: Results from the US GEOTRACES North Atlantic Zonal Transect (GA03)
Particulate Trace and Minor Element Behavior in a Dust-dominated Basin: Results from the US GEOTRACES North Atlantic Zonal Transect (GA03)
Abstract:
We present and interpret the distributions of the total particulate concentrations of 13 trace and minor elements (Al, Ba, Cd, Co, Cu, Fe, Mn, P, Pb, Th, Ti, V, Y) from the US GEOTRACES North Atlantic Zonal Transect (GA03) between Woods Hole, USA and Lisbon, Portugal. Principal components analysis (PCA) and factor analysis were used to determine the main modes of variability for these elements in this basin. We find that the first four principal components (PC) explain 85% of total dataset variance for these 13 elements, and can be used to understand the primary types of particulate trace and minor element behavior. The greatest variability (i.e., the first principal component) can be explained by the behavior of elements with a strong lithogenic component in this basin: Al, Fe, Th, Ti, V, and Y. The second principal component describes the behavior of biogenic elements that have a surface maximum, and strong decrease with depth, as exemplified by P. The third principal component describes elements with a shallow subsurface maximum such as Ba. The fourth principal component describes the behavior of Cu, whose horizontal and vertical distributions are unlike any other element considered here. Co has a strong biological component in the upper water column, but is more similar to Mn and Pb below 500 m. Similarities between the distributions of Co, Pb and Mn are consistent with a role for Mn oxyhydroxides in the scavenging of Co and Pb. Cd has a significant biological component (high loading of PC2), but its distribution is otherwise more similar to that of Ba (high loading of PC3), with a strong, shallow subsurface maximum, especially in the oxygen minimum zone region in the eastern tropical Atlantic. We will discuss hypotheses for these distributions.