HR: 0830h
AN: B21F-0777    [PDF]
TI: Trace-Element Evidence for an Aqueous Atmospheric Origin of Desert Varnish: implications for the aqueous atmospheric input flux into the ocean
AU: * Thiagarajan, N
EM: nivedita@rice.edu
AF: Rice University, Department of Earth Sciences, MS-126 6100 Main St, Houston, TX 77005 United States
AU: Lee, C
EM: ctlee@rice.edu
AF: Rice University, Department of Earth Sciences, MS-126 6100 Main St, Houston, TX 77005 United States
AB: Desert varnish is a slow-growing dark patina commonly found on rock surfaces in arid environments. Varnishes consist of about 30% Mn and Fe oxides accompanied by oxides of Si, Al, Mg, K and Ca, which occur primarily in the form of clays. Although it is generally agreed that varnishes have an atmospheric origin, their exact formation mechanism remains highly debated. Two endmember hypotheses are gradual accumulation of wind-blown dust followed by diagenesis, and direct chemical precipitation of dissolved elements from atmospheric aerosols. To rule out one of these hypotheses, we investigated the trace-element systematics of varnishes, in particular, focusing on those elements that have contrasting solubilities in aqueous environments. If our trace element analyses are consistent with the varnishes being derived from dissolved atmospheric constituents then the data can be used to quantify the paleofluxes of the soluble fraction of atmospheric aerosols to various depositional environments. For example, this will have implications for the transport of metals to the ocean that are immediately biologically available. We collected varnishes deposited on smooth basaltic lava flow surfaces in the Cima Volcanic Field (Mojave Desert) and in Death Valley, California. The chosen lava flows retain original flow surface structure and are topographical highs; the effects of erosion are hence minimal. Varnishes were scraped off with a quartz rod to minimize trace element contamination and the trace element compositions were then determined by ICP-MS using an external synthetic standard for calibration. Our analyses show that the rare-earth elements (REEs), Co, Ni, and Pb are enriched 1.5 to 10 times relative to the upper continental crust (UCC) and that Nb, Ti, Ta, Hf, Th, Rb and Cs are depleted to varying degrees relative to UCC and the REEs. These fractionations can be explained by their differing chemical behaviors in aqueous environments. The extreme depletion in Rb and Cs reflect their high solubilities and tendency to be progressively leached out by rain water. Nb, Ti, Ta, Hf and Th are present only in detrital concentrations, reflecting their high insolublities and their probable depletion in the Fe- and Mn-rich components of the varnish. Co, Ni, Pb and Ce are soluble but readily coprecipitate with Mn oxides hence their 10-fold enrichments. Enrichments caused by diagenesis of dust accreted on the varnish substrate cannot achieve the 10-fold enrichments of some elements observed here, indicating that the aqueous component must be derived directly from the atmosphere. Remarkably, we find that ferro-manganese crusts produced by hydrogenous processes in the marine environment have trace-element abundance patterns nearly identical to those of varnishes. Relative to the upper continental crust, they are enriched in REEs, Co, Ni, and Pb, depleted in Nb, Ti, Ta, Hf, Th, Rb and Cs and are anomalously high in Ce. These unexpected similarities provide additional evidence that desert varnishes represent the direct precipitation of aqueous components in the atmosphere. It may be possible to estimate the aqueous atmospheric input of such trace elements as the REEs into the ocean. For example, multiplying the Nd/Fe and Nd/Mn ratios of the varnishes by estimates of modern day Fe and Mn wet deposition inputs to the ocean yields an oceanic input of 4 to 15 x 107 moles of Nd/year. This is slightly larger than the amount of dissolved Nd entering the oceans each year (2.4 x 106 moles/yr) via rivers, hence, there is a significant atmospheric input of REEs into the ocean in aqueous form.
DE: 1020 Composition of the crust
DE: 1050 Marine geochemistry (4835, 4850)
DE: 4251 Marine pollution
DE: 4805 Biogeochemical cycles (1615)
DE: 4808 Chemical tracers
SC: Biogeosciences [B]
MN: 2003 Fall Meeting