 |
| Titel |
Resolving environmental signatures from a paleovalley sedimentary sequence from arid northwest Australia |
| VerfasserIn |
Alexandra Rouillard, Grzegorz Skrzypek, Shawan Dogramaci, Pauline Grierson |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250094770
|
| Publikation (Nr.) |
 EGU/EGU2014-10200.pdf |
|
|
|
|
|
| Zusammenfassung |
| Sediments from paleolakes can retain invaluable archives of past environmental conditions.
However, deciphering a depositional signal from digenetic processes can be challenging in
arid environments owing to extremely variable rainfall and saline groundwaters, which result
in aggressive chemical conditions that often limit the preservation of traditionally used
proxies. We investigated the development of hydroclimatic proxies based on sediment
geochemistry from the Fortescue Marsh, in the arid Pilbara region of northwest Australia.
The Marsh lies in a paleovalley that acts as a terminal basin for the upper part of the
Fortescue River and consists of a ~1000 km2 contiguous floodplain with freshwater pools
episodically inundated during intense rainfall events. The paleovalley is bound by mountain
ranges that contain some of the most Fe-ore rich and ancient deposits on Earth, which we
expected to confer unique geochemical characteristics to the sediments. We used
a sonic rig to retrieve a 25 m core from one of the deepest sedimentary sections
of the Fortescue Marsh (86 m to bedrock). We combined δ34S and δ18O stable
isotopes analyses with scanning μXRF and reflectance spectroscopy to quantitatively
map the elemental and mineralogical composition of the sedimentary sequence
and to identify underlying mechanisms relating to paleoclimate. We found that Fe,
Ca and Sr were the most abundant elements identified by μXRF. Typically, layers
of up to 1 m that were almost exclusively Fe-dominated alternated with layers of
0.3-2.4 m thickness dominated by Ca and/or Sr, with at least five intervals with
distinct peaks in Sr. As expected, the hyperspectral characterization confirmed that
Fe oxides were most abundant during the Fe-rich intervals. While clay minerals
including kaolinite and montmorillonite were also indicated from the spectral data, this
assessment is contradicted by the low relative abundance of Al and Si. Peaks in Sr don’t
appear to reflect carbonates nor Sr sulfates, thus further analyses are required to
confirm their origin. Hyperspectral mapping and strong correlation between S and Ca
confirmed that the intervals with high Ca corresponded to gypsum (CaSO4). We
suggest that this gypsum is an authigenic sulfate mineral that formed within the
top-most sediment due to groundwater fluctuation. Water chemistry measurements
taken down the profile show that the conditions are close to saturation of gypsum
throughout, and it is likely that such high concentration of crystals would form
during prolonged “dry” periods following very “wet” periods. Under this scenario,
current conditions are “dry” but there have been significant “wet” periods during
which detrital materials, such as Fe, Rb and other low abundance metals here, have
accumulated in the system via transport from the catchment. δ34S and δ18O analyses
will help confirm the origins of formation of the crystals. Overall, our findings
demonstrate that wetting and drying cycles and de-dolomitization processes are the main
mechanisms influencing local geochemistry. Establishment of geochronology for
this profile is currently underway. We are also examining other proxies including
δ87Sr and particle size. Collectively, a multi-proxy and process-based approach will
be used to characterize long-term hydroclimatic change in northwest Australia. |
|
|
|
|
|
|