GP22A-01 INVITED
On the interpretation of Chinese loess as a paleoclimate indicator
The records of wind blown dust, or loess, in China and elsewhere are some of the most important terrestrial records of past climate changes, stretching back over the last ten million years. In the paleoclimate literature, intervals of increased dust generation have been almost always interpreted as being associated with more intense or prolonged wintertime conditions. Here we show that, in accordance with modern observations, dust outbreaks in Asia are predominantly springtime phenomena. During spring, frequent cyclogenesis in the lee of the Mongolian Altai, and the passage of strong cold fronts produce the intense windstorms that loft and entrain dust into the air. The meteorology governing such outbreaks is likely to be robust in past climates. Contrary to the common paleoclimate presumption, it is actually the breakdown of the Siberian High that permits the dust- producing windstorms to occur. The importance of cold fonts in generating such windstorms suggests that cooling of high latitude climate over the Miocene, or during glacial intervals, might play a significant role in the signal recorded in the loess deposits.
GP22A-02 INVITED
Comparison of Magnetic, Geochemical and Biological Proxies Signals in a ca. 2,000 yr Record from the Tropical Lowlands of Eastern Mexico.
Pollen, diatoms, geochemical, magnetic and non-magnetic mineral analyses were conducted on a lacustrine sequence from a maar lake on the tropical lowlands of eastern Mexico. Chronological framework for this lake is based on age determinations by 210-Pb, 137-Cs and 14-C. The studied sequence covers the last ca. 2000 yr, a time of important environmental transformations in the area due to climatic variability as well as human impact since the early Olmec societies until the recent forest clearance of the 20th century. Through these analyses we investigated the processes that affected the magnetic mineralogy in order to construct a model of past environmental changes, and compare it with the biological proxy records (diatoms and pollen) and the archeological record. Inferred climatic changes for this area are further compared with the documented climatic changes in the northern hemisphere of tropical America. Volcanic activity has played a major influence on sediment magnetic properties, as a purveyor of Ti-magnetites/Ti-maghemites, and as a factor of instability in the environment. Moisture availability has been determinant for the diatom and pollen records, and human impact is mostly reflected in the pollen and geochemical records. Direct observations of magnetic minerals and ratios of geochemical (Fe, Ti), and ferrimagnetic (χ f ) and paramagnetic (χ p) susceptibility (χ) data, are used as parameters for magnetite dissolution (χ p/χ, Fe/χ f ), and precipitation (χ f/Ti) of magnetic minerals. Evidence of agricultural practices associated with increased erosion, deforestation, higher evaporation rates, lower lake levels, anoxia and reductive diagenesis in non-sulphidic conditions are inferred for laminated sediments between A.D. 20-850. This deposit matches the period of historical crisis and multiyear droughts that contributed to the collapse of the Maya civilization. Dissolution of magnetite, a high organic content, framboidal pyrite and a change in the diatom assemblage point to anoxic, sulphidic conditions and higher lake levels after A.D. 850. Higher lake levels in Lago Verde coincide with a recovery in the forest cover, the same lake and vegetation signals are present in nearby lake Pompal, allowing to infer increased precipitation. This signal is coeval with the increased moisture documented during the Medieval Warm Period (A.D. 950-1350) in the northern tropical and subtropical regions of the American continent. For the Little Ice Age (A.D. 1400-1800) data are suggestive of relatively mistier conditions, with a deeper lake and highest vegetation cover, in concordance with the glacial advances recorded in central Mexico and tropical Andes. Higher erosion rates reflect destruction of the rainforest over the last 40 years.
GP22A-03 INVITED
Does Rainfall Always lead to Magnetic Enhancement in Topsoils of the World?
We propose a soil geochemical model that aims to explain both magnetic enhancement and depletion found in the topsoil in different climate regimes of the temperate zone. A high degree of correlation has been shown by B. Maher to exist between mean annual rainfall and magnetic enhancement in modern soils developed over loess in China and Russia. However, questions have also been raised about a poor correlation at some other locations such as Argentina, where many areas do not show enhancement. To reconcile both types of observations we propose a soil geochemical model that appeals not to total rainfall but to Potential Water Storage (PWS) of the soils and the presence of organic compounds in the environment. PWS, or total precipitation minus evapotranspiration, plays the major role in our model but other parameters of the soil microenvironment are also important. These are presence of organic ligands, clay and/or silica content, presence of carbonates and local drainage conditions. We compare the different environments in selected sites and find that an effective balance between iron oxide dissolution and authigenesis, driven by PWS and soil chemistry, can explain magnetic enhancements in China and Russia as well as magnetic depletion in Argentina. Rigorous, quantitative testing of our model would go a long way towards a comprehensive understanding of the magnetic effects of pedogenesis in diverse climatic environments.
GP22A-04 INVITED
Quantifying the diagenetic pathway of iron oxides in soils: a modeler's perspective
The first observation of enhanced magnetic susceptibility in topsoils by LeBorgne in 1955 has driven several studies on the transformation pathway of magnetic iron oxides in soils, whose complexity became immediately evident. The principal factor responsible for the magnetic enhancement (ME) is ultrafine (5-100 nm) pedogenic, cation-deficient magnetite (PCDM), whose low concentration - sometimes well below 1% - makes magnetic measurements the only reliable quantification method. It is now widely accepted that PCDM originates from the partial reduction of ferrihydrite, however, the detailed diagenetic path and the final fate of magnetite or maghemite nanoparticles is not clear. Nevertheless, an empirical correlation is observed between ME and the climate in which modern soils from selected regions of the World were formed. Unfortunately, this correlation is not as good as we wish for using magnetic measurements as a precise climatic proxy. What is the reason for the scatter observed in the ME of soils formed under similar conditions? Is it entirely due to the existence of "hidden variables", such as pH, Eh, vegetation and soil age? To address this question in a bottom-up approach, we first need to test the equation ME = PCDM, which implicitly underlies many studies on the magnetic properties of soils. For example, the definition of ME as the difference between magnetic measurements of the topsoil and of the "unaltered" parent material does not allow us to discriminate the accumulation of diagenetic minerals from other processes, such as weathering of magnetic minerals contained in the parent material, or changes of the dust accumulation rate. Furthermore, hypothetical differences in the grain size distribution of PCDM would be "seen" as concentration changes, because bulk magnetic parameters are grain size dependent. How can we deal with the complex magnetic properties of ferrimagnetic nanoparticles and a plethora of concomitant geochemical processes related to the iron cycle in soils? To address this question, we explored a rigorous modeling approach to (1) isolate the magnetic contribution of PCDM and other magnetic iron oxides, (2) quantify the concentration and grain size distribution of each diagenetic mineral, and (3) test the hypothesis that PCDM is characterized by specific physical and chemical "fingerprints", regardless of the degree of ME, which can be used to develop simple quantification methods. This approach produces a clear definition of ME and sets precise constraints on the diagenetic pathway of pedogenic iron oxides. Preliminary results from a selected group of soils show a dramatic improvement in the correlation between rainfall and our estimates of the PCDM concentration. Other pedogenic minerals, such as hematite and goethite, display a less clear correlation with climate and a far smaller enhancement that is not confined to topsoils.
GP22A-05
Paleomagnetic and Environmental Magnetic Studies of Eltanin Core 27-21 from the Ross Sea Sector, Antarctica
Due to low core recovery, stratigraphic discontinuities, and poor microfossil abundances, it has proven difficult to obtain high-resolution sedimentary records from Antarctica with unambiguous age-models and uncomplicated magnetostratigraphies. In addition, carbonate dissolution due to corrosive bottom waters makes it difficult to use the record of oxygen isotopes as a clear climatic proxy for variations in ice volume and/or temperature. We have been conducting new paleomagnetic and environmental magnetic studies of old cores from the Ross Sea sector of Antarctica. Here we present results from 682 discrete samples from Eltanin 27-21, an 18-meter long piston core recovered in 1968 by the USNS Eltanin as part of Operation Deep Freeze. We believe the discrete samples were collected by Norm Watkins and Jim Kennett shortly after the core was returned to the United States and that those samples which were measured for paleomagnetism were only demagnetized to 15 mT. We find that after removal of a low-coercivity overprint, most samples show a strong remanent magnetization that is stable between about 25-50 mT. The stable characteristic remanent directions of the samples define four normal and five reversed polarity zones. We have used these zones to develop a high-resolution magnetostratigraphy that encompasses all polarity intervals from the Brunhes Chron to the Reunion Subchron. This correlation is consistent with a low resolution study of radiolaria and foraminifera done by Fillon (1975). Average rates of sedimentation for individual polarity zones range from 0.39 cm/ky to 1.46 cm/ky. Various environmental magnetic parameters, including magnetic susceptibility, anhysteretic remanent magnetization and isothermal remanent magnetization, show features that are suggestive of orbitally-induced cyclicity. These parameters are being used to create an astronomically-tuned age model that encompasses the last 2.15 million years of sedimentation in the Ross Sea. The environmental magnetic data, coupled with the chronostratigraphic results, will provide new insights on the timing and nature of paleoclimate events in the Ross Sea sector, especially those that affect the availability of sediment and its transport into the basin. We also believe that this record can serve as a "Rosetta Stone" for correlation and dating of other old (and new) cores from the Ross Sea sector, even including cores that do not go back as far as the Brunhes/Matuyama boundary.
GP22A-06
Magnetic Mineralogy as Indicator of dry Conditions in Lacustrine Sediments From Santa María del Oro, Nayarit, Central Mexico
Combined magnetic and geochemical analysis were conducted on laminated sediments from Santa Maria del Oro, a crater lake in Nayarit (Mexico), to build up a model of paleoenvironmental conditions for the late Holocene. The occurrence of a severe drought at the end of the archeological Classic period (100 - 900 AD) has been documented in sites of central Mexico (Zirahuen lake and Lerma basin), the Gulf of Mexico coast (Los Tuxtlas) and the Yucatan peninsula. The effects of this climatic event are considered to have stressed the social and political situation in the Yucatan area and other sites in Mesoamerica, and resulted in the "collapse" of the Maya civilization. Santa Maria del Oro sediments between ca. 600 - 1140 AD are characterized by repeated sequences of ocher silt laminae with high inorganic carbon content, authigenic siderite, and low concentration of SD magnetic minerals, followed upward by an increase of concentrations of fine grained SD and SP ferrimagnetic minerals in brown silt laminae. This sequence is considered to represent dissolution-precipitation cycles of magnetic minerals in low erosion, concentrated waters and anoxic water-sediment interface environments. Dissolution of magnetite occurs in reductive conditions, which are considered as warmer and dryer periods. Above the ocher silt, precipitation of fine grained magnetite occurs when conditions change to oxic environments. Ostracode C and O isotopy document a negative precipitation/evaporation balance during this time period.