NEW HIGH-RESOLUTION ALKENONE RECORD OF LAST GLACIAL TO HOLOCENE SEA-SURFACE TEMPERATURE CHANGE IN THE EAST-EQUATORIAL SOUTH ATLANTIC OCEAN

ABSTRACT. A new high-resolution, alkenone-de-rived record of sea-surface temperature (SST) change covering the last 26000 years was obtained from the east-equatorial Atlantic off the Congo River. Temperature fluctuations correspond to climate change recorded in other marine and terrestrial archives of the region. The maximum temperature difference between 26000 years BP ago and the Holocene climate optimum around 6000 years BP was 4.3°C, corroborating other SST estimates from the same area. The coldest conditions were followed by a warming that began at 24000 calendar years BP, a time when Northern Hemisphere ice sheets were still at their maximal position. This comparatively early warming is in agreement with previous findings from the east-equatorial South Atlantic. After a relatively stable period between 21500 and 14500 calendar years BP, a second warming began at 14500 calendar years BP which coincides with the onset of the African Humid Period. A cool period at 11500 calendar years BP halted this warming but the trend of increasing temperatures began again at 10000 calendar years BP, this time through large-scale oscillations. The warmest time, around 6000 calendar years BP, was followed by a modest cooling that coincides with the end of the African Humid Period and the onset of Neoglaciation on the African continent. Following this the record shows two distinct warming-cooling cycles during the late Holocene.     

Zebra dolomitization as a result of focused fluid flow in the Rocky Mountains Fold and Thrust Belt, Canada

Discordant zebra dolomite bodies occur locally in the Middle Cambrian Cathedral and Eldon Formations of the Main Ranges of the Canadian Rocky Mountains Fold and Thrust Belt. They are characterized by alternating dark grey (a) and white (b) bands, forming an ‘abba’ diagenetic cyclicity. These bands developed parallel to both bedding and cleavage. Dark grey (a) bands consist of fine (< 300 μm) non-planar crystalline impure dolomite. The white (b) bands are composed of coarse (up to several millimetres) milky-white pure saddle dolomites (b1) which are often covered by pore-lining zoned dolomite (b2). The b phases often possess a saddle-shaped morphology. In contrast to the replacement origin of the a dolomite, the zoned b2 dolomite rims are interpreted as a cement formed in open cavities. The b1 dolomite is interpreted as the result of recrystallization with diagenetic leaching of non-carbonate components. All the zebra dolomites studied are (nearly) stoichiometric and are characterized by enriched Na and depleted Sr concentrations. Fe and Mn concentrations in these dolomites differ depending on the sample locality. Fluid inclusion data indicate that the dolomites formed from relatively hot (T_H = 130–200 °C), saline (20–23 wt% CaCl₂ eq.) fluids. A diagenetic high temperature origin is also supported by depleted δ¹⁸O values (−20 to −14‰ VPDB). A contribution of ⁸⁷Sr-enriched fluids is reflected in the ⁸⁷Sr/⁸⁶Sr values (0·7091–0·7123). Zebra dolomite development is explained by focused fluid flow, which exploited areas of structural weaknesses (e.g. basin-platform, rim areas, faults, etc.). Expulsion of hot basinal brines in a tectonically active regime generated overpressures, which explains the development of secondary porosity during zebra dolomitization as well as the intra-zebra fracturing at decimetre to micrometre scale.     

Sea Level Rise

No immediate prospect is in sight for a technically credible and unified approach to assessing the importance of sea level rises to New Zealand's coast environments. This review examines the state of disarray in the sea-level rise literature and traces reasons for such a state of affairs.     

Coarse-grained sediment gravity flow facies in a large supraglacial lake

Near Williams Lake, in the central interior of British Columbia, the Fraser River exposes long sections of late Pleistocene glaciolacustrine sediments selectively preserved within a bedrock trough. The dominant facies types are thick, normally graded gravels and sands that occupy steeply dipping multistorey channels up to 300 m wide and several tens of metres deep. Channels appear to have been simultaneously cut and filled by high density turbidity currents in a glacial lake floored by stagnant ice. Fining upward sediment gravity flow sequences up to 50 m thick may be the product of quasi-continuous ‘surging’ turbidity flows triggered by catastrophic meltwater discharges into the trough or retrogressive failure of ice-cored sediments. Large-scale post-depositional deformation structures, such as synclinal folds, normal faults, sedimentary dyke swarms and dewatering structures, record gravitational foundering of sediment and pore-water expulsion caused by the melt of underlying glacier ice. Melting of buried ice masses along the floor of the trough appears to have controlled the flow paths of turbidity currents by producing sub-basins within the overlying sediment pile. An idealized model of ‘supraglacial’ lacustrine sedimentation is developed that may be applicable to other glaciated areas with similar bedrock topography.     

FINDING CAUSES OF OUTLIERS IN MULTIVARIATE ENVIRONMENTAL DATA

Multivariate outliers in environmental data sets are often caused by atypical measurement error in a singlevariable.From a quality assurance perspective it is important to identify these variables efficiently so thatcorrective actions may be performed.We demonstrate a procedure for using two multivariate tests toidentify which variable‘caused’each outlier.The procedure is tested with simulated data sets that havethe same correlation structure as selected water chemistry variables from a survey of lakes in the WesternUnited States.The success rates are evaluated for three of the variables for sample sizes of 50 and 100,significance levels of 0.01 and 0.05 and various amounts of mean shift.The procedure works best forhighly correlated variables.     

Altimeter Signal-to-Noise for Deep Ocean Processes in Operational Systems

The ocean signal for this study is the sea surface height due to the slowly varying (greater than 5-day) ocean processes, which are predominantly the deep ocean mesoscale. These processes are the focus of present assimilation systems for monitoring and predicting ocean circulation due to ocean fronts and eddies and the associated environmental changes that impact real time activities in areas with depths greater than about 200 m. By this definition, signal-to-noise may be estimated directly from altimeter data sets through a crossover point analysis. The RMS variability in crossover differences is due to instrument noise, errors in environmental corrections to the satellite observation, and short time period oceanic variations. The signal-to-noise ratio indicates that shallow areas are typically not well observed due to the high frequency fluctuations. Many deep ocean areas also contain significant high frequency variability such as the subpolar latitudes, which have large atmospheric pressure systems moving through, and these in turn generate large errors in the inverse barometer correction. Understanding the spatial variations of signal to noise is a necessary prerequisite for correct assimilation of the data into operational systems.     

A polygon reduction algorithm for enhancing graphics performance with application to fast rendering of geophysical and ionospheric model data

Many scientific applications employ computer graphics as a means of interactively visualizing two-dimensional geophysical data sets. This paper presents an algorithm for improving rendering performance by using an image segmentation approach to reduce the number of rendered polygons while maintaining complete display fidelity. Though applied here specifically to the output from a numerical ionosphere model, the algorithm is broadly applicable to many other two-dimensional cellular data sets, geophysical and otherwise. In an ionospheric context, this algorithm demonstrates the ability to reduce the number of rendered polygons by 95 per cent or more with a corresponding reduction in rendering time.     

Mineralogy and Geochemistry of the Younger Volcanic Islands of Tonga, S.W. Pacific

The Tongan Islands comprise a double island chain, and lie atthe northern end of an active island arc extending NNE fromnorthern New Zealand. The eastern chain are limestone covered,with an underlying pre-Upper Eocene gabbro-basalt-andesite seriesexposed on one island. The active volcanoes, the subject ofthis paper, form the western island chain, and lie some 100km above a westward dipping seismic zone. Basaltic andesitedominates the islands of Hunga Ha'apai and Late (and also Tofuaand Kao, which are not described in this paper). Dacite is themain lava of Fonualei and the 1967–68 Metis Shoal eruption.Acid andesites occur only sporadically. Chemically, the lavas are moderately iron-enriched, with relativelylow concentrations of K, Rb, Ba, Zr, REE, Pb, and U. K/Rb ratiosare 500–700. The lavas are not typically ‘calcalkaline’.Phenocryst minerals (normally poorly zoned) are dominated bybytownite with subordinate orthopyroxene, augite, and sometimespigeonite in the basaltic andesites. Titanomagnetite occursonly in the more silicic andesites and dacites. The pyroxenesshow a limited iron enrichment through the series. Coexistingpyroxenes exhibit equilibrium distributions of Mg, Mn, V, Sc,and Ni. Pyroxene partition coefficients for V, Sc. Mn, and Coincrease with increasing iron enrichment. Groundmass mineralsinclude intermediate plagioclase, pigeonite, hypersthene, magnetite,plus potash feldspar and quartz in the dacites. The Metis Shoallava consists of rhyolitic glass with xenocrysts of forsteriticolivine and enstatite and phenocrysts of bytownite, hypersthene,augite, and titanomagnetite. The basaltic andesites are interpreted to be the product ofdirect peridotitic fusion within the mantle, presumably in theregion overlying the seismic zone, and modified by olivine fractionationduring ascent. Least squares numerical calculations, using majorand trace element data, support the derivation of the more siliceouslava compositions by low pressure crystal fractionation of basalticandesite Sr⁸⁷/Sr⁸⁶ ratios exhibit a small range of variation(0.7036–0.7043), possibly indicative (except for the MetisShoal lava) of minor limestone (?) contamination within themore silicic lavas during fractionation.     

Mid-Cretaceous basin margin carbonates, east-central Mexico

Isolated, high relief carbonate platforms developed in the intracratonic basin of east-central Mexico during Albian-Cenomanian time. Relief on the platforms was of the order of 1000 m and slopes were as steep as 20–43°. Basin-margin debris aprons adjacent to the platforms comprise the Tamabra Formation. In the Sierra Madre Oriental, at the eastern margin of the Valles-San Luis Potosi Platform, an exceptionally thick (1380m) progradational basin to platform sequence of the Tamabra Formation can be divided into six lithological units. Basinal carbonate deposition that preceded deposition of the Tamabra Formation was emphatically punctuated by an allochthonous reef block 1 km long by 0·5 km wide with a stratigraphic thickness of 95 m. It is encased in Tamabra Formation unit A, approximately 360 m of peloidal-skeletal wackestone and lithoclastic-skeletal packstone that includes some graded beds. Unit B is 73 m of massive dolomite with sparse skeletal fragments and intraclasts. Unit C, 114m thick, consists of structureless skeletal wackestone passing upward into graded skeletal packstone. Interlaminated lime mudstone and fine grained bioclastic packstone with prominent horizontal burrows are interspersed near the top. Unit D is 126 m of breccia with finely interbedded skeletal grainstone and burrowed or laminated mudstone. The breccias contain a spectrum of platform-derived lithoclasts and basinal intraclasts, up to 10 m in size. The breccias are typically grain supported (rudstone) with a matrix of lightly to completely dolomitized mudstone or skeletal debris. Beds are up to several metres thick. Unit E is 206 m of massive, sucrosic dolomite that replaced breccias. Unit F is approximately 500 m of thick bedded to massive skeletal packstone with abundant rudists and a few mudstone intraclasts. Metre scale laminated lime mudstone beds are interspersed. The section is capped by El Abra Formation platform margin limestone, consisting of massive beds of caprinid packstone and grainstone with many whole valves. Depositional processes within this sequence shift from basinal pelagic or peri-platform sedimentation to distal, platform-derived, muddy turbidity currents with a large slump block (Unit A); through more proximal (coarser and cleaner) turbidity currents (Unit B?, C); to debris flows incorporating platform margin and slope debris (Units D, E). Finally, a talus of coarse, reef-derived bioclasts (Unit F) accumulated as the platform margin prograded over the slope sequence. Interspersed basinal deposits evolved gradually from largely pelagic to include influxes of dilute turbidity currents. Units containing turbidites with platform-derived bioclasts reflect flooding of the adjacent platform. Breccia blocks and lithoclasts were probably generated by erosion and collapse of the platform during lowstands. Laminated, black, pelagic carbonates, locally cherty, are interbedded with both breccias and turbidites. At least those interbedded with turbidites may have been deposited within an expanded mid-water oxygen minimum zone during relative highstands of sea level. They are in part coeval with mid-Cretaceous black shales of the Atlantic Ocean.