Kerry Gallagher

Evolving temperature histories from apatite fission-track data

A procedure for quantifying thermal histories from apatite fission-track data is presented. Given an appropriate annealing algorithm, or forward model, the problem is to determine a best-fit thermal history and quantify the resolution of this solution. Also, a priori geological information should be readily incorporated. Due to the non-linear nature of the problem, conventional inversion methods such as iterative least squares are inappropriate. The procedure described here relies on an initial stochastic search of a broad range of potential thermal histories using a genetic algorithm (GA). These techniques are extremely efficient in defining the regions of time-temperature space where good data-fitting solutions occur. However, the best model found with a genetic algorithm is generally non-optimal. Therefore, a multi-dimensional direct search method is used to update the best GA solution. A variety of standard statistical methods are considered to define confidence regions around this refined best thermal history.

Partial melting of mafic (amphibolitic) lower crust by periodic influx of basaltic magma

Abstract We present the results of numerical simulations to explore the effects of periodicity of basalt intrusion on the degree and timescales of melting of mafic lower (arc) crust. Melt production as a function of temperature was determined from parameterisation of published (0.7–1.6 GPa) fluid-absent partial melting experiments on amphibolites. We focus on a periodicity ranging from 20 to 30 000 years, with a total of 1 km of basic material intruded. Emplacing new basalt intrusions on top of earlier ones maximises the amount of silicic melt generated in the overlying protolith, and reduces greatly the heat loss through the base of the pile. Except for the case of a single intrusion, average partial melt temperatures generally exceed 900°C, and maximum temperatures may exceed 1000°C. The degree of partial melting is governed by the initial intrusion temperature and the periodicity, and yields a maximum predicted average melt fraction of 0.38. There is a time lag between maximum melt production and the maximum height reached by the melt column, with melt fractions in excess of 0.2 generated only where the time interval between each new intrusion is ≤200 years. Predicted maximum melt thicknesses do not exceed ca. 100 m, implying that large granitic magma chambers may not develop in the source region during partial melting of mafic lower crust. The ratio of the period of intrusion (τi) and the characteristic timescale for heat loss (τd) defines an important variable R that can be used to assess the thermal behaviour of the melting column, with both melt temperature and average (maximum) melt fraction maximised where R=1. The ratio R is used as an indicator of composition (expressed through changes in REE content) of a hypothetical partial melt, with smaller numbers of thicker intrusions (R→0) resulting in higher modelled (La/Yb)N ratios. Partial melting of continental crust beneath intrusions may provide a means of mixing mantle and crustal components at source prior to magma emplacement. Alternatively, if crustal melt freezes at the base, strong compositional (mafic–felsic) layering may result. During partial melting by periodic, multiple intrusion, the lower crust is a dynamic environment, with the amount of silicic (granitic) melt available for extraction in the source region rising and falling in the crustal column with time.

Calc-alkaline magmatism, lithospheric thinning and extension in the Basin and Range

Most of the volcanic rocks in the Colorado River Trough (CRT) and the Mogollon-Datil Volcanic Field (MDVF) in the Basin and Range exhibit calc-alkaline major element trends and relatively low high field strength element abundances, similar to those erupted from the volcanoes of Aso and Towada in Japan. Such features are widely regarded as characteristic of subduction-related magmatism, and yet the rocks in the Basin and Range were generated in response to lithospheric extension. The preextensional to synextensional rocks of the CRT and the MDVF have higher Na2O, K2O, and TiO2, in the range 47–55% SiO2, and relatively low Al2O3, and overall, they tend to have higher Sr contents and Zr/Y and La/Nb ratios than those from Aso and Towada. In addition, the basalts in the Basin and Range tend to be more aphyric than those in Japan, consistent with more rapid movement of magma through the crust during extension in the Basin and Range, and the rate of melt generation appears to have been significantly less in the Basin and Range than along recent destructive plate margins. The geochemical differences are attributed to smaller degrees of partial melting in the Basin and Range and to source regions that had been enriched in incompatible elements since the Proterozoic, resulting in parental magmas with higher alkali contents than those commonly observed in subduction-related calc-alkaline suites. Within the CRT the subsequent calc-alkaline trend was due at least in part to mixing with crustal derived melts, whereas in the MDVF such trends reflect both crustal contamination and fractional crystallization involving magnetite and amphibole. The small volumes of magma with minor and trace element features similar to oceanic basalts indicate that relatively little melt was generated in underlying asthenosphere. Thus it is inferred that magmatism in the Basin and Range was not associated with a significant increase in temperature, such as might be attributed to a mantle plume, but rather it was in response to lithospheric extension. Calculations are presented which demonstrate that the magma volumes and inferred source regions, extension, present-day heat flow, and topography are consistent with a model of convective lithospheric thinning after thickening in the Laramide and Sevier orogenies.

Transdimensional inversion of receiver functions and surface wave dispersion

[1] We present a novel method for joint inversion of receiver functions and surface wave dispersion data, using a transdimensional Bayesian formulation. This class of algorithm treats the number of model parameters (e.g. number of layers) as an unknown in the problem. The dimension of the model space is variable and a Markov chain Monte Carlo (McMC) scheme is used to provide a parsimonious solution that fully quantifies the degree of knowledge one has about seismic structure (i.e constraints on the model, resolution, and trade-offs). The level of data noise (i.e. the covariance matrix of data errors) effectively controls the information recoverable from the data and here it naturally determines the complexity of the model (i.e. the number of model parameters). However, it is often difficult to quantify the data noise appropriately, particularly in the case of seismic waveform inversion where data errors are correlated. Here we address the issue of noise estimation using an extended Hierarchical Bayesian formulation, which allows both the variance and covariance of data noise to be treated as unknowns in the inversion. In this way it is possible to let the data infer the appropriate level of data fit. In the context of joint inversions, assessment of uncertainty for different data types becomes crucial in the evaluation of the misfit function. We show that the Hierarchical Bayes procedure is a powerful tool in this situation, because it is able to evaluate the level of information brought by different data types in the misfit, thus removing the arbitrary choice of weighting factors. After illustrating the method with synthetic tests, a real data application is shown where teleseismic receiver functions and ambient noise surface wave dispersion measurements from the WOMBAT array (South-East Australia) are jointly inverted to provide a probabilistic 1D model of shear-wave velocity beneath a given station.

Paraná magmatism and the opening of the South Atlantic

Abstract New chemical and isotope results are presented on dyke rocks associated with the Paraná CFB, together with preliminary laser 40Ar/39Ar analyses on selected Paraná basalts. Dyke rocks from the Ponta Grossa Arch are similar to the Pitanga and Paranapanema magma types in the Paraná lavas, but dykes from the Santos-Rio de Janeiro section include samples with compositions not observed in the overlying lavas. Rather their minor and trace elements are strikingly similar to basalts recently erupted on Tristan de Cunha, and thus these late stage dykes may represent the first direct evidence for the involvement of typical plume-related OIB in the Paraná province. Laser 40Ar/39Ar analyses of two Gramado low Ti basalts have yielded preferred isochron ages of 132.4 ± 1.4 and 132.9 ± 2.8 Ma. These indicate a short eruption time for at least the Gramado magma type, and that magmatism took place several million years after the species extinction in the Tithonian (c. 141 Ma). The majority of basalts and basaltic andesites in the Paraná CFB have distinctive trace elements ratios (low Nb/La and Nb/Ba), and relatively enriched Sr, Nd, and Pb isotope compositions. Since such features are not commonly observed in oceanic basalts, and they occur in CFBs which have been screened for the effects of crustal contamination, they are typically attributed to old, incompatible element enriched source regions in the continental mantle lithosphere. In some models the minor and trace element ‘mantle lithosphere’ component was introduced in small degree melts (lamproites) added to asthenosphere derived magmas. However, such models appear to be inconsistent with the data from low Ti CFB, and they also require that the asthenosphere derived magmas have very low incompatible element contents, in marked contrast to the high Nb/La late stage dykes in the Paraná. Alternatively some CFBs may have been generated within the mantle lithosphere in the presence of small amounts of water. The results of preliminary calculations indicate that in the presence of a mantle plume up to 5 km of melt may be generated entirely from within the mechanical boundary layer, for β values of less than 1.2.

The denudation history of the onshore continental margin of SE Brazil inferred from apatite fission track data

A suite of 68 new apatite fission track analysis results from the onshore rift margin in southeast Brazil are presented. Some of the results reflect resetting related to the eruption of the Parana continental flood basalt (∼130 Ma), and four others are directly related to Late Cretaceous magmatism (∼80 Ma). The majority of the results were obtained from Precambrian basement rocks and sediments from the Parana Basin. The fission track ages broadly increase inland from 60–90 Ma on the coastal plain to >300 Ma in the continental hinterland. The track length distributions show a characteristic trend, being negatively skewed at the coast and showing increasing bimodality inland, except for the oldest samples which are more unimodal. These trends do not reflect rift-related heating and resetting but are the result of protracted denudation since the opening of the South Atlantic. The results are consistent with more than 3 km of exhumation on the coastal plain but little more than 1 km in the hinterland. The topographic morphology in the northern region of the study area is complex and is clearly influenced by structure and bedrock lithology. For the region further south, model calculations of the isostatic response to the denudation suggest that isostatic rebound makes a minor contribution to the long-wavelength topography, unless the flexural rigidity is more than 1025 N m. In this case the initial elevation of the rift margin would need to be >2.5 km in order to accommodate the amount of denudation inferred from the fission track data. However, it is possible that the effective flexural rigidity varies across the margin, becoming greater in the continental interior. Alternatively, the broad scale wavelength topography may be the result of magmatic underplating related to the Parana continental flood basalt event. If so, estimates of the volume of magma generated would need to be revised upward by as much as a factor of 5.

Mantle plumes, flood basalts, and thermal models for melt generation beneath continents: assessment of a conductive heating model and application to the Paraná

Growing evidence suggests that there is more than one type of continental flood basalt (CFB). Many CFB, such as the Deccan, were probably derived by decompression melting of asthenospheric peridotite in a mantle plume resulting in high eruption rates (∼1 km3 yr−1) and geochemical signatures of uncontaminated basalt which are similar to ocean island basalts. However, several geochemical studies have concluded that other CFB were derived from the subcontinental lithospheric mantle. This requires that melting occurs via heating of lithospheric peridotite above a mantle plume instead of melting within the plume itself. The feasibility of this model is tested with a simple one-dimensional, time-dependent thermal model. The results show that 1–2 km thick CFB may be derived from the lithospheric mantie without melting occurring within the underlying mantle plume if (1) the mantle potential temperature of the plume is between 1380 and 1580°C; (2) the lithospheric mantle is composed of volatile-enriched peridotite; (3) the overlying lithosphere is >100 km thick; (4) eruption occurs over 10–15 Myr; and (5) melting occurs over an area similar to the surface distribution of basalt. Small volumes of alkalic basalt may precede the eruption of tholeiites and if higher plume temperatures prevail, or extension/thermal erosion leads to lithospheric thicknesses ≤100 km, melts from the plume will rapidly dominate those from the lithosphere and eruption rates will increase (magmatic underplating may reduce the strength of the lithosphere sufficiently to initiate or focus rifting if the lithosphere is already under tension). These predictions can be used to discriminate between different CFB and are illustrated by application to the Parana CFB using published data along with new geochemical data from a borehole through the thickest section of basalt.

Genetic algorithms: a powerful tool for large-scale nonlinear optimization problems

Genetic algorithms represent an efficient global method for nonlinear optimization problems, that are encountered in the earth sciences. They share the favorable characteristics of random Monte Carlo over local optimization methods in that they do not require linearizing assumptions nor the calculation of partial derivatives, are independent of the misfit criterion, and avoid numerical instabilities associated with matrix inversion. The additional advantages over conventional methods such as iterative least squares is that the sampling is global, rather than local, thereby reducing the tendency to become entrapped in local minima and avoiding a dependency on an assumed starting model. In contrast to random Monte Carlo, however, they also share a desirable characteristic of the local methods in that they assimilate and take advantage of information collected during the sampling of the model space, resulting in an extremely efficient and robust optimization technique. This paper describes the basic genetic algorithm, briefly highlights some recent applications in the earth sciences and concludes that, in this field, the methodology should have many applications.

Shaping the Australian crust over the last 300 million years: insights from fission track thermotectonic imaging and denudation studies of key terranes

Apatite fission track thermochronology is a well‐established tool for reconstructing the low‐temperature thermal and tectonic evolution of continental crust. The variation of fission track ages and distribution of fission track lengths are primarily controlled by cooling, which may be initiated by earth movements and consequent denudation at the Earths surface and/or by changes in the thermal regime. Using numerical forward‐modelling procedures these parameters can be matched with time‐temperature paths that enable thermal and tectonic processes to be mapped out in considerable detail. This study describes extensive Australian regional fission track datasets that have been modelled sequentially and inverted into time‐temperature solutions for visualisation as a series of time‐slice images depicting the cooling history of present‐day surface rocks during their passage through the upper crust. The data have also been combined with other datasets, including digital elevation and heat flow, to image the denudation history and the evolution of palaeotopography. These images provide an important new perspective on crustal processes and landscape evolution and show how important tectonic and denudation events over the last 300 million years can be visualised in time and space. The application of spatially integrated denudation‐rate chronology is also demonstrated for some key Australian terranes including the Lachlan and southern New England Orogens of southeastern Australia, Tasmania, the Gawler Craton, the Mt Isa Inlier, southwestern Australian crystalline terranes (including the Yilgarn Craton) and the Kimberley Block. This approach provides a readily accessible framework for quantifying the otherwise undetectable, timing and magnitude of long‐term crustal denudation in these terranes, for a part of the geological record previously largely unconstrained. Discrete episodes of enhanced denudation occurred principally in response to changes in drainage, base‐level changes and/or uplift/denudation related to far‐field effects resulting from intraplate stress or tectonism at plate margins. The tectonism was mainly associated with the history of continental breakup of the Gondwana Supercontinent from Late Palaeozoic time, although effects related to compression are also recorded in eastern Australia. The results also suggest that the magnitude of denudation of cratonic blocks has been significantly underestimated in previous studies, and that burial and exhumation are significant factors in the preservation of apparent ‘ancient’ features in the Australian landscape.

The onshore record of passive margin evolution

The onshore region of passive margins has been relatively neglected because there is no unambiguous record of the denudation chronology and processes. However, apatite fission track data are sensitive to temperatures <150°C, appropriate to the cooling interval relevant during denudation of passive margin topography. In order to understand the meaning of the measured fission-track age, it is crucial to consider the track-length distribution as this is where most information concerning the thermal history is available. Over the last 10 years or so, apatite fission-track analysis studies have been undertaken from a variety of passive margins and we consider data from southern Africa, Brazil, Yemen, western India and southeast Australia. These data suites are consistent with greater amounts of section removed on what is now the low elevation coastal plain, with less material lost from the hinterland. In general there is little evidence remaining of the break-up process because of protracted denudation during the subsequent evolution of the margin. Each passive margin shows different morphologies and amounts of denudation reflecting both local geomorphic and geodynamic conditions. In addition, regional structures (e.g. shear zones) show evidence of reactivation long after break-up, but still related to large-scale plate processes.