Crustal thickness, discontinuity depth, and upper mantle structure beneath southern Africa: constraints from body wave conversions

The technique of receiver function analysis is applied to the study of crustal and upper mantle structures beneath the Kaapvaal craton in southern Africa and its surroundings. Seismic data were recorded by the seismic array of 82 sites deployed from April 1997 to April 1999 across southern Africa, as well as a dense array of 32 sites near Kimberley, in operation from December 1998 to June 1999. Arrival times for phases converted at the Moho are used to determine crustal thickness. The Moho depth in the south–western section of the craton was found to vary between 37 and 40 km, except for one station that recorded a depth of 43 km (SA23). Farther north along the western block of the craton (into Botswana) the depth increases up to 43 km. The depth increases even further in the north–eastern section of the craton, where results vary from 40 to 52 km. Just north of the Kaapvaal craton, in the neighbouring Zimbabwe craton, the crustal thickness drops significantly. The results obtained there varied from 36 to 40 km. For the Kimberley area, using the dense array, the Moho depth was found to be 37.3 km. Arrivals of the Ps and Ppps phases were used to determine the Poisson’s ratio in the region. This was found to be 0.26±0.01. Arrivals of phases from the 410 and 660 km mantle discontinuities are used to interpret the relative positions of these discontinuities, as well as for comparison of mantle temperatures and seismic velocities in the region with global averages. In the Kimberley area the 410 and 660 km discontinuities were found at their expected depth, implying that mantle temperatures in the region are close to the global average. The seismic velocities above the ‘410’ were found up to 5% faster than the averages from the global iasp91 model, which is fast even by Precambrian standards. In other sections of the Kaapvaal craton, the velocities are also faster than global averages, but not as fast as beneath Kimberley. In these sections, the ‘410’ is also slightly elevated, while the ‘660’ is depressed, which implies a slightly lower mantle temperature relative to the global average. Beneath the Kaapvaal craton we find evidence suggesting the presence of a zone with a reduced wavespeed gradient at an upper bound of approximately 300 km, which may mark the lower chemical boundary of the craton.     

Reproducibility of river water quality measurements: inter-agency comparisons for quality assurance

We assessed the reproducibility of river state-of-environment (SoE) water quality measurements in the Wellington Region, New Zealand (NZ). Field staff from GWRC and NIWA conducted 29 side-by-side water sampling and in-situ measurements at six river sites of diverse water quality for 12 variables measured routinely in river SoE monitoring across NZ. Field measurements of water temperature, dissolved oxygen, electrical conductivity and visual clarity agreed closely with strong numerical similarity (within 10%). Numerical similarity ranged widely for laboratory measurements, from strong for nitrate-nitrite-nitrogen to weak for turbidity, dissolved reactive phosphorus, and ammoniacal-nitrogen. Numerical agreement was very weak for laboratory pH (which is problematic) and E. coli–which is ‘tolerable’ for many applications given good correlation (R?=?0.94) over a 2000-fold concentration range. The findings of our inter-agency comparison have contributed to quality assurance recommendations in the NZ National Environmental Monitoring Standard (NEMS) for water quality.     

Two-dimensional modelling of subduction zone anisotropy with application to southwestern Japan

We present a series of 2-D numerical models of viscous flow in the mantle wedge induced by a subducting lithospheric plate. We use a kinematically defined slab geometry approximating the subduction of the Philippine Sea plate beneath Eurasia. Through finite element modelling we explore the effects of different rheological and thermal constraints (e.g. a low-viscosity region in the wedge corner, power law versus Newtonian rheology, the inclusion of thermal buoyancy forces and a temperature-dependent viscosity law) on the velocity and finite strain field in the mantle wedge. From the numerical flow models we construct models of anisotropy in the wedge by calculating the evolution of the finite strain ellipse and combining its geometry with appropriate elastic constants for effective transversely isotropic mantle material. We then predict shear wave splitting for stations located above the model domain using expressions derived from anisotropic perturbation theory, and compare the predictions to ∼500 previously published shear wave splitting measurements from seventeen stations of the broad-band F-net array located in southwestern Japan. Although the use of different model parameters can have a substantial effect on the character of the finite strain field, the effect on the average predicted splitting parameters is small. However, the variations with backazimuth and ray parameter of individual splitting intensity measurements at a given station for different models are often different, and rigorous analysis of details in the splitting patterns allows us to discriminate among different rheological models for flow in the mantle wedge. The splitting observed in southwestern Japan agrees well with the predictions of trench-perpendicular flow in the mantle wedge along with B-type olivine fabric dominating in a region from the wedge corner to about 125 km from the trench.     

Numerical comparison of different convergent plate contacts: subduction channel and subduction fault

At convergent plate boundaries, the properties of the actual plate contact are important for the overall dynamics. Convergent plate boundaries both mechanically decouple and link tectonic plates and accommodate large amounts of strain. We investigate two fundamental physical states of the subduction contact: one based on a fault and the other based on a subduction channel. Using a finite element method, we determine the specific signatures of both states of the subduction contact. We pay particular attention to the overriding plate. In a tectonic setting of converging plates, where the subducting plate is freely moving, the subduction channel reduces compression relative to the fault model. In a land-locked basin setting, where the relative motion between the far field of the plates is zero, the subduction channel model produces tensile stress regime in the overriding plate, even though the amount of slab roll-back is small. The fault model shows a stronger development of slab roll-back and a compressive stress regime in the upper plate. Based on a consistent comparison of fault and channel numerical models, we find that the nature of the plate contact is one of the controlling factors in developing or not of backarc extension. We conclude that, the type of plate contact plays a decisive role in controlling the backarc state of stress. To obtain backarc extension, roll-back is required as an underling geodynamic process, but it is not always a sufficient condition.     

Seismic microzonation of Dehradun City using geophysical and geotechnical characteristics in the upper 30 m of soil column

The understanding of geotechnical characteristics of near-surface material is of fundamental interest in seismic microzonation. Shear wave velocity (Vs), one of the most important soil properties for soil response modeling, has been evaluated through seismic profiling using the multichannel analysis of surface waves in the city of Dehradun situated along the foothills of northwest Himalaya. Fifty sites in the city have been investigated with survey lines between 72 and 96 m in length. Multiple 1-D and interpolated 2-D profiles have been generated up to a depth of 30–40 m. The Vs were used in the SHAKE2000 software in combination with seismic input motion of the recent Chamoli earthquake to obtain site response and amplification spectra. The estimated Vs are higher in the northern part of the study area (i.e., 200–700 m/s from the surface to a depth of about 30 m) as compared to the south and southwestern parts of the city (i.e., 180–400 m/s for the same depth range). The response spectra suggest that spectral acceleration values for two-story structures are three to eight times higher than peak ground acceleration at bedrock. The analysis also suggests peak amplification at 3–4, 2–2.5, and 1–1.5 Hz in the northern, central, and south-southwestern parts of the city, respectively. The spatial distributions of Vs and spectral accelerations provide valuable information for the seismic microzonation in different parts of the urban area of Dehradun.     

Possibilities of soil spectroscopy for the classification of contaminated areas in river floodplains

During the past decades, large amounts of diffuse contaminated soil material have been deposited in the floodplains of the river Rhine in the Netherlands. The dynamic character of this river causes a large spatial variability in the contamination level of its floodplain soils. Characterisation of the spatial variability exclusively based on soil sampling and analysis is often insufficient and expensive. Hyperspectral images can provide additional spatial information for a proper characterisation of the contamination situation of river floodplains. This paper describes the possible application of soil spectroscopy to estimate metal concentration levels in river floodplains. Soil reflectance spectra in the visible-near infrared region (VNIR) were measured in the laboratory for soil samples taken from two river floodplains along the river Waal, the main tributary of the river Rhine in the Netherlands. A multivariate calibration procedure using partial least squares (PLS) regression was applied to establish a relationship between reflectance spectra in the visible-near infrared (VNIR) region and spectrally active soil characteristics (organic matter and clay content) that are intercorrelated with concentration levels of Cd and Zn. Results of the analysis of two river floodplains are summarised and the influence of scale-level and sub soil material on the prediction capability is discussed.     

Sensitivity and cost considerations for the detection and eradication of marine pests in ports

Port surveys are being conducted in Australia, New Zealand and around the world to confirm the presence or absence of particular marine pests. The most critical aspect of these surveys is their sensitivity-the probability that they will correctly identify a species as present if indeed it is present. This is not, however, adequately addressed in the relevant national and international standards. Simple calculations show that the sensitivity of port survey methods is closely related to their encounter rate-the average number of target individuals expected to be detected by the method. The encounter rate (which reflects any difference in relative pest density), divided by the cost of the method, provides one way to compare the cost-effectiveness of different survey methods. The most cost-effective survey method is site- and species-specific but, in general, will involve sampling from the habitat with the highest expected population of target individuals. A case study of Perna viridis in Trinity Inlet, Cairns, demonstrates that plankton trawls processed with gene probes provide the same level of sensitivity for a fraction of the cost associated with the next best available method-snorkel transects in bad visibility (secchi depth=0.72 m). Visibility and the adult/larvae ratio, however, are critical to these arguments. If visibility were good (secchi depth=10 m), the two approaches would be comparable. Diver deployed quadrats were at least three orders of magnitude less cost-effective in this case study. It is very important that environmental managers and scientists perform sensitivity calculations before embarking on port surveys to ensure the highest level of sensitivity is achieved for any given budget.