Effects of pore fluid pressure on the seismic response of a fractured carbonate reservoir

An effective medium model for the stress-dependent seismic properties of fractured reservoirs is developed here on the basis of a combination of a general theory of viscoelastic waves in rock-like composites with recently published formulae for deformation of communicating and interacting cavities (interconnected pores/cracks and fractures at finite concentration) under drained loading. The inclusion-based model operates with spheroidal cavities at two different length scales; namely, the microscopic scale of the pores and (grain-boundary) cracks, and the mesoscopic scale of the fractures (controlling the flow of fluid). The different cavity types can in principle have any orientation and aspect ratio, but the microscopic pores/cracks and mesoscopic fractures were here assumed to be randomly and vertically oriented, respectively. By using three different aspect ratios for the relatively round pores (representing the stiff part of the pore space) and a distribution of aspect ratios for the relatively flat cracks (representing the compliant part of the pore space), we obtained a good fit between theoretical predictions and ultrasonic laboratory measurements on an unfractured rock sample under dry conditions. By using a single aspect ratio for the mesoscopic fractures, we arrived at a higher-order microstructural model of fractured porous media which represents a generalization of the first-order model developed by Chapman et al. (2002,2003). The effect of cavity size was here modelled under the assumption that the characteristic time for wave-induced (squirt) flow at the scale of a particular cavity (pore/crack vs. fracture) is proportional with the relevant scale-size. In the modelling, we investigate the effect of a decreasing pore pressure with constant confining pressure (fixed depth), and hence, increasing effective pressure. The analysis shows that the attenuation-peak due to the mesoscopic fractures in the reservoir will move downward in frequency as the effective pressure increases. In the range of seismic frequencies, our modelling indicates that the P-wave velocities may change by more than 20% perpendicular to the fractures and close to 10% parallel to the fractures. In comparison, the vertical S-wave velocities change by only about 5% for both polarization directions (perpendicular and parallel to the fractures) when the effective pressure increases from 0 to 15 MPa. This change is mainly due to the overall change in porosity with pressure. The weak pressure dependence is a consequence of the fact that the S waves will only sense if the fractures are open or not, and since all the fractures have the same aspect ratio, they will close at the same effective pressure (which is outside the analysed interval). Approximate reflection coefficients were computed for a model consisting of the fractured reservoir embedded as a layer in an isotropic shale and analysed with respect to variations in Amplitude Versus Offset and aZimuth (AVOZ) properties at seismic frequencies for increasing effective pressure. For the P-P reflections at the top of the reservoir, it is found that there is a significant dependence on effective pressure, but that the variations with azimuth and offset are small. The lack of azimuthal dependence may be explained from the approximate reflection coefficient formula as a result of cancellation of terms related to the S-wave velocity and the Thomson’s anisotropy parameter δ. For the P-S reflection, the azimuthal dependence is larger, but the pressure dependence is weaker (due to a single aspect ratio for the fractures). Finally, using the effective stiffness tensor for the fractured reservoir model with a visco-elastic finite-difference code, synthetic seismograms and hodograms were computed. From the seismograms, attenuation changes in the P wave reflected at the bottom of the reservoir can be observed as the effective pressure increases. S waves are not much affected by the fractures with respect to attenuation, but azimuthal dependence is stronger than for P waves, and S-wave splitting in the bottom reservoir P-S reflection is clearly seen both in the seismograms and hodograms. From the hodograms, some variation in the P-S reflection with effective pressure can also be observed.     

NICMOS imaging search for high-redshift damped Lyα galaxies

We are engaged in a programme of imaging with the STIS and NICMOS (NIC2) instruments aboard the Hubble Space Telescope ( HST ), to search for the galaxy counterparts of 18 high-redshift z >1.75 damped Ly α absorption lines and five Lyman-limit systems seen in the spectra of 16 target quasars. This paper presents the results of the imaging campaign with the NIC2 camera. We describe the steps followed in reducing the data and combining in mosaics, and the methods used for subtracting the image of the quasar in each field, and for constructing error frames that include the systematic errors associated with the PSF subtraction. To identify candidate counterparts, which are either compact or diffuse, we convolved the image and variance frames with circular top-hat filters of diameter 0.45 and 0.90 arcsec respectively, to create frames of summed S /N within the aperture. For each target quasar we provide catalogues listing positions and aperture magnitudes of all sources within a square of side 7.5 arcsec centred on the quasar, detected at S / N >6 . We find a total of 41 candidates, of which three have already been confirmed spectroscopically as the counterparts. We provide the aperture magnitude detection limits as a function of impact parameter, for both detection filters, for each field. The average detection limit for compact (diffuse) sources is H _AB =25.0 (24.4) at an angular separation of 0.56 (0.79) arcsec from the quasar, improving to H _AB =25.5 (24.8) at large angular separations. For the brighter sources we have measured the half-light radius and the n parameter of the best-fitting deconvolved Sersic-law surface brightness profile, and the ellipticity and orientation.     

Geochemical processes and solute transport at the seawater/freshwater interface of a sandy aquifer

Geochemical processes occurring at a seawater/freshwater interface were studied in a shallow coastal siliclastic aquifer containing minor amounts of calcite. Data were collected from 106 piezometers in a 120-m transect from the coastline and landward. In the first 40 m from the coastline, a wedge of saltwater is intruding below the freshwater aquifer. The aquifer is strongly reduced with mineralization of organic matter by methanogenesis in the freshwater aquifer, and sulfate reduction dominating in the most seaward part of the saline aquifer. The spatial separation of cations in the aquifer indicated a slow freshening process where Ca²⁺ from freshwater displaced the marine cations Na⁺ and Mg²⁺ from the exchanger complex. The resulting loss of Ca²⁺ from solution decreases the saturation state for calcite and possibly causes calcite dissolution. A storm-flooding event was recorded where pulses of dense seawater sank through the fresh aquifer. As a result, the terminal electron accepting process switched from methanogenesis to sulfate reduction. The pulses of sinking seawater also triggered cation exchange reactions where Ca²⁺ was expelled from the exchanger by seawater Na⁺ and Mg²⁺. The released Ca²⁺ is being flushed from the aquifer by groundwater flow, and this export of Ca²⁺ will, in the long term, cause decalcification of the sediment. The water composition in the aquifer is in a transient state as the result of various processes that operate on different timescales. Oxidation of organic matter occurs continuously but at a rate decreasing on a geological time scale. The freshening of the aquifer operates on the timescale of a few years. The episodic flooding and sinking of seawater through the aquifer proceeds in the course of days to weeks, but occurs irregularly with years in between.     

The reactivity of iron oxides towards reductive dissolution with ascorbic acid in a shallow sandy aquifer (Rømø, Denmark)

The pool of iron oxides, available in sediments for reductive dissolution, is usually estimated by wet chemical extraction methods. Such methods are basically empirically defined and calibrated against various synthetic iron oxides. However, in natural sediments, iron oxides are present as part of a complex mixture of iron oxides with variable crystallinity, clays and organics etc. Such a mixture is more accurately described by a reactive continuum covering a range from highly reactive iron oxides to non-reactive iron oxide. The reactivity of the pool of iron oxides in sediment can be determined by reductive dissolution in 10 mM ascorbic acid at pH 3. Parallel dissolution experiments in HCl at pH 3 reveal the release of Fe(II) by proton assisted dissolution. The difference in Fe(II)-release between the two experiments is attributed to reductive dissolution of iron oxides and can be quantified using the rate equation J/m₀ = k′(m/m₀)^γ, where J is the overall rate of dissolution (mol s⁻¹), m₀ the initial amount of iron oxide, k′ a rate constant (s⁻¹), m/m₀ the proportion of undissolved mineral and γ a parameter describing the change in reaction rate over time. In the Rømø aquifer, Denmark, the reduction of iron oxides is an important electron accepting process for organic matter degradation and is reflected by the steep increase in aqueous Fe²⁺ over depth. Sediment from the Rømø aquifer was used for reductive dissolution experiments with ascorbic acid. The rate parameters describing the reactivity of iron oxides in the sediment are in the range k′ = 7·10⁻⁶ to 1·10⁻³ s⁻¹ and γ = 1 to 2.4. These values are intermediate between a synthetic 2-line ferrihydrite and a goethite. The rate constant increases by two orders of magnitude over depth suggesting an increase in iron oxide reactivity with depth. This increase was not captured by traditional oxalate and dithionite extractions.     

Explaining NDVI trends in northern Burkina Faso

Many studies have shown a ‘greening of the Sahel’ on the basis of analysis of time series of satellite images and this has shown to be, at least partly, explained by changes in rainfall. In northern Burkina Faso, an area stands out as anomalous in such analysis, since it is characterized by a distinct spatial pattern and strongly dominated by negative trends in Normalized Difference Vegetation Index (NDVI). The aim of the paper is to explain this distinct pattern. When studied over the period 2000–2012, using NDVI data from the MODIS sensor the spatial pattern of NDVI trends indicates that non-climatic factors are involved. By relating NDVI trends to landscape elements and land use change we demonstrate that NDVI trends in the north-western parts of the study area are mostly related to landscape elements, while this is not the case in the south-eastern parts, where rapidly changing land use, including. expansion of irrigation, plays a major role. It is inferred that a process of increased redistribution of fine soil material, water and vegetation from plateaus and slopes to valleys, possibly related to higher grazing pressure, may provide an explanation of the observed pattern of NDVI trends. Further work will focus on testing this hypothesis.     

New Reference Materials,Analytical Procedures and Data Reduction Strategies for Sr Isotope Measurements in Geological Materials by LA-MC-ICP-MS

Laser ablation multi-collector mass spectrometry (LA-MC-ICP-MS) has emerged as the technique of choice for in situ measurements of Sr isotopes in geological minerals. However, the method poses analytical challenges and there is no widely adopted standardised approach to collecting these data or correcting the numerous potential isobaric inferences. Here, we outline practical analytical procedures and data reduction strategies to help establish a consistent framework for collecting and correcting Sr isotope measurements in geological materials by LA-MC-ICP-MS. We characterise a new set of plagioclase reference materials, which are available for distribution to the community, and present a new data reduction scheme for the Iolite software package to correct isobaric interferences for different materials and analytical conditions. Our tests show that a combination of Kr-baseline subtraction, Rb-peak-stripping using βRb derived from a bracketing glass reference material, and a CaCa or CaAr correction for plagioclase and CaCa or CaAr + REE²⁺ correction for rock glasses, yields the most accurate and precise ⁸⁷Sr/⁸⁶Sr measurements for these materials. Using the analytical and correction procedures outlined herein, spot analyses using a beam diameter of 100 μm or rastering with a 50–65 μm diameter beam can readily achieve < 100 ppm 2SE repeatability ("internal") precision for ⁸⁷Sr/⁸⁶Sr measurements for materials with < 1000 μg g^-1 Sr.