Deep neural network application for 4D seismic inversion to changes in pressure and saturation: Optimizing the use of synthetic training datasets

In this work, we tackle the challenge of quantitative estimation of reservoir dynamic property variations during a period of production, directly from four-dimensional seismic data in the amplitude domain. We employ a deep neural network to invert four-dimensional seismic amplitude maps to the simultaneous changes in pressure, water and gas saturations. The method is applied to a real field data case, where, as is common in such applications, the data measured at the wells are insufficient for properly training deep neural networks, thus, the network is trained on synthetic data. Training on synthetic data offers much freedom in designing a training dataset, therefore, it is important to understand the impact of the data distribution on the inversion results. To define the best way to construct a synthetic training dataset, we perform a study on four different approaches to populating the training set making remarks on data sizes, network generality and the impact of physics-based constraints. Using the results of a reservoir simulation model to populate our training datasets, we demonstrate the benefits of restricting training samples to fluid flow consistent combinations in the dynamic reservoir property domain. With this the network learns the physical correlations present in the training set, incorporating this information into the inference process, which allows it to make inferences on properties to which the seismic data are most uncertain. Additionally, we demonstrate the importance of applying regularization techniques such as adding noise to the synthetic data for training and show a possibility of estimating uncertainties in the inversion results by training multiple networks.     

Seagrass ecophysiology meets ecological genomics: are we ready?

In March 2011, the Ecophysiology and Genetics Working Groups of the European Science Foundation COST Action ES 0906, entitled Seagrass Productivity: From Genes to Ecosystem Management, organized an exploratory workshop entitled “Linking Ecophysiology and Ecogenomics in Seagrass Systems”. The goal of the workshop was to discuss how to integrate comparative gene expression studies with physiological processes such as photosynthetic performance, carbon and nitrogen utilization and environmental adaptation. The main questions discussed for integrative research related to mechanisms of carbon utilization, light requirements, temperature effects and natural variation in pH and ocean acidification. It was concluded that the seagrass research community is still in the nascent stages of linking ecophysiology with genomic responses, as carbon and nitrogen metabolism of seagrasses have not been sufficiently well studied. Likewise, genomic approaches have only been able to assign meaningful interpretations to a handful of differentially expressed genes. Nevertheless, the way forward has been established.     

Significance of river–aquifer interactions for reach-scale thermal patterns and trout growth potential in the Motueka River, New Zealand

To assess whether reaches of the Motueka River (New Zealand) that gain water from groundwater were likely to represent significant cold-water refugia for brown trout during periods of high water temperatures, water temperature was monitored for more than 18 months in two gaining reaches of the Motueka River and three reaches that were predicted to be losing water to groundwater. These data were used to predict brown trout (Salmo trutta) growth in gaining and losing reaches. Groundwater inputs had a small effect on water temperature at the reach-scale and modelling suggests that the differences observed were unlikely to result in appreciable differences in trout growth. Several coldwater patches were identified within the study reach that were up to 3.5°C cooler than the mainstem, but these were generally shallow and were unlikely to provide refuge for adult trout. The exception was Hinetai Spring, which had a mean water temperature of close to 16°C during the period January–March, when temperatures in the mainstem regularly exceeded 19°C. Trout were observed within the cold-water plume at the mouth of Hinetai Stream, which would allow them to thermoregulate when mainstem temperatures are unfavourable while still being able to capitalise on food resources available in the mainstem.     

Geochemistry of groundwater in front of a warm‐based glacier in Southeast Greenland

Groundwater in front of warm‐based glaciers is likely to become a more integrated part of the future proglacial hydrological system at high latitudes due to global warming. Here, we present the first monitoring results of shallow groundwater chemistry and geochemical fingerprinting of glacier meltwater in front of a warm‐based glacier in Southeast Greenland (Mittivakkat Gletscher, 65° 41′ N, 37° 48′ W). The groundwater temperature, electrical conductivity and pressure head were monitored from August 2009 to August 2011, and water samples were collected in 2009 and analyzed for major ions and water isotopes (δD, δ¹⁸O). The 2 yrs of monitoring revealed that major outbursts of glacier water during the ablation season flushed the proglacial aquifer and determined the groundwater quality for the next 2–8 weeks until stable chemical conditions were reached again. Water isotope composition shows that isotopic fractionation occurs in both groundwater and glacier meltwater, but fractionation due to evaporation from near‐surface soil moisture prior to infiltration has the most significant effect. This study shows that groundwater in Low Arctic Greenland is likely to possess a combined geochemical and isotopic composition, which is distinguishable from other water sources in the proglacial environment. However, the shallow groundwater composition at a given time is highly dependent on major outbursts of glacier water in the previous months.     

Numerical Modeling of Atoll Island Hydrogeology

We implemented Ayers and Vachers' (1986) inclusive conceptual model for atoll island aquifers in a comprehensive numerical modeling study to evaluate the response of the fresh water lens to selected controlling climatic and geologic variables. Climatic factors include both constant and time-varying recharge rates, with particular attention paid to the effects of El Niño and the associated drought it brings to the western Pacific. Geologic factors include island width; hydraulic conductivity of the uppermost Holocene-age aquifer, which contains the fresh water lens; the depth to the contact with the underlying, and much more conductive, Pleistocene karst aquifer, which transmits tidal signals to the base of the lens; and the presence or absence of a semiconfining reef flat plate on the ocean side. Sensitivity analyses of steady-steady simulations show that lens thickness is most strongly sensitive to the depth to the Holocene-Pleistocene contact and to the hydraulic conductivity of the Holocene aquifer, respectively. Comparisons between modeling results and published observations of atoll island lens thicknesses suggest a hydraulic conductivity of approximately 50 m/d for leeward islands and approximately 400 m/d for windward islands. Results of transient simulations show that lens thickness fluctuations during average seasonal conditions and El Niño events are quite sensitive to island width, recharge rate, and hydraulic conductivity of the Holocene aquifer. In general, the depletion of the lens during drought conditions is most drastic for small, windward islands. Simulation results suggest that recovery from a 6-month drought requires about 1.5 years.     

Sensitivity Kernels for Time-Distance Inversion

Inversion of local-area helioseismic time-distance data has so far only been done in the ray approximation (Kosovichev, 1996). Since this is a high-frequency approximation its applicability can be questioned for the solar case. Bogdan (1997) showed that for a simple solar model the localized wave packets do follow, but are not confined to, the ray path. We use an approximation based on the first Fresnel zone that has been developed in geophysics by Snieder and Lomax (1996) to go beyond the ray approximation in the inversions. We have calculated sensitivity kernels using both approximations. To test them we use a finite-difference forward modeling of the whole wave field in an acoustic medium reminiscent of the Sun. We use the finite-difference modeling to calculate sensitivity kernels for the full wave field and compare this with the other kernels. The results show that the Fresnel-zone-based kernels are in good agreement with the sensitivity obtained from the modeling. Thus these new kernels represent a significant step forward in the inversion of time-distance data.     

Watson: A new link in the HE iron chain

Abstract— Watson, which was found in 1972 in South Australia, contains the largest single silicate rock mass seen in any known iron meteorite. A comprehensive study has been completed on this unusual meteorite: petrography, metallography, analyses of the silicate inclusion (whole rock chemical analysis, INAA, RNAA, noble gases, and oxygen isotope analysis) and mineral compositions (by electron microprobe and ion microprobe). The whole rock has a composition of an H-chondrite minus the normal H-group metal and troilite content. The oxygen isotope composition is that of the silicates in the HE iron meteorites and lies along an oxygen isotope fractionation line with the H-group chondrites. Trace elements in the metal confirm Watson is a new HE iron. Whole rock Watson silicate shows an enrichment in K and P (each ～2X H-chondrites). The silicate inclusion has a highly equilibrated igneous (peridotite-like) texture with olivine largely poikilitic within low-Ca pyroxene: olivine (Fa₂₀), opx (Fs₁₇Wo₃), capx (Fs₉Wo₄₁) (with very fine exsolution lamellae), antiperthite feldspar (An_1–Or₅) with <1 μm exsolution lamellae (An_1–3Or_>40), shocked feldspar with altered stoichiometry, minor whitlockite (also a poorly characterized interstitial phosphate-rich phase) and chromite, and only traces of metal and troilite. The individual silicate minerals have normal chondritic REE patterns, but whitlockite has a remarkable REE pattern. It is very enriched in light REE (La is 720X C1, and Lu is 90X C1, as opposed to usual chonditic values of ～300X and 100–150X, respectively) with a negative Eu anomaly. The enrichment of whole rock K is expressed both in an unusually high mean modal Or content of the feldspar, Or₁₃, and in the presence of antiperthite. Whole rock trace element data for the silicate mass support the petrography. Watson silicate was an H-chondrite engulfed by metal and melted at > 1550 °C. A flat refractory lithophile and flat REE pattern (at ～1x average H-chondrites) indicate that melting took place in a relatively closed system. Immiscible metal and sulfide were occluded into the surrounding metal host. Below 1100 °C, the average cooling rate is estimated to have been ～1000 °C/Ma; Widmanstätten structure formed, any igneous zoning in the silicates was equilibrated, and feldspar and pyroxene exsolution took place. Cooling to below 300 °C was completed by 3.5 Ga B. P. At 8 Ma, a shock event took place causing some severe metal deformation and forming local melt pockets of schreibersite/metal. This event likely caused the release of Watson into interplanetary space. The time of this event, 8Ma, corresponds to the peak frequency of exposure ages of the H-chondrites. This further confirms the link between HE irons and the H-chondrites, a relationship already indicated by their common oxygen isotope source. Watson metal structures are very similar to those in Kodaikanal. Watson, Kodaikanal and Netschaëvo form the young group of HE meteorites (ages 3.7 ± 0.2 Ga). They appear to represent steps in a chain of events that must have taken place repeatedly on the HE parent body or bodies from which they came: chondrite engulfed in metal (Netschaëvo); chondrite melted within metal (Watson); and finally melted silicate undergoing strong fractionation with the fractionated material emplaced as globules within metal (Kodaikanal). Watson fills an important gap in understanding the sequence of events that took place in the evolution of the IIE-H parent body(ies). This association of H-chondrite with HE metal suggests a surface, or near surface process-a suggestion made by several other researchers.     

Sediment accumulation rates in Conowingo reservoir as determined by man-made and natural radionuclides

The Susquehanna River is the major contributor to sediment loadings in the Chesapeake Bay. Because many environmental contaminants are associated with suspended particulates, the degree of particle retention within the reservoirs of the lower Susquehanna River is an important consideration in evaluating contaminant loadings to the Chesapeake Bay. Profiles of weapons-test Cs-137, nuclear power plant-related Cs-134 and Cs-137, and naturally-derived Pb-120 were used to estimate rates of sediment accretion in the conowingo Reservioir, an impoundment of the Susquehanna River along the Maryland-Pennsylvania border. Net accretion rates ranged from about 2 cm yr^?1 downstream of a nuclear power plant cooling discharge to a high of about 7 cm yr^?1 at the mount of an incoming creek. Slight, but consistent, increases in the annual rate of accretion since the creation of the reservoir in 1928 are apparent. The current net average annual sediment load reatined by the reservoir is estimated to be 0.4×10⁶ to 1.5 × 10⁶ metric tons yr^?1. The retained sediment load represents about 8–23% of the long-time average sediment input to the reservoir.