Saliency-Map Guided Salt Prediction by a Multi-Channel Convolutional Neural Network

One of the major challenges in seismic imaging is accurately delineating subsurface salt. Since a salt boundary has strong impedance compared with other sediments, we build a saliency map with intensity and orientation to create a pixel-level model for salt interpretation. In this abstract, we train a saliency-map as an additional attribute to combine with the original seismic to predict salt bodies. We also train a saliency-map to classify multiple geological facies in a multi-channel convolutional neural network with residual net architecture to help build subsurface velocity models. Two examples are shown which demonstrate that a saliency-map-plus-seismic model successfully improves the accuracy of salt prediction and reduces artifacts.

The response of groundwater to multiple concerning drivers and its future: a study on Birbhum District, West Bengal, India

Groundwater and its upcoming crisis are the present-day concern of the scientist. This research mainly focuses on responses of groundwater dynamicity to some important drivers, viz. agricultural yield, groundwater irrigated area, groundwater draft, landuse/landcover, and stage of development. The result of this study has been done under three sections. In the first section, the spatiality of groundwater has been discussed where it has been noticed that the western side of the district groundwater level is near the surface due to low drafting and low agricultural yield. Moreover, hard rock geology in the western part disappoints the drilling process. On the eastern part, rich alluvial soil influences high agricultural yield hence groundwater level lowering down rapidly. In the second section, the nature of groundwater levels has been analyzed through the boxplot, and cluster diagram, where boxplots have been drawn over different geological facies, which depicts groundwater is highly fluctuating in hard clay geology. For example, high agricultural intensity and high groundwater draft is the characteristic feature of hard clay geology. The dendrogram in cluster analysis represents a homogeneous groundwater level fluctuating station in three different time series. Last section deals with the future of groundwater level where an artificial neural network (ANN) model has been applied to extract the predicted groundwater level for 2030. This type of environmental analysis, such as groundwater fluctuations in relation to different sensitive parameters and the use of a machine learning model, would aid potential researchers and communities in making wise groundwater use decisions.

Last Interglacial Sea-Level Indicators in the Western Indian Ocean

<p>With global average temperatures 2°C higher than pre-industrial and eustatic sea-level ranging between 5 and 9 m above present, the Last Interglacial is often regarded as a good process-analogue for a future warmer climate.  Large uncertainties are associated with Last Interglacial eustatic sea-level estimations. To quantify these uncertainties through standardization of sea-level metadata, the World Atlas of Last Interglacial Shorelines (WALIS) provides a community-wide standard for documenting the geological context of sea-level indicators and their chronology. By applying this standard, WALIS allows for the quantitative cross-comparison between previous studies, often times separated by decades.</p><p> </p><p>We use WALIS to review published sea-level indicators for the Last Interglacial within the Western Indian Ocean basin. Located in the far field with respect to past glaciations, the Western Indian Ocean has the potential to provide precisely measured and dated sea level proxies, enabling a reliable estimation of maximum eustatic sea level for the Last Interglacial. This, in turn, would allow to better constrain upper boundaries of melting within ice-sheet models. Furthermore, this review highlights localities that should be revisited based on the presence of geological facies indicative of former highstands where not enough detail has been reported or where advanced dating and geodetic techniques can increase the accuracy of metadata.</p>