AIs future may be limited not by chips, but by the power to run them. Eric Schmidt highlights how data centers fueling AI models are consuming record amounts of water and electricity, risking an environmental crisis. As big tech races toward superintelligence, the looming question is whether our energy grid can handle the load.
Category: futurism
Integrated geophysical prospecting for deep ore detection in the Yongxin gold mining area, Heilongjiang, China
Geophysical exploration techniques play a pivotal role in enhancing the accuracy of mineral prospecting predictions. However, relying solely on individual methods often introduces uncertainties. This study presents a case study from the Yongxin gold deposit, where we integrated audio-frequency magnetotelluric (AMT) methods with gravimetric surveying and high-resolution magnetic profiling to overcome this challenge. Advanced three-dimensional modeling techniques were utilized to precisely delineate lithological variations and deep-seated mineralization features inherent to the area. The inversion and interpretation of cross-sectional AMT data provided insights into the subsurface structure down to a depth of 1.5 km. This enhanced data reliability was achieved through an integrated interpretation constrained by multiple datasets, enabling a more accurate inference of the deeper geological framework. Furthermore, by amalgamating various datasets, we uncovered characteristics of deep mineralization, the three-dimensional configuration of mineralization-related rock masses, and the spatial orientation of known ore deposits. This holistic approach facilitated a comprehensive understanding of the deeper geological formations. A detailed analysis of ore-controlling structures and exploration markers led to the development of a tailored geological-geophysical model for mineral exploration within the study area, serving as a valuable reference for future deep exploration efforts.
(Note China found over $80 billion worth of gold with the help of new mineral prospecting technological advances)
Scientific Reports -Sci Rep 15, 7258 (2025). https://doi.org/10.1038/s41598-025-92108-3
Controlled-Source Audio-frequency MagnetoTelluric methods from the near- to the far-field: theory and applications
Controlled-Source Audio-frequency MagnetoTelluric (CS-AMT) is an active electromagnetic (CSEM) sounding technique, sensitive to the resistivity variations of the subsoil from a few meters to a few kilometers depth. It derives from the magnetotelluric (MT) method, which is based on measurements of natural electrical and magnetic fields related by induction; the interaction between natural EM signals at the Earth surface is controlled by the underground resistivity. With MT methods, the sources of these signals are distant enough from the measurement station, so that the EM waves are considered to behave as “far-field” waves. In this configuration, the Cagniard formula expresses the apparent electrical resistivity of the sensitive area as a function of the squared ratio of the electrical field amplitude to the magnetic induction amplitude. Because natural sources are irregular, AMT data acquisition and processing can be improved by the use of controlled source (CSEM). In most of the active geophysical techniques, and amongst others for CSEM, the input signal is voluntarily transmitted close to the measurement location, in order to maximize the signal to noise ratio. This implies that the measurements are sensitive not only to the area underneath the measurement location, but also to the area in between the transmitter and the receiver. This area needs therefore to be modelled. In the special case of CS-AMT, the source (transmitter that is either a horizontal electric dipole or a vertical magnetic dipole) is located as far as possible from the measurement location, so that the measured signals comply to the requirement of far-field waves. This assumption allows using the standard processing used for the MT methods, ignoring the effect of diffusion pattern and change of direction and intensity of the EM waves from the transmitter to the receiver. In practice, the distance between the transmitter and the receiver is often limited by the decrease in signal strength when moving away from the transmitter, and because the choice of the source location is also affected by accessibility or environmental constraints. For these reasons, the measured signals are more often in a “transitional domain” where the behavior of the signal corresponds either to the far-field or to the near-field, depending on its frequency. We propose a reformulation of the Cagniard formula to the interpretation of the ratio of the electrical to magnetic fields in the case of near-field magnetotelluric signals. We illustrate the use of these atypical formulations with applications to the granitic catchment of the Strengbach (Vosges mountains, North-East of France), and to the Séchilienne landslide, a micaschist instability in the Alps (South-East of France), where CSAMT data have been acquired in the near field, the transitional-field and the far-field.