.png){kind=icon}
.png){kind=icon}
Return
Advances and perspectives in modern geostatistics for energy, natural resources, and environmental studies
.png){kind=icon}
PRE
AI
D
L
J
J
A
J
J
A
V
A
DOI：10.1016/j.earscirev.2026.105611.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Geostatistics plays a central role in exploration and management of energy, natural resources, and environment, where sparse and/or indirect measurements must be translated into predictive models of the subsurface. Despite decades of advances, the field faces a set of pressing challenges that demand conceptual and methodological developments. This work summarizes recent advances over the past decade and discusses perspectives and research directions in the field of modern geostatistics, which refers to the integration of classical spatial statistics with data-driven and physics-informed learning methods. The growing availability of data and computational tools promotes the integration with artificial intelligence and machine learning and it offers new opportunities, but it also raises questions of interpretability and robustness. Capturing geological structures, heterogeneity, and complexity remains a fundamental issue, particularly when using data-driven methods that must balance realism with computational tractability. Emerging hybrid workflows, combining classical geostatistics, deep generative models, and physics-informed learning, hold promise but require rigorous approaches to uncertainty quantification to be trusted in decision making. An additional challenge lies in bridging the gap between scales of a broad range of data, reconciling high-resolution models with the heterogeneous spatial representativity of measurements. These range from sparse point samples to densely sampled borehole data, broad-scale geophysical surveys, and full-coverage remote sensing outputs. This diversity of data types and applications shapes which geostatistical methods are applicable and how uncertainty must be represented. Beyond methodological advances, there is a need to promote geostatistical thinking in new application domains, from carbon storage and hydrogen energy systems to sustainable mining, to environmental and climate studies. Addressing these challenges will define the trajectory of geostatistics in the coming decade, positioning it as a key tool of sustainable resource and environmental management.
Journal
E
IF:
10
Papers:
3.8K
Citations:
4.2W
Organization
Cited Papers