Special Issue: Geostatistics and Machine Learning

Mathematical Geosciences - Recent years have seen a steady growth in the number of papers that apply machine learning methods to problems in the earth sciences. Although they have different...     

Comment on “Indicator principal component kriging” by V. Suro-Pérez and A. G. Journel

Conclusions The indicators covariance is equivalent to the bivariate distribution of the class index. The factor decomposition of the indicator covariance proposed by SPJ is an isofactorial model of the corresponding bivariate distribution. However, the choices of cumulative indicators and of principal component analysis produce unacceptable inconsistencies. In LL, a correspondence analysis of the bivariate distribution was used to produce more satisfactory empirical factors. These were used in the procedure of identification of discrete isofactorial models, with improved consistency, and the benefit of change of support models.     

A Special Issue on Geomathematics for Real-Time Mining

Mathematical Geosciences -     

A Special Issue on Data Science for Geosciences

Mathematical Geosciences -     

A Special Issue on Geothermal Energy and Its Application

Mathematical Geosciences -     

Guest Editorial to the Special Issue: Lessons Learned from post-2008 Wenchuan Earthquake Community Recovery

Identifying the spatial extent of volcanic ash clouds in the atmosphere and forecasting their direction and speed of movement has important implications for the safety of the aviation industry, community preparedness and disaster response at ground level. Nine regional Volcanic Ash Advisory Centres were established worldwide to detect, track and forecast the movement of volcanic ash clouds and provide advice to en route aircraft and other aviation assets potentially exposed to the hazards of volcanic ash. In the absence of timely ground observations, an ability to promptly detect the presence and distribution of volcanic ash generated by an eruption and predict the spatial and temporal dispersion of the resulting volcanic cloud is critical. This process relies greatly on the heavily manual task of monitoring remotely sensed satellite imagery and estimating the eruption source parameters (e.g. mass loading and plume height) needed to run dispersion models. An approach for automating the quick and efficient processing of next generation satellite imagery (big data) as it is generated, for the presence of volcanic clouds, without any constraint on the meteorological conditions, (i.e. obscuration by meteorological cloud) would be an asset to efforts in this space. An automated statistics and physics-based algorithm, the Automated Probabilistic Eruption Surveillance algorithm is presented here for auto-detecting volcanic clouds in satellite imagery and distinguishing them from meteorological cloud in near real time. Coupled with a gravity current model of early cloud growth, which uses the area of the volcanic cloud as the basis for mass measurements, the mass flux of particles into the volcanic cloud is estimated as a function of time, thus quantitatively characterising the evolution of the eruption, and allowing for rapid estimation of source parameters used in volcanic ash transport and dispersion models.

     

A critique of A. Journel's “The Deterministic Side of Geostatistics”

Mathematical Geosciences -