Value of Information in the Earth Sciences: Integrating by Jo Eidsvik, Tapan Mukerji, Debarun Bhattacharjya

By Jo Eidsvik, Tapan Mukerji, Debarun Bhattacharjya

Amassing the correct and the correct amount of knowledge is essential for any decision-making approach. This ebook provides a unified framework for assessing the worth of power information accumulating schemes by means of integrating spatial modelling and determination research, with a spotlight in the world sciences. The authors speak about the worth of imperfect as opposed to ideal details, and the worth of overall as opposed to partial info, the place basically subsets of the knowledge are got. options are illustrated utilizing a collection of quantitative instruments from selection research, comparable to choice bushes and effect diagrams, in addition to types for non-stop and discrete based spatial variables, together with Bayesian networks, Markov random fields, Gaussian techniques, and multiple-point geostatistics. certain in scope, this booklet is of curiosity to scholars, researchers and execs within the Earth and environmental sciences, who use utilized data and selection research thoughts, and especially to these operating in petroleum, mining, and environmental geoscience.

Show description

Read or Download Value of Information in the Earth Sciences: Integrating Spatial Modeling and Decision Analysis PDF

Similar mining books

Handbook of Gold Exploration and Evaluation

Designed for geologists and engineers engaged in particular within the look for gold deposits of all kinds and as a reference for lecturers in larger colleges of studying, guide of gold exploration and evaluate offers ideas and unique reasons that underpin the right kind interpretation of day by day adventure within the box.

Text Mining and its Applications: Results of the NEMIS Launch Conference

The realm of textual content mining is concurrently a minefield and a gold mine. textual content Mining is a speedily constructing functions box and a space of clinical study, utilizing options from well-established medical fields resembling info mining, desktop studying, info retrieval, average language processing, case-based reasoning, information and data administration.

Hydraulic fracturing impacts and technologies : a multidisciplinary perspective

Hydraulic Fracturing affects and applied sciences: A Multidisciplinary point of view serves as an advent to hydraulic fracturing and gives balanced insurance of its merits and capability unwanted effects. featuring a holistic evaluate of hydraulic fracturing and its environmental affects, this ebook chronicles the background and improvement of unconventional oil and gasoline construction and describes the dangers linked to using those applied sciences.

Value of Information in the Earth Sciences: Integrating Spatial Modeling and Decision Analysis

Accumulating the correct and the correct amount of knowledge is important for any decision-making strategy. This publication offers a unified framework for assessing the price of power information amassing schemes by means of integrating spatial modelling and selection research, with a spotlight on this planet sciences. The authors speak about the price of imperfect as opposed to excellent info, and the price of overall as opposed to partial details, the place simply subsets of the knowledge are bought.

Extra resources for Value of Information in the Earth Sciences: Integrating Spatial Modeling and Decision Analysis

Sample text

This graph illustrates a lag-1 or first-order Markov chain model. In the simplest setting, the variables may be binary, xi ∈{0,1}, i = 1,…, n, with stationary conditional probabilities p( xi +1 = l | xi = k ) = P(k, l ). These one-step probabilities can be organized as a Markov transition matrix as shown next:  P ( 0, 0 ) P ( 0,1) P=  . 15 Bayesian networks are characterized by a set of nodes and directed edges between the nodes. This figure shows two examples.  Below: the situation defines a first-order Markov chain model where each node only connects with the nearest nodes.

The problem, of course, is that we observe some data and have to use that as a basis for finding a useful probability model. 4) where the integral is replaced by a sum over the sample space for the discrete setting. The entropy is large when there is notable uncertainty. We briefly present some common discrete univariate pdfs. A key element for many discrete distributions is the indicator variable, which is 1 if an event occurs (success) and 0 otherwise (failure). We assign a probability p of success.

8 in Sequences 1 and 2, respectively. The empirical means and standard deviations (in parentheses) for P-wave velocity are 2740 m/s (220) in Sequence 1 and 3150 m/s (140) in Sequence 2. Findings from such histograms are useful for building realistic models based on rock physics and statistical assessments. 3 Histogram of well data. The top stratigraphic layer is in the upper row, while the deeper stratigraphic layer is at the bottom. Gamma ray (left) and P-wave velocity (right). 4 Oxide grade data observed along boreholes in a mine.

Download PDF sample

Rated 4.23 of 5 – based on 7 votes