Litho-flow facies prediction in an alluvial fan/fluvial system, central North Sea

Abstract

An existing generic facies scheme consisting of a channel, sheetflood and floodplain facies was available from the cored wells. Bivariate and multivariate analysis of various wireline and core data indicated that each generic facies exhibits wide variations and significant petrophysical overlap with other facies. This scheme cannot therefore reliably be applied using wireline data, negating its application in uncored intervals through discriminant analysis. Due to the large variation in porosity and permeability of each facies, the existing scheme does not permit adequate determination of the probable reservoir performance of the sediments.In order to address these problems a new 'litho-flow' facies scheme has been constructed which can clearly be identified from wireline data thereby enabling it to be reliably extended beyond cored intervals. Each facies in the scheme exhibits a discrete range in porosity and permeability by which the probable flow characteristics of the sediments can be assessed and has successfully been used by reservoir engineers as input to a probabilistic reservoir modelling package. Finally, the predicted facies are valid in terms of the environment of deposition, assisting geologists in the correlation of units between wells for the establishment of a depositional model. The new facies scheme was formulated from a careful selection and processing of data to achieve the above. This included core descriptions, core porosity and permeability measurements, mineral volume analysis and porosity estimation from wireline data, permeability predictions and electrofacies from cluster analysis.

Towards prediction of saturated-zone pollutant movement in groundwaters in fractured permeable-matrix aquifers: the case of the UK Permo-Triassic sandstones

“…There are many practical problems with this approach: for example, quantifying uncertainty, dealing with qualitative data, appropriate definition of facies and calibration of resulting models (e.g. Wong et al 1997;Vaughan et al 1999;Biver et al 2002;Liu & Oliver 2005;Martinius & Naess 2005). Nevertheless, codes that include explicit accounts of bedforms and facies are starting to be used not only in research but also in practical applications (e.g.…”

The physical scale modelling of braided alluvial architecture and estimation of subsurface permeability

“…Using the six‐fold facies classification scheme (Table 3), between five and 10 samples were taken from each facies type to characterise their grain size distribution and ensure that the grain size distribution accounted for intra facies variation, which has been recognised as important in previous field studies (e.g. Dreyer et al ., 1990; Vaughan et al ., 1999).…”

“…Porosity and permeability are probably the most important quantifiable controls in reservoir fluid flow simulations and studies have shown that they are primarily a function of grain size and sorting (Freeze & Cherry, 1979; Brayshaw et al ., 1996; Robinson & Friedman, 2001), sedimentary structure (Scheibe & Freyberg, 1995) and diagenesis (Jacobson & Rendall, 1991; Hornung & Aigner, 1999). Many studies assume that the permeability distribution in an alluvial deposit is a function of sedimentary facies and depositional process (Alexander, 1993), and thus the link between petrophysical properties and fluvial sedimentology has been made in deposits from both reservoir wells (Begg et al ., 1989; Atkinson et al ., 1990; Lui et al ., 1996; Eschard et al ., 1998; Tye et al ., 1999; Vaughan et al ., 1999) and outcrop (Ravenne & Beucher, 1988; Dreyer et al ., 1990; Miller et al ., 1990). Clearly, there is an untapped potential for using aggrading physical models of alluvial sedimentary basins to provide information on the spatial distribution and geometry of different sedimentary facies and then relate this to the petrophysical properties.…”