AFES 2017 Seminar - Net Definition & Discussion Seminar

All barrels are not born equal! The Net definition dilemma

Wednesday September 27th 2017, Station Hotel, Guild Street

 

  • Net: Its Evolution in a Dynamic World

  • by Mike Webster, Production Petrophysics Ltd

Abstract

The definition of net depends on a number of factors in included in well architecture, fluids and the rock.  As the field evolves, the definition of net needs to evolve to help describe the active volumes of rock at each stage in the life cycle of the field.  This presentation explores some of these factors attempts to show that multiple net definitions may be in play for different user groups.

You can download this paper from here.

Biography

Mike Webster is an independent petrophysical consultant with over 36 year of oil industry experience.  He holds a BSc in Geology from Aberdeen university and an M.Eng in Petroleum Engineering from Heriot Watt.  He recently retired from BP after over 34 years of service and was BP’s first Director of Petrophysics.

  • Why You Cannot Map Net Sand

  • by Andy Beckly, formerly Principal Geologist at Lloyds Register

Abstract

Why you cannot map Net Reservoir. This talk argues that the definition of Net depends on the rock and on the fluids that you want to get out of it. Based on a large core data set, the talk shows the impact of different cutoff techniques in relation to NTG and associated averages. Also, an interesting overview of petrophysical averages is presented of the oil fields in the North Sea. On top of that, different rocks require different cutoff techniques. In the ‘world-according-to-Andy’, hydrocarbon volumes would be classified per habitat and per fluid type.

You can download this paper from here.

Biography
Andy has just retired as Principal Geologist at LR. Andy was with BP from 1985 to 1999 Andy’s 31 years of experience cover the entire life-cycle of the upstream business from frontier exploration through field development, late field-life management and Cessation of Production. He has extensive UKCS experience

  • The difference between net pay and net reservoir, how to pick them and their application to reservoir modelling

  • by Steve Cuddy, Baker Hughes

Abstract

Net reservoir and net pay have many definitions. Net reservoir is generally thought of as the cut-off that identifies reservoir rock capable of storing hydrocarbon. Net pay is sometimes defined as reservoir rock which will produce commercial quantities of hydrocarbon or to determine intervals to perforate. Net pay is a subset of net reservoir and is much more difficult to define. The net reservoir cut-off is required to upscale porosity, permeabilities and water saturations to initialise the reservoir’s 3D model.
Net reservoir and net pay can be determined using a variety of core and log techniques. Core fluorescence is a useful pointer and several electrical log crossplots can quickly define net reservoir. Water saturation vs. height functions (SWHF) are used in understanding the distribution of hydrocarbons in the 3D model, but must accurately upscale the log and core derived water saturations to the reservoir model cell sizes. The SWHF shows how net reservoir varies as the function of the height above the free water level.

Log derived permeabilities across the net reservoir tend to be low compared to core data and well performance. Calculation of permeability from upscaled porosities also leads to low values. This is usually because petrophysical averages fail to capture the full range of layer permeabilities. It is essential that the upscaled permeability have the same dynamic range as the core data. This can be achieved using a two-step prediction method. First facies types are predicted and then these are used to estimate the permeability. Fuzzy logic is one way of doing this.

You can download this paper from here.

Biography
Steve Cuddy is a Principal Petrophysicist with Baker Hughes and an Honorary Research Fellow at Aberdeen University where he holds a PhD in petrophysics. He has 40 years industry experience in formation evaluation and reservoir description. He has authored several SPE and SPWLA papers and is a SPWLA Distinguished Speaker for 2017/18.

  • A sedimentologist’s guide to Vsh and heterogeneity: An example of net sand identification in high sediment density mass flow deposits

  • by John Ford, CGG Robertson

Abstract

The degree of ‘shale distribution’ i.e. Vsh or reservoir heterogeneity, is a common risk in terms of reservoir quality discrimination. Clay can be distributed throughout the reservoir in several ways; as pore-lining clays within relatively clean sands, as thin beds of shale between sandstone beds, or as clay clasts intermingled with clean sand. The interpretation of Vsh curves (and other petrophysical logs) by geologists in the absence of supplementary core or image log data can cause substantial risk for exploration, and/or production from, “shaly sand” reservoirs. From a reservoir quality perspective, SCAL and petrographic data can accurately measure grain/pore size parameters and flow capacity of the reservoir. These analyses are, however, restricted to single point scale analyses, commonly >3ft apart with such spacing highly variable, being largely dependent on budget requirements. Log-scale analysis provides a macro-scale approach, however this is restricted by both hole condition and tool resolution.

An intermediary analytical scale must be applied to fully understand reservoir heterogeneity. Traditional core description techniques begin to bridge this scale-gap through the identification of lithological and lithofacies units as well as the identification of component mud blocks or curves. However, such component curves are qualitative, subjective to the individual geologist and produced with a relatively high degree of uncertainty. Furthermore, core descriptions are commonly scale dependent, typically determined by either budget, time constraint or both.
The entrainment of mud-clast volume in high density, sediment laden mass flow deposits can result in a heterogeneous character not observable at the micro-scale or macro log-scale. The intensity of mud-clasts within certain intervals could easily be misidentified as continuous mud beds from petrophysical logs if cored intervals were absent, leading to an underestimate in net sand. Alternatively some of these mud-clast zones may be beyond the resolution of conventional open hole logs which could lead to an over estimate of net sand. To address this issue, quantitative assessment of standard core-photographic data was undertaken in an attempt to compute a more accurate Vsh through the measurement of individual mud-clast long axis as well as the thickness of intermittent mudstone beds. Applying an aspect ratio to the long axis of identified mudclasts allowed an area to be calculated for each individual record. Summing the total of the mud-clast areas (>2000 measurements within ~200 m of core) with the total area of individual mudstone beds generated a combined area that could be subtracted from the total area of the core. The result was a Vsh that took into account randomly disseminated and randomly sized mudclasts within an otherwise relatively aggradational, coarse sandstone package. Finally, extraction of mud-clast long axes at discrete depths allowed frequency distribution data to be calculated for individual and combined facies blocks, ultimately providing a better insight into the macro-scale/log-scale heterogeneity of both ‘reservoir’ and ‘non-reservoir’ facies. These data, combined with pressure/production data would provide greater detailed inputs for reservoir risking workflows

You can download this paper from here.

  • Using Core to Define Net in a Complex Heterogeneous Reservoir

  • by Daryl Jackson, BP

Abstract

Conventional wisdom is that net is defined simply as rock that either contains, or is capable of containing hydrocarbons. A further refinement to this is that the rock containing hydrocarbons are within a connected pore system, and they are capable of responding to changes in pressure and hence moving or expanding. Translated into the petrophysical realm this corresponds to a certain amount of porosity, not exceeding a certain amount of shale content, with a certain water saturation being easily defined as net … in theory.

The Clair Field reservoir highly variable, heterogeneous sandstone dominated matrix, with varying amounts of dispersed clays of different mineralogy, and highly variable reservoir quality. The large variability in the dispersed clays and changeable sand mineralogy makes calculating shale volume from Gamma Ray and Neutron-Density methods very challenging; without a field wide marker bed for normalisation the calculations vary well-to-well. This makes using Vshale as a cut off impractical. The large matrix variability also makes the measurement of water saturation relatively uncertain (~+/- 20%), which again is unhelpful for defining net. On top of the uncertainty in saturation itself, the complex fill and spill history of Clair means that logged hydrocarbons are not always mobile and may represent paleo, irreducible, oil saturations. The “homogenously heterogenous” nature of the Clair field has necessitated an alternate approach to the conventional net definition.

Clair is fortunate in that it has extensive amounts of core (~6 km) which have been extensively and consistently described. Establishing a core-oil stain classification for all core has allowed the most direct and reliable measure of net, in such a heterogeneous reservoir. The core staining has been integrated with the conventional core measurements allowing a stain defined net cut-off to be accurately matched to permeability. The value of the cut off was investigated over a range of values and compared back to the net-to-gross seen in core to accurately define both the cut off, and the uncertainty in the value.

You can download this paper from here.

Biography

Daryl has a PhD in Physics and has spent
7 years at BP working on a variety of UK and Norwegian assets. He is currently working on the Clair field, and his specializations include production petrophysics and reservoir description.

  • Net Definition Dilemma: Challenges in Upscaling Log Defined Net to Geomodels

  • by Alan Johnson, Interactive Petrophysical Solutions Ltd

Abstract

The requirement to distinguish net from non-net is driven to a large extent by the averaging or “upscaling” required in applying petrophysical results in geomodels. In the early days, reservoir models were built on a zonal basis and the input of petrophysical properties was relatively simple; being based on zonal sums and averages generated by the petrophysicist as one of the deliverables of his analysis.

Modern geomodels are based on a finer layering structure within the zones, reservoir zones, designed to capture fine scaled reservoir architecture. The petrophysical input to these models is no longer the zonal average values calculated by the petrophysicist, but the log scale calculated rock property curves themselves. The averaging of these curves, now termed “upscaling”, now gets performed within the geomodelling software.

The geologist, and hopefully the petrophysicist, are usually presented with a number different options when performing the upscaling step, and not all will preserve the net rock and pore volume, as calculated at the log scale. This can lead to major volumetric discrepancies in the final models.
This presentation will highlight the potential volumetric deficiencies resulting from a number of standard upscaling strategies and propose a QC methodology and approaches to upscaling to ensure the preservation of pore volumes in layered geomodels.

You can download this paper from here.

Biography

Alan Johnson is a Petrophysical Consultant with Interactive Petrophysical Solutions Ltd. Until April 2016 he was Principal Petrophysicist with Shell UK based in Aberdeen, and has worldwide experience.

  • Poster: What do Geomodellers do with Net Logs?

  • by Claire Imrie & Moira Belka, Lloyds Register

Abstract

The 3D modelling of NTG is always a good topic for discussion. The ‘best approach’ does seem to depend on several variables, such as: geology of the reservoir; architecture of the reservoir; fluids in place; purpose of the model; number of wells/amount of data; and last but not least the preferences of the geologist/reservoir engineer

You can download this poster from here.

Biography

Claire is an experienced geomodeller with expertise in 3D geocellular modelling, geostatistics and uncertainty analysis in the field of CO2 storage as well as hydrocarbons. She has experience in multivariate geostatistical simulation, advanced facies modelling, experimental design, proxy modelling, fracture network modelling, upscaling for dynamic simulation and using seismic attributes in geomodelling.

Moira has 20 years’ experience as a Geologist in the oil industry. She is highly experienced in the construction and population of 3D geological models incorporating geostatistical techniques and conducting uncertainty analysis. She has worked on shallow marine, fluvial, submarine fan/conglomerate, deep water fan/turbidite, reefal and carbonate ramp reservoirs.

  • Are You Afraid of Water?

  • by Craig Lindsay, Core Specialist Services Ltd

Abstract

Quantification of water saturation distribution in hydrocarbon bearing intervals is a primary objective of petrophysical studies and of profound importance in estimating hydrocarbon in place. In addition, the mobility of water controls the viability of the reservoir – will it produce dry hydrocarbon or will water be mobile? Can we increase / define / constrain net pay if we fully understand the controls water saturation and mobility on a pore level?

Firstly, we look at some concepts and definitions with regard to water saturation and capillarity – and debunk some myths along the way;

  • Free water level
  • Hydrocarbon water contact / entry pressure
  • Transition zones
  • Irreducible water
  • Reservoir seals

Does high water saturation prohibit economic hydrocarbon production? Should we be afraid of water?

We look at core capillary pressure data that provides clear and unambiguous detail on what controls porosity and water saturation distribution within reservoir rock on a pore scale.

Examples are presented to show whether we should be afraid of water or not!

You can download this paper from here.

Biography

Craig has 36 years of experience in the core analysis industry. After 18 years working in the lab in a wide variety of roles, Craig joined Helix RDS consultancy in 2002 and founded Core Specialist Services in 2010. Craig served as President of the Society of Core Analysts from 2012-13.

  • Thin beds and net:gross from borehole image logs

  • by Stephen Morris, Baker Hughes

Abstract

Evaluating net:gross from image logs is dependent primarily upon the resolution of the log measurements. Variable filtering and thresholding methods can be used to isolate features of any particular wavelength in the image log that may or may not be desired in the analysis. In addition, new techniques for thin bed evaluation are explored that use a structural earth model and resistivity inversion for enhanced log responses in high-deviation wells. Examples of low and high-resolution borehole image logs with core comparison and different methods of calculating N:G are shown.

Biography

Stephen Morris has a PhD in sedimentology, from Cardiff University awarded 1998. Since then he has worked as a consultant geologist before joining BHGE in 2001 as a borehole image log interpreter. He has specialised in LWD borehole image logs and has worldwide experience.

  • Poster: Log and Core Resolution – Quantifying what’s inside the Seismic Loop

  • by Andrew Winter, Repsol Sinopec Resources UK

Abstract

Formation evaluation in thin bedded shale-sand sequences continues to be a major challenge to the industry. Thinly laminated pay zones are frequently overlooked when using conventional formation evaluation techniques because deep-water turbidite sequences and their productivity are affected clay characteristics, mineralogy, low resistivity contrast and sometimes by the unconsolidated nature of the sediments themselves.

Thin bedded sand-shale sequences are often difficult to detect on logs because the resolution of most logging tools is larger than the thickness of the beds themselves. Additionally because the tools sample at half foot increments there is a smoothing effect which results in an under estimation of the true formation properties.

In order to fully quantify thin bedded reservoirs beyond log and seismic resolution requires full integration of log, core, reservoir engineering and seismic information.

You can download this poster from here.

Biography

Andrew is Petrophysical Advisor to Repsol Sinopec Resources UK and has spent 25 years in the oil and gas industry. He has worked for several major oil companies around the world.