QUANTITATIVE CHARACTERIZATION OF CARBONATE PORE … · observing carbonate pore system is a must development. VII. ACKNOWLEDGEMENT Authors would like to acknowledge to Geological

PROCEEDING, SEMINAR NASIONAL KEBUMIAN KE-8 Academia-Industry Linkage

15-16 OKTOBER 2015; GRHA SABHA PRAMANA

165

QUANTITATIVE CHARACTERIZATION OF CARBONATE PORE SYSTEMS ON

JONGGRANGAN FORMATION USING DIGITAL IMAGE ANALYSIS (DIA)

Ari Wibowo1*, Nana Higiana Hasanah2 1Department of Geological Engineering, UPN “Veteran” Yogyakarta

2Department of Geophysical Engineering, UPN “Veteran” Yogyakarta

JL. SWK 104 (Lingkar Utara) CondongCatur, Yogyakarta 55283 *corresponding author: [email protected]

ABSTRACT Jonggrangan Formation is characterized by major occurrence of reef limestone and small portion of

bioclastic limestone that were well exposed at Samigaluh Area, Kulon Progo Regency, Daerah

Istimewa Yogyakarta Province. These carbonates have a wide variety of pore systems that imprint

different petrophysical properties, which are more difficult to predict than in siliciclastics. Digital

Image Analysis (DIA) can be applied to response the carbonates complexity by calculating pore value

and also characterizing the pore shape in digital form. The method is based on images from rock thin

sections taken under an optical microscope (OM) and also core analysis results to be compared.

Sixteen rock thin section samples have been analyzed with several observation representatively in

order to mitigate the effect of area selection problems. Blue-dyed liquid also has been added to those

thin section samples, so the pore could be identified clearly. Based on further examination, both

primary and secondary pore systems are well developed. There are six pore shapes are identified

within different pore values. Crossplots of pore values indicate the pore variety depends on several

rock parameters. In summary, understanding characterization of carbonate pore system on

Jonggrangan Formation by using DIA method is fast and accurate, useful to encourage and enrich

carbonate petrophysical analysis concept.

I. INTRODUCTION

The carbonates are influenced greatly by sea

level fluctuation which recorded on rock

textures. Pore system in carbonate is much

more complex than siliciclastics, as a result of

overhelming biological origin of carbonate

sediments that reflected by the occurance of

porosity within grains, growth framework

porosity within reefs, and the common

development of secondary porosity due to

pervasive diagenetic processes (Choquette

and Pray, 1970). In addition to traditional

descriptive and qualitative porosity

evaluations, there exists a need for

quantitative methods that characterize the

various aspects of pore space and enable a

quantitative assessment of the distribution of

porosity and other physical properties

(Anselmetti et al., 1998). This Paper discussess

about evaluating pore characteristic

quantitatively on carbonates using digital

image analysis (DIA) which well-established

method of quantifying pore space from images

of thin sections. Numerous researches were

conducted over time to classify pore structure.

Anselmeti et al. (1998) quantified the pore

shape in carbonates from digital images

obtained by optical microscope and Scanning

electron microscope. Castro (2013)

determined Gamma parameter (ratio between

pore perimeter and area) and correlated it

with permeability measurements. In this study,

authors try to enlight about relationship

several parameters toward cabonate facies

and pore type based on thin section and

supported by core analysis.

Jonggrangan Formation is one of many

carbonate facies that well exposed in

Indonesia, particularly in Java island, has been

elected to be analyzed. Carbonate reef on

Jonggrangan Formation may lead us to

understand more about the development of

carbonate pore system.

II. REGIONAL GEOLOGY

The study area belongs to Kulonprogo Sub-

basin, the eastern part of South Central Java

Basin. This basin is located on the southern



166

portion of modern Java Volcanic Range and

could be identified as well as Basin lies on

Magmatic Arc. Volcanic activity which is

occured was the result of subduction beneath

the Hindia Ocean. Thus, this basin is classified

as fore arc-basin since it's location is in front of

Java Volcanic mountain range. The oldest

Formation in this area is Middle Eosen

Nanggulan Fm. which is well exposed on

Nanggulan Village. This Formation is typified

by the occurance of intercalation of sandstone,

shale, claystone and lignit which is deposited

on paralic depositional environment. Lies

conformably above Nanggulan Formation is

Kaligesing Fm. consisted of volcanic breccia

and intrusion. Since no fossil could be attained

from this formation, it is assumed as Oligosen

Age formation. Another formation which is

deposited on the approximate same time with

this formation is the polymict breccia Dukuh

Fm. Dukuh Formation consists of polymict

calcareous breccia with interbedded

sandstone and claystone, interpreted as

shallow marine sediments. Dukuh Formation

and Kaligesing Formation has interfingering

stratigraphic relation. Lies unconformably

above Kaligesing Formation are two distinctive

carbonate formation, Sentolo

Fm. and Jonggrangan Fm. Sentolo

Fm, typified as clastic carbonate formation,

consist of intercalation of calcarenite,

calcilutite, and marl. The abundant fossil

content in this formation allowed a firm

depositional environment interpretation of

this formation, as well as the Age of this

formation. Sentolo is Early Miocene carbonate

formation deposited on Neritic-upper Bathyal

environment. On the other hand Jonggrangan

Fm., the focus of this study is characterized by

the occurence of reefal limestone which

depict various carbonate facies. Jonggrangan

Fm. is interpreted as a patch reef deposited on

shelf area. Sentolo and Jonggrangan have

interfingering relation. (Pringgoprawiro and

Riyanto, 1968).

III. SAMPLE AND RESEARCH METHOD

Sixteen rock thin section samples have been

analyzed with several observations

representatively in order to mitigate the effect

of area selection problems. Blue-dyed liquid

also has been added to those thin section

samples, so the pore could be identified

clearly. The DIA (digital image analysis)

method is based on images from rock thin

sections taken under an optical microscope

(OM) and also core analysis results to be

compared. The method comprises of three

steps: image acquisition, segmentation, and

calculation of pore parameters. Image

acquisition takes approximately 10 – 15

minutes per sample, with additional time

needed for image analyses and statistical

evaluations. The segmentation process is the

separation of a specific feature from the

background and it was performed on colored

digital images acquired in standard RGB (red-

green-blue) which were converted to a binary

8-bit BW (black and white) format to identify

respectively, the pore and the matrix/cement.

In this study, the object feature of

segmentation was the rock pore that was

quantified and its geometry parameterized. In

addition, the image analyses program authors

used (Image J) is freely available as public

domain software. Then calculation process

uses mathematical programs in order to

create crossplots for several parameters.

IV. DATA AND ANALYSIS

Several parameters defined in the DIA to

quantify pore space. Two DIA parameters,

Dominant pore size (DOM) and Perimeter over

area (PoA), are proved to best describe several

aspects of pore system, namely facies, pore

type, porosity (Φ) and permeability (K).

Theoretically, dominant pore size (DOM) is

determined as the upper boundary of pore

sizes of which 50% of the porosity on a thin

section composed. Half of the pore space of

an area is composed of pores as big as or



167

smaller than the DOM. Whereas, Perimeter

over area (PoA) is the ratio of the sum of total

perimeter of the pores and total pore area

identified in the thin section. It describes how

complex is the porous system independent of

total porosity.

Based on further examination from outcrop,

core (polished slab) and thin section data,

eight lithofacies have been determined (Figure

1 &2). Six pore types also have been observed

which are Vuggy, Moldic, Intraparticle,

Interparticle, Channel, and Growth-Framework.

The quantitative results show that visible

porosity (Φ) varied between 1.87 – 34.7 % and

permeability (K) varied between 44.68 –

151.68 md (Table 1). Based on crossplot both

visible and measured porosity vs perm, there

is no relationship between these parameters.

DOM varied between 157 - 1100 µm2.

Crossplot of porosity vs permeability with

according to DOM values in figure 3 shows

that for a given porosity, the high DOM values

indicate the presence of large pores in the thin

section as observed in Figure 4.A, where a

grainstone facies with vuggy pore type

displays DOM 1100 µm2. At the same porosity,

the smaller value DOM, smaller pore size is

and the larger number, larger the pore size is.

The PoA varied between 75.5 - 166 mm-1.

Figure 5 shows crossplot of porosity vs

permeability with according to PoA values.

Figure 4.B shows the high PoA value, 139.5

mm-1, for a thin section of a framestone facies

with growth-framework porosity. At the same

porosity, the smaller PoA, the simpler pore

geometry is and the larger number, the more

intricate pore geometry.

V. DISCUSSION

Linkage of measured pore with visible pore

Both visible porosity and measured porosity show the different values on several samples (Figure 6). It is result of the effect of area selection for 2 cm x 2 cm thin section samples, does not represent the whole porosity characteristic of carbonate rock. The

difference value is a possibility, because on carbonate facies, the secondary porosity (eg: vuggy, moldic, and channel) are subsequently well developed on Jonggrangan Formation. Simply put, scale of observation and area selection take an important play.

Linkage of facies with porosity and permeability

Facies is known as one factor which control petrophysical values such as porosity and permeability. However the extent of facies control on porosity is constraint by several aspects. In order to prove the hypothese, the cross-plot of lithofacies toward porosity and permeability has been done (Figure 7). It shows that the lithofacies influence on porosity and permeability is not linear. Simply put, there are several samples of the same lithofacies type which possessed different petrophysical values. For example, grainstone (red symbol) present the lowest porosity and lowest permeability but the other grainstone present higher result with different value significantly. These condition, by far are controlled by diagenetic processes. A carbonate rock has its own sedimentation history both syn depositional or post depositional. The petrophysical values can be better when dissolution happen but it can be worst while recrystallization occur, such as porosity, will be enhance by dissolution process, while recrystalisation does the opposite.

Linkage of pore type with DOM and PoA

In general, grainstones with large secondary pores (vuggy) had high DOM and low PoA. Otherwise, fine grained facies which are packstones and wackestones supported by microporosity had high PoA and low DOM. Figure 8 shows a summary of these result, crossplot of DOM vs PoA plot for all thin sections analyzed and a few images of the sections indicating the respective PoA and DOM parameters.

VI. CONCLUSIONS

This study concludes that:



168

Carbonate pore system is more complex than siliciclastics.

Digital image analysis method is useful to encourage and enrich carbonate petrophysical analysis concept.

The geometric DIA parameters that varied better able to quantify the differences in pore type of the thin sections study are dominant size (DOM) and perimeter of area (PoA) which could be used to identify high and low porosity.

Integration of all available data while observing carbonate pore system is a must because its vast secondary porosity development.

VII. ACKNOWLEDGEMENT

Authors would like to acknowledge to

Geological Engineering Department of UPN

“Veteran” Yogyakarta for the support and also

to Sedimentology Lab. for facilitating the

laboratory analysis.

REFERENCES Anselmetti, F.S., Luthi, S., Eberli, G.P., 1998. Quantitative Characterization of Carbonate Pore Systems

by Digital Image Analysis. AAPG Bulletin, V.82, No. 10 p. 1-15.

Castro, D.D., Rocha, P.L.F., 2013. Quantitative Parameters of Pore Types in Carbonate Rocks. Revista

Brasileira de Geofisica.

Choquette, P.W. dan Pray, L.C. 1970, Geologic Nomenclature and Classification of Porosity in

Sedimentary Carbonates. The American Association of Petroleum Geologist Bulletin, Vol. 54, No. 2,

Ferbruary, 1970. p.222 – 224.

Ditya, A., Eberli, G.P., McNeil, D., Klaus, J., 2015. Segregating Porosity-Permeability Transform Using

Velocity and Pore Structure In Carbonate Rock. Proceedings Indonesian Petroleum Association 39th

Annual Convention & Exhibition. Jakarta.

Pringgoprawiro, H and Riyanto, 1968. Geologi Regional KulonProgo. ITB, Bandung.

Ramadhan, G.C., 2013, Organism Variety Effect on Carbonate Rock Porosity of Jonggrangan

Formation: Alternative Approach To Predict Porosity Complex. MajalahGeologi Indonesia, Vol. 28, 1

April 2013. p.29-40.

Widijana, A.R.T., Latief, F.D.E., Anissofira, A., Fatkhan, Handoyo, 2015. Modelling of The Sandstone

Ngrayong Formation Using Digital Image Analysis. Proceedings Indonesian Petroleum Association

39th Annual Convention & Exhibition. Jakarta.

Zhao, L., Nasser, M., Han, D., 2013. Quantitative Geophysical Pore-Type Characterization and Its

Geological Implication in Carbonate Reservoir. European Association of Geoscientists & Engineers,

Geophysical Prospecting p. 1-15.

TABLES Table 1. The summary of analysis result

Sample

Code Lithofacies

3D Φ

(%)

Digital Image Analysis

Visible Φ (%)

Main

Pore Type

DOM (µm2)

PoA (mm-1)

Perm

(md)

Ari 1 Wackestone 5,7 15,63 Vuggy 202 144 49,98



169

Ari 2 Framestone 5,07 3,44 Moldic 704 75,5 151,68

Ari 3 Packstone 3,59 8,96 Intraparticle 110 166 84

Ari 4 Framestone 3,18 22,59 Growth-Framework 366 137,5 69

Ari 5 Bafflestone 2,58 11,25 Growth-Framework 575 116.5 52,79

Ari 6 Floatstone 2,47 1,87 Intraparticle 476 136 76,55

Ari 7 Floatstone 5,1 6,28 Vuggy 353 138,75 46,29

Ari 8 Grainstone 3,16 2,50 Intraparticle 157 175 44,68

Ari 9 Marl 6,43 22,38 Interparticle 580 106,5 -

Ari 10 Packstone 6,43 13,91 Moldic 445 114,75 83,92

Ari 11 Grainstone 8,87 34,70 Vuggy 1110 105 75,43

Ari 12 Packstone 6,6 10,74 Channel 345 139,5 118,26

Ari 13 Rudstone 7,16 8,25 Intraparticle 262,5 97, 5 93,22

Ari 14 Floatstone 7,66 8,49 Intraparticle 272,5 121,5 120,56

Ari 15 Bafflestone 2,77 13,12 Interparticle 840 92,5 100,48

Ari 16 Packstone 4,94 4,87 Intraparticle 483 145,75 -

Φ: porosity; DOM: dominant pore size; PoA: perimeter over area; Perm: permeability.

FIGURES

Figure 1. (From left to right) Integration of outcrop, core (polished slab) and thin section data.

Figure 2. Core-plug sample and tools for measuring porosity analysis to support DIA result.



170

Figure 3. Crossplot of porosity vs permeability with according to DOM values. The high DOM values

indicate the presence of large pore size, and lower DOM values indicate the opposite.

Lithofacies = Grainstone Main Pore Type = Vuggy Avr. Porosity = 34.7 %

DOM = 1101µm2 PoA = 105 mm-1 K = 75.43 md

Lithofacies = Framestone Main Pore Type = Growth-Framework

A

B



171

Avr. Porosity = 22.59 % DOM = 366 µm2

PoA = 137.5 mm-1 K = 69 md

Lithofacies = Packstone Main Pore Type = Channel Avr. Porosity = 10.74 %

DOM = 345 µm2 PoA = 139.5 mm-1 K =118.26 md

Figure 4. Photomicrographs show three samples section of pore system (left) in which porosity is shown as blue, and digital image analysist result in BW-image (right) which porosity is shown as black and matrix is white. (A) Grainstone facies, (b) Framestone facies, and (c) Packstone facies.

Figure 5. Crossplot of porosity vs permeability with according to PoA values.

C



172

Figure 6. Comparison of measured porosity values obtained from core-plug analysis and porosity

values obtained from digital image analysis for various pore type . There is no linear caused by

nonrepresentative observation area of view under optical microscope.

Figure 7. Crossplots of measured pore vs permeability (left) and visible pore vs permeability (right) for various facies. Both crossplots ilustrates that the extent of facies control on carbonate pore system is constraint by several aspect.



173

Figure 8. Crossplot of dominant size (DOM) vs perimeter of area (PoA) plot for all thin sections analyzed and a few images of the sections indicating the respective DOM and PoA parameters.

Documents

QUANTITATIVE CHARACTERIZATION OF CARBONATE PORE … · observing carbonate pore system is a must development. VII. ACKNOWLEDGEMENT Authors would like to acknowledge to Geological