Determination of Total organic Content Using Alternating Conditional Expectation Method in Cambay Basin for Shale Gas Project Oriented Interim Report Submitted By

ASHISH MUYAL (Roll No. 07CL6003)

Under the supervision of

Dr. A. CHAKROBORTY

Dr. U.S. PRASAD

Supervisor CORAL IIT Kharagpur

Co-Supervisor ONGC (KDMIPE) Dehradun

Centre for Oceans, Rivers, Atmosphere and Land Sciences Indian Institute of Technology Kharagpur - 721302 2008

0

Abstract Cambay basin situated in the north-west part of Indian Craton got importance when huge generation and accumulation of large amount of hydrocarbon generation and accumulation is explored from Cenozoic sediments. Now researches also indicate that it also holds good potential of shale gas reservoir and can be exploited in future. Shale gas is essentially natural gas contained within sequences of formations of fine grained sedimentary rock dominated by shale. The term shale gas has been recognizing as unconventional resource as it was virtually impossible to produce gas in commercial quantities by conventional technologies. . Geochemical studies of Cambay shale shows that it posses all favorable conditions for gas generations. An attempt has been made using non-parametric method called Alternating Conditional Expectation (ACE) to get mathematical structure between various geo-chemical parameters to derive total organic content (TOC) for the Cambay basin as there is no functional relationship among these parameters.

Key Words: Craton, Alternating Conditional Expectation, Total Organic Content

1

1.

Introduction Unconventional resources were primarily based on economics. In the broadest sense

unconventional natural gas is gas that is more difficult, and less economically sound to extract usually because the technology to reach it has not been develop truly or too expensive. Conventional gas recourses are bouncy driven deposits, occurring as discrete accumulation in structural and startigraphic traps, where as unconventional gas resources is generally not bouncy- driven accumulation, most economically independent of structural or startigraphic traps.

1.1

SHALE GAS SYSTEM Shale gas is essentially natural gas contained within a sequence of predominantly

fine grained rocks by shale. The gas is stored within the shale sequence in two ways. 1. An adsorbed gas phase into kerogen, in this respect it is similar to natural gas in coal. 2. Free gas in the matrix porosity and fractured of shale rock mass. Commonly the gas is stored within the silty and fine grained sandy lenses that are interbedded with the shale dominated beds, or in the natural fractures system that in places may be pervasive in the shale rock section. Gas is compressible and this regards similar to conventional gas reservoirs. Thus different storage mechanism affects the speed and efficiency of gas production. Unlike conventional hydrocarbon targets, gas shale acts both as the source rock and reservoir rock for the generation and storage of the gas. The natural gas is either is biogenic in origin, formed by the action of biological organism by breaking down organic material within the shale, or of thermogenic origin, formed at depth and higher temperature. 2

Gas shale is considered continuous type natural gas plays, they are pervasive across large geographic areas and the reservoirs generally have along production life. Most gas shale has very low permeability, production rates are usually low and recovery factors are a fraction of conventional reservoirs.

1.2

SHALE GAS FORMATION There are two theories as to how natural gas is formed. The most widely accepted

theory, the organic theory, maintains that natural gas formation begins with photosynthesis, where plants use energy from the sun to convert carbon dioxide and water into oxygen and carbohydrates. The remains of these plants and the animal forms that consume them are buried by sediment and as the sediment load increases, heat and pressure from burial converts the carbohydrates into hydrocarbons. Natural gas formation takes place in fine-grained, black, organic, shale source rocks. Continued pressure from burial forces most of the natural gas to migrate from the organic shales into more porous and permeable rock such as sandstone and limestone. The natural gas remaining in the shales is termed shale gas. The other theory of natural gas formation is the inorganic theory which speculates that hydrocarbons did not originate from buried plant and animal material, but instead were trapped inside the Earth as it formed. This theory is most likely not applicable to shale gas.

3

Fig 1: Depositional Environment for Shale Gas

1.3

SHALE GAS EXPLORATION Exploration for gas shales is similar to exploration for conventional reservoirs

which, for an unexplored basin, usually includes: 1. review of existing information 2. aerial surveys to gather data regarding magnetic fields, gravity and radiation 3. seismic surveys to locate and define subsurface structures capable of trapping natural gas 4. exploration drilling to test subsurface structures for the presence of hydrocarbons 5. logging the wells to determine porosity, permeability and fluid composition

4

6. In the case of shale gas, the primary targets are shale formations with interbedded porous and permeable fine-grained sediments and natural fracture systems. 7. Down-hole tools used to find fractures include density compensation, caliper and temperature logs, and formation micro-scanner imaging. 8. Low-altitude, airborne multispectral imaging is a new tool used to locate subsurface micro-fractures and prospectivity of shale formation.

5

1.4

CAMBAY BASIN Exploration for hydrocarbon in Cambay basin has been continuing since 1958 and

today, exploration in the basin is in near mature stage. It is an Intracratonic N-S trending rift graben which developed in the northwest part of Indian peninsula and covers an area of approx. 58700 sq. km. including the area within the Gulf of Cambay. From the gulf area, the basin extends 450 km inland to the north of Sanchor where it is separated from the Rajasthan than basin by a basement swell. In the south, it merges with the Surat depression of the Bombay offshore basin across the Gogha-Aliabet basement arc.

Fig 2: Generalized Stratigarphy Of Cambay Basin

6

1.5

EVOLUTION OF CAMBAY BASIN The development of Cambay graben initiated following the last phase of effusion

of deccan basaltic lava which formed the floor of the Cenozoic sediment. the graben developed in the three stages : early rift stage , late rift stage and post stage . these three stages are co relatable to internal fractural(IF1),stages1; , (IF2),stage2; (IF3),stage 3 of Kingston ,s al. (1983 ) three stages mono cyclic basin . During the earliest ,Paleocene the western margin of India and its northern extension was subjected to upliftment , possibly on account of mantle upwelling .the consequent crustal stressing produced a rift stages crust with the development of tensional cracks to the east of Bermer Rabhanpur –Wadhwan gravity high along the trend of Dharwar strike .these tensional cracks formed a zone of weakness in the crust which culminated in rifting and subsidence .the ne –SW Delhi-Aravali and ENE-WSW Satpura fold Ridges , possibly, offered passive resistance to subsidence through accommodation of strain and sedimented the graben into a number of tectonic block. The early rift stages was characterized by fault dominated

subsidence and

deposition of basalt derived material has cones and fans (olpad formation) under aerial to fluvial environments, .argillaceous facies ,developed locally under lacustrine condition in the depression , particularly in the Mehsana-Patan–Sanchor area area, show good sources rock characteristics. The end of the rift stages (paleocene) was marked by block movement, tilting and erosion of elevated surfaces. Erosional products deposited at the flanks of Paloegeomorphic high may have formed reservoir facies within the Olpad formatio The early rift stages was succeeded by late rift stages subsidence (early Eocene) caused by thermal cooling and sediment loading in addition to faulting. And wide spread marine transgression during the late paleocence–middle Eocene with the deposition of the cambay shale group containing the major petroleum source bed in the basin .towards the later part of this started (middle Eocene),when tectonic subsidence started waning regression set in the northern cambay basin where the deltaic kadi formation ,an 7

important reservoir facies, was deposited . Reservoir facies within the cambay shale, particularly in the middle and the upper parts, also developed sporadically under paleoshoal environments as in linch area, and as on lapping still stand sand along the eastern basin margin. The rift stage fault-dominated subsidence was succeeded by post rift flexural subsidence due to sediment loading and thermal cooling in the middle part of Middle Eocene .marine regression punctuated by transgressive pulses , which set in during early middle Eocene , spread all over the basin in which all the major sandstone reservoir facies ( Kalol and Hazad formations ) with intervening transgressive shales were deposited . the shale acted as source rocks in depressions and also as local cap rocks. During the late Eocene to early Oligocene, in conformity with the global eustatic changes, there was a relative rise of the sea level in the basin having very low relief resulting in widespread transgression in which the Tarapur shale was deposited. Towards the later part of the early Oligocene when the sea started receeding, sedimentation took place under prograding deltaic condition (Ardol and Dhadhar units ) in the southern Cambay Basin . The above event was followed by withdrawal of the sea at the end of the early oligocene which resulted in the formation of a basin wide unconformity at the top of the Tarapur shale. The basin again witnesses marine transgression during the early Miocene when the Kathana formation was deposited in the southern cambay basin. During the middle Miocene, when the Indian plate was undergoing active subduction beneath the Eurasian plate, there, possibly, developed beneath oblique intraplate stresses along the son-Narmada lineament locally developed transcompressional stresses reactivated the major primary (listric normal) faults of the Narmada block and subsequently transformed them to up thrusts in the post middle Miocene time . Most of the hydrocarbons generated in the basin, accumulated in various plays during this phase of development of the basin. During Mio-Pliocene, the basin 8

behaved as an unstable platform with the deposition of shallow marine to continental deposits.

1.6

BASIN ARCHITECTURE

Mathur et al. (1968) divided the n-s trending linear graben into four morphotectonic unit namely Narmada, Jambusar-Braoch, Cambay-Tarapur and Ahemdabad-Mehsana block based on cross lineaments. Subsequently, markevich (1976) identified two more tectonic blocks, viz, Tharad and Sanchor in the northern part of the basin. Deep sounding seismic data shows that the basin is divisible into several blocks (kaila et al 1980, 1989). These cross- lineaments are interpreted to represent a transfer zone that is zones of structural overlapping between the tectonic units, which could accommodate strain through internal structural adjustment in an extensional setting.

Fig 3: Cambay Basin Cross Sectional Profile

9

2

SOURCE ROCK EVOLUTION Any source-rock evaluation should attempt to answer three questions: What are

the quantity, type, and maturity of the organic matter present in the rocks? Satisfactory methods are available in most cases to answer all these questions. In some areas one technique may fail completely or may be only partially successful. Whenever possible, therefore, we should not rely on a single analytical technique; rather, we should attempt to corroborate the measured data by other analyses. Interpretation of source-rock data on a basic level is quite simple. With increasing experience one can also learn to derive important information on thermal histories, unconformities and erosional events, and organic facies. We should always attempt to extrapolate our measured data over as large an area as possible. To do this intelligently we must have the ability to develop regional models of organic facies and thermal maturity.

2.1

DEFINITION OF SOURCE ROCK Much of modern petroleum geochemistry depends upon accurate assessment of

the hydrocarbon- source capabilities of sedimentary rocks. Although the term source rock is frequently used generically to describe fine-grained sedimentary rocks, that usage is a bit too broad and loose. For better communication, the following distinctions can be made. Effective source rock: any sedimentary rock that has already generated and expelled hydrocarbons Possible source rock: any sedimentary rock whose source potential has not yet been evaluated, but which may have generated and expelled hydrocarbons. Potential source rock: Any immature sedimentary rock known to be capable of generating and expelling hydrocarbons if its level of thermal maturity were higher.

10

2.2

QUANTITY OF ORGANIC MATERIAL The amount of organic material present in sedimentary rocks is almost always

measured as the total-organic carbon (TOC) content. This simple, quick, and inexpensive analysis serves as the first and most important screening technique in source-rock analysis. Analysis normally requires about one gram of rock, but if the rocks contain abundant organic matter, much smaller amounts can be analyzed. The quantity actually measured in the laboratory is always G, the remaining source capacity and not the original capacity (Go) . 2.3

MATURITY OF ORGANIC MATERIAL Knowing a rock's remaining source capacity G solves only one part of the puzzle;

it is also necessary to know what level of thermal maturity is represented by that particular G value. For example, if G is very low, is it because the rock never had a high initial source capacity, or is it because the rock is "burned out" (i.e., over mature, in which case virtually all the initial hydrocarbon-source capacity has already been used up)? The exploration implications of these two scenarios are, of course, very different. A substantial number of techniques for measuring or estimating kerogen maturity have been developed over the years. All the methods have strengths and weaknesses, and none can be applied in all cases. The feeling of most workers today is that there is no single maturity indicator that tells the whole story unerringly all the rime. All the techniques discussed are useful and probably reasonably accurate if the analytical work is carefully done. The key to using maturity parameters effectively lies in evaluating the measured data carefully (and sometimes with skepticism) and, whenever possible, in obtaining more than one maturity parameter. The most commonly used maturity parameters today are spore colour (Thermal Alteration Index, or TAI), vitrinite reflectance, and pyrolysis temperature. Less commonly used are fluorescence and conodont colour (CAI). A few of these parameters will briefly be discussed.

11

Vitrinite reflectance (Ro): Vitrinite-reflectance techniques were developed for measuring the rank of coals, in which the vitrinite maceral is usually very common. The method is based on the fact that with increasing thermal stress, the reflectance value of vitrinite increases. Vitrinite-reflectance measurements begin by isolating the kerogen with HCl and HF, and then embedding the kerogen particles in an epoxy plug. After the plug is polished, the microscopist shines light on an individual vitrinite particle. The fraction of the incident beam that is reflected coherently is measured and recorded and stored automatically on a computer. If enough vitrinite particles can be found, between 50 and 100 measurements will be taken. At the end of the analysis a histogram of the collected data is printed, along with a statistical analysis of the data. Results are reported as Ro values, where the o indicates that the measurements were made with the plug immersed in oil. Reflectance values are normally plotted versus depth in a well. If a log scale is used for the reflectance, the plot is a straight line. There are many problems with vitrinite reflectance as applied to kerogens. In many rocks vitrinite is rare or absent. Because what is present is often reworked, its maturity is not related to that of the rock in which it is found. Reworked vitrinite is, in fact, far more common in shale’s than in coals, leading to frequent difficulties in establishing which vitrinite population is indigenous. The ideal histogram of reflectance values is therefore rather rare; more common are histograms showing few vitrinite particles or multiple modes as a result of first-cycle vitrinite contaminated with reworked vitrinite or caving of less-mature material from uphole. Such histograms are quite often difficult or impossible to interpret, unless surrounding samples help us determine the indigenous vitrinite population. Other macerals or solidified bitumens can often be misidentified as vitrinite. Because each maceral type increases in reflectance in a slightly different way as thermal stress increases, misidentification of macerals can cause problems, even for experienced workers. Despite its weaknesses, vitrinite reflectance is the most popular technique today for estimating kerogen maturity. In many areas it is easy to use and valuable. In other rocks, however, paucity of first-cycle vitrinite renders vitrinite-reflectance measurements essentially worthless. In all cases it is worthwhile to supplement vitrinite with other measures of maturity; in some cases it is essential.

12

Thermal Alteration Index (TAI): TAI measurements are made on the same slides prepared for microscopic kerogen-type analysis. The darkening of kerogen particles with increasing thermal maturity can be used as an indicator of maturity. In order to minimize differences in color caused by changes in the type or thickness of the kerogen particles, TAI measurements are carried out on bisaccate pollen grains whenever possible. If no pollen can be found, TAI values are estimated, with lower confidence, from amorphous kerogen. Each laboratory has reference slides so that microscopists can continually compare the color determinations they are now making with those they and their colleagues made in the past. A careful worker can reproduce earlier work with excellent precision, thus defusing to a large degree the criticism that TAI is too subjective to be valid. Although TAI determinations are subjective, use of careful standards and the same type of palynomorph in each analysis greatly aid reproducibility. TAI measurements are therefore often quite accurate and correlate very well with results from other techniques. The chief problems arise with inexperienced workers, lack of proper standardization, or most commonly, the absence of spores and pollen in the samples. When palynomorphs are absent, TAI values must be estimated from amorphous debris, which can vary greatly in its chemical and physical properties. TAI values estimated from amorphous material are always suspect and should be corroborated by other analyses. Conodont Alteration Index (CAI): Conodonts are isolated, most commonly from fossiliferous carbonates, by removing the mineral matrix with acetic or formic acid. Colors of the specimens thus obtained are determined under a binocular microscope and compared with standards. The technique is simple and quick and can be done even by inexperienced workers. Although conodonts are composed of carbonate apatite, changes in conodont color are apparently due to carbonization of inclusions of small amounts of organic matter during catagenesis and metagenesis. One advantage of CAI over other maturity parameters is that because conodonts existed as early as the Cambrian, they offer a means of measuring maturity in rocks that do not contain pollen grains or vitrinite. Furthermore, conodonts are plentiful in carbonate rocks, where pollen and vitrinite are 13

often absent. Thirdly, the CAI scale is most sensitive at levels of maturity much higher than can be measured by TAI, and thus helps expand the range over which maturities can be measured. Finally, CAI is inexpensive and easy to measure and, with the help of color charts can be carried out by inexperienced personnel. One disadvantage of CAI measurements is that CAI values can be dramatically increased in the presence of hot brines, leading to an inaccurate assessment of kerogen maturity. Other disadvantages overlap with some of the advantages. Conodonts do not occur in rocks younger than the Triassic, and thus are of no value in many areas. Conodonts are not very sensitive indicators of maturity within the oil generation window, where most of the interest is. Finally, because the organic metamorphism displayed by conodonts is not related to hydrocarbon generation or destruction, CAI is only an indirect indicator of hydrocarbon maturity. Carbon Preference Index (CPI): The first maturity indicator applied to sediments was the Carbon Preference Index. Early investigations showed that immature rocks often had high CPI values (> 1.5), whereas those of oils were almost always below 1.2. This discovery led to the use of CPI as an indicator of maturity. Later it was realized that the decrease in CPI with increasing maturity depends upon the type of organic matter originally present as well as on maturity. In particular, rocks deposited in pelagic environments, in which the input of terrestrial lipids was very limited, have low CPI values even when immature. Furthermore, in the last decade kerogen analyses have replaced bitumen analyses as the routine procedure in source-rock evaluation. As a result, fewer CPI determinations are made now.

2.3

ESTIMATION OF ORIGINAL SOURCE CAPACITY Of the three major methods of determining kerogen type, only microscopic

analysis is relatively unaffected by maturity. As long as kerogen particles are not completely black, they can usually be identified with reasonable confidence. The exception to this rule is with amorphous material, where the fluorescence that enables us to distinguish between oil-prone and non-oil-prone disappears toward the end of the oil14

generation window. Pyrolysis yields are, of course, strongly affected by maturity. The most common method for taking maturity effects into account in evaluating pyrolysis data is to use a modified van Krevelen diagram to backcalculate the original hydrogen index. This method works fairly well if the kerogen is still within the oil-generation window. It breaks down at high maturity levels, however, because all kerogens have low pyrolysis yields. Without additional information, therefore, it is impossible to determine which maturation path brought it to that point. Like pyrolysis, atomic H/C ratios measure the present day status of the kerogen rather than its original chemical composition. Atomic H/C ratios must therefore be corrected for the effects of maturation by using a van Krevelen diagram. These immature H/C ratios can then be used to calculate Go.

3

INTERPRETATION OF QUANTITY OF ORGANIC MATERIAL

Almost all measurements of the amount of organic matter present in a rock are expressed as TOC values in weight percent of the dry rock. Because the density of organic matter is about one-half that of clays and carbonates, the actual volume percent occupied by the organic material is about twice the TOC percentage. Those rocks containing less than 0.5% TOC are considered to have negligible hydrocarbon-source potential. The amount of hydrocarbons generated in such rocks is so small that expulsion simply cannot occur. Furthermore, the kerogen in such lean rocks is almost always highly oxidized and thus of low source potential. Rocks containing between 0.5% and 1.0% TOC are marginal. They will not function as highly effective source rocks, but they may expel small quantities of hydrocarbons and thus should not be discounted completely. Kerogens in rocks containing less than 1% TOC are generally oxidized, and thus of limited source potential. Rocks containing more than 1% TOC often have substantial source potential. In some rocks TOC values between 1% and 2% are associated with depositional environments intermediate between oxidizing and reducing, where preservation of lipid-rich organic matter with source potential for oil can occur. TOC values above 2% often indicate highly reducing environments with excellent source potential. Interpretation of TOC values therefore does not simply focus on the quantity of organic matter present. A rock containing 3% TOC is likely to have much more than six times as much source capacity 15

as a rock containing 0.5% TOC, because the type of kerogen preserved in rich rocks is often more oil-prone than in lean rocks. We therefore use TOC values as screens to indicate which rocks are of no interest to us (TOC < 0.5%), which ones might be of slight interest (TOC between 0.5% and 1.0%), and which are definitely worthy of further consideration (TOC > 1.0%). Many rocks with high TOC values, however, have little oilsource potential, because the kerogens they contain are woody or highly oxidized. Thus high TOC values are a necessary but not sufficient criterion for good source rocks. We must still determine whether the kerogen present is in fact of good hydrocarbon-source quality.

3.1

TYPE OF ORGANIC MATTER

Microscopic kerogen-type analysis describes the proportions of the various macerals present in a sample. In interpreting these observations we normally divide these macerals into oil-generative, gas-generative, and inert. The oil-generative macerals are those of Type I and Type II kerogens: alginite, exinite, resinite, cutinite, fluorescing amorphous kerogen, etc. Gas generative kerogens are mainly vitrinite inertinite is considered by most workers to have no hydrocarbon-source capacity. Smyth (1983), however, has dissented from this pessimistic view, claiming, on the basis of deductive reasoning, that at least some Australian inertinites can generate significant amounts of oil. Nevertheless, the direct evidence for such a statement is rather meager. Pyrolysis results are normally reported in two ways. Raw data (S1, S2, and S3) are expressed in milligrams of hydrocarbon or carbon dioxide per gram of rock sample. As such these quantities are a measure of the total capacity of a rock to release or generate hydrocarbons or carbon dioxide. These raw data are then normalized for the organiccarbon content of the sample, yielding values in milligrams per gram of TOC. The normalized S2 and S3 values are called the hydrogen index and the oxygen index, respectively. Because variations in TOC have been removed in the normalizing calculation, the hydrogen index serves as an indicator of kerogen type. Measured hydrogen indices must be corrected for maturity effects by using a modified van 16

Krevelen diagram as outlined above. Interpretation of hydrogen indices for immature kerogens is straightforward. Hydrogen indices below about 150 mg HC/g TOC indicate the absence of significant amounts of oil generative lipid materials and confirm the kerogen as mainly Type III or Type IV. Hydrogen indices above 150 reflect increasing amounts of lipid-rich material, either from terrestrial macerals (cutinite, resinite, exinite) or from marine algal material. Those between 150 and 300 contain more Type III kerogen than Type II and therefore have marginal to fair potential for liquids. Kerogens with hydrogen indices above about 300 contain substantial amounts of Type II macerals, and thus are considered to have good source potential for liquid hydrocarbons. Kerogens with hydrogen indices above 600 usually consist of nearly pure Type I or Type II kerogens. They have excellent potential to generate liquid hydrocarbons.

3.2

MATURITY

Kerogen Parameters: Determination of the oil-generation window in a particular section is the objective of most maturity analyses performed on possible source rocks. A second, less common application is to decide whether oil will be stable in a given reservoir. The limits of the oil generation window vary considerably depending upon the type of organic matter being transformed. Nevertheless, for most kerogens the onset of oil-generation is taken to be near 0.6% Ro. Peak generation is reached near 0.9% Ro, and the end of liquid-hydrocarbon generation is thought to be at about 1.35% Ro. The ultimate limit of oil stability is not known for certain, but in most cases is probably not much above 1.5% Ro. Because vitrinite reflectance is the most popular method of determining maturity, most other maturation parameters are related to Ro values. The correlations among maturity parameters have been fairly well established, but there are still some minor variations from one laboratory to another. It is particularly difficult to generalize about TAI values because the numerical values of TAI scales have not been standardized among laboratories. Thus, if you are using TAI determinations determined by an analytical laboratory, make sure that you have a copy of their equivalency between TAI and Ro. Although Tmax values are determined objectively, because they vary with kerogen type as well as maturity, a unified scale for comparing them with Ro values has 17

not been adopted. Some laboratories put the onset of maturity at 435° C; others use 440°. Conodont Alteration Index (CAI) values ranging from 1 to 5 were tied loosely to vitrinite reflectance and fixed carbon content of coals. CAI can actually measure high-grade metamorphism, with CAI of 8 reached in a marble. The main variables important for the identification of geo-chemical parameters are the following; 1. TOC:

which percentage of total organic carbon of rock sample.

2. S1:

which is free hydrocarbon present in rock ( mg/g).

3. Tmax

which is temperature maximum at peak of S2, a rock eval thermal maturity parameter

4. H I:

which is hydrogen index [ (S2/ TOC )x 100 (mg/g) ].

5. PI:

which is production index denoting ratio of free hydrocarbon to total hydrocarbon [(S1 / (S1+ S2)].

6. S2:

which is remaining generation potential of rock (mg/g).

18

The standard values of those variables are shown in Table 1.

GENERATIVE POTENTIAL Poor Fair Good Very good

TOC Wt% 0.0 – 0.5 0.5 – 1.0 1.0 – 2.0 >2.0

S1 mg HC/g 0.0 – 0.5 0.5 – 1.0 1.0 – 2.0 >2.0

S2 mg HC/g 0.0 – 2.5 2.5 – 2.5 5.0 – 1.0 >10

MATURITY

PI

Tmax

Immature Mature Post mature

<0.10 .10 - .40 -

< 435 435 – 470 > 470

TYPE

HI

Gas Gas & Oil Oil

0 -150 150 -300 > 300 TABLE 1: EVALUTION OF GEOCHEMICAL PARAMETERS

4.

ALTERNATING CODITIONAL EXPECTATION (ACE)

In the present work a methodology is presented which employs ACE algorithm to construct mathematical structure between various geochemical parameters to derive TOC. ACE takes into consideration non-parametric regression. Conventional multiple or Parametric regression assumes a functional relationship between dependent and independent variables. But, because of uncertain nature of relationship between these variables it may not be possible to identify such functional relationship. On the contrary, nonparametric regression proposed by Breiman and Friedman (1985) and refined by Xue, et al (1997) does not require a priori knowledge of the functional relationship between dependent variable ‘y’ and independent variables ‘x1, x2, x3,……….. xn’ and it leads to establishing the least error relationship between the dependent and independent variables (Duolao Wang et al 2004). As no functional relationship can be assumed between HI 19

and PI S1, S2, TOC, and Tmax, non-parametric transformation may lead to some solution which is derived from the data sets of dependent and independent variables.

4.1

APPLICATION OF ACE ALGORITHM IN TOC ESTIMATION

The non-parametric regression approach proposed by Briemann and Friedmann [1] and refined by Xue , et al. [2] , provides a non-biased mechanism for the purpose of establishing minimum error

relationship between the dependent and independent

variables. In non-parametric regression priori knowledge of the functional relationship between dependent variable y and independent variables x1, x2,…,xn is not required. In fact, one of the main results of non-parametric regression is determination of the actual form of this relationship. A model predicting the value of y from the values of x1, x2… xn -1

generic form y = f (Z) Where Z = ∑Zi and Zi = fi ( xi ) The procedure for this approach is given by Calculate the data transform: Zi = fi ( xi ) i = 1,2, …….., n Calculate the transform sum: Z = ∑ Zi , i = 1,2,……,n Calculate the inverse transform: y = f-1 (Z)

20

is written in the

Given n observation points, we wish to find the best transformation functions f1(.), f2(.), …………., fn(.) , but not as algebraic expressions, rather as relationships defined point wise. The method of ACE constructs and modifies the individual transformations to achieve maximum correlation in the transformed space. Graphically this means that the plot of z = ∑ zi

against z' = f (measured) should be as near to the 45o straight line as

possible. The resulting individual transformation are given in the form of a point by point and/or table, thus in any subsequent application (graphical or algebraic) interpolation is needed to obtain the transformed variables and to apply the inverse transform to predict y. Naturally, the smoother the transformation the more justified and straightforward the interpolation is, therefore, some kind of restriction on smoothness is built into the ACE algorithm. In other words, based on the concept of conditional expectation, the correlation in transformed space is maximized by iteratively adjusting the individual transformations subject to a smoothness condition. An attempt has been made by employing the ACE algorithm to establish a relationship between TOC and five parameters explained in Table 1. Application the ACE yielded following relationship: HI_Tr = -1.4838E-05x2 + 5.9391E-04x + 1.1217E-01 PI_Tr = 1.1351E+00 x2- 1.9890E+00x + 2.5403E-01 SONE_Tr = -2.1687E-02 x2 + 1.2107E-01x - 3.3104E-02 STWO_Tr = 5.5876E-04 x2 + 1.5914E-01x - 5.8834E-01 TMAX_Tr = 1.2005E-03 x2 - 1.0616E+00x + 2.3463E+02 TOC = -7.2503E-02 SumTr2 + 2.6753E+00 SumTr + 2.7566E+00

21

Optimal transformation constructed to obtain above mentioned TOC – transform are presented in Fig. 5a to Fig. 5f.

RAW DATA STASTICAL ROCK EVAL PARAMETERS OF CAMBAY BASIN IN CAMBAY SHALE FORMATION WELL NAME W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 W16 W17 W18 W19 W20 W21 W22 W23 W24 W79

TMAX 434 438 436 442 440 437 438 432 447 438 446 440 440 438 445 442 437 448 447 450 441 449 450 458 438

S-ONE 0.28 0.28 0.06 0.08 0.12 0.13 0.07 0.07 1.17 0.46 0.5 0.08 0.03 0.15 0.26 0.24 0.06 0.77 0.33 0.35 0.8 0.17 0.32 0.62 .09

PI 0.19 0.1 0.06 0.05 0.14 0.17 0.11 0.13 0.31 0.05 0.18 0.1 0.07 0.13 0.23 0.19 0.05 0.24 0.11 0.17 0.15 0.16 0.14 0.31 .08

22

TOC% 1.2 5.2 2.1 1.9 1.16 0.4 1.7 1.5 2.4 2.9 2.1 1.9 1.1 1.6 2.4 1.5 2.2 2.1 3.6 2.3 3.6 1.7 2.9 2.6 1.9

S-TWO 2.5 5.69 0.99 1.7 1.6 0.68 0.59 0.7 3.02 5.86 2.28 0.98 0.51 1.21 0.86 0.93 1.33 2.34 2.58 1.59 3.57 0.89 2.97 3.42 1.13

HI 136 77 63 85 21 163 40 126 117 67 110 54 40 58 52 91 70 112 95 84 87 79 67 54 60

RESULTS AND DISCUSSION 8

8

7

7

6

6

5

5

TOC %

TOC %

5.

4

4

3

3

2

2

1

1

0

0 0

0.5

1

1.5

2

2.5

3

3.5

4

0

0.2

0.4

8

8

7

7

6

6

5

5

4

3

2

2

1

1

435

440

445

450

455

460

0

50

100

7 6 5 4 3 2 1 0 5

10

150

HI

8

TOC %

1

0 430

TMAX

0

0.8

4

3

0 425

0.6

PI

TOC %

TOC %

S ONE

15

20

S TWO

Fig 4: Scatter plots of observed data.

23

200

250

300

Optimal Transform 4.00E-01 2

y = -1.4838E-05x + 5.9391E-04x + 1.1217E-01 2 R = 9.3847E-01

2.00E-01

0.00E+00

HI_Tr

-2.00E-01

-4.00E-01

-6.00E-01

-8.00E-01

-1.00E+00 0.00E+00

5.00E+01

1.00E+02

1.50E+02

2.00E+02

2.50E+02

3.00E+02

3.50E+02

6.00E-01

7.00E-01

HI

Fig 5a: Optimal transform of HI_Tr Optimal Transform 3.00E-01

2.00E-01 2

y = 1.1351E+00x - 1.9890E+00x + 2.5403E-01 2 R = 9.9187E-01 1.00E-01

PI_Tr

0.00E+00

-1.00E-01

-2.00E-01

-3.00E-01

-4.00E-01

-5.00E-01

-6.00E-01 0.00E+00

1.00E-01

2.00E-01

3.00E-01

4.00E-01

5.00E-01

PI

Fig 5b: Optimal transform of PI_Tr 24

Optimal Transform

2.00E-01

1.50E-01

1.00E-01

SONE_Tr

5.00E-02

0.00E+00

-5.00E-02

-1.00E-01 y = -2.1687E-02x 2 + 1.2107E-01x - 3.3104E-02 R2 = 5.7838E-01

-1.50E-01

-2.00E-01 0.00E+00

1.00E+00

2.00E+00

3.00E+00

4.00E+00

5.00E+00

6.00E+00

7.00E+00

8.00E+00

-2.50E-01 SONE

Fig 5c: Optimal transform of S1_Tr

Optimal Transform 9.00E+00 8.00E+00

2

y = 5.5876E-04x + 1.5914E-01x - 5.8834E-01 2 R = 9.9862E-01

7.00E+00 6.00E+00

STWO_Tr

5.00E+00 4.00E+00 3.00E+00 2.00E+00 1.00E+00 0.00E+00 -1.00E+00 0.00E+00

5.00E+00

1.00E+01

1.50E+01

2.00E+01

2.50E+01

3.00E+01

3.50E+01

4.00E+01

-2.00E+00 STWO

Fig 5d: Optimal transform of S2_Tr

25

4.50E+01

5.00E+01

Optimal Transform

3.00E-01 y = 1.2005E-03x2 - 1.0616E+00x + 2.3463E+02 R2 = 7.7580E-01

2.50E-01

2.00E-01

TMAX_Tr

1.50E-01

1.00E-01

5.00E-02

0.00E+00

-5.00E-02

-1.00E-01 4.25E+02

4.30E+02

4.35E+02

4.40E+02

4.45E+02

4.50E+02

4.55E+02

4.60E+02

TMAX

Fig 5d: Optimal transform of Tmax_Tr

Optimal Inv Transform 2.50E+01

2

y = -7.2503E-02x + 2.6753E+00x + 2.7566E+00 2 R = 9.9534E-01 2.00E+01

TOC

1.50E+01

1.00E+01

5.00E+00

0.00E+00 -2.00E+00 -1.00E+00 0.00E+00 1.00E+00 2.00E+00 3.00E+00 4.00E+00 5.00E+00 6.00E+00 7.00E+00 8.00E+00 9.00E+00 TOC_Tr

Fig 5e: Optimal Inverse transform of TOC_Tr

26

Fitted Stdev = 0.6098

25

HI_Tr= -1.4838E-05x2 + 5.9391E-04x + 1.1217E-01 PI_Tr= 1.1351E+00x2 - 1.9890E+00x + 2.5403E-01 SONE_Tr= -2.1687E-02x2 + 1.2107E-01x - 3.3104E-02 20

STWO_Tr= 5.5876E-04x2 + 1.5914E-01x - 5.8834E-01 TMAX_Tr= 1.2005E-03x2 - 1.0616E+00x + 2.3463E+02

From Fit

15

10

5

0 0.00E+00

TOC= -7.2503E-02 SumTr2 + 2.6753E+00 SumTr + 2.7566E+00

5.00E+00

1.00E+01

1.50E+01

2.00E+01

2.50E+01

TOC_Meas

Fig 6: Fitted standard deviation Using this transform, TOC-values in the 16 locations of Cambay basin were estimated and is shown in the Table 2. WELL NO. W#1 W#2 W #3 W #4 W #5 W #6 W #7 W #8 W #9 W #10 W #11 W #12 W #13 W #14 W #15 W #16

ESTIMATED VALUES 2.131972382 4.468453885 4.032998371 1.707398868 2.180092009 2.353108452 1.837916231 0.830580021 2.038260225 1.888477412 2.269525982 1.759567135 3.039179375 2.224399075 3.772538396 2.920592623

Table 2: TOC amount at different stations. 27

6.

CONCLUSSIONS Satisfactory match between measured and estimated value of Total Organic

Content (TOC) shows that Alternating Conditional Expectation methodology have been successful in establishing a relationship between TOC and other five variables S1, S2, HI, PI, Tmax. This relationship can be used in the same basin for TOC evaluation.

7.

FUTURE WORK 1- To locate or to find sweet spots of shale gas, this can be done in terms of porosity & permeability. 2- To find porosity & permeability from log data using ACE. 3- To generate 3D variation of porosity &permeability to locate sweet spots using Geostatistical Tool viz GRIDPROSTAT.

28

References 1. Breiman, L, and Friedman, J,H.: Estimating Optimal Transformations for multiple regression and correlation .Journal of the American statistical Association ( September 1985 ) 580. 2. SPE (35412), Akhil Dutta – Gupta, SPE, Peter Vailo, SPE, and Tom Blasingame, SPE,Texas

A&MU:

Optimal

Transformations

for

Multiple

Regression:

Application to Permeability Estimations From well logs. 3. A Report on Assessment of Gas Desorption Studies Cambay basin UCR group KDMIPE ONGC. 4. Dhar P.C. and Bhattacharya S.K, “Proceedings of the second seminar on petroliferous basins” page 1-32, vol. 2,

December 1991, KDM Institute of

petroleum Exploration. ONGC, Dehradun.. 5.

Raju A.T.R, and Srinivasan S. “Cambay Basin – Petroleum Habitat” page 33-78, vol. 2, December 1991, KDM Institute of petroleum Exploration. ONGC, Dehradun..

6. http://en.wikipedia.org/wiki/Shale_gas 7. http://www.centreforenergy.com/generator2.asp?xml=/silos/ong/ShaleGas/shaleG asOverview01XML.asp&template=1,2,4 8. http://www.slb.com/media/services/solutions/reservoir/shale_gas.pdf 9.

http://www-odp.tamu.edu/publications/tnotes/tn30/tn30_11.htm

29

30