PetroPulse Magazine | Issue 7

Life is unfair, every way you take in your life will be full of troubles and challenges that try to pull you backward and weaken your own resolve. There is a quote that says “The only way of finding the limits of the possible is by going beyond them into the impossible.” That means you are the only one that can know your abilities and stamina. Therefore, never underestimate your own abilities.

The beginning of the story is not like its end, someday you will get what you want. So, start now to correct your life path. Break your comfort zone, determine your destination, keep in touch with your fears, be realistic in dealing with your problems and care about being surrounded by people who believe in you. Shove your way by a strong will, persistence is the vehicle you arrive in, you can break all expectations, just keep going ahead.

The common thing between successful people and the ordinary ones is the 24 hours per day, but the difference lies in how each of them uses their time. Each second passes is a chance and you are in charge of taking the advantage of it. Every second may contribute in changing the world, that is what successful ones believe in. The reasons for success are unlimited. Good planning, hardworking, strong will, and cooperative partnership are reasons, but if you looked deep inside each reason, you would find that it is all about you again and again. You are the controller of your life. You are the only one who knows your abilities, and according to that, you can dream, plan, work, and choose the best partners to help you succeed.

In searching about the unique value, I joined AAPG SU SC as a Magazine Editor. Now, I am the Editor-in-Chief there. It is a great journey, the journey of the main target of enhancing myself. As known, “The journey of a thousand miles begins with one step,” the journey of enhancing my own self has begun by joining AAPG SU SC, as well. Everything began to be better. Everything changed, from A to Z such as how I see everything around, how I can face the obstacles that I meet in life, and many other things. This is an unforgettable stage in my life that added to me a lot of value and helped in finding the right path of success, I think it is the best ever.

Finally, there is no sanctuary for being successful. So, always remember that your life is a closed loop and you are the primary engine for it, a battle and you are the only champion. So, accept it bravely, you deserve the best of everything.

In talking about PetroPulse, I feel proud to be part of this great project. As known, PetroPulse plays a vital role in reducing the gap between the academic life and the world of petroleum industry through its articles written by highly-experienced engineers and its interviews with perfect-profile experts, all of this will enhance the reader with the required knowledge. It is a great feeling to be in charge of such a great magazine.

On the other hand, we have a non-technical magazine, Aspire, concerned with many topics such as Translation, Time Management, Human Resources, and many other topics. Aspire started its journey last season, at Es3a 4. The 2nd issue saw the light in March 2019, at Es3a 5. It aimed at raising the success level of the 1st issue. I hope the new issue of PetroPulse takes the same track and meets all the expectations.

At the moment we succeed, I will not forget to thank who helped in paving the impassable. Therefore, I would like to thank everyone contributed in preparing the 7th issue of PetroPulse and special thanks to the editors and designers teams for their fabulous efforts in trying to add new fascinating ideas to create a higher level of professionalism and open new horizons for more and more creativity and inspiration. They stood as one hand to achieve this target as we learned at AAPG SU SC. I am sure, they are the key factor for the success of PetroPulse.

2

Believing in what you do, even if you see no direct return on yourself, is the key for perfectly carrying it out. “Volunteers are not paid not because they are worthless, but because they are priceless” Sherry Anderson. Later, you will discover the reflection of your efforts on your knowledge, skills, character, and especially on the community that you are meant to serve.

I have lived this worthy lesson through my three years of volunteering in AAPG Suez.

Many undergrads look through a narrow scope especially when they first join the college; having good technical knowledge in their major field is more than enough for them to be hired, and it was what I thought by the way. However, nothing counts only on the technicality and no one will invest in a zero-skills person. Lucky ones will later find out the right scope, our duty in AAPG Suez is to aware them, that they have to develop themselves and the first step is to engage. Neither hidden talents nor lack of experience would be discovered if you remain in your comfort zone.

If you choose to be the life lived for others and the one who is responsible to boost your community, you should be initially able to carry this responsibility out through having self-awareness of your abilities, never stopping learning and always understanding that you need more. You should be well-armed with knowledge and soft skills. And as I said later, while you are carrying it out, you will gain much more experience.

“Doing nothing for others is the undoing of ourselves” Horace Mann. That was the first and foremost thing we were thinking about when we started the tenth season and it spontaneously was reflected on our slogan; “Pave The Impassable”.

We work as a one-man army to make the way easier for everyone, empower students, to help graduates achieve their goals, and to make a smile on the elderly and the orphanage in our community.

For me, being recruited in AAPG Suez is the whole experience that I’m proud to have. I thought I was good with my skills but once I engaged with some experienced members I actually found that I am far away under the minimum limit. Step by step until I am currently the President of this great entity. No words can never explain how this place changed me and enhanced my skills. Generally, I believe that just dealing with calibers here in the chapter or there in the top management of companies that we deal with, would be enough to develop me.

In AAPG Suez, we could make it and solve the hard equation of ensuring quality besides quantity. Although only sustaining our annual events and projects is a challenge itself, we never stop aspiring for more, we never stop planning, and we always set higher goals. Nine years of experience are our legacy for the tenth season throughout sustainable development.

Celebrating the 10th anniversary at Suez University, I will give you a brief about some of our achievements this year.

Regarding the technical part, we have held the third Oil and Gas Industry conference that provided exciting topics and competition by an elite of petroleum engineers and managers from reputable companies. We have also made PetroUp the fourth, concerning with some new technologies. Moreover, sessions, field and yard trips to petroleum companies and to geological places in Egypt. On the other hand, the annual non-technical mega event, Es3a – Seek The Peak Conference the fifth which witnessed the first innovation competition in AAPG Suez for start-ups with more than three hundreds and fifty attendees. In addition, we take care of young generations from high school through Minute Schools project. Besides undergrads, graduates, and high school students, we are keen to serve the community through charity visits.

AAPG Suez has launched the first Chat Bot and further enhanced its website, Android app, and blog. PetroPulse is launched together with a non-technical magazine, Aspire. The members’ development programs in AAPG Suez have been a brand for three years. Finally, we seek to conduct this knowledge to other student activities in Egypt through Activation program.

All of this would have never happened unless we had an extraordinary high board, an unprecedented board, and a great family; able to develop, plan, execute and make a change! Believing in what they do, believing that they could always make a change to the community and to themselves, have made AAPG Suez proudly crowned as the Outstanding International Student Chapter.

3

Let us start with your career and role at ENI? - I have started to work in the National Research Institute of Astronomy and Geophysics for almost two years. My studies were related to geophysics and its tools specially for potential tools like seismic graph, magnetic, and so on. I studied well the seismology and earthquakes. This approach is almost the start to all petroleum exploration as we use the seismic wave technology, generate a wave, send it into the earth, and receive it back by recording sensors, these received waves tell us about the possible subsurface structure and tectonic regime in that region.

- Then, I moved to Petrobel and stayed there for almost ten years working all over Egypt specially in the Mediterranean and Gulf of Suez.

- After that, I went to Tharwa Petroleum Company for almost two years and there was a big acreage in the Western Desert that added to my experience. Beginning from here, my vision started to increase. When I moved to Tharwa, a first real exploration company working outside the region you know, you suppose to have a bigger regional vision including whole of north Africa and some countries in the Middle West.

- When I was localized in my work, I focused only on where I were working but when it became more regional, my vision increased a lot including the basins. And then focus toward development lease which we can start production. But the real exploration started from regional view, this is the point. After this, I moved to ENI international, I stayed almost six years in Pakistan, South West Asia.

It was the first time to work in real collision regime like what we see in the mountains like in front of Himalayas and Soliman Mountains.

4

For me this could be the start for real exploration outside of our area. Finally, I came back to Cairo in 2013 to be an exploration project manager for Gulf of Suez and the western desert for IEOC. So, this is the whole of my story for about 22 years.

Is being a part of ENI was your ambition?

- Sure, ENI represents one of the biggest companies in oil & gas industry and it is quiet interesting for me. I think it is a goal for explorationists in Egypt as ENI is a leading company, wellestablished, and well-organized. It is a big organization that will certainly add to you a great experience.

What about situations that act as turning points in your career?

In fact, there was more than one: - First of all, when I was in Petrobel after 5 years, I had to work with a team and then started to understand the meaning of team work and my vision completely changed. Now, I am not only a seismic interpreter but also I have to be aware of all other sectors of business itself. As you know, I am an explorationist, I am supposed to focus on how to explore and add reserve to the company but in the same time, I have to work in production, drilling, and other sectors as well. Of course, it is interesting to know about other sections and that made me understand what is going in the whole business and gave me a capability to control and make a decision when I am in a position of managing.

- The second one was when I moved to Tharwa Petroleum Company. It was a good chance to increase my scope from an area having limited extension with localized development lease to an area looking for exploring in and outside Egypt. When I was in Tharwa, I have attended a lot of global work like in Cyprus which was supported by the minister of petroleum. The ministry created a team from Egypt to study what is going up there and I was chosen to be part of this team and we succeeded in evaluating and predicting what was going on the East Mediterranean as the whole project.

- The third one was the moving from local to international work at ENI International in Pakistan at South West Asia. Surely when you work over a certain area, you start to study to improve your knowledge and experience to be capable of understand what is happening around. But it was completely different to go from one environment to another like the Indian passive margin that parallel with collisions between plates. That gave me great experience.

- Finally, when I came back to Cairo in 2013, I really wanted to give my country a part of my success by making a great discovery in my own country. And being part of ENI gave me the chance to do that when we worked on Zohr field, I was one of who brought this project into real.

What is ENI looking for when it comes to invest in Egypt?

- Now, there is a great improve in our internal security in recent times which in turn encourage not only ENI but also any other investigating company to come and start its business in Egypt. So, ENI are widening its investment in Egypt by taking a new concession in Delta and Western Desert. This investment is due to the good environment and well-substructure. Also, our good reserve of hydrocarbons makes a good reason to come and start a business.

What do you think about the future of Oil and Gas industry in Egypt?

- As it is clear, Egypt is one of the lucky countries in energy field as it has a quite good reserve of oil & gas and we see that in the great recent discoveries in Gulf of Suez and the Western Desert.

These discoveries provide Egypt a sustained energy sources which supply a great part of factories and cities. And now, we are going to self-sufficiency and in near future, we will export. This of course will save a great amount of hard currency for Egypt.

How is petroleum exploration important in Oil and Gas industry?

- Petroleum exploration is the key of the whole process. It is very important to add reserve to your company to maintain its life. In oil and gas industry, every section has his role from exploration to drilling then production then development, all of these teams form an integrity to keep the company. But the future of any company depending on the exploration team whose job is to add reserve and maintain company life.

Zohr field is considered to be one of seven record breaking projects for ENI, could you tell us more about it?

- Surely, Zohr field was previously localized in the north-east Mediterranean concession, and it was a large concession under the management of Shell. After Shell left the project, ENI was lucky having that site and they saw the area as a good potential so we are encouraged to get it, we started to do all studies about this site then started to drill. There was a risk, but sometimes you should take it, that decision was a good one that revealed that great discovery. We are talking about one of the biggest discoveries; about 30 TCF and up to now still valid with a good production rate.

This project we were calling it the mission impossible as we should discover, drill, start production, and development in just two years that was the great challenge.

We were able to finish a project with a time frame of five or six years in only two years and that shocked others. That how we accomplished such a project -deep water 1500 m- and to build the facilities about 150 km from the site to port-said. This was a really success based on great effort and team work to accomplish this mission impossible.

At ENI, what are the main qualifications that you are looking for?

- First of all, hardworking of any student is shown by his grade.

Sure we are looking for good personality, soft skills, language is very important point as you deal with multi-national company. So, you should work hard on yourself improving your technical knowledge, attitude and personality as well. Also, being professional in dealing with others makes a difference, so you should start to improve you own, learn a language like Italian, German, and of course English.

Hardworking and only hardworking will lead you to get what you want.

What is your impression about student activities and specially about AAPG SU SC? Student activities now are totally different and very useful. And as it clear, the new technologies of communication and media have a big role in that improvement. For AAPG, I think that is a good thing to have such reach and make such conferences by only students this very impressive. It is a good point that only students take the responsibilities and have the passion to make these events from A to Z, that really widen your horizons and increase your knowledge.

5

Could you tell us about yourself and the whole As you discussed, the whole journey was in journey of your career?

Schlumberger. So, was it your dream?

- My name is Karim Badawi, I joined Schlumberger in 1996, so almost 22 and a half years ago. I studied mechanical engineering at the American University in Cairo and started first as a Field Engineer and wireline in Indonesia. I stayed there throughout my field career for three and a half years which was a very good experience, getting to know different cultures, people and different working environments. Then I got transferred to Aberdeen where I worked on offshore fields. After that, I became a Field Service Manager. I got promoted to Wireline Training Center manager in Egypt followed by Wireline Recruiting, Training and Development Manager worldwide, it was an interesting position as I was involved in setting up the recruiting strategies and the training facilities.

- In 2006, I came to Egypt as the Operation Manager for wireline for Egypt and East Africa. Then, I moved to Houston, in a new role, worked with the headquarters to look after all the different business systems and how can you have these business systems be tailored to enhance the efficiencies of operations.

- Then, I moved to Russia in 2010 and in Russia, I had four different positions. I worked as IT Manager and a year and a half later, I became the Vice President of shared services for Russia and Central Asia. After that, I became the Vice President of Testing services for 2 years and then, I became the Reservoir Characterization Manager for Russia and Central Asia, overall my experience there was for about 8 years and I was very lucky to have different exposure from the business perspective and interactions. And then almost exactly a year ago, I moved to my current position which is the Managing Director for Schlumberger for Egypt and East Mediterranean.

6

Absolutely, I was studying Mechanical Engineering and I initially thought that I would be working for a factory but the ambition started from the time I saw the presentation about what is Schlumberger and what is the life of the field engineer and what is the career that Schlumberger offers and breadth of global footprint that exist and that you can succeed in Schlumberger, no matter what nationality you have. So, it was my dream to join Schlumberger.

You must have faced a lot of difficulties to get this position, could you tell us more about?

And how did you overcome them?

I would not call them difficulties; I would call them opportunities.

When you look at my journey, I have worked for different product lines and different countries. I think it was a fantastic opportunity to be exposed to different cultures, to meet different people from different backgrounds, to be able to contribute to the business.

I overcame those challenges by having a positive mindset and a mindset which is open for change and having a mindset of always wanting to learn and to contribute and to really provide value to whatever role I am doing, in whatever place I am being positioned, and by putting a lot of focus on teamwork and collaboration and have the determination to contribute and add value wherever you are.

What is Schlumberger looking forward to in Egypt in the future?

How is Information Technology important in the oil and gas sector?

Schlumberger is very motivated about its presence in Egypt.

We have always made sure that Egypt receives all the latest technologies that we have. We always seek to help our clients to achieve their goals and Schlumberger in Egypt is a very important partner in the oil and gas sector to bring the highest level of safety and the highest level of technology and flawless service. We have been fortunate to be a key partner to the oil and gas sector in major projects such we have been part in the team of the discovery of Zohr field, for example. Our reservoir wireline tool has a special role to be able to take the first sample of the fluid from Zohr.

It is very critical. IT is taking a new dimension in the oil and gas sector. Schlumberger has been at the forefront of the digitalization of the oil and gas sector and this is very important because now we have the capabilities of using data to be able to make better decisions. Schlumberger is investing a lot in the digitalization front a lot of alliances that Schlumberger has with Google and Microsoft in terms is artificial intelligence and how we can develop smarter technologies which can actually leverage and have better insight based on technology and all of this helped to reach a higher level of efficiency to the industry but more importantly, it will really help the industry to accelerate the time to discovery and accelerate the time between discovery and first gas or first oil. It will also maximize the whole production process and optimizing the full potential of the reservoirs.

What is the role of Schlumberger in new discoveries in Egypt?

At Zohr, Schlumberger delivered the early production facility in a record time to be able to have this first gas in 2017 and up until today. We are operating this production facility and we are very proud of our presence in big projects like this and the BP Atoll field as well. We have managed to break world records in terms of deliveries if you look from time of discovery to first gas which is something unheard in any part of the world and a real benchmark in the industry.

The other part I feel proud of working in the western desert with the Apache group, Shell and many other operators there to help in increasing the discoveries and the production of oil from the western desert. We also worked in the Gulf of Suez which should be a new area for exploration for Egypt.

As known, Schlumberger plays a vital role in the development process in the oil and gas sector in Egypt, could you explain that?

What would you say to the students to be as successful?

It is a difficult question, but I think all students need to work hard, to be passionate about what they do, to always have a positive mindset. Do not wait for things to come to you, be proactive and whatever opportunity that you have do the best you can. Learn the most you can from it and try to contribute the most to it. Always think of how you can develop yourself, how you can contribute to the organization that you are working for, and how you can be useful for this organization. Never stop learning and continue your learning path to have innovation, advancements in technology and organizations. Always remember that success in the world relies on teamwork and collaboration.

We are working on training and development of three categories of people; students through engagement with student chapters by giving the exposure to the students to linking the academia to the oil and gas real technologies. The second category is the fresh grads who have not yet been employed, We offer them training and exposure So, they get hired in the oil and gas sector.

The third category is the management program from the Ministry of Petroleum and Mineral Resources. So, we are proud to be contributing to people development.

Finally, what is your impression about student activities and AAPG Suez especially?

Student activities are very important, and I am very supportive of the Schlumberger team to lead activities for the students. I encourage you to fully engaged in extracurricular activities because it is how you develop other than the academic part and it will be very beneficial. I think that you are an excellent pool of talent and at the same time I think you should feel very proud of your activities, keep up your activities, keep up your visibility and always think of how can you contribute to the oil and gas sector before graduating, once you graduate and after you graduate.

7

Flowing Material Balance; Simplicity & Power Hesham Mokhtar Ali Senior Reservoir Engineer at General Petroleum Company

Dynamic or Flowing Material Balance (FMB), as a branch of the Production Data Analysis (PDA), basically means the use of production data (flow rates & flowing pressure) to perform some analyses or diagnostics aiming to have information about the reservoir characteristics, reserve volume, hydrocarbons initially in place, etc.

PDA has been developing through time starting from the explicit concept of: - PA (Production Analysis) - PTA (Pressure Transient Analysis) - RTA (Rate Transient Analysis) - FMB (Flowing Material Balance)

The introduction of FMB mainly converted the constant bottom-hole flowing pressure (BHFP), production scenario into constant flowing rate production conditions. The constant BHFP assumption basically utilized in the conventional “empirical” decline curve analysis (Arps Equation) and also Fetkovich advanced DCA, that combined the previously Arps empirical solution for boundary dominated flow and analytical solution of transient flow equation. Determination of hydrocarbons initially in place (HIIP) and reserve is a fundamental process for Reservoir Engineering and the Subsurface Team. The reservoir estimation techniques can be classified into two basic groups; volumetric calculation (initial estimate) which is commonly applied in the early stages of oilfield development. Then, when sufficient production data is available, the second group (performancebased estimates) can be applied. Where the traditional or static material balance (MB) method uses actual reservoir performance data, therefore it is generally accepted as the most accurate procedure for estimating HIIP due to the involved assumptions that the MB can only view the reservoir connected volume.

To generate a traditional MB analysis, the wells are needed to be shut-in regularly several times throughout their production life to obtain accurate estimation for the average reservoir pressure. However, this is usually impractical in the actual oilfield operations due to production losses, and the duration of the shut-in is often not long enough to obtain pressure equilibrium or stabilized pressure measurements specially in tight reservoirs. Which considered the main limitations or pitfalls in performing traditional MB analysis. FMB, on the other hand, uses the concept of boundary-dominated flow (BDF) or pseudo-steady state (PSS) flow, to handle this scenario. The reservoir pressure distribution typically follows the transient flow period then the stabilized conditions or BDF period which means the reservoir boundaries have been reached. That actually the interest of reservoir volume determination and reserve calculations.

In BDF condition, the reservoir pressure will decline throughout the

8

entire reservoir at the same rate (Figure 1). Therefore, the pressure drop recorded at the well at certain time will represent the pressure drop in the reservoir at the same time.

The procedure of FMB is simply involving conversion of well flowing pressure to equivalent average reservoir pressure which facilities the application of MB while the wells are producing (no shut-ins are required).

FMB analysis plots a normalized-rate versus normalized cumulative production, on a linear scale to get STOIIP (N), where no type-curves are plotted. FMB methodology utilizes the concepts of material balance time (MBT; te) to convert variable rate/variable pressure production scenarios into equivalent constant rate production scenario. MBT, by definition, is the time needed to produce this cumulative production amount with the instantaneous flow rate value (Figure 2). In gas reservoir, where the gas properties are highly dependent on pressure, MBT will be defined as material balance pseudo-time.

MBT;

te =

Np qo

From definition isothermal compressibility of oil,

co = −

1  ∂V  1  Np  =   V  ∂P T N  pi − pr

   T

Where the initial volume is N (STOIIP), and net change in volume will be the cumulative production (Np). Incorporating this definition with PSS flow equation, we will get;

In the previous equation, two different pressure losses; pressure loss due to depletion (pi-pr) and pressure loss due to inflow (pr-pwf). For simplicity, the second term will be replaced by PSS constant (bpss) and substitution for MBT;

Therefore, by plotting rate-normalized pressure (dp/q) vs. MBT (te); the resultant straight line will has an intercept of bpss (Figure 3). Then, the average reservoir pressure will be estimated as;

For determination of STOIIP (N), the equation will be re-arranged as;

Applying the previous equation as shown in Figure 4, X-axis intercept will be the value of N, and Y-intercept will be the well’s productivity index.

Fig 4: Estimation of N by FMB Procedure

Field application of an oil well with production data for about 10 days (reservoir limit test) is shown in Figure 5.

Fig 1: Typical Reservoir Pressure Distribution

Fig 5: Production Data for Oil Well; Flowing Pressure and Flow Rate By applying the previous procedure of FMB on that well, the results showed very high agreement between the estimated STOIIP by FMB for ONLY 10 days with results of traditional MB for about 2 years of production! A reliable estimate of the productivity index is also given at early time of the field development.

Fig 2: Concept of Material Balance Time (MBT)

Fig 3: Estimation of bpss from BDF Period

Finally, application of FMB is following a very simple procedure with a reliable results; 1. No shut-in is required for wells; only production data is needed.

2. Accurate estimate for STOIIP is achieved with very short production periods.

3. STOIIP estimation from FMB is considered one the powerful performance-based techniques, where the connected volume is only considered.

4. Early and accurate determination of well’s productivity index (PI) is easily determined from the reciprocal of PSS constant (bpss)

9

Status of Shale Gas Reservoirs; Technologies, Challenges and Developments

The contribution of hydrocarbons from shale to the global energy mix has significantly increased in the past few years and is slated to increase further in the near future. This increase can largely be attributed to the following techniques: 1. Horizontal drilling and accurate well placement within the resource play 2. Advances in resource characteristics, formation evaluation and completion modeling to determine the quality and quantity of the hydrocarbon actually stored 3. Improved horizontal fracture design, monitoring, optimization and cost-effective methods for fracture placement.

Shale formations contain a variety of hydrocarbon such as for this variety and a combination of these resources can be seen in Eagle Ford Shale play, Texas, or in the Marcellus Shale in the Eastern United States to name a few.

Some of the important factors in shale gas development are: 1. Socio-political favorability 2. Technological advancement 3. Expertise 4. Supply-chain management 5. Reservoir 6. Infrastructure Amount of recoverable reserves and drilling technology are the biggest factors for the economical development of shale reservoirs, where drilling, completion and fracture operations must maintain very high rates of efficiency with minimum downtime (non-productive). Along with high heterogeneity, since each shale reservoir has different intrinsic reservoir properties (geochemical, geomechanical, petrophysical, etc.), every reservoir requires specific customized evaluation analyses and modeling.

To suffice economic and environmental constraints while meeting current and future demand for hydrocarbons, integral factors that development and production plans must take into account include fracturability and producibility.

There are shale gas reservoirs distributed all around the world as described

by EIA report of 2016. According to the EIA report there is about 7795 tcf of gas recoverable from shale plays across the world. Some of the leaders in shale gas include: 1.

2.

3.

4.

5.

6.

7.

8.

9.

United States of America – 1161 tcf China – 1115 tcf Argentina – 803 tcf Algeria – 707 tcf Canada – 573 tcf Mexico – 545 tcf Australia – 437 tcf Russia – 285 tcf Brazil – 245 tcf

10

Shale Gas Stimulation Advancements in horizontal drilling techniques and multi-stage fracturing designs have allowed for the economic exploitation of unconventional resources. To create massive and complex hydraulic fracture networks, necessary to connect the tight shale gas formation to the wellbore, horizontal wells are drilled in the direction of minimum horizontal stress.

Fracture and Completion Design. Optimization of fracture and completion design is the key factor for maximizing the shale gas reservoirs productivity. Multistage fracturing has been advances significantly over the past the decade to create complex fracture networks with maximum contact area with the reservoir. Well spacing, fracture spacing, perforation cluster spacing, fracture geometry, and limited entry perforation are among many factors that should carefully considered when designing multi-stage fractured horizontal wells in shale gas reservoirs.

Well Spacing. Well spacing controls, the total developed reservoir area and the total development cost. It defines the total number of wells, drilling and completion schedules and the field production curve (Suarez and Pichon 2016). Well spacing is highly dependent on geomechanics, petrophysics, stimulation design, hydraulic fracture geometry, and reservoir fluids.

well is producing and the fluid is leaking off from the microfractures, their width and conductivity lower to near zero. Acid does not have any physical restriction in accessing microfractures.

Fracture Spacing. The stress shadow is controlled by the fracture spacing in the horizontals wells. Increase of fracture spacing minimizes the effect of stress shadow and hence better control over the trajectories of the fractures. The lag time between successive fractures is also important to allow fractures close to the propped fracture width. Stress shadow effect can significantly reduce the horizontal stress contrast which leads to more fracture complexity when creating multiple fractures from a single, horizontal wellbore. This helps to improve the productivity of shale gas.

Fracture Geometry in Shale. Enhancing gas productivity may require decreasing the fracture spacing and increases the number of stages.

However, this is expensive and may cause interferences and reduction in gas production.

CO2 Injection for Enhanced Gas Recovery. Alnoaimi and Kovscek (2013) studied the capability of shale gas reservoir for carbon dioxide sequestration. They conducted physical and numerical pressure transient, pulse decay experiments on Eagle Ford shale cores. They found that CO2 has strong sorption on shale gas and that CO2 can desorb methane from shale formations. Hence, CO2 can be used to increase the productivity of shale gas reservoir by desorbing methane from kerogen and clays. CO2 injection into gas shales could be highly beneficial economically and environmentally: incremental recovery of adsorbed methane, and secure CO2 storage Godec et al. (2014). Fathi and Akkutlu (2014) developed a mathematical model based on Maxwell-Stefan formulation to simulate CO2 injection into shale gas reservoirs. The model shows that due to the counter diffusion and competitive adsorption in the micropores, CH4 production is significantly enhanced at the reservoir scale.

Reservoir Modeling and Simulation in Unconventional Plays As unconventional resource is becoming more important for world energy supply, the tool with the ability to predict well performance remains a big challenge for the industry. The need of hydraulic fracturing that creates complex fracture network in reservoir, gas transport system from ultralow matrix permeability system, and complex porosity system that need to be incorporated make modeling work in reservoir simulation or empirical solution are more complicated.

Fracturing Fluids and Its Impact on Fracture Geometry. Selection of fracturing fluid is a critical point in shale reservoir completion process(Chong et al. 2010). Hydraulic fractures in shales are usually complex networks of fractures with various lengths, heights, and widths. However, the mechanical properties of shale can control the degree of complexity of the fracture network. Ductile shale can produce a more conventional biwing fractures. Thus, ductile shale requires a fully packed fracture wings to improve conductivity. Thus, it requires a fracturing fluid with higher proppant carrying capacity and can evenly distribute the proppant across the fracture. However, as the shale becomes more brittle due increase of calcite and/or quartz contents, it can be fractured easily (King 2010). In this case slickwater can be efficiently used to create a complex fracture network. Gelled, crosslinked gels, and foams can be used to increase the proppant carrying capacity and fracture width (Sayed et al. 2015).

Suarez and Pichon (2016) studied two different fluid systems; hybrid fracturing treatment using a combination of low viscosity fluid (slickwater) and high viscosity fluid (crosslinked gel). The use of high viscosity crosslinked gel as a fracturing fluid allows to pump higher proppant concentration thus reducing fluid volume which results in a lower standard deviation in simulated length. Consistent hydraulic fracturing geometry between stages allows a more even drainage for a given well spacing.

Shale Acid Fracturing. Many shale reservoirs accommodate a significant amount of carbonates, minerals that can be easily dissolved in acid solutions suggesting acid hydraulic fracturing as an enhanced-recovery stimulation technology (Igor and Espinoza 2017). Calcite and dolomite in shales can be either embedded into the rock matrix or deposited and localized within natural fractures (so-called veins). Extending fracture network and reducing the closure of natural fractures are desirable to increase and maintain reservoir fracture permeability. Combining hydraulic fracturing with acidizing can be an enhanced stimulation technique in calcite-rich shales.

Wu and Sharma (2017) suggests that shale acidizing can create a network of channels along fracture faces that result in permeability enhancement.

Hydraulic fracturing also creates many microfractures which can be preexisting natural fractures or induced unpropped fractures branching from the main propped fractures. They are too small to be accessed by most commonly used proppants. Microfractures are thus likely to be closed during production, and considerable productivity may be lost because these microfractures contribute in a substantial way to production. As the

From all the work that has been done in numerical simulation, the advancement of computer technology has increased rapidly allowing models to better capture more complex phenomena, from dual to quad porosity system or from simple SRV to complex fracture network model. But even though many modeling approaches have been explored, most of them have their own limitations to simulate the entire hydraulics fracturing process. The key is that engineers must understand the limitations for each work and apply the model appropriately based on engineering judgment.

Obtaining good reservoir characterization is the most important step to reduce uncertainty in reservoir simulation to get better production forecast.

11

Offshore Well Testing; Challenges and Its Mitigation

In a typical offshore well testing operations, the drill stem test tool string has a downhole shut-in valve, packer to isolate the zone of interest, downhole gauges to measure annulus and tubing pressures, temperature and screens for the fluid inlet. At the surface, we have the choke manifold to control the rates. Heave compensator at the top of the tool string, used for compensating the vertical motion of the rig.

The operational sequence consists of a clean-up period at the start, followed by the first build-up, followed by single or multi-rate tests and finally, we have sufficient final build-up which could be interpreted to derive reservoir parameters. In this case study, we showcase some challenges faced in offshore well test data acquisition such as rig heave movement, tidal effects, gauge movement and downhole screen plugging.

Case 1: Rig Heave Movement Figure 1 shows pressure vs time plot for the entire sequence of a DST job. During this job, due to operational constraints, the well could not be shut-in using the downhole valve (IRDV). Therefore, the well was shut at the choke manifold. If we focus on the final build-up data, we realize that it is affected by noise, which can be seen in both tubing and annulus gauge.

Figure 2: Noise Investigation Due Rig Heave Movement To mitigate the noise, close the master valve on the flow head or the lubricator in the subsea to cut the communication between the moving choke and the downhole gauges.

Figure 1: Pressure vs Time Plot for the Entire Sequence of a DST Job The investigation was carried out to know the reason for the noise shown in the Figure 2, in semi-subs or drill-ships, even though we have compensators to prevent the DST string from moving due to heave, the choke manifold will be moving up and down, changing the hydrostatic pressure at the bottom of the DST string. To validate this hypothesis, a simple calculation was performed to calculate the rig movement (change in hydrostatic column height) by using the pressure change we observed with the gauges and the density of the fluid in the column.

Rig movement came to around 12-14 ft which correlated closely with the reported rig heave by the operation team.

Case 2: Tidal Effects The tidal effects are not simply some noise in the data, but a composite of eight or more signals with different periods and amplitudes. As a result, this problem cannot be removed by “smoothing� the data but by using an algorithm and a reference tidal signal, usually from a seabed gauge. In the case study discussed here, reference signal data is not available. So, a software algorithm-generated reference signal is used for this purpose.

Figure 3: Recorded Pressure Data Showing Raw and Corrected Tidal Trend

12

Figure 3 shows recorded pressure gauge data along with tidal correction.

It is recommended to run a seabed gauge as a reference to remove the tidal signal. It is sometimes possible to generate an approximate seabed pressure response using tidal data from databases which in our case was provided by the commercial PTA software.

The Figure 4 shows the derivative plots of the main build-up where the tidal correction was applied. The oscillations on the derivative seen on the original gauge data start from about 4 hours and last to the end of the tests. This is due to the tidal effects. The low amplitude tidal effects may become dominant at late times as the DP changes are getting smaller and smaller. The tidal signal has been identified in our case and removed from the pressure data.

The data is affected by possible gauge movements which have masked the infinite acting horizontal radial flow response. (Figure 7) Therefore, in order to estimate a range of reservoir properties, the below technique has to be applied: 1. Estimation of reservoir properties, assuming the position of IARF and fixing the Pi value seen at 2 different periods of 20 hrs and at end of 24 hrs.

2. Establish a potential range of reservoir parameters (sensitivity analysis).

Figure 4: Pressure Derivative Plot of the Main Build-up Before and After the Tidal Correction

Case 3: Gauge Movement Multi-rate test was carried out at different chokes followed by the main buildup of 24 hrs. Figure 5 below shows the pressure and rate plot.

Figure 7: Log-Log and Pressure History Plot

Case 4: Downhole Screen Plugging Figure 8 shows pressure vs time plot for the entire sequence of a DST job.

Figure 5: Pressure Rate Plot Pressure fluctuation of about 0.4 psi was observed during the first 20 hrs of build up time. This behavior has similar noise and amplitude during these 20 hrs attributed to possible gauge movement. Below seen behavior was observed on both downhole gauges. After 22 hrs into the Buildup, pressure fluctuation decreases, which may be an indication of possible reservoir response. Pressure stabilization after 22 hrs from the beginning of the PBU could be due to the positioning of the gauge at the desired location, the increase in the pressure could be a result of the liquid level falling below the gauge depth. (Figure 6)

During this job, after the well was perforated and opened to flow at the choke manifold, the well ceased to flow. If we look into the pressure data during this well activation period, we will see that both the tubing and the annulus gauges are not following each other, meaning that the tubing and the annulus are not communicating with each other. And the only source of communication between the tubing and the annulus is the screen. This suggests that the screens are getting plugged after the well was perforated.

Now, to investigate the cause behind screen plugging, we started off with root cause analysis i.e. list out all the possible causes(screens, completion fluids & unconsolidated formation) behind screen plugging, analyze them and pick the most prominent one.

Figure 8: Pressure vs Time Plot for the Entire Sequence of a DST Job

Figure 6: Zoom of The Main Buildup for Possible Gauge Movement Investigation

13

Reserve Booking

Petroleum is the world’s major source of energy and a key factor in the development of world economies.

It is essential for future planning that governments and industry have a clear assessment for the quantities of petroleum available for production and quantities anticipated to become available within a practical timeframe through additional field development, technological advances, or exploration. Achieving that requires industry to adopt a consistent nomenclature for assessing current and future quantities of petroleum expected to be recovered from underground accumulations. Such quantities are defined as reserves.

Reserves Estimation Estimating hydrocarbon reserves is a complex process that involves integrating geological and engineering data. Depending on the amount and quality of data available, one or more of the following methods may be used to estimate reserves:

1. Volumetric Estimation of OOIP and OGIP are based on a geological model that geometrically describes the volume of hydrocarbons in the reservoir.

N = OOIP (STB) • • • •

2. Material Balance This technique mathematically models the reservoir as a tank. It uses limiting assumptions and attempts to equilibrate changes in reservoir volume because of production. This method can account for aquifer support and gas cap expansion. Its generic equation is: Volume In = Volume Out Pore Volume Change = Fluid Volume Change

G = OGIP(SCF)

A = area of reservoir (acres) from map data h = height or thickness of pay zone (ft) from log and/or core data ø = porosity (decimal) from log and/or core data Sw = connate water saturation (decimal) from log and/or core data • Boi = formation volume factor for oil at initial conditions (reservoir bbl/STB) • Bgi = formation volume factor for gas at initial conditions (RES ft3/ SCF) However, due to gas evolving from the oil as pressure and temperature are decreased, oil at the surface occupies less space than it does in the subsurface. Conversely, gas at the surface occupies more space than it does in the subsurface because of expansion. This necessitates correcting subsurface volumes to standard units of volume measured at surface conditions.

14

Estimating the recovery factor is a key point here and directly influence the volumetric calculated reserve. These reserves generally referred to as «static reserves».

This equation assumes thermodynamic equilibrium between oil and gas, a uniform pressure distribution, and a uniform saturation distribution in the reservoir.

3. Production History DCA It is used to estimate economic ultimate recovery and the expected economic life of a reservoir. The rate of production and at any point in time can also be estimated. This method relies on historical production data to extrapolate future production performance. A variety of curves can be used, the most common is a semi-log plot of rate of production versus time (exponential decline). Three models are described.

4. Reservoir Simulation Reservoir simulators use material balance as well as fluid flow equations to model the reservoir as a group of interconnected tanks.

Different commercial simulators are all common in one thing; they solve the diffusivity equation for the grid blocks created.

The advent of powerful computers has made the use of numerical simulation common for estimating reserves and recovery as well as initial volume in place.

Since reservoir simulation can account for performance history through history matching, this method incorporates facets of all the techniques discussed. With sufficient data, this method provides the best recovery estimates for complex reservoirs.

“Probable” or “possible” reserves are lower categories of reserves, combined and referred to as “unproved reserves” with decreasing levels of technical certainty.

Probable reserves are volumes that are defined as “less likely to be recovered than proved, but more certain to be recovered than Possible Reserves.”

Possible reserves are reserves which analysis of geological and engineering data suggests are less likely to be recoverable than probable reserves.

5. Analogy This method directly compares a newly discovered or poorly defined reservoir to a known reservoir have similar geological or petro-physical properties (depth, lithology, porosity, and so on). While analogy is the least accurate of the methods presented, it is often used early in the life of a reservoir to establish an order-of-magnitude recovery estimate.

Analogy should always be used in conjunction with other techniques to ensure that the results of the more computationally intensive methods make sense within the geological framework.

Petroleum Resources Management System (PRMS)

PRMS provides updated definitions and the related classification system for petroleum reserves and resources. These definitions establish a universal language, which can be used for estimating and classifying quantities of oil and gas discovered in a reservoir.

PRMS incorporates a central framework that categorizes reserves and resources according to the level of certainty associated with their recoverable volumes (horizontal axis in the figure below), and classifies them according to the potential for reaching commercial producing status (vertical axis).

Reserves

Oil and gas reserves and resources are defined as volumes that will be commercially recovered in the future. Unlike the inventory of a manufacturing company, reserves are physically located in reservoirs deep underground and cannot be visually inspected or counted, but estimated based on the evaluation of data that provides evidence of the amount of oil and gas present.

The estimation of reserves volumes is generally performed by highly-skilled individuals who use their experience and professional judgment in the calculation of those volumes.

Reservoir engineers usually call the process of estimating and agreement on the reserves “Reserve Booking”.

The highest valued category of reserves is proved reserves. They have a “reasonable certainty” of being recovered, which means a high degree of confidence that the volumes will be recovered. To be clear, reserves must have all commercial aspects addressed. It is technical issues which separate proved from unproved categories.

15

Importance of Attentive Reservoir Surveillance and Immediate Response in:

Commingled Completion to Maximize Recovery

Reservoir Surveillance is a dynamic activity of monitoring reservoir inflow and wells performance, including production and injection wells, through different tools and techniques. Several factors can provide a direct or indirect indication of well performance including, pressure, temperature, water rate, and the most important is the oil rate. The tools that may be used to facilitate surveillance activity range from sensors that might be attached to the wells to tools that might be run in wellbores beside the required software that manage or provide the interpretation of the required information. The second part of the surveillance would be the reaction from the team and the decision to execute a specific job to overcome a certain issue or to prevent its occurrence if the team would be adequately proactive.

Reservoir Background Zubair reservoir is a lower cretaceous formation which is one of the main producing reservoirs in the south of Iraq, Figure (1). It is comprised of oil bearing sandstone interbedded with shale sequences with different extension and connectivity between the layers. Some of these shale layers are totally impermeable; therefore, they divide the reservoir in three major units with different formation quality and oil properties: A, B, and C units. Units A and C has similar quality and oil properties (API = 24), while unit B has lighter oil of 32 API and located in the middle of the units A and C. Moreover, oil reserve in unit B was estimated to be 72% of the total reserve in Zubair reservoir, whereas unit A was estimated to have 12 % and the remaining 16% in unit C. Reservoir and aquifer connectivity were observed to be relatively poor in units A and C while this weak aquifer support totally disappeared in unit B.

Figure 1 - Geological Sequence of South of Iraq

20

Reservoir Development The early development plan for this reservoir proposed drilling vertical wells to initiate oil production. The first three-production wells were drilled about 40 years ago, whereas the main drilling campaign to totally develop the reservoir was started in 2011. This campaign included drilling 20 production wells in the crest of the reservoir and targeted all the three major sandstone units. However, the main target was to produce unit B and commingle the other units if their formation quality and oil saturation were found to be profitable. All the 20 vertical production wells were cased and production was initiated after performing perforation jobs with no need for any stimulation treatment or artificial lift techniques.

Wells Performance and Production Sustainability The new drilled wells within the new campaign were all perforated in unit B while some of them also have unit A or C or both depending on formation quality and oil content. When the production was started, the majority of the wells have a daily oil production ranging from 2 to 5 KBD with no production of the associated water. This performance continued for a couple of years before some wells have severe decline in the production and a rapid depletion in the reservoir pressure. Over time, few wells went cease and all the attempts of bringing them back to production did not succeed. Using the production logging tools showed that the highest depletion in the pressure occurred in the unit B while the other units have a pressure close to the initial reservoir pressure. Figure (2) shows zonal oil production for all the units in well-X6 and well-X19.

Figure 2 - Zonal Contribution of Wells X6 and X19 in 2013

Production Decline Issues and Diagnosis After 5 years of production, about half of the wells stopped flowing or suffered significant production drop while the others have sustained oil production with the same original rate even though there was a noticeable drop in the reservoir pressure. The results from the production logging surveys along with wells location, oil properties, as well as which units were perforated in each well were utilized to diagnose the reason of the production deterioration. After reviewing all the available data and investigating all the possible scenarios which may be the cause of the production decline, it was found that the key drive behind the issue is the unit C. During the investigation, wells were classified into categories based on their producing units. Wells with unit C were found to have lower productivity and shorter production time compared with the other wells that do not have unit C in them.

Also, the production history of the wells shows two distinguished trends of production. Wells with unit B only could sustain high oil production rate for long time even the reservoir pressure dropped significantly. Whereas, wells that have unit C commingled with unit B could not sustain the high production rate for more than three years while the other wells stopped flowing completely. Figure (3) depicts the two production trends due to differences in the completed units.

Finally, it was found there was a strong relationship between completing unit C and the cumulative oil production. Figure (4) depicts the cumulative oil production for the wells as well as their perforated units. It is obvious that the presence of unit C perforation resulted in significant drop in cumulative oil production in each well. This issue was overcome in well-X1 when unite C was isolated from early time of production which led to achieve quite high cumulative oil production with no issues to sustain well production until now.

Figure 3 - Production History of Wells With Different Units

Figure 4 - Oil cumulative production from each well with the corresponding perforated units.

Remedies and Recommendations To overcome the issues that were caused by unit C presence in the production wells, a bridge plug was used to isolate the units in some wells in order to diminish unit C impact and stop the upward water cross flow to unit B. Also, for the new drill wells, a proposal was made to not to commingle units B and C in the same well until water injection is initiated and adequate support is available to enhance unit B pressure. The other producers with both units B and C which are still producing normally were set to be monitored closely to make a quick decision to do any required well intervention to sustain oil production as the case with well-X1. Late work overs and remedies will not be adequate to solve the issues or to retrieve the same production rate unless artificial lifts techniques to be utilized and that was the case of few wells with unit C.

Enhancing oil production rate for a period of time by commingling more units might result in significant decline in the recovery if the properties of units are not well understood. Attentive reservoir surveillance and quick response to production changes play a key role in enhancing production and supporting its sustainability.

Late well interventions might not be able to bring well performance back to its initial conditions.

21

Delivering Photorealistic Borehole Images in NonConductive Mud: Latest Advancements Since the introduction of the first micro-electrical imaging tool in 1986, wireline resistivity images have proven to be an invaluable tool for geological and petrophysical formation evaluation in wells drilled with conductive water-based drilling mud (WBM). However, until recently, wellbore images acquired in non-conductive mud had been met with some less success due to poor borehole coverage, relatively low image resolution, and electrical artifacts.

In 2014, an OBM-adapted imaging tool was introduced. The new tool was designed to provide improved resolution and borehole coverage as well as geological representativeness of the images. From operations prospective, tool sonde and hardware were designed to increase robustness and ease of logging for logging engineers, improve operational efficiency, and use less rig time in consideration of high spread rates of deep-water drilling rigs and the overall high costs of deep-water wells.

Tool Design and Measurement Physics The sonde design with two sets of pads supported by spring loaded arms allows both logging down and logging up of the tool to minimize logging time. Unlike previous imaging tools, pads are applied to the formation using spring load and not pad pressure to minimize stick-slip of the tool. Pads are fully gimballed and are free to tilt, rotate around its axis to enable maximum contact with the borehole wall. See (Figure 1) for tool diagram, tool pad, and cross-sectional projection of the pads.

include coarse-grained basal conglomerate, rip-up clasts, and large clay clasts, debrite, dewatering and flame structures, dish structures, internal injectite structures, pyrite nodules/streaks, and deformed faces, see (Figure 2). In addition to the qualitative interpretations of geologic features and formation texture, the high-resolution image can be used for a wide range of quantitative image analyses such as net pay computation, textural attribute extraction, as well as other quantitative and semi-quantitative answer products, see (Figure 3).

As for the measurement physics, high frequency is sent into the formation which reduces the non-conductive mud electrical impedance. Amplitude and phase of this current are measured and is used in the processing to create electrical impedivity measurement. In order to cover the full range of formation resistivities, two frequency ranges are used. After the acquisition, composite processing technique is used in which amplitude and phase measurement from the two frequencies is processed to generate a final impedivity image that is a function of formation resistivity and dielectric permittivity.

Case Study: Imaging Turbidites of West Africa The first case study presented in this article is Oligocene-Miocene age deep-water turbidite deposited on the passive margin of West Africa and comprise a complex of channels and sheet sand with localized intense faulting and tilting due to salt tectonics and diapirism.

The high-resolution image enabled high-confidence tracing and classification of geologic features. The variety of geologic features ranges from fine-scale laminations and syn-depositional micro-faults with a displacement of few centimeters to variable-scale injectite features and erosive surfaces. Also, A wide variety of formation textures that represent turbidite channel and levee facies are observed that

22

Figure 1. Adapted Imaging Tool; upper and lower pad sets offset by 45 deg resulting in pad arrangement shown on the bottom

elongation (NW-SE) which was parallel to the strikes of erosive surfaces and channel-margin slumps (Figure 6). Current dips (pink tadpoles) and over-bank deposits (green tadpoles) were parallel and perpendicular to channel elongation, respectively with current bedding representing channel propagation (SE).

A mud-filled channel was interpreted according to the following criteria: 1. High angle dips in the absence of faulting, 2. High GR profile with a basal erosive surface, 3. Orthogonal relationship between erosive surface and bedding dips (Figure 7). These features may imply by-passing of sediment downslope and deposition as fan lobes.

Figure 2: Distinctive facies observed in this case study: (A) Rip-up clasts in channel sequence. (B) Possible calcareous concretion.

(C) Large, imbricated clay clasts in channel sequence. (D) Floating clay clasts in debrite facies in the upper part of channel sequence. (E) Erosive surface overlaid by basal conglomerate facies. (F) Flame structures in alternating sand-shale sequence.

(G) Deformed facies with multiple micro-faults. (H) Clay clasts in sand bed. (I) Erosive surface overlaid by conglomeratic facies.

(J) Laminated facies with streaks and nodules of pyrite. (K) Dish structures due to dewatering of sandstone. (L) Injectite internal structure.

Figure 3: Lithofacies, textural and borehole shape analysis from the new OBM-adapted imaging tool. Note the increase of sorting index next to rip-up clast zone. Grain size index may be used as qualitative indicator grain size variations on a very fine scale. This interval represents a deep-water channel sandstone with a variety of coarse-grained and fine-grained sediment and a fining-up levee section overlying the channel sequence

Figure 4:- Sedimentary structures and facies observed included: (A) water-escape structures (B) Load (right) and flame (left) (C) Large, imbricated clay clasts at channel base (D) Contorted Bedding (E) Laminated Facies (F) Pebbly Sandstone (G) Slumped Bedding

Borehole images shown in this case study demonstrate very highquality images acquired in non-conductive mud environments.

The innovative design of the tool with spring-loaded eight-pad configuration with the ability of logging down has proven to be very efficient for the logging operations and helps improve image quality.

After five years of its introduction, this technology has been used in nearly every part of the world delivering outstanding results.

It is very realistic to say that the amount of detail revealed by these borehole images surpasses any other non-conductive mud imagers by an order of magnitude enabling the delivery of high confidence interpretation.

Case Study: Imaging Turbidites of Offshore Nile Delta The second case study is a facies analysis of turbidite deposits in Egypt’s Offshore Nile Delta. The objectives of this study were to differentiate between the lobe and channel architectural elements and compute an accurate net-to-gross for petrophysical evaluation.

According to the facies assemblage encountered in the well, it was interpreted that the sand bodies penetrated were slope channels (Figure 4). The high proportion of deformed and slumped facies were used to enhance the petrophysical evaluation of the sequence as they were interpreted to have a minor contribution to production due to their low connectivity (Figure 5).

Dip patterns were interpreted to determine the dominant channel

Figure 5:- Facies oriented sand count excludes contorted bedding (C) and sand pods (P) to provide a more accurate net-to-gross. These heterogeneous sands could have a high hydrocarbon saturation but due to poor connectivity would result in low ultimate recovery.

Figure 6:- Channel-levee sequence showing relationship between erosive surfaces (red circles) and channel-margin slumps (blue square) as well as current and overbank beds (blue rectangles). A mud-filled channel (red rectangle) with basal scour (blue circle) was observed.

23

Predicting Sweet Spots in Shale Plays by DNA Fingerprinting and Machine Learning

The patented technology produces a >70% accurate predictive map of the highest producing areas in a shale play, so called sweet spots, using DNA analysis of surface soil samples. The DNA analysis contains information on the mix of microbial species in the soil samples which is a direct indicator for the presence of vertical micro-seepage from hydrocarbon accumulations in the subsurface. For exploration, it paves the way to highly effective exploration drilling and investment decision- making. For production, it leads to significant cost reduction by optimizing drilling scenarios and increase development speed of a play. The accuracy of the predictive map is iteratively increased from about 70% before drilling to up to 85% by incorporating information from the drilled wells with local, field specific microbial information.

Methodology 1. DNA Analysis to Get DNA-fingerprints

The methodology is based on identification of bacteria, present in soil samples, based on their 16S ribosomal DNA sequence (hereafter 16S sequence). The 16S sequence of bacteria is generally used as a genetic fingerprint of bacteria and consists of conserved and variable regions. The sequence of one such variable region is determined by illumina sequencing. Similarities and differences in the 16S sequence are translated to bacteria. In Figure 1 a schematic representation of the steps required for the DNA-based analysis of soil samples is shown.

Figure 1: Workflow for getting DNA fingerprints for soil samples: 1) Soil samples preparation for DNA analysis 2) Extraction of bacterial DNA from soil sample 3) Amplification of the 16S rDNA by Polymerase Chain Reaction 4) and 16S Sequence analysis using next generation Illumina MiSeq 5) Processing raw data from MiSeq to verified 16S sequences 6) Processing verified 16S sequences back to individual soil samples and 7) Interpretation of 16S sequence data: translation to bacterial genera (families); the steps 2 and 3 are tuned by Biodentify to get maximum information on species.

Analyzing the composition of the microbial population present in the soil samples requires extraction of microbial chromosomal DNA from the samples. This is performed by mechanical disruption of the bacteria using small beads and vigorous shaking.

After amplification and sequencing the processing is finalized in a series of special designed quality control steps which together are called the ‹pipeline› (Figure 2). Initially, the number of DNA segments equals the number of unique sequences (>35 million sequences).

During the processing, the total number of sequences (blue bars) slowly decreases to ~29 million sequences. However, the number of unique sequences (crimson bars) diminishes, finally resulting in a

24

fraction (~340 thousand) of the initial number of sequences. These unique sequences form the basis for the identification of biomarkers as they are translated from sequences to families or genera. Figure 3 shows a representation of the relative abundance of bacterial families or genera within 7 different samples.

Figure 2: Schematic representation of the effect of the various proprietary filtering steps within the ‹pipeline›. Please note, at this scale, the number of unique sequences at the final stage is hardly visible (value ~340 thousand).

Figure 3: Schematic representation of the bacterial diversity and relative abundance within samples. Each colored bar represents a bacterial genus or family. Please note that the legend on the right only shows a minor fraction of all bacterial diversity.

2. Using Our Database With Samples from Drilled and Known Plays, to Correlate DNA in Soil Samples with Production Data When predicting the productivity of undrilled, target locations where only DNA fingerprints of shallow soil samples is available, it is necessary to have a correlation model that connects the new fingerprints of the new target location with a prospectivity index. This correlation model is produced by using our database, containing DNA fingerprints and production data from executed projects. Currently (May 2017), we have well over 2000 samples with known production data from six different and varying shale plays in our database (Figure 4):

Figure 4: Map of The Sampled Shale Gas and Shale Oil Plays Each sample contains up to hundreds of thousands counted different species. The correlation of each of these samples (a vector x) to the newly sampled DNA fingerprints is a measure for the similarity of the DNA fingerprints in the database with the new one. This correlation, or more precisely a “co-occurrence measure”, is estimated by unsupervised k-nearest neighbor techniques (KNN) addressing the high dimensionality of the problem. The co- occurrence is expressed by the cosine similarity. By selecting the samples from the database that have the highest correlation with the DNA fingerprints of the new samples, a training set is generated, where fingerprints and productivity are known. This training set is used to find the correlation model with Machine Learning algorithms to predict the prospectivity at new locations (see next step 3 described in detail below). Because the database has samples from different and varying plays (hydrocarbon source, climate, geology), creation of a training set with a sufficiently strong correlation between samples from drilled areas and new locations can be realized.

To find the most robust and reliable predictive biomarkers, we adapted the above described algorithms in a triple-loop validation/prediction procedure (developed for medical applications by TNO; Figure 5). To ensure that the model correctly captures the information contained in the dataset, samples are randomly shuffled, based on the input data matrix of microbial abundances. Next, the reshuffled matrix is split into a calibration subset (randomly 70% of the training set) and a validation subset (30% of the training set). The calibration set then is subjected to N-cross folds to estimate the modeling parameters N-times. The final selected parameter set minimizes the average misfit in the predictions.

These parameters are used on the separately reserved validation set to confirm the validity of the model and the identified biomarkers. Since variations inevitably occur within this type of computational modeling, 1000 repetitions are run, thus selecting only the most stable biomarkers over all 1000 x N simulations (second loop in Figure 5). The model is aimed at reducing the number of biomarkers until an optimum in prediction accuracy is achieved where the prediction is at its optimum and the number of biomarkers is at its minimum (expressed in a ROC curve; Figure 6). Only then a prediction can be made for a new area, and a model can be improved with new measurements. The price for calculating a robust predictive model is the compute time needed for calibrating 1000 x N models (to achieve the best final prediction).

This is solved economically by computing in the cloud using Hadoop technology (32,000 processors) and using GPU clusters in- house to accelerate the highly parallelizable calculations.

Figure 5: Schematic representation of the machine learning ‹triple loop› modeling procedure.

Figure 6: Representation of the performance of a binary classifier system (prospective vs. non-prospective).

3. Modeling and Validating The Microbes That Determine The Sweet Spots The next step is to correlate the results of the sequenced analysis (the bacterial diversity in the DNA-fingerprints of all soil samples) with production data in the selected training set. The goal is to find correlations between production and presence or absence of specific bacteria, typically about 50-200 out of the 340,000 species in our database, the so-called biomarkers.

To build a model that consists of the 50 to 200 differentiating biomarkers is a real Big Data challenge. It aims to find both those species that are more present above sweet spots because they use the seeped colloids of hydrocarbons for their metabolism, and those species, that are less present above sweet spots because the population is partially eliminated by the hydrocarbons. The number of possible solutions is a power law of the hundreds of thousands of possible species, resulting in an almost infinite solution space. A special Machine Learning algorithm has been developed to perform this task. State of the art methods were used to deal with ‹sparse modeling› issues like non- linearity, the influence of noise and to prevent overfitting (Gaussian Kernels for non-linear problems (Cortes, 2012) and L1-based regularization methods (Mosci, 2011).

To generate an accurate and predictive map with sweet spots highlighting the most prospective areas, the following steps are followed: 1. Soil sampling in the field, followed by DNA analysis to get DNA-fingerprints of the microbes 2. Selecting a training set from our database with sample data from earlier drilled and known areas. The training set has similar DNA fingerprints compared to the new samples and can correlate DNA from soil samples with production data 3. Modeling and validating the microbes that determine the sweet spots 4. Mapping and analyzing the prospectively / sweet spots / low production areas of the target play

25

Reservoir Stimulation Reservoir Engineering is one of main branches of petroleum engineering. The main target of this branch is to get maximum recovery with lowest cost and highest NPV. To make good reservoir study, three phases must be conducted.

The first phase including data gathering, data QC and analysis phase. In this phase all available techniques should be applied in all types of data. Results of each technique applied should be gathered and integrated together to get conclusion which lead you to a bright vision before the second phase of your study.

The second phase is the calibration phase of your data if it is analytical (such as material balance or nodal analysis) or numerical (history matching phase in modeling) and in which all previous conclusions will be an input to this phase.

The last phase is the prediction phase according to its decision is taken.

The third phase is prediction phase you assume different scenarios and show different in results.

Before starting any study, the main objective of it should be clear. In our study, the main objective was definition of optimum development strategy for one of the main reservoirs in this field.

Characteristic of Field Our field is X-FIELD field. X-FIELD field discovered in Oct-2014, X-Field1X showed pay intervals in AEB-3D, Simulated Reservoir & Shiffa reservoirs. Twenty well were drilled to appraise the structure and develop the reserves for these multi target field. The estimated OOIP for Simulated Reservoir only about 42 MMSTB. The main reservoir to be discussed in our case is Simulated Reservoir. The reservoir has different sand qualities but the major is the high quality (high permeability reservoir) Simulated Reservoir produces under partial water drive mechanism (both bottom and edge drive mechanism are available). The produced oil can be classified as black oil with low GOR (around 100 scf/stb). Initial reservoir pressure is 4500 psi. Our reservoir is heterogeneous sand with good sand quality (PI ranges from 7 to 56 bbl/psi/d). Cum oil till now about 5 MMSTB with daily rate 9500 BOPD. Our main concern would be the water breakthrough. This concern of the water movement moved us on the conservative side in the reservoir management in order not to over produce the wells and have bypassed oil in the reservoir, here comes the role of the reservoir simulation, the simulation target was to develop good understanding of the physics of the reservoir and the nature of the multiphase flow, evaluate the present potential to increase the production, how it should be done, drilling new wells or use current wells, best location of proposed wells, the possible risks based on the present uncertainties, the operational requirements and finally study the impact of several strategies in the economical point of view.

26

Phase-I Data Availability and Sources This is the first phase of the study. During this phase, we checked the different sources of data, QC on its quality and determine the degree of confidence of it to be used in our study. The following table highlights data availability and its validity to be used:

Data

Availability

Comment

PVT

No

Used offset field data

Core Data

No

Used offset field data

Pressure Data

Yes

RFT, DST & pressure surveys

Production Data

Yes

Good high frequency data

OHL

Yes

Phase-II: Model Initialization & History Matching In this phase, we try to build the model using the previous data with static model made by geo-modeler. The main target of this phase is to make model simulate actual production data, actual change in different phase saturations and change in pressure on static under actual production constrains and operating conditions.

During this phase, more than one realizations of history matching were achieved (according to several combinations of SCAL and facies modeling). These different matching scenarios were used to be different base cases for future development strategies prediction. The following figure shows field oil and water rates history matching.

Figure 3: The Effect of Adding Single Well Individually to Base Case on Ultimate Recovery Incremental.

Figure 1: Field Oil and Water Rates History Matching.

Phase-III: Prediction & Economics

In this case, prediction conducted on 3 levels of evaluation.

1. No further action (Base case): in this case we try to predict the recovery factor and ultimate recovery to be used as a comparison case with any future development plan 2. Proposed wells: in this level we test each proposed well individually to define its effect on ultimate recovery and NPV

For the first level:

Figure 4: The Effect of Adding Single Well Individually to Base Case on NPV Incremental.

The expected recovery factor in the most pessimistic case was 37.6% with NPV of 364 MM$. These values are essential to determine the impact of any change in production conditions or constrains.

Figure 2: The Impact of Any Change in Production Conditions.

Level 2 Proposed wells (well by well) In this level, we have four proposed wells, the target was definition of priority of these wells. The following plot shows the effect of adding single well individually to base case and its effect on ultimate recovery incremental and NPV incremental.

- Matching model was not complicated but its prediction is very sensitive - Acceleration with drilling new wells or recompletion of some wells (producing from another reservoirs) have more positive impact in EUR & NPV than increasing rate from existing wells.

- Multi-target reservoir need more reservoir studies to minimize cost (such as drilling) and maximize NPV

27

Predicting Sweet Spots in Shale Plays by DNA Fingerprinting and Machine Learning

The development of recent DNA analysis techniques makes the microbial quantification of species comprehensive and economically feasible. The millions of counted species in thousands of soil samples by applying 16SrDNA techniques creates an enormous amount of input data (terabytes in size) that must be correlated with the presence of hydrocarbons. This is a serious big-data mathematical and computational challenge. The progress made in supercomputing makes it possible to construct robust and reliable predictive models by applying machine learning techniques derived from pharmaceutical applications. The strict experimental design rules in this field are used to find a complex composition of microbes, not only those who flourish by the micro-seepage, but also those who are terminated by it and that have, therefore, lower concentrations above sweet spots. The latter category could never be found by the earlier adopted microbial exploration techniques, however have proven to be essential for identification of sweet spots.

Validation Case in Haynesville A grid with 362 locations were sampled, 20-50 cm below surface elevation (Figure 1c on the right). The productivity map (Figure 1b in the middle), from actual well production data, is used to validate the generated model.

a) pilot Haynessville shale

b) drilled wells and contoured production (2 years)

highly productive area in the North East is clearly predicted, but there are also some probably false sweet spots (sweet spots predicted by our model but not seen in the validation/productivity map) in the South East.

c) locations of soil samples

Figure 1 orange = top 10% of producers in this play

1. Generate a model with the triple loop modeling procedure applying selected sample data from the database that show sufficient similarity with the local DNA fingerprints from the target area.

2. Predict the chance of a sweet spot at grid sample point location by using the model from the training set and the local DNA-fingerprint data only, and generate a first sweet spot prediction map by interpolation of the predictions. In this case, successful DNA analyses were generated at 314 locations. For those locations, the predictions are shown as colored points on the map which are subsequently contoured (Figure 2). Since this area is used as a validation case, the actual drilled wells with production rates are known (background map of Figure 2) and these are used to benchmark the DNA based predictions. As mentioned above, the model is trained with data from our database from other shale areas, and the model productivity predictions are made solely by using local soil sample DNA fingerprints (bright colored contours at foreground of Figure 2). The map in Figure 1 provides the predicted sweet spot map before drilling, with 72% accuracy. The

28

Figure 2: Prediction Map of Sweet Spots Without Using Local Production Data (can be seen as predicted map before drilling).

3. Wells will be drilled targeting the predicted sweet spots. When drilled, production information from these wells will be used to increase prediction accuracy by adding these locations with known data to the training set, then re-model. This is how we generated our final map, using ‘new’ and local field information from the validation data. Applying this principle, a new map with increased accuracy, now 86% (Figure 3) was produced. The mismatches on individual isolated locations reflect the ‘noise in the data’, always present in nature, but as there is no correlation between these points, it will not impact the decision-making based on this predictive map.

In reality, 97 wells were drilled in this area, with 27 high producers (table with Figure 3). These numbers showcase the value of the methodology: through smart and prioritized drilling, based on updated predictive maps, more hydrocarbons can be produced even while drilling significantly fewer wells.

Actual Drilled Wells

With DNA Information

Low Producing

70

15

High Producing

27

31

Total

97

46

Figure 4: Predicted Sweet Spot Maps of Two Areas in the Netherlands: One of the Geverik and One Over the Posidonia Formation.

Discussion and Outlook

We propose that sweet spot prediction, with an accuracy of 70% prior to drilling, by using DNA fingerprints from shallow soil samples is a valid additional exploration tool. Our case studies show reproducible results supporting this. The presence of vertical micro-seepage by buoyancy through microcracks is a difficult process to prove because it cannot be measured directly (as opposed to macro-seepage). Although many geochemical anomalies have been detected for conventional fields and distinctive differences in biomarkers above sweet spots in shale are found as described in this article, proof of micro-seepage remains an indirect case.

Some Additional Remarks: Figure 3: Final estimated sweet spot map after ‘virtually drilling’ wells and therefore iteratively increasing the prediction accuracy. By using the DNA fingerprints the area is (virtually) produced with significantly fewer wells yet with a higher absolute number of wells in sweet spots

Undrilled Pilots in the Netherlands The second case is carried out prior to drilling. In the Netherlands, two pilot areas were sampled and analyzed on DNA fingerprints to produce predictive maps. As a training set, again, data from the US database is used, with sufficient, validated similarity in microbial DNA. The maps, from two different formations (Geverik and Posidonia) are showing a significant difference (Figure 4): in the Geverik, no shale gas is predicted while in the Posidonia, there is a clear indication of sweet spots. Both maps are generated using the same trained model from our database. The absence of shale gas in the Geverik map confirms the existing theory that the circumstances for the formation of shale gas were unfavorable because it is on top of Visean carbonate structures (Harings, 2012). Naturally, the outcome can only be verified after drilling, but the plausibility of the maps is further demonstrated because the DNA fingerprints in the Geverik area were positive in the small zone above a conventional prospect (prospect mapped based on seismic data and geological interpretations), whereas the known and produced (now empty) fields in the Posidonia area were not captured.

These findings again support the earlier explained concept of microseepage with a relatively high velocity of buoyancy where the signal in the DNA fingerprint disappears within a few years after production because of a decrease in reservoir pressure.

1. What to do in case of stacked layers where also conventional reservoirs are produced in the same area? A good example is the Avalon/ Bonespring formations in the Permian basin. Conventional reservoirs are mostly known and already produced; its projected areas on the surface (usually less than a few % of the total area) can be excluded from the shale grid sampling, but the different sweet spots in stacked shale layers may still overlap. The method in this paper can delineate the summed lateral spreading of the stacked sweet spots. Additional geological (drilling/seismic/other) information or interpretations is needed to determine the vertical origin of the hydrocarbon.

2. The quality of completions clearly has a major effect on the productivity of the well. For samples in our reference database, completions influence the actual production rates and thus influence the classification used in the training set that we correlate with.

Therefore, classification of well productivity in our database is done on a relative base, e.g. the top 10% producing wells compared to wells in the same area/play, that are drilled in the same period (using comparable technology).

Despite these remarks, the outlook of this methodology is very positive.

The accuracy in predictability that can be reached constitutes to very significant cost savings. And with more data becoming available, our database increases in value as it is a learning system: every new sample point and drilled well increases the information content that can be used to predict new wells drilled in the future.

29