GIS Visión País (VIPA):
An Information System for decision making in Exploration
Esri User Conference, San Diego, July 2001


Taylhardat A., Violeta
Hernández, Gustavo CONTENTS:
ABSTRACT
INTRODUCTION
SCOPES
BACKGROUND
METHODOLOGY
DATA
Classification according to processing degree
Raw Data
A. Source
B. Layers
Processed data
A. Source
B. Layers
Interpreted data
A. Source
B. Layers
Quality
Technical Specifications
Metadata
PRODUCTS
Thematic Maps
Quality data maps
Spatial analysis products
FINAL CONSIDERATIONS
ACKNOWLEDGMENTS

Annex # 1:  WELL SELECTION PROCESS
Annex # 2:  DATA VALIDATION MATRIX EXAMPLES
Annex # 3:  EXAMPLES OF GIS VIPA LAYERS
Annex: # 4:  EXAMPLE OF A SPATIAL ANALYSIS PRODUCT
REFERENCES
 
 

GIS VIPA: An Information System for decision making in Exploration
Taylhardat, Violeta; Hernández, Gustavo
Visión País, PDVSA Exploration, June 2001
Esri User Conference, San Diego, July 2001

Abstract
An information management tool for Hydrocarbon Exploration Planning has been considered a fundamental element in Petroleos de Venezuela S.A.
VIsión PAís, a technical group of PDVSA Exploration, is responsible for the identification and characterization of new hydrocarbon opportunities nationwide.
Validation of data in terms of disciplines and reliability is necessary before loading the information.
The entire process is carried out centered upon a GIS containing geological, geophysical, geochemical, petrophysical, production and environmental data. It is capable of rendering a three dimensional view of exploration targets, oriented towards the support of risk and economic evaluation for recognized opportunities, linked to installed interpretation facilities. The main objective is to establish exploration strategies and profit from the synergy with neighboring countries.
Products, including Spatial Analysis for specific needs, are based on the fact that the area covers more than 1 million square kilometers, meaning that the scale of all maps must have a range from 1: 500.000 to 1: 2.000.000.
 

Introduction
An effective information management during the Exploration Planning Cycle is considered a fundamental element for the success in the subsequent stages of oil discovering and exploitation processes.
During the last 8 decades, the Venezuelan oil industry has always been characterized for the acquisition and maintenance of large volumes of information amongst its assets. However, its dispersion and the lack of specialized tools for the analysis have made difficult, in some cases, the conception of the Exploration Plan in the long and short term.
Also, the technical data obtained during the information acquisition stage, once it has been safeguarded, processed and analyzed under a global perspective, might help in obtaining additional information in order to improve plans and strategies for the oil exploration and exploitation.
Visión País (VIPA), one of the main functions of the Exploration Unit of Petroleos de Venezuela s.a. (PDVSA), is carrying out the gathering, validation and processing of selected geological, geophysical, economical, environmental and production information of the Corporation. The goal is to integrate it in a System that allows its visualization and analysis to obtain, as the result of spatial analyses, the identification of those areas qualified as "best exploration opportunities", under a technical and business perspective. Additionally, VIPA must engage in the preparation of the documental base required to facilitate the prioritization of the potentially prospective areas, during the exploration planning stage.
The VIPA Geographic Information System has been conceived with the vision of supporting the Exploration Process in its whole extension, to canalize the efforts and aims towards an opportune delivery of information and a proper alignment during the Preparation of the Plans and Strategies of the Corporation Exploration Business Unit.
 

Scopes

  • To organize and maintain available under a unique computing platform, the required updated information for oil exploration activities.
  • To shorten the specialist tasks of information gathering at the exploration opportunities evaluation moment.
  • Facilitate the risk/opportunity evaluation and decision making processes, by providing an objective and global vision of the exploration opportunities of the country.
  • To have the exploration information evaluated and ranked under standard data quality control parameters.

    •  

     

    Background
    There have been some intents of GIS implementation in PDVSA. Until now, we have achieved good results handling cartographic, environmental and cultural information, but not as far as geosciences data is concerned, i.e., geology, geophysics, petrophysics, geochemistry, sedimentology and bioestratigraphy.
    The goal to integrate geoscientific information, jointly geo-referenced with cartographic information and satellite images in the same system as well as permitting its global visualization for countrywide analysis has been a challenge in PDVSA during the last ten years. This has been an arduous task despite of disposing of sophisticated technology.
    It has not been sufficient, because GIS technology development has been partially divorced from Geosciences tools development. One of the main handicaps has been the 3D information management, taking into account that rocks, geologic time, crude characterization, etc., exist for the same Cartesian point but may vary with depth.
    Another important feature is that, even for the same Cartesian point, some characteristics can vary in time, for example the oil production, and this has to be considered and included in the Database for the posterior analysis.

    Methodology
    In the following lines, the procedure to incorporate the geoscientific information into the system is described in a general way.
    The GIS implementation began structuring the data, information and processes of each discipline. The data validation and processing tasks that every specialist carry out was studied in detail, beginning with the data gathering until the final delivery, according to technical and administrative specifications from users or Corporate Data Base Managers. The established processes define the data that must be gathered and provided. It is very important to define how they have to be loaded in the layers: alone or jointly with other data in the same layer. The level of complexity of the information must be the lowest possible.
    This applies for every category of data, even if it is raw or provided by interpreters. In this case, a layer is created for every type of data, which constitutes basic information for the spatial analysis.
    It must be noted that with the GIS standard software used in the Corporation, 3D-information management has a significant degree of difficulty. Nevertheless, this is compensated by the analysis capability that can be achieved. Those geo-related attributes generated by specialized applications can be referred to planes at different depth values. In Figure # 1, the average porosity f2 of Interval 2 is referred to plane P2 associated to depth h2.

    Figure # 1: 3D Information Management Schema

    Data
    Data, in terms of the Cartography French Committee is: "fact, phenomenon or notion that can be represented in a conventional way appropriated to the communication, interpretation or processing.5" From this point of view, "data is transformed to information while it is pertinent for the user".
    Data or information required for the oil exploration comes from different disciplines and for every one of them, there generally exist, many types, formats, mediums and quality levels. Often, this data in its original version can not be used by specialists in the formats and media in which it is originally acquired. For that reason, it must be previously processed to be suitable or appropriate.

    Classification according to processing degree
    The following table shows an example of some of the information commonly used by specialists of petrophysics. The content of the referred table has been classified as "raw data", "processed data" and "interpreted data" depending if the data was generated during the acquisition time, by means of some intermediate processing, or when the specialists have generated new data and/or information through an interpretation process. This data, independently of the stage of processing, is gathered, loaded in the system and used for the analysis. The difference is mostly based on the necessity of upgrading and improving the data through time.
    The classification must be done for every discipline data. Raw and processed data of each discipline can be loaded and managed by the Database Management System. There is an interpretation environment directly linked to this DBMS, where raw and processed data is handle with the geosciences specialized software and transformed to layers easy to manage and visualize.
    In the next lines, there is a brief description for the different types of data and the source and layers that the system contemplates as essential, without restricting the incorporation of new layers in the future.
    DISCIPLINE
    DATA
    TYPE
    FEATURE
    DESCRIPTION
    Well (id) Raw Text Well id, in which the sampled was taken
    Well spatial location (N,E; ?, ?) Raw Point Coordinates of the well
    Log type Raw Text Log name according to the standard nomenclature
    Top Raw Point Top depth in feet
    Base Raw Point Base depth in feet
    Curves Raw Line
    Well logs Processed Point, Line
    Petrophysics Deviation Log Raw Point, Line Record of the well drilling trajectory
    Technical Report Reference Raw Text Reference of the technical report in which the samples were analyzed
    Reservoir Average Oil Saturation Interpreted Polygon Oil Saturation average calculated for reservoir (So). Units: %
    Reservoir Average Resistivity Interpreted Polygon Average resistivity for specific reservoir. Units: ohms-m
    Reservoir Average Porosity Interpreted Polygon Average porosity calculated or measured in cores and logs for reservoir, Units: %
    Reservoir Average Permeability Interpreted Polygon Average permeability, calculated or measured in cores. Units: darcys or milidarcys
    Reservoir Thickness Interpreted Polygon Net oil sand thickness or net oil limestone thickness resulted by petrophysical evaluation.

    Units: feet or meters

    Table # 1: Examples of petrophysics data, classified according to processing degree


    Raw Data
    A. Source
    The largest part of data contained in the VIPA GIS, comes from the Corporate Database.  The original data pass through a data quality validation process, and loaded afterwards in the VIPA’s Database.  Each specialist validates its own discipline data by a validation matrix according to previously established parameters for each discipline.  Once this is accomplished, the data is transferred to the GIS environment.
    B. Layers
    Only the previously validated and geo-referenced data is the one to be loaded in the GIS as a layer. The separation of the information in layers permits a better visualization and analysis of all the information to obtain maps and other required products for the exploration studies.
    Annex # 3: EXAMPLES OF GIS VIPA LAYERS, shows some of the layers created for the VIPA’s Geographic Information System, as well as a brief description of the contents. It must be noticed that each layer has in the System its own metadata (see Metadata). This table must be used only as reference and it must not be taken as a definitive chart representing all the information of the GIS.
    The raw data of the VIPA GIS can be augmented (adding new data) while geological and geophysical interpretations along the country are being performed. But we try not to modify the original data in order to preserve it as the basis of the original interpretation and monitor changes through time.
    Processed data
    In the case of oil geosciences, some data is not found in the appropriate conditions to be incorporated directly to the interpretation environment. The incorporation depends on the processing that must be carried out to convert it in effective and suitable for the analysis that will be performed. A. Source
    Processed data come from either corporate archives (DataBase) or from individual DB preserved by specialists. Data from Corporate DB do not represent a problem for gathering and loading, since a standard and known Data Model has been established.
    The problem arises when the data is not in standard databases and they have not been "modeled" before. For example, the standard data model used by our Corporation does not embrace fields for Geochemical and Bioestratigraphical data. In this case, the parameter values measured in the samples are loaded in particular databases, to be able to represent them as maps or charts.
    B.Layers
    This data is placed in layers, and it has the same treatment as the raw data. In the Annex # 3, we can see some of the layers corresponding to processed data, loaded until now in the system.         Quality
    Data quality is a property that can be measured under different aspects or parameters. When dealing with Geographic Information Systems, there are at least certain characteristics that data must fulfil to be appropriate for the posterior utilization needs: "Data quality has the following measures: geo-reference, data attribute accuracy, consistency between geometry and attributes, topological consistency, data completeness and actualization, source"2.
    In the case of PDVSA, there exists information acquired many years ago (1920’s), preserved in analogical or digital media. We’re concerned about this ancient data, because in many cases it is data with low quality degree data, taking in account that the acquisition tools of past time, are not the same now neither they have the high quality degree from today’s ones.
    Although the biggest part of the information acquired nowadays is preserved in digital media, this does not guarantee its highest confidence, for this reason a sample quality control practice before processing is always advisable.
    As the quality control is an essential prerequisite for a GIS implementation, it increases generally the cost and the cycle time of any project by a significant amount. For this reason, it is necessary to find a balance between the required data quality for a system and the implied costs. Generally there has to be a compromise between both.
    Additionally, the information management processes are sometimes a factor that adds insecurity to the user in the case they are not well established and audited. For this reason, we need to have very well defined processes for data gathering, validation and actualization, in order to be consistent and more efficient in the data management and avoid uncertainty and criteria divergence about data quality. Annex # 1: WELL SELECTION PROCESS, shows one of the general processes established in order to assure that data coming into the system has been examined and validated by the specialist.
    The VIPA GIS has established validation matrixes for each class of data (See examples in Annex # 2: DATA VALIDATION MATRIX EXAMPLES). These matrixes contain parameters that permit to evaluate integrity, completeness and consistency in a color scale (called "traffic light" because of the colors: green, yellow and red), that help to diminish subjectivity during the evaluation of the data and generates homogeneous validation processes. Before incorporating the data to the System, it must pass by this quality control and the specialist in charge of this qualification process must document the results.
    Amongst the very useful products are the data quality maps, which can give the user a quick reference of how good or bad the data is distributed in a particular area. This helps by far the specialist not to underestimate information subjectively and to be able to provide recommendations about data.     Technical Specifications
    Technical specifications establishment for the data administration and quality control is a topic that deserves a complete document. However, in this paper we will mention the essential parts that must be included in the processes to guarantee the quality data of the VIPA products.
    For a GIS it is very important to set up, before the data loading process begins, the specifications that data must accomplish to be processed by the system. These specifications should be well documented and guarantee its accomplishment in a "natural" way during the process.
    The specifications must be documented for each type of data. For example, to "accept" the well data, the name must have a number specific of characters according to what the organization has settled, the coordinates must be in a particular coordinates system, and so on.
    For the final products, the specialist must take in to account the client requirements about what he/she needs for its own process and deliver a product that fulfill expectations. For example, a map with a specific scale or resolution to work with and allowing to obtain the required information. The legend, feature symbols and colors must also be previously established to avoid overwork.     Metadata
    The documentation of data is one of the processes that adjoin more value to an Information System. Metadata is the complementary information about the data, i.e. data about data.
    Data has a value directly associated to its source, age, accuracy, scale, among others. When this information is loaded in a System, the decision making process cycle time decrease favorably because of the confidence that it provides to the decision-maker.
    VIPA GIS metadata consist in two types of information. The first one is the information inherent to data, as creation date, scale, source, etc. The second is information provided by the analyst or interpreter and can be some comments or an own opinion about the data evaluation and quality control.
    It is necessary to emphasize that all the information loaded in the GIS can contain metadata, which remains associated to the base/source data, even this last one is copy or moved to different directories of the System.     Products
    The aim is to document the exploratory models generated from the analysis of geological and geophysical information in order to help the exploration decision making process.
    A preliminary consideration is that VIPA approach for the analysis of these models is based on expressing the sedimentary and tectonic events occurred in the past, using Sequence Stratigraphy Concepts. This implies that the subsurface is broken up in intervals bounded by tops and bottoms. These tops and bottoms result from the interpretation of seismic and well log profiles that are part of the most relevant raw data loaded into the system. Consequently, any feature characterizing or describing a specific sequence interval of the geological record may be projected on a map representing the base of the sequence involved. Figure # 2 illustrates this situation.
    On the other hand, since all the information is being properly geo-referenced, the z-value, which locates the data in depth, indicates automatically (as long as the interpretation has been carried out) which sequence is being referred to. Therefore, any geo-related attribute associated to a z-value is directly linked to a specific sequence and sequence maps might be prepared for any attribute of interest.
    Other VIPA GIS products consist in graphic or numeric material. Among these documents we can find reports, statistical charts and others.
    In this paper we will deal only with the cartographic products (i. e maps), which we classified in threedescriptive blocks according to their contents, nature and utilization. In the following lines, we describe them out and focus mostly on the spatial analysis products, one of the more relevant results of the VIPA GIS.
      Thematic Maps
    Thematic maps for each sequence consist in a series of layers containing parameters of interest, such as, porosity, Total Organic Content (TOC), sedimentary environments, etc. All features (points, lines, polygons) describing these parameters are stored in layers within the system.

    Figure # 2: Sequence information representation for products

    Quality data maps
    Some maps, representing the data quality for each discipline were built. These are maps where the well or seismic information is represented with specific symbols according to its certainty degree. This degree of confidence is derived from the data validation process together with the interpretation reliability.     Spatial analysis products
    In VIPA, spatial analyses are made to determine those areas with the best exploration potential, taking into account the resulting exploratory models as well as the data quality and interpretation confidence.
    Some maps, tables or reports are generated as a result of queries made to the system as a function of the stored data. The result of this process, called in the GIS "Spatial Analysis", contain the source data and information resulting from the specific query.
    The process for spatial analysis require previous activities, that consist in structuring the required data and information in such a way that allows the query to be done and obtaining interpretation and results in a easy manner.
    To find the best or worst exploration areas we have to ask to the system questions like:
  • Which are the areas where the interpreted sequence has the most confident information?
  • For a reservoir age X which are the nearest known reservoirs within the Sequence N?
  • Which are the reservoirs with better/worst productivity levels?
    •  These are some of the questions that can help to improve exploratory plans, to select a proper information depending on specific project needs and to prepare appropriate documentation depending on requirements. In Annex # 4, an example of an area selection under specific conditions is shown.


    Final considerations
    This work has lead us to perceive relevant aspects for implementing a GIS in order to plan a phase very important for hydrocarbon exploration as the analysis of data and information for decision-making in hydrocarbon exploration processes.
    Obviously, to create a GIS is always an arduous task. Especially if it deals with a wide variety of data, coming from different disciplines, indulging a diversity of clients, following specific organizational norms, procedures and products that must guarantee additional expectations in terms of exploration opportunities.
    A useful recommendation that can be withdrawn from this work is to establish a technique to allow an appropriate separation of system elements. This can help to define many of the features of the system as for example which data we can manage, the process of upgrading, etc.
    Likewise, establishing clear and logical processes from the beginning eases the task of "clarifying the natural path" for each data and any alteration or modification that may occur during manipulation. This gives double reward, because it is not only a question of establishing a process for people to illustrate how a system works. We also will be able to detect any alteration or detour during the process that might affect the system and, evidently, the final results.
    Another very important feature that we can stand out from this GIS is the data validation technique. Before loading into the DBMS, any data must be previously validated for specialists who, definitely, takes the final decision about this issue giving the eventual interpreter or user of the information, the needed confidence for carrying out his or her job. At this point, one of the main considerations is to establish the loading process and its responsible team. This is because, even though we mentioned that, the data validation must be done by the geo-related specialist, but the process of loading and its regular actualization must be done by the group in charge of administrate the GIS data and this task can never be delegated.
    We are sure that many of these premises will be improved through time. However, this is something that was considered during the launch that it is sometimes better to take risks than not try it at all.
    Our final propose of having an effective information management system that will be used for the Exploration Planning Cycle will be reached in not to much time, adding great value to the Corporation.
     

    Acknowledgments:
    The realization of this paper has been possible due to the initiative of Corporation for motivating and permitting the documentation and publication the daily work, especially that referred to new technologies applications.
    The excellent interdisciplinary teamwork of Visión País (VIPA) has been the fundamental key to accomplish this work, not only because of the provided knowledge for the realization of it, but the interest and advises given during the design and specifications establishment. All the team feels the challenge that this project means for the whole Corporation.Finally, the interaction among the members of the GIS group has given the final touch, especially with concerns to integration of cartography base and geosciences attributes, which is one of the strategic issues of this project.
     

    Annex # 1: WELL SELECTION AND VALIDATION PROCESS

     
     

    Annex # 2: DATA VALIDATION MATRIX EXAMPLES


     
     

    Annex # 3: EXAMPLES OF GIS VIPA LAYERS
     
    LAYER NUMBER
    NAME
    DESCRIPTION
    SOURCE
    1 to 5 Wells VIPA Key Wells (with TD)
    Well with sequence tops
    Wells with core information
    Wells with geochemistry data
    Wells with bioestratigraphy data    		 VIPA Integration
    VIPA Visualization
    6 to 8 Geochemistry COT values (grid)
    Madurity Values (grid)
    Pollution values (grid)    		 VIPA Integration
    VIPA Visualization
    9 to 12 Geology Sequence Tops
    Sequence Bases
    Sedimentary Environments
    Faults    		 VIPA Integration
    VIPA Visualization
    13 to 15 Geophysics 2D Survey Lines
    3D Survey Polygons
    Transects    		 VIPA Integration
    VIPA Visualization
    16 to 18 Production Leases
    Wells with "x" production
    Wells with API Crude "X"    		 PDVSA Production Department
    19 to 26 Cartography Country boundaries
    State boundaries
    County boundaries
    Main rivers
    Main roads
    Topography
    Bathymetry
    City names    		 Environment Minister
    27 to 28 Remote Sensing Radar Images
    Landsat Images    		 Exploration Technical Services
    29 to 43 Areas under special regimes ABBP, ZPL, ZAA, RF, MN, ZPCH, ACPT, RCPE, RFS, RB, RH, ZPL, PN, ARDI, RESFS Environment Minister

     

    Annex # 4: EXAMPLE OF A SPATIAL ANALYSIS

    References
    1. Chevallier, Jean Jacques; Systèmes d’information à référence spatial. Notes de cours; 1994.
    2. Detreköi, Äkos; Data Quality in GIS Environment; Budapest.
    3. Lilliu, Antonello et al; GIS Enabled Data Quality in Large Areas. AAPG, June 2001.
    4. Mitchell, Andy; The Esri Guide to GIS Analysis, Volume 1; 1999.
    5. Pornon, Henri; Les SIG, Mise en œuvre et applications.
     

    Author Information:
     
    Name: Violeta Taylhardat 
    Title: Geodesic Engineer, Zulia University, Venezuela
    Geodesic Sciences GIS MSc, Laval University, Canada.
    Company: Petroleos de Venezuela (PDVSA) s, a.
    Exploration Unit; Visión País (VIPA) Department.
    Address: Edif. PDVSA E, P & M; Av. La Estancia, Urbanización Chuao. Z. P. 1061
    Apdo. 829, Caracas, Venezuela
    Telephone: 58 - 11 - 0212 – 9084368
    Fax: 58 - 11 - 0212 – 9082053
    e-mail address: taylhardatv@pdvsa.com
    vtaylhardat@cantv.net


    BACK TO CONTENTS