A national 3D geological model of Denmark: Condensing more than 125 years of geological mapping

Transcription

1 A national 3D geological model of Denmark: Condensing more than 125 years of geological mapping Peter B.E. Sandersen Thomas Vangkilde-Pedersen Geological Survey of Denmark and Greenland Flemming Jørgensen Ministry of Energy, Utilities and Climate Richard Thomsen Jørgen Tulstrup Johnny Fredericia 9 th 3D Geological Mapping Workshop. October 31, Baltimore, Maryland

2 Introduction More than 125 years of geological mapping at GEUS We are expected to solve more and more complicated geological problems From from 2D to 3D: Data and interpretations are seen from new angles New insights that could lead to revision of earlier interpretations New types of challenges GEUS strategy: A national, digital 3D geological model for the Danish area The current interpretation of the subsurface geology The initial considerations and discussions on how to make a national geological model for Denmark Early sketches presented at the Workshop in Denver two years ago (Jørgensen et al. 2013) 2

3 Presentation outline Model considerations Initial considerations and open questions (GEUS work groups and workshops) The components of the national model GEUS consensus based on the initial discussions Challenges Challenges in connection with the modelling process and how we plan to meet them! Perspectives and strategy Where are we a few years from now? 3

4 Model considerations - 1 Model area GEUS is surveying the geology of both Denmark and Greenland The area of Denmark is very small compared to the area of Greenland Geographically Greenland and Denmark are separated, so two (or more) separate models will be needed Up till now we have worked mostly on the model for the Danish area 4

5 Model considerations - 2 Depth range Which parts of the subsurface should be included in the model? How to cope with the general problem of varying data coverage and data resolution? The deep part (> m) vs. the shallow part (< m) Higher data density in the shallow part The Quaternary succession is very complex and requires a large amount of data in order to be mapped Is it necessary to split into an upper and a lower part and therefore deal with separate models? Shallow subsurface: Groundwater investigations Raw materials/minerals exploration Soil contamination investigations Geotechnical projects Geothermal projects Deep subsurface: Oil/ gas exploration Geothermal projects Storage 5

6 Model considerations - 3 Building on existing data and knowledge GEUS: National 2D maps of specific subsurface layer boundaries and surface geology maps 2D/3D models from different mapping projects are difficult to merge: different ways of modelling, varying quality, different purposes When building a new national geological model it is necessary to collect and evaluate existing data and models A range of public databases are hosted by GEUS (raw data and model interpretations): the backbone of the national model Updating these databases is very important 6

7 Apologies to different internet sources Model considerations - 4 The end-users and the national model A wide range of end-users with different model needs Purpose-specific mapping and modelling does not apply for the national model If we construct a model that is not purpose-specific there is a risk that we will end up with a model that is either too sketchy or too heterogeneous to be attractive to the end-users 7

8 Model considerations - 5 Model detail Can we map the detail we want with the data we have? More data needed? Complex geology requires a large amount of good data Data is geographically clustered: Detailed model within the data clusters? Less detailed model outside the clusters? Model scale Should the model be able to manage multiple scales? In 3D modelling software there are no zoom limitations: the data and model can be viewed and evaluated at any scale It is very important to describe the scale limitations of the model to the end-user 8

9 The components of the national model - 1 The geological elements: The model will be a 3D framework model: Surfaces: Tops and bottoms of defined geological formations, stratigraphic complexes or other types of spatially recognizable units Surfaces: Erosional surfaces, stratigraphic markers and transgressional surfaces Formations are defined in the legend of the Danish subsurface The surfaces will be defined and controlled by interpretation points, lines, polygons etc. together with an interpolated grid or triangulation Detailed geological information between the mapped surfaces (volumetric cells: lithology and lithofacies or parameters such as porosity) 9

10 The components of the national model - 2 Model strategy The model will contain selected surfaces in the beginning It shall be possible to include varying levels of detail within the national model. Defined standards and procedures Platform-independent: individual model elements can be made using different software, but with a standardized export format The model will firstly include the Danish onshore, with the offshore areas and Greenland being added later 10

11 The components of the national model - 3 The model surfaces: 13 surfaces encompassing the deep succession from top of Pre-Zechstein (Permian) to the top of the Chalk Group (Danian; Paleocene) exist in a preliminary version Two tertiary surfaces and the top of the Pre-Quaternary surface is planned The complex geology of the Quaternary will be mapped locally and regionally using layer-boundaries and volumetric cells 11

12 Chr. Brogaard, GEUS The components of the national model - 4 Databases and visualization: The model database is currently in the planning phase Focus on constructing an architecture that can contain all model elements Future changes should be possible Different visualization tools are being evaluated but no decisions have been made at this point 12

13 Glacial tectonics in Røgle Cliff (Larsen 2002) Complex geology Challenges m Modelling complex geology The Quaternary succession is generally highly complex A limited amount of detailed data and the data is often clustered Small areas with high resolution and large areas outside with low resolution Enough data to map in high detail? This challenge can be met with a model that can handle different scales with different degrees of detail Not necessarily full coverage with interpretations A model like this will show the status of the mapping Visualization of where our knowledge is good and where it is poor Mapping local detail Data clusters 13

14 Challenges - 2 Keeping the model up-to-date Most likely the model will be outdated even before it is finished Fast development in computer power and modelling software Improvements in modelling and mapping methods Continuous emergence of new data Keeping our methods and procedures up front and include the newest data No updates? The end-users will lose interest in the model and seek information elsewhere The model should be dynamic and regularly updated (versioning) The update procedure should be as easy as possible 14

15 Challenges - 3 Handling uncertainty Model uncertainty has been heavily discussed for many years When going from 2D to 3D there is a growing demand for an uncertainty assessment along with the model This issue is highly complex and the challenges are numerous Uncertainties of each dataset Uncertainties of a combined dataset How do we handle both the quantitative and the qualitative aspects? The uncertainty assessment concept must be tailored especially for the national model and apply to all model elements 15

16 Challenges - 4 Meeting the end-user needs Our models will be used for very different purposes There may be specific user needs that standard model-outputs cannot meet Distinguishing between standardized offthe-shelf products and individually tailored products (based on the same framework model) A standard product from the 3D Model Department Store could for instance be a suite of nation-wide surfaces to be used in a project dealing with regional or national assessments A tailored product from the 3D Custom Shop could be a number of specific surfaces in a small urban area supplied with lithology in volumetric cells in that specific area 16

17 Challenges - 5 Organizing the work A national 3D geological model will require an organization that is capable of supporting a project of this size and complexity for a long period of time GEUS is an institution with highly specialized researchers and the organization will have to provide the project with the required man-power and skills Focus on project management 17

18 Challenges - 6 Funding Stable funding for a long period of time This matter is still not fully resolved and different scenarios have been discussed A national model may be difficult to sell before is has been made especially when the construction process spans several years 18

19 Perspectives and strategy - 1 A long but necessary pre-modelling phase We have a large amount of data and numerous existing geological models that can serve as a platform for the model building We need to merge existing knowledge, old and new data, old and new models into a coherent, national 3D geological model that can be used by a wide range of end-users We are slowly progressing, but at this point we have decided on type and content of the national model and a strategy for the coming years has been outlined 19

20 Perspectives and strategy - 2 The short-term strategy (4 years): Established an organization around the national model Established at least 15 key surfaces Initiated the work on local/regional elements in the shallow parts Launch of the database and a betaversion of the web interface Standards and procedures in a number of guidelines Established a dialogue with end-users Established a long-term financing plan The long-term strategy (10 years): Finished modelling the major mapped areas with local/regional surfaces Included the Danish offshore areas in the model 20

21 Geological Thank Survey of Denmark you and for Greenland listening! Ministry of Energy, Utilities and Climate 9 th 3D Geological Mapping Workshop October 31, Baltimore, Maryland