Geostatistics for the Oil & Gas Industry

• Depth Conversion

Shooting a seismic campaign is too expensive not to use the results optimally when building structural models. Using geostatistics correctly to perform time-to-depth conversions means that the resulting structural model not only ties to the wells but also reflects the inherent spatial behavior of the seismic information. This is possible because the geostatistical working framework allows the user to take the natural correlation between seismic and well data into account. We offer depth conversion studies that optimally integrate the available seismic information (like TWT or stacking velocities) using state of the art geostatistical techniques.

• Gridding

Gridding techniques are used to interpolate between known data points (e.g. the wells) in order to get a map that should show the main structural features of the reality (e.g. trends or specific areas of high values). Geostatistics has a very important advantage over other gridding techniques. The map obtained using the geostatistical interpolator (called kriging) directly incorporates the inherent spatial continuity of the variable under study. This spatial continuity, modeled by the so-called variogram, determines the weights assigned to each datum in the geostatistical gridding algorithm.

When more information is correctly incorporated in the gridding algorithm the result will be more interpretable. Using geostatistics means that we can correctly integrate seismic or other auxiliary information when gridding petrophysical characteristics. This can be done by using the one of the many variants of the co-kriging algorithm. Likewise, the map obtained by kriging or co-kriging correctly reproduces the discontinuities induced by faults. These are just two of the numerous possibilities that make geostatistics the most suitable gridding algorithm when dealing with natural resources.

• Seismic Attribute Analysis

A good analysis of the different seismic attributes should not only extract the maximum amount of information from your seismic data. It must also define which is the most relevant information. Any decision you make must be based on concise pertinent information without the extra volumes of superfluous data.

We firstly extract the maximum from your seismic attribute data. Then using multivariate analysis we can assess the real significance of the individual attributes. We finally provide you with easily read classification maps that help you to make the right decision.

• Volumetrics

Any business decision made about a field’s potential is based on volumes. Unfortunately, you never know the exact volume of a reservoir. Moreover, one estimate of the volume is just that, an estimate without any indication of its precision. How can you make the right business decision if you don’t know the uncertainty associated with the estimated volume? To maximize your chances of making the right decision you need to quantify the risk associated with the volume calculation whatever the volume considered (e.g. GRV, HCPV, IGIP, STOIIP, HIIP ...).

We use state of the art geostatistical simulation techniques to produce different possible images of the reservoir each of which is consistent with both well and seismic data. Applying the OWC to each of these plausible images means that we can obtain a series of plausible volumes for the reservoir and finally their distribution curve. This allows you to make your decision according to a fully quantified risk calculation. And doesn’t knowing the odds increase your chances of success?.

• Reservoir Characterization

To successfully evaluate the performance of a reservoir you need the most appropriate numerical model. Using the latest geostatistical techniques we firstly produce a structural model based on the complete integration of seismic, geological and well data. So as to obtain the most representative model, the choice of the geostatistical method is largely determined by the depositional characteristics of the field. We then fill the structural model with the petrophysical characteristics in accordance with their particular spatial behavior (e.g. trends or anisotropies) and any correlation with seismic attributes. The complete package means that we offer you integrated reservoir modeling solutions that are consistent not only with all numerical data but also the geological interpretation. In this way your investigation of multiple drilling and production scenarios for future field development will be based on a totally consistent reservoir model.

• Risk Assessment

You understand that risk is present everywhere in oil & gas projects. But simply knowing it is there doesn’t help you to make the right decision. We can quantify the risk associated with the different aspects of a project by using the most appropriate geostatistical simulation technique. After having quantified these individual risks, you can incorporate them in a global risk analysis framework to obtain the probability of success associated with the project. Consider the following simple examples:

 Dry Wells
Drilling a dry well onshore or, even worse, offshore is a disaster. Using a series of equally likely geostatistically generated structural maps we can evaluate the probability of drilling a dry well throughout the entire region of interest. This can help you to define a drilling program that must consistently intersect the reservoir while providing as much information as possible about its lateral extension.

 Producing Wells
You cannot assign a degree of certainty to one gridded map of the petrophysical characteristics. While it will show certain zones of high and low values, one map does not allow you to evaluate the possible range of values at any potential well location. Using geostatistical simulations we can calculate the distribution curve, and hence probability maps, for the petrophysical characteristics. So we can help you maximize your chances of success when deciding where to drill a new well.

 Spill Point Locations
Spill points can have a tremendous influence on the oil & gas volumes you think you have. Only correctly applied geostatistical simulation techniques allow you to produce images of your reservoir that show all the potential spill points. A gridded map only shows you the most obvious spill points that are not necessarily the ones that have the greatest financial impact on your project. We integrate all sources of information in the most appropriate geostatistical technique to obtain the most realistic structural images of the reservoir to help you decide what a difference the spill points make.

• Multi-Layer Modeling of Layer Cake Reservoirs

Using a standard sequential approach to modeling layer cake reservoirs (going from the top down for example) means that the resulting reservoir model does not correctly reflect the inherent correlation between the layers. This information can only be incorporated in the model via a global multi-layer approach that incorporates all the information (i.e. deviated and vertical wells and seismic data) simultaneously. However standard geostatistical techniques have not been designed to correctly deal with deviated wells. This is unfortunate as they have become an important tool in the industry and as such provide a wealth of information that must not be ignored. For this reason Geovariances has developed and successfully applied a new multi-layer modeling technique that not only uses all your data in a consistent way but also correctly reproduces the structural discontinuities resulting from any existing fault system. So we can provide you with the most realistic model of your layer cake reservoirs.