Thin Beds and Anomaly Resolution in the Niobrara

Thin Beds and Anomaly Resolution in the Niobrara

By Rocky Roden
March 2018

Identifying thin beds using machine learning

In March of 2018, a study was conducted of the Niobrara using machine learning for multi-attribute analysis. Geophysical Insights obtained 100 square miles of seismic data covering the Niobrara in northeast Colorado from the Geophysical Pursuit and Geokinetics multi-client seismic library.  Using Paradise®, a multi-attribute learning application, geoscientists generated and classified seismic volumes, the results revealing thin beds below seismic tuning and anomaly resolution not practical with traditional interpretation methods. In total, the results of that classification demonstrated dramatically improved stratigraphic resolution and anomaly isolation within the Niobrara Formation and associated reservoir and source rock units.


Outline of Niobrar Phase 5


Figure 1: Outline of Niobrara Phase 5. 100 square miles within this survey were selected for analysis.

 Seismic Volume Classification Results

The result set, or classification in multi-colors, shows the continuity of facies and the continuity/discontinuity of anomalies in the greater Niobrara section. Each voxel represents 1 millisecond in the volume or approximately 15 feet. The white dashed correlation line shows anomalies on the B bench and the brackets indicate anomalies at the Codell level. Two horizontal boreholes are shown, however, neither seems an optimal penetration. The results demonstrate that machine learning in Paradise enables a sample-based thin bed analysis.

Classification - Instantaneous Attributes

Figure 2a: Classification based on eight Instantaneous attributes.

Figure 2b: The original amplitude data.

Figure 3: Self-Organizing Map (SOM) classification with low probability (<10%) anomalies (white) in the greater Niobrara section. The results may indicate concentrations of hydrocarbons in organic-rich shales. Anomalies also may show migration up section along faults from the Niobrara into the Sharon Springs. Niobrara B anomalies are best developed in elevated areas. Additional well data can corroborate anomaly interpretations.

Interactive 2D Map

Figure 4: Interactive 2D Colormaps use transparency (left and see inset) to highlight and extract anomalous behavior or particular facies. Note that the wellbore intersects a fault (white arrow) and misses the bracketed anomaly.

Principal Component Analysis (PCA) was used to identify and quantify the key attributes in the seismic volumes. The SOM process was then applied on these attributes to learn and classify the data. The neural topology, shown in the Paradise 2D Colormap, establishes the number of classes in the resulting seismic volume. The classification volumes show geologic features and anomalies that can aid in well location and development planning.


Top Niobrara Map - Structures and Associated Fractures

Figure 5: Map of structural features identified in this Niobrara volume. Near vertical graben-style faulting predominates in the upper Niobrara..

Using Self Organizing Maps to Expose Direct Hydrocarbon Indicators

Using Self Organizing Maps to Expose Direct Hydrocarbon Indicators

Analysis of offshore Gulf of Mexico – Class 3 AVO setting

The case study highlighted below is an offshore oil/gas field in 470 feet of water on the continental shelf of the Gulf of Mexico. The eld has two producing wells that were drilled on the upthrown side of a normal fault and into an amplitude anomaly. The normally-pressured reservoir is approximately 100 feet thick. The hydrocarbon-filled sandstone reservoir has low impedance compared to the encasing shales, indicative of a Class 3 AVO environment.

Applying both traditional interpretation methods and the Self-Organizing Maps (SOM’s) capability within Paradise, Mr. Rocky Roden, a Sr. Geophysical Consultant with Geophysical Insights, conducted an analysis on the 3D volume over this field. A group of seismic attributes were selected that would best expose Direct Hydrocarbon Indicators (DHIs).

Using known conditions, including AVO Class, reservoir characteristics, well logs, and seismic response data, the Rose and Associates DHI Consortium has developed a methodology for quantifying the risk of DHI-related prospects. The Consortium has compiled a database of several hundred wells and has been able to identify the most important DHI Characteristics.

Data quality, bed resolution, or seismic attribute algorithm may or may not allow an interpreter to nd any of the following examples, even if they are physically present in an ideal natural setting. The existence and/or variance in any of these DHI characteristics, as expressed in the data, enable the extent of risk to be better understood, and more so in a group of prospects, to be far better appreciated and quantified. According to the DHI Consortium, some of the most important DHI characteristics are:

  1. Amplitude conformance to downdip structural closure
  2. Internal consistency of amplitude
  3. Phase, frequency, or character changes at edge of reservoir
  4. Flat spots
  5. Attenuation

A SOM probability display using flat spot attributes


Results – Reduced risk through greater insights

The key findings of the study are as follows:

  • The SOM analyses of the seismic attributes associated with Flat Spots helped define two levels of hydrocarbon contacts in this dataset.
  • The SOM analyses on Attributes for Attenuation amplified apparent attenuation features, especially within the reservoir sand.
  • The SOM analyses of the seismic attributes named Ten Attributes 10% was especially helpful in defining amplitude conformance to downdip closure, and provided confidence in the internal consistency of the reservoir.
  • All three combinations of seismic attributes analyzed by SOM analysis revealed phase and character changes near the edge of the hydrocarbon reservoir.

SOM Analysis proved to complement and enhance the conventional interpretation by providing a second, completely independent method of exposing DHIs. This application of the SOM method increased confidence that insightful DHI characteristics are truly evidenced in the appraisal area. The results of this case study demonstrate that applying the DHI methodology with the SOM analysis engine in Paradise on selected seismic attributes can dramatically reduce uncertainty in the interpretation, thereby decreasing exploration risks in this geological setting.

A SOM classification of attenuation attributes


Seismic data provided courtesy of Petroleum Geo-Services (PGS)

Using Self-Organizing Maps to Define Seismic Facies

Using Self-Organizing Maps to Define Seismic Facies

Offshore South America Seismic Facies Analysis

This analysis and application of Paradise involved the evaluation of a 3D volume offshore
South America. A well had been drilled and encountered unexpected high pressures which prevented drilling to the desired deeper target. Traditional methods to identify pressure did not readily reveal a high-pressure zone before drilling. The challenge was to identify the high-pressure zone employing Self-Organizing-Maps (SOMs).


Initial evaluation of the 3D seismic volume suggested there may be facies and stratigraphic variations in the high-pressure zone.

After an interpretation of the local geology and putting this into a regional context, the anomalous high-pressure area appeared to be associated with a slope facies as interpreted from the conventional stacked seismic data.

Therefore, five different combinations of seismic attributes were applied in a SOM analysis to help define the seismic facies in the zone of interest. One specific combination of six seismic attributes clearly defined the seismic slope facies and associated high-pressure region. With the use of 2D colorbars in the 3D Viewer in the Paradise software, the highlighting of specific neurons enabled the visualization of the high pressured seismic facies.


A SOM Classification that exposes the region of high pore pressure using the Paradise 2D neural color map (right)




Analysis Results – Exposing High Pore Pressure Region


  • Based on pressure readings from a single well, the increase in pressure seems to be associated with a hummocky, wavy, and at times chaotic seismic reflection character.
  • This reflection character is typically associated with a slope setting where there are rapid facies changes, discontinuous siltstone and mudstone beds and at times channelized sands with interchannel mudstones.
  • Dozens of seismic attributes were generated to help define this seismic facies associated with pressure in the well.
  • Five different sets of seismic attributes were selected for SOM analysis to define this pressure associated seismic facies.
  • All of the SOM Classification volumes and to some degree Probability volumes defined components of this seismic facies (e.g., top, bottom, internal seismic reflection character, reflection character above and below, etc.).
  • A specific set of seismic attributes effectively isolated the pressure zone through a SOM analysis


A SOM probability display showing the high pore pressure region as an anomaly in Paradise


Visualization and Characterization of Paleozoic (Ordovician-Devonian) Tight Carbonate Reservoirs, Oklahoma, Part 1

Visualization and Characterization of Paleozoic (Ordovician-Devonian) Tight Carbonate Reservoirs, Oklahoma, Part 1

The results documented in this Application Brief are from a proof-of-concept study that culminated in the acquisition of a Paradise license by the client. The evaluation was intended to demonstrate the effectiveness of the Paradise multi-attribute analysis software.

Self-Organizing Maps (SOMs) were developed from PCAs that utilize all of the attributes for the first 4 Eigenvectors. Other SOMs were comprised of a “recipe” of the top 11 attributes from the first 4 Eigenvectors:

EV1: MM Trace Envelope and MM Sweetness
EV2: MM Imaginary Part, MM RAI, MM Instantaneous Phase
EV3: Instant. Frequency, Thin Bed, Smoothed Frequency
EV4: MM Normalized Amplitude, MM Enhanced Migrated Stack (input data used for pruning), Phase Breaks / MM Paraphrase


The following issues were examined for the Viola Chocolate Brown platform carbonates:

  • Evaluation of a Test location for a well;
  • Compare and contrast low cumulative productivity Well #1 with high productivity Well #2.

For the Hunton / Sylvan:

  • Identify play area(s) for the Hunton;
  • Explain moderate cumulative production in key Well #3.

Additionally, this work identified an uncalibrated play area in the Hunton and demonstrated an ability to identify and solve correlation issues around karst features. 

The results of the project demonstrated to the client that the technology could enable better well planning, help identify more productive perforation intervals, and aid in the retention of critical leases. In summary, the project was a success!

To convey a sense of the scope, deliverables, and project timeline, the following results were accomplished in 30 days:

  • Extracted platform and basinal carbonate play areas
  • Optimized potential TEST location
  • Distinguished good from poor productivity areas (note limited productivity data)
  • Helped assess efficacy of perforations for different play types
  • Identified under-perforated zones
  • Enabled easy high-grading of leasehold for three plays
  • Results exposed critical aspects “hidden” in Client’s seismic data regarding
    Structure: Revealed unknown karst features
    Stratigraphy: Enhanced key geometries
    Rock properties: Resolved flat spots in Hunton, Viola, and Simpson that indicated likely facies contrasts, possible diagenetic changes, and/or fluid effects

Viola Findings

Figure 1 is a structure map on the Wichita unconformity. Synthetic T/D functions from one Viola well (north) and one Hunton well (south), indicated by black triangles, were used to tie 58 wells to this map level. Labeled in white is the proposed Test location and three key wells whose productivity was evaluated. The most prominent feature of this map is the red horst block bounded by faults. 

Figure 2 shows the horizons on conventional seismic opposite the instantaneous phase attribute which was used to guide the interpretation. Three interpolated horizons – Top Hunton (peak), Top Viola (trough), and Base Viola (peak) – were repicked on every line. Three additional events – the Wichita unconformity (Peak), Top Sylvan (peak) and Top Arbuckle (trough) – were picked from scratch on every line.

Visualizing reservoirs with machine learning 01

Visualizing tight carbonate reservoirs 02

The platform carbonates can be extracted from the volume so that its key features can be described when the 2D Colormap in Paradise makes the non-platform facies transparent (Figure 3a). There are as many as eight SWNE fault bounded segments located by black gaps and one NW-SE frontal segment that extend nearly end to end in this area. The Test location can be seen to be too far downdip to get a good perforation perimeter so the client was encouraged to move it updip as there was no danger of penetrating the unconformity. The only well with productivity information in this facies, #1, has a moderate 38,000 BO (Figure 3b). On the multi-attribute extraction map, its location can be seen to be less than optimal and at the edge of a null zone.

Visualizing tight carbonate reservoirs 03

tight carbonate reservoirs 04

With the basinal carbonates back in view in Figure 4a, one can see the location of the better performing Well #2. In vertical section (Figure 4b), the red highlighted and white-circled perforation zone shows a position on the updip, lapout trap. At 92,595 BO, this well’s productivity from the Viola Chocolate Brown was nearly three times that of the platform carbonates.

tight carbonate reservoirs 05

A conventional seismic section (Figure 5) shows potential trap configurations fairly well but does not discriminate the key objectives as well as results from multi-attribute analysis do.

tight carbonate reservoirs 06

Note that SOM results (Figure 6a and 6b) of the well’s penetration seem to reveal the extent of perforation efficacy in the basinal Viola. Also the low probability facies in white overlay suggest that it is possible that perforation of the upper most Viola Chocolate Brown might have also yielded hydrocarbons.

tight carbonate reservoirs 07

Figures 7a shows a regional NW-SE Northern Arb line in which the black circle represents a difficult correlation area that was not unraveled until it was examined via a multi-attribute section.

tight carbonate reservoirs 08

Figure 7b reveals a large karst feature in the updip Viola. Note that the patterns make correlation around the feature straightforward. Our client was very interested in this result because of another play in Kansas. We have also imaged karsts in the Austin Chalk.

SOM of tight carbonate reservoirs

Figure 8 highlights flat spots via red ellipses that represent the facies change from platform to basinal facies. This relationship persists in each of these three topologies, which represent increasing discrimination from 8x8, 10x10, and 12x12. Refer back to Figure 2 at approximately 1.5 seconds for a similar relationship in the Simpson.

SOM for tight carbonate reservoirs 02


Visualization and Characterization of Paleozoic (Ordovician-Devonian) Tight Carbonate Reservoirs, Oklahoma, Part 2

Visualization and Characterization of Paleozoic (Ordovician-Devonian) Tight Carbonate Reservoirs, Oklahoma, Part 2

Hunton Findings

This Application Brief is a companion piece to another case study on findings in the Viola (Paradise Application Brief PC-1). Together, these two pieces describe a proof-of-concept project conducted to demonstrate the efficacy of the Paradise® multi-attribute analysis software. For this section, a structure map on the Wichita unconformity is used as an Index Map to all of the other figures.

Visualizing reservoirs with machine learning 01

Visualizing reservoirs with machine learning 02

Figures 2a and 2b show the remarkable isolation of a Hunton objective above the Top of the Sylvan horizon and sequence boundary (black), another immediately below the Sylvan with a possible flat spot, and another updip of the well above the Top Viola with an apparent flat spot.

In Figures 3a and 3b transparency flip shows only the stratigraphy that was previously highlighted by white low probability. On the left, the Top Silvan horizon bisects the anomalous areas; its location is indicated by white arrow on the right. The extent of the anomalies is shown by the large white ellipse. The small ellipse indicates the location of a small anomaly that sits above the Top Viola. “Mining” the data set with the same multi-attribute result is as simple as moving the inline and timeslice through the volume till another area lights up (Figure 4). One can see that another anomaly is located downdip and further to the Northwest of the anomalies in Figure 3.

Visualizing reservoirs with machine learning 03

Null zone is the Top Sylvan Horizon Pick

Visualizing tight carbonate reservoirs


Key Well #3 was regarded as a “Type” Hunton well with a modest hydrocarbon recovery. However, the well was perforated in the Sylvan to exploit a smallish truncation trap. Figure 5 makes it abundantly clear that the perforations were taken pretty far downdip and would not produce much from the tapered reservoir. A take point further updip might have been planned that would still keep a fair distance from the unconformity. Also, if the well had been deepened it might have encountered a target at the base of the Sylvan that has very similar properties to the Test objective at the base of the Viola complete with pseudo flat spot at the base of its platform (Figure 6b).

Visualizing tight carbonate reservoirs

Figure 6a has an overlay of the objective section at Well #3 that shows a low probability feature that is in the Hunton section.

Visualizing reservoirs with machine learning

Figure 7a shows the red anomaly lit up within the volume and Figure 7b shows that trend in two parts and comprised of four red neurons. It is underlain by the Top Sylvan horizon. Note that this early result predated the well corrections so they are used here only as locaters on the trend to show that this anomaly may not have been fully exploited.

unconformities and anomalies 01

A summary of the workflow outlined in Figure 8 shows how quickly and effectively Paradise can be applied to hi-grade a leasehold!

unconformities and anomalies 02