[First Hit]

Datapages, Inc.Print this page

Risk Reduction through Neural Networks Chimney Analysis: Frontier Exploration in East African Rift Basin

Baranova, Valentina; Mustaqeem, Azer; Karaja, Francis; Mburu, Danson


Application of advanced methods of Reservoir Characterization for frontier exploration area [East African rift basins specifically South Anza Basin, Kenya] is discussed. Application of neural network, especially chimney analysis, has led to reducing risk in frontier exploration and identifying new targets by analyzing the vertical migration pathways and fault seal integrity.


Chimney Analysis is a type of neural network which is an intelligent mix of a number of seismic attributes related to the effects of vertical hydrocarbon migration. Each Previous HitattributeNext Hit carries a partial correlation to the vertical migration. By integrating the attributes the software manages to create a stacking effect where these Previous HitattributeNext Hit reinforce each other. The examples of attributes used for the Chimney analysis in this project are: Similarity, Simple Chimney Previous HitAttributeNext Hit, Polar Dip, RMS (Root Means Square), Filter Residual, Average Frequency, Frequency Wash-Out Ratio, Average Frequency. The results obtained then are compared with the chimney interpretation model which shows the type of chimneys and the risk associated to each trap type.

Case Study

There are many indicators of hydrocarbon vertical migration in this dataset. On the seismic lines chimneys often exhibit conical shapes of vertical effects and show as incoherent or hummocky signature on Previous HitcoherencyTop/similarity sections. There are lots of chimney-looking areas on almost every seismic line we had received which shows an active migration system in the uplifted and reactivated fault zones.

The results of chimney analysis are compared with the chimney interpretation model which shows the type of chimneys and the risk associated to each trap type. We have strong evidence of fault related chimneys for this basin.

Ranking of the leads have been divided into three categories: 1. Good leads, leads supported by good evidence of charge and seal; 2. Questionable leads, when some questions about structure, charge, or seal could be confirmed by 3-D seismic data; and 3. High risk leads, where chimney results indicate high risk of seal or charge failure. Based on chimney analysis for South Anza Basin we narrowed down the areas for further 3-D seismic acquisition and drilling.

In conjunction with burial curves and geological insight the target areas with migration pathways and possible source rocks were remodeled in association with the chimney results.


AAPG Search and Discovery Article #90163©2013AAPG 2013 Annual Convention and Exhibition, Pittsburgh, Pennsylvania, May 19-22, 2013