HART, D. M., R. S. BALCH, W. W. WEISS, S. WO, Petroleum Recovery Research Center, New Mexico Tech, Socorro, NM

ABSTRACT: Time-to-Depth Conversion of Nash Draw "L" Seismic Horizon using Seismic Attributes and Neural Networks

Well log and 3-D seismic data were used to construct three depth maps to the top of the target L horizon of the Nash Draw field in southeastern New Mexico. Two depth maps were made using Landmark TM software packages TDQ and Z-map. The third depth map was made using a multilayer perceptron (MLP) neural network to regress for velocity at each seismic bin. Conventional geostatistics reliably interpolated depths only in the region defined by well control. The MLP approach used the best three of 28 statistically ranked seismic attributes to predict the average velocity field to the top of the L horizon. Each map was constructed using 15 wells as control points with three wells excluded for testing. Test wells one and two were located away from the control wells and have anomalous average velocities.

Accurate depth maps are useful for reservoir development, particularly for stratigraphic and structural trap location, drilling depth and reservoir modeling.

The three test wells were used to compare the robustness of the computed depth maps, and all depth predictions were compared to the true depths determined from gamma ray logs for each well.

Geostatistical methods underestimate depths to the top of the L for the test wells lying outside the central clustering of control wells, while the MLP solution calculates a relationship that should be valid in each seismic bin in the field. Further refinements in the data are expected to yield a higher degree of accuracy between the real and predicted depths using MLP.

AAPG Search and Discovery Article #90915©2000 AAPG Rocky Mountain Section, Albuquerque, New Mexico