Summary
We discuss the use of nuclear-magnetic-resonance (NMR) logging in the
petrophysical evaluation of thin sand/shale laminations. NMR helps detect thin
beds, determine fluid type, establish the hydrocarbon type and volume if
hydrocarbon is present, and, finally, determine the permeability of the sand
layers (as opposed to that of the sand/shale system).
Experiments were conducted on samples of 100% sand, 100% clay, and sand/clay
layers with an NMR-logging tool at surface to verify the characteristic
T2 bimodal relaxation distribution often observed in NMR logs
that are acquired in thin beds. From the bimodal distribution, it is often
possible to determine a cutoff to separate the productive sand layers from the
shale layers and, with it, the porosity fraction of each component.
Subsequently, the sand fraction, or net/gross ratio, can be estimated assuming
that the 100%-sand porosity is known. Because gas, oil, and water have
different NMR properties, fluid-typing techniques such as 2D NMR offer useful
insights into the fluid type and properties in thin-layer sands. Because the
laminations thickness is often less than the antenna aperture, the estimated
permeability of the sand/shale system will undercall the true permeability of
the sand layers only. In this case, their permeability can be estimated quickly
from Darcy's fluid-flow model.
We show examples of thin sand/shale laminations that are oil-bearing and
gas-bearing. In each case, the NMR detection was verified against
borehole-imaging logs, and the fluid type in the sands was determined from
multidimensional NMR analysis. The derived hydrocarbon volume was then compared
with the results estimated from a triaxial induction tool. Permeability of the
sand layers was also computed and compared to that of nearby thick sands. Core
data in one well was used to validate NMR detection, porosity, permeability,
and net sand thickness.
© 2010. Society of Petroleum Engineers
View full textPDF
(
2,617 KB
)
History
- Original manuscript received:
7 June 2006
- Meeting paper published:
25 September 2006
- Revised manuscript received:
6 May 2010
- Manuscript approved:
26 May 2010
- Published online:
21 October 2010
- Version of record:
17 June 2011
View Cart
