Register      Login
The APPEA Journal The APPEA Journal Society
Journal of Australian Energy Producers
RESEARCH ARTICLE

A PROCESS FOR PREDICTING POROSITY AND PERMEABILITY IN DEEP EARLY JURASSIC/TRIASSIC TARGETS, AUSTRALIAN NORTH WEST SHELF

P.C. Smalley, D. Jablonski and I. Simpson

The APPEA Journal 38(1) 759 - 775
Published: 1998

Abstract

In deep exploration prospects, reservoir quality is often a key risk. We describe a hybrid empirical-theoretical approach to minimise this risk:

Use available regional petrographic-sedimentological data to tune theoretical depth-porosity-permeability curves.

Verify that the model correctly represents the controlling geological processes by comparing these curves to core analysis data.

Re-tune the model to the expected conditions in the deep prospects, using empirical quartz cementation predictions, regional depositional models and pressure prognoses from basin modelling.

Use the re-tuned model to extrapolate porosity to the new depth, then predicting permeability from porosity.

Eight studied wells in the Early Jurassic/Triassic, Dampier Sub-basin, provided an understanding of regional diagenetic style and the major reservoir quality controls. BP's PermPredictor model was used to construct regional, zone-specific depth-porosity-permeability curves from the petrographic and sedimentological data: sand ductile grain content, grain size, sorting and quartz cementation. Quartz cement correlates with burial depth, beginning at ~2,200 m and increasing by seven per cent (± two per cent) per km.

The regional modelled depth-porosity-permeability relations agree well with the core analysis dataset, indicating model reliability. The modelled curves were then re-tuned to the predicted conditions in two notional exploration prospects, with top-structure depths of 4.7 km (Prospect 1) and 4.4 km (Prospect 2), the latter of these being overpressured. Predicted porosities were 5−11 per cent in Prospect 1 and ll−17per cent in Prospect 2, with permeabilities of 30−250 mD and 400−1,000 mD respectively assuming a clean sand composition. A dirty sand model (less likely) predicts <0.01mD in Prospect 1 and 0.1−35mD in Prospect 2, illustrating the huge impact of sand ductile content.

https://doi.org/10.1071/AJ97050

© CSIRO 1998

Committee on Publication Ethics


Export Citation

View Dimensions