Abstract
The objective of this paper is to describe a new methodology for estimating the volume and distribution of hydrocarbons-in-place in the Deep Coal Play (depth>3,000 feet) of the Gidgealpa Group in the Cooper Basin of Australia. Huge volumes of gas, at very high concentrations, are present in the source rock reservoirs of the Deep Coal Play. However, there are no publicly available detailed resource estimates largely because, until recently, the coals have not been considered as potential reservoirs and therefore the majority of the routinely gathered well data are not applicable for estimating coal gas volumes.
To circumvent this problem, a new methodology for estimating the hydrocarbon gas content of Deep Coals was developed from the existing mudlog dataset. These data are available for all coal intervals in almost every well drilled in the basin. Normalized (for drilling rate-of-penetration) mud hydrocarbon gas readings from coal layers obtained during the drilling process were compared to total hydrocarbon gas content estimates derived by desorbing coal core samples. This comparison showed a consistent proportional relationship indicating that the mud gas dataset could be used to estimate the hydrocarbon gas content for the Deep Coal Play.
This analysis focussed on coals likely to be completed and contribute to production. Thus, Deep Coals greater than 10 feet thick were selected and named Target Deep Coals (TDCs). The average hydrocarbon gas content and some 20 other parameters such as GR, caliper, thermal maturity and rate-of-penetration were documented for all TDCs in 1,300 Cooper Basin wells.
For the hydrocarbons-in-place estimate, the Gidgealpa Group was subdivided into 10 lithostratigraphic zones across the Cooper Basin. The individual TDC parameters in each zone were combined to provide per well and zone TDC hydrocarbon content and thickness as inputs for the hydrocarbons-in-place estimate.
TDC reservoirs have enough lateral continuity to apply a conventional gridding process for creating 3 realizations (low, deterministic, high) to map the coal thickness. In contrast, the TDC reservoir gas content varies significantly over short distances. This high degree of spatial variability, combined with the large variability in well spacing, required the use of a conditional simulation technique to estimate ranges (P90, P50, P10) of the gas resource. These methods are discussed in the paper, with the Toolachee T1 zone used to illustrate the process.
Volumes are world-class. Total hydrocarbons-in-place for the Deep Coal Play for just the Toolachee T1 zone is estimated to be 82 Tscf of gas with 3.9 billion bbl of hydrocarbon liquids.
The high gas content values (400-1,000+ scf/ton) and large aggregate thickness (up to 500 feet) of the Cooper Basin coals make them a prime target for exploitation. Work is currently underway by several companies to "crack the code" that will enable economic production. Assuming this becomes a reality, estimates of the in-place gas resource and the location of volumetric "sweet spots" provided by the new process will be of critical importance.
