This page contains a Flash digital edition of a book.
LONDON TECHNICAL MEETINGS ROUNDUP London


‘Optimised shale resource development: proper placement of wells and hydraulic fracture stages’


Mark Wilson, European Oil & Gas Equity Research Analyst with Macquarie Securities Group, summaries the SPE London September presentation by SPE Distinguished Lecturer Usman Ahmed, of Baker Hughes.


The concept of recovering hydrocarbons from shale reservoirs is theoretically not that difficult - the combination of two technologies - horizontal drilling and downhole hydraulic fracturing - has allowed the process to become commonplace in the USA.


The global shale gas resource potential is huge: some 44,400 Tcf of shale gas in place. The Baschenov Shale in Russia was illustrated as being 80 times the (resource) size of the Bakken play in the USA! “No wonder integrated companies are investing in that area of Russia.”


However, the development of shale gas can be “not very profitable - you have to watch production and cost”. The reason for this stems from the common misconception that shales are homogenous (ie constant in properties), whereas they are in fact highly heterogenic (they vary in properties significantly and sometimes within small areas). Hence, the key question companies must answer is not simply ‘How many wells can be drilled or fracs performed?’, but, more importantly, ‘Where do you place those wells and where do you place those fracs within the well?’


The data


A graph showing maximum production rates from individual shale gas wells in the USA, from 1981 onwards, showed first a gradual increase in the number of wells drilled during the late 90s, as lateral well design took over from vertical wells. Then, from the mid-2000s, as fraccing was combined, a significant increase in both the number of wells and the maximum production rates was achieved. However, along with the increase in number of wells and more wells with higher maximum production rates, there were also just as many dots on the chart showing recent wells that produced low, or no, gas rates.


From 2007 to 2011, the average length of the horizontal section has increased from 2,500ft to 4,500ft and the average number of frac stages per well from 14 to 22. The average maximum peak production rate was also shown to have gone up, but still the significant decline in the first six months of a well’s producing life was evident. Even with all this progress, it is estimated that 70% of all unconventional wells fail to reach their production targets, 60% of all frac stages are ineffective and 73% of operators say they do not know enough about the subsurface they are dealing with.


The goal: maximise return on investment (ROI)


The broad concept of unconventional development is no different from conventional: maximise reservoir contact (of the well) and manage production, in order to maximise overall recovery. However, the important subsurface parameters differ, ie the four key components of a conventional reservoir are porosity, permeability, water saturation and reservoir pressure, whereas the four key components of an unconventional reservoir are: total organic content (TOC), vitrinite reflectance (VR), thermal maturity and reservoir pressure - the one key parameter common to both.


Find the sweet spot


Finding locations with an acceptable combination of these key parameters within a shale formation, or any unconventional formation, is the goal to start maximising ROI. However, those reservoir parameters may just satisfy one of three broader catagories which must combine to


08


produce a true sweet spot for production: the geochemical parameters (of TOC, VR, etc) may be acceptable in a certain area, but to find a true sweet spot, we also want the geomechanical parameters (stress regime, fractures, faults, brittleness) and the geological parameters (depth, thickness, etc) to also be ideal.


Sweet spots can also be very localised: analysis has shown that sweet spots are not contiguous. A case study on a five-lateral well development within a small area of the Barnett shale illustrated how, across five horizontal wells, with 10-12 stage fracs in each, ‘micro-seismic’ showed extremely effective ‘rock movement’ in the areas around the wells during the frac procedures (ie, good fracs?). However, when production logging tools (PLTs) were run in all five wells, they showed that only the middle three-to-four frac zones in just two of the wells were actually producing the development’s gas.


In such scenarios, a traditional view could be to “blame the fracs” (not powerful enough, not the right proppant etc). However, subsequent seismic analysis of the area, combined with log and core data, suggested an extremely localised sweet spot combining high TOC, suitable brittleness, but also - very importantly - the presence of natural fractures in the rock, and that was the area fracced by the ‘good’ (~25%) fracs.


The pre-drill ability to locate sweet spots suggested TOC could be estimated using ‘seismic volume’ data and a drilling plan to ‘follow the high TOC path’ (a Haynesville shale case study was presented). It is virtually impossible to generalise on this, however, due to the “significantly different lithologies across different shale formations/basins”. The important takeaway was that cost control up front, ie in the early stages of a shale gas exploration/appraisal/development project, may not be cost-effective in the long run. The logs (all logs - particularly fracture imaging logs and shear wave accoustic logs for stress regimes), the cores, the tests all need to be taken in multiple wells to understand the lithology and the heterogeneity of that lithology as best you can. Then the same process may have to be repeated at more localised areas within a basin.


How to place fracs within a well


One main takeaway here is that placing fracs at a constant spacing along a well bore will, according to the overall production data, reduce ROI by being inefficient use of capital versus the production achieved. However, the best places to locate a frac are ‘natural fractures zones’, which have been shown from cores and/or image logging tools. Fault zones must be avoided at all costs.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16