Since the inception of 3-D seismic, geological and geophysical disciplines have dominated exploration software advancement. Progress in seismic interpretation, mapping and the like has been plentiful along the way. However, software for analyzing and producing unconventional plays, where success is not dependent upon geological characterization alone, has not reached full potential. Incremental software improvements and a cookie-cutter approach to drilling and completions do not allow the oil and gas industry to optimize performance in tight gas fields. Exploiting unconventional resources requires breakthrough technology and a shift in the engineering role as it pertains to the geological, geophysical and engineering equation.
As the race to recover tight gas in North America heats up and everyone searches for the next Barnett Shale or San Juan Basin, it makes sense to remember that the vast majority of past learning came at a high cost over many years. More often than not, knowledge was gained through the drill bit. And learning through the drill bit is more costly and takes longer than any other alternatives. Although operators can't unlock all of the mysteries of a given area without drilling, making the most of the acquired data makes great sense. Leveraging today's technology can allow users to test competing drilling and completion scenarios before, during and after the drill bit hits the ground. Once the next Barnet or San Juan is found, industry professionals should be ready to dramatically compress the time required to "unlock the play."

Unique technology conquers tight gas challenges
The challenges of reservoir characterization in these unconventional formations are many, but most involve one or some combination of the following, all of which reduce operator's abilities to predict and obtain high return on investment:
• Complex geology of rock properties - difficulty mapping rock quality;
• Difficulty distinguishing between completion performance and reservoir performance; and
• Lack of understanding regarding depletion or bypassed reserves.
These challenges hinder good decision-making regarding spacing, drilling/completion design and best practice development. Decisions are therefore often made based on the drill bit and the pump truck and take a long time to reach.
Conventional wisdom suggests that it generally takes 3 to 5 years to establish best practices in planning and execution. Technical professionals have had neither the technology required to meet some of these challenges nor the time to apply "reservoir science" in the face of rapidly moving rig schedules. Object Reservoir, a technology company focused on tight gas plays, has developed a modeling tool that allows users to apply a level of rigor to their work that was deemed impractical, if not impossible, until now.
Resolve, a proprietary application by the company, enables rapid fluid flow modeling with the built-in ability to change geometry and reservoir properties without "regridding." This allows testing of all possible geologic and completion scenarios and diminishes uncertainty. This software accounts for static information and transient behavior of wells and reservoirs in a single model, yielding earlier understanding of reservoir performance.
The system enables operators to distinguish between reservoir and frac performance in tight gas, where higher accuracy is needed since the reservoir behavior is driven by near-well flow, reservoir properties and hydraulic fracturing.
By employing an unstructured, automatic mesh (as opposed to a grid) and applying finer resolution near the well bore and frac, the tool enables users to mine transient data and get the best possible understanding of drainage patterns and reservoir performance. This high resolution also allows the user to distinguish between reservoir performance and completions performance, which is critical in managing tight gas assets. An unstructured mesh enables model flow from the reservoir to the frac and from within the frac to the well bore within a single flow model as opposed to introducing gross approximations with orthogonal grids and local grid refinement.
Once existing production is characterized, this same auto-generating mesh makes it simple for users to make changes to reservoir properties and geometry, insert new wells, and test various drilling and completion scenarios. For each scenario (and there can be hundreds) users can generate a recovery forecast, compare results and integrate with their economic analysis to make decisions. In effect, the software becomes a very reliable development planning tool that can help operators increase the visibility around each key decision.
Workflows have been developed to allow users to model large multiwell reservoirs in a technically sound and efficient manner. Hundreds of competing development plans can be evaluated against a range of possible geologic realizations. The results yield a better understanding of drainage patterns based on well performance and leads to "right-spacing" decisions as opposed to merely "down-spacing."
The majority of tight gas reservoirs have been developed on uniform spacing, resulting in high numbers of wells.

Unconventional assets
Traditional modeling and simulation tools have struggled with unconventional assets for a variety of reasons, but the main issue is that they lend themselves to "full-field model." When an area may include hundreds if not thousands of wells, the traditional technology simply lacks the right mix of computational horsepower and resolution. Conquering this challenge will require innovation through process and perhaps a new way of thinking.
Arden McCracken, chief reservoir engineer for the software company, explained, "The first thing we need to recognize is that, while many of these reservoirs are made up of large contiguous sand bodies, the level of heterogeneity throughout the sands creates a lot of internal barriers to flow. Once people realize this, they can understand and appreciate an approach which takes advantage of high resolution around the well bore and frac but, more importantly, uses single-well, mechanistic models to derive key reservoir properties that are the basis for multiwell sector models. With this technology and process, we can help people bring science to the onshore market in an effective and efficient way."
Gene Ennis, chief executive officer for the company, added, "The only thing worse than being without the tools you need to do your job is having a set of tools that you think is right for the job and then applying them in the wrong way. Whether it's a wrench that a rig hand tries to use as a hammer or a traditional simulator that an engineer tries to apply to tight gas fields, the results are rarely worth the pain of the effort."
One variation of this tight gas workflow, recently applied to identify infill opportunities, is outlined below.
The area of interest was drilled on 40-acre spacing. The operator was contemplating downspacing to 20s and possibly even 10s. All wells were fracture-stimulated with variable frac half-lengths and three producing layers. Within the three layers existed a great deal of variation in permeability, porosity and net-to-gross ratio.
As a starting point, a section - 640 acres with 16 producing wells - was selected for analysis. Within the section, three wells were selected for detailed modeling. Selection criteria included location across the section and rate profile differences. Initially, single-well, 40-acre models were used and fine-tuned to honor the first 2 years of each well's producing history (pressure and rate). As is the case with most tight gas wells, the performance of each single-well model proves highly sensitive to permeability and fracture half-length. Based on three wells, these key parameters were mapped across the entire section and used to generate a 16-well, single-layer model. A first-pass forecast was run for all wells and compared to actual behavior.
The frac half lengths were then estimated from the treatment volumes using single-well models, which are much easier and quicker to work with, ultimately producing a predictive model in much less time than modeling this same system using traditional methods. The performance was decent for replicating the behavior of the wells but not deemed acceptable for selecting locations.
For further refinement and to increase confidence in the predictive nature of the model, two additional wells were studied individually, and the key parameters were fine-tuned. Now, based on five single-well models over a 640-acre section, a revised permeability map was generated. The 16-well model was calibrated based on the change in properties. Additional forecasts were generated and compared to actual performance. At this point the model honored both the initial production as well as the late-life production (which indicates no significant influence from offset wells).
Armed with the confidence of a 16-well predictive model, which was based on detailed models for five of the 16 wells, it was time to analyze the virtual wells. Using automatically generated mesh without regridding, new wells were inserted at various locations. Multiple forecasts were generated for comparison and to optimize future well placement. This efficiency of both technology and process can by applied to wells previously untouched by comparable analytics. The high resolution of reservoir/wellbore connection allows increased detail in understanding key parameters such as permeability and effective half-length. In this case, well placement was improved dramatically by discovering that down-spacing to 20-acre units was the way to go (but only across half of the section). Gaining an accurate picture of the reservoir in a short period of time positioned the team to select optimal well locations, ultimately reducing the total number of wells required to sweep the zone. Increased understanding of the reservoir optimizes recovery and influences future development decisions.

The economic impact
Exploration and mature asset exploitation have held very different meanings to the oil and gas industry over the years.
However, the technology and process outlined above bring a much higher level of sophistication to a group of assets that have traditionally foregone this step for a large percentage of operators. Separating the performance of the rock and the completion brings success in the form of lower expenditure and higher return. Put another way, understanding the interaction between rock and the well allows lower cost coupled with higher recoveries. The example outline above saved eight wells in just one section. In today's commodity price environment, these eight wells could be used to increase production elsewhere. Drilling rigs are difficult to obtain, so every well bore must count.
Higher gas prices got us to this point. Better technology could take us where we need to go without burdening operators' organizations. It is time to interpret the pressure data as closely as the seismic or log data. Breakthrough technology essential to successful exploration in tight gas environments will allow companies to make decisions and establish best practices about drilling and completion strategies by rapidly combining geological, reservoir and near well/near frac properties.

For more information, please visit www.objectres.com.