A model of the data management "production line." Alternating white and gray bands show six major functions in the data flow. Orange boxes are spatially dependent activities in each functional area; the pink shows areas where the company has only indirect control of spatial quality. Green boxes show systematic data review throughout an asset data cycle, and the blue lines indicate inter-functional relatedness and data movement. (Images courtesy of Devon Energy)

There’s a scene in “Raiders of the Lost Ark” where the heroes are attempting to divine the location of the Ark of the Covenant by using an ancient headpiece from a staff. The Nazis are also seeking this treasure, but they’re missing half of the information. Our heroes, realizing this mistake, gleefully exclaim, “They’re digging in the wrong place!”

It’s not fun and games when you’re drilling a US $100 million deepwater well in the wrong place, however.

Unfortunately, according to Jon Stigant, geodetic operations coordinator for Devon Energy, that’s exactly what happens when operators don’t pay attention to geospatial information. Surely an industry that drills for hidden targets must have some idea of where those targets are. Yes, but when different data sources are integrated into an interpretation project or a reservoir model, location mistakes can happen that can cause well bores to miss their targets, sometimes by hundreds or even thousands of feet.

“Geodesy is a way of describing the Earth that allows us to make accurate measurements,” Stigant said. “In geological or geophysical terms, it’s an earth model. But it’s an earth model that is entirely related to knowing where you are on the Earth and how that location relates to other locations.”

But a map is a map, right? Not necessarily. “The mapping process is fundamentally dependent on a mathematical model called an ellipsoid,” he said. “That mathematical model is attached to the Earth at a point.” In North America, that reference point is Meade’s Ranch in Kansas. The North American reference system is known as NAD 27. Other parts of the world use different reference systems, and GPS satellite data is referenced as WGS 84. Its reference point is the Earth’s center of mass.

Stigant likens these different reference systems to measuring the distance to the corner of a desk from two different corners in the same room. “It’s a different set of measurements, from difference origins or reference points, but the same point that is being measured,” he said. “It’s a simple concept. It’s just that the ‘corner’ we’re talking about in each case is the center of a different ellipsoid. The measurements are the latitude and longitude of the point. So the same point on the surface of the earth can have different latitude and longitude depending on the reference origin or datum.”

The difference, he said, is usually about 500 to 600 ft (152.5 to 183 m) but can be as much as 2,000 ft (610 m), “enough to cause us heart failure if we’re drilling wells and we want to be plus or minus 100 ft (30.5 m).”

The problem comes from merging different datasets that might have been mapped on different reference systems. If the plat for the prospect and the seismic survey are both referenced one way and the well log data is referenced differently, those wells may not be where they appear to be in the seismic cube. This problem is compounded by the industry’s slow cycle time — by the time production begins, the data may have changed hands several times, and referencing corrections may not make it all the way through the process. Often the applications used for interpretation and reservoir analyses are also manipulating the data incorrectly.

The hard sell

So why isn’t anyone doing anything about this? Actually, people are trying — Stigant has helped form a joint industry partnership (JIP) with other companies to try to address the problems in software, and he works in many forums within the industry to address appropriate spatial data handling. But the education challenge is huge.

“Geodesy is important,” Stigant said. “Unfortunately, most people don’t even know the word, let alone what it actually means and the impact it can have on their operations.”

He added that it’s likely that the industry is aware, at least at some level, of the problem. “Not many people actually want to say anything about it, and they often skirt around the problem,” he said. “Unfortunately, it’s costing our industry a fortune every year.”

If a system was in place to prevent these mistakes, it could make an enormous difference. “Some geologists spend 80% of their time trying to figure out where the well bores are and 20% of their time developing new prospects,” he said. “If those numbers were switched, it would represent a 300% improvement in staff efficiency. It would dramatically reduce the cycle time and aid in development of new wells in that field.” It would also aid in rig scheduling and better preparation for subsurface faults and hazards when drilling.

How to fix the problem

Stigant has a two-fold approach to solving this problem. One is procedural. While it’s difficult to correct historical data because of lack of support data and the sheer volume, it’s entirely possible to make sure that new data coming into the database is spatially correct.

One way is through contracting. “It becomes a supply-side management issue,” he said. “If you’re buying data, you put specifications in contracts that say this is the standard that I expect you to apply, and you have to be able to tell me that it’s the standard that you’ve applied, and if your people don’t understand that, you need to train them.”

Another is software. The explosion of Geographical Information System (GIS) software in recent years has, in Stigant’s view, led to serious issues due to a lack of awareness of and training in geodetic concepts by the developers of that software. In a Society of Exploration Geophysicists paper, he wrote, “Low-level training in geodetic principles, distributed data sources and computer applications, and a generally low quality in implementation of spatial sciences in both, can produce some serious errors in operation implementation as well as some serious mismatching of data upon which many major and expensive decisions are made.”

The JIP is evaluating software applications and developing a set of standards to offer to software developers to ensure that their applications take geospatial issues into account.

Lastly, Stigant is working on a system within Devon that he eventually hopes will become a standard workflow for any new data entering the database. The methodology treats the process like a production line with data quality-checks at frequent stages throughout the life of the asset (Figure 1). These checks on the data are incorporated into a GIS system that provides an authoritative information source for reviewing projects for an E&P asset throughout its lifecycle.

Stigant noted that audit checks will help with several issues:

  • Checking all third-party supplied data and providing them to users from a centrally managed and verified store;
  • Sending qualified and properly briefed professionals to oversee the spatial data management of all geophysical acquisition;
  • Rigorously reviewing incoming speculative and multiclient data and their internal transmission between teams;
  • Auditing all proposed well locations in relation to the original data from which they are derived;
  • Sending qualified and experienced personnel to oversee the work of directional and horizontal drilling crews; and
  • Auditing software from application vendors, especially those used for cartography, interpretation, directional drilling and reservoir management.

Within Devon, Stigant’s team helps out on a project-by-project basis and is working to develop more systematic support across all disciplines. At many other companies it’s probably not even being addressed. For this, Stigant also has a suggestion.

“Spatial management is a bit like legal planning; accounting; diversity; or health, safety and environment (HSE),” he said. “These are things the corporation takes care of. The operational divisions don’t see them as primary issues that they should be taking care of from a profit point of view.

“In order to do HSE properly, management has to tell them, ‘There’s a policy in place, and that policy is that you will observe all of the HSE protocol to avoid killing people or setting fire to refineries or blowing up rigs. It’s an acceptable expense against your business unit, and we expect you to spend that money to make sure we do things safely.’”

Similarly, geospatial mismanagement is a risk the industry need not take. In the digital world, problems with spatial data are invisible but build up like “hardening of the arteries,” insidiously reducing the efficiency of the enterprise. “It is often easier to explain a dry hole by saying oil and gas were not there than to admit the well was drilled in the wrong place,” Stigant said.

If the industry applied the same standards to good geospatial management as it does to HSE, the rewards would be tremendous. “We can actually do this,” Stigant said. “The people in this industry, in my opinion, are some of the most creative and inventive people in the world when they see a need. They’re some of the best at solving problems once they recognize them. The fact that this issue is not well recognized, particularly at the corporate management level, doesn’t alter its impact on the bottom line.

“I think if they could understand that this is actually affecting their production and their bottom line profits to the degree it is, they would be strongly motivated to implement changes to address this issue.”