BISMARCK, N.D. — Scientists who helped calculate oil spilled from a broken BP well into the Gulf of Mexico are questioning the methodology used to estimate the amount of crude that recently leaked from a ruptured pipeline into a wheat field in northwestern North Dakota.

Tesoro Corp. said it came up with its more than 20,000-barrel spill estimate using ground analysis. But oil spill experts say a more accurate assessment likely would come from calculating how much crude went into the pipeline versus what was supposed come out at its terminus.

Farmer Steve Jensen discovered the North Dakota oil spill the size of seven football fields while harvesting wheat Sept. 29. Tesoro first estimated the spill at its underground pipeline near Tioga at 750 barrels.

About a week later, the San Antonio, Texas-based company increased the estimate to 20,600 barrels, or some 865,000 gallons, making it one of the largest spills in North Dakota history.

Tesoro said in a statement to The Associated Press that it based its spill-size calculations on “data collected during our field investigation work. This investigation included a thorough examination of the site spill characteristics including factors such as surface area and depth of soil impacted, and soil porosity.” The company would not elaborate.

The company said its “site investigation was developed based on well-established and recognized American Petroleum Institute, Geologic Society of America and American Institute of Professional Geologists standards.”

Jack Hess, executive director of the Geologic Society of America and Bill Siok, executive director of the American Institute of Professional Geologists, said their groups have no such standards or guidelines for estimating an oil spill.

“We’ve never issued any guidelines over oil spills,” Hess said. “That’s not the kind of business we are in and something we wouldn’t get into.”

Siok, of the American Institute of Professional Geologists, said, “I’m stumped. I kind of suspect they made an incorrect reference.”

Purdue University engineering professor Steve Wereley said Tesoro’s calculation of how much oil it released in the North Dakota wheat field likely is “at best, a guess.”