While deputies and volunteers searched last month for a missing 13-year-old in the Sprague River area, Bill Jennings was crunching numbers, trying to determine where the teenager could be.
KLAMATH FALLS — While deputies and volunteers searched last month for a missing 13-year-old in the Sprague River area, Bill Jennings was crunching numbers, trying to determine where the teenager could be.
The Klamath Community College math professor was able to determine that Cody Jones-Barnard likely walked south.
He also determined that Cody stuck to roads, and being young, he probably stopped to take a break.
That information put the lost teenager inside a 6.2-mile radius from his Sprague River home.
Jennings was right on.
Authorities found Cody under a tree just off the road a little more than a mile from his house.
Jennings made his determinations using information from Robert Koester's book, "Lost Person Behavior," and other resources.
The book wasn't the only tool used to find Cody, but the information Jennings gleaned from it certainly contributed, said Klamath County Sheriff Tim Evinger.
The sheriff plans to use the information — with Jennings' help — more often.
"In many jurisdictions, this has become the search manager's Bible," Evinger said.
Using statistics and probability to help track down lost people caught Jennings' attention last fall when Evinger was taking one of his classes.
Evinger told the professor he saw potential application in using statistics for search and rescue. He loaned Jennings a copy of "Lost Person Behavior." In it, author Koester amassed data on numerous searches across the U.S. to make probability tables for how far missing people had traveled from their last known locations, known in search and rescue circles as the initial planning point.
Age, mental and physical status, terrain, time of day, and other factors were taken into consideration.
"He's giving us a lot to work with," Jennings said.
Jennings saw a practical application in the book, and he later used some of the data on final exams.
"'When am I ever going to use this?' This is an example of that," Jennings said about the recent search for Cody.
The method does not pinpoint an exact location, he said, but it can help narrow the search based on the already-recorded data.
Jennings was called on to use the collected data in a real-life scenario soon enough.
Zoey Dorsey, a 4-year-old living outside Brookings, disappeared from her home the afternoon of March 17.
The area surrounding the house was hilly and wooded. Evinger arrived the following day to assist in the search.
Koester's book laid out the case. There was a high probability Zoey had not gotten far because of the steep terrain.
There were additional variables to consider because of Zoey's age. Was she hungry? Was she sleep-deprived? Was she frightened? What could all these potential variables do to her mental status?
Jennings considered all the variables.
He contacted a nurse and asked her about the likelihood that Zoey, now probably hungry, cold and confused, had gone into shock and was no longer on the move. The nurse said the probability was high.
"I think the big thing here is not being afraid to use your resources," Jennings said.
Zoey was found, crying, the following day about 1,800 feet from her home in the woods. The data that Jennings presented to Evinger closely matched the location where searchers found the girl.
"The information he put forward was very accurate," Evinger said.
The same proved to be true when Cody went missing. In this case, Jennings, Evinger and searchers had to consider additional variables.
They knew Cody was a high-functioning autistic and probably wouldn't respond to verbal calls from searchers. He also had told his parents he wanted to go to California to see his birth mother.
"We made informed decisions," Evinger said. "We didn't just throw a dart at a map."