By Jayson Jacoby
A 10-liter water sample taken Sunday from Elk Creek, one of several sources Baker City taps from its watershed, contained 913 cryptosporidium oocysts, a much higher number than any previous sample test reported, Michelle Owen, the city's public works director, said Thursday.
Seven water samples collected on July 31 - none of them directly from Elk Creek - contained from zero to three oocysts.
Eight other samples that were also collected on Sunday contained 0, one or two oocysts.
Owen said when city officials learned Wednesday morning between 9 a.m. and 10 a.m. about the test results from Elk Creek, workers immediately shut off the diversion, blocking water from that creek from entering the city's distribution system.
The city had been using Elk Creek water, along with water from several other sources in the watershed, throughout the summer up until Wednesday morning, Owen said.
The Elk Creek water sample deepens the mystery of how, when and where enough crypto oocysts entered the city's water to cause widespread illness among city residents, an outbreak that apparently started around mid-July.
Mark Bennett, a Baker County commissioner, said during Wednesday's commission meeting that based on door-to-door surveys and other anecdotal evidence, an estimated 300 to 400 people have been infected with crypto and showed symptoms, chiefly persistent watery diarrhea and stomach cramps.
It's not uncommon for infected people to show no symptoms, however.
Lab tests of stool samples have confirmed crypto infection in 15 people, but officials have said repeatedly that the confirmed number will likely be considerably smaller than the actual total because most people who have symptoms aren't tested.
One major question the Elk Creek water sample raises has to do with dilution - specifically, how the crypto concentration changes as a result of multiple water sources being mixed in the city's pipeline.
For instance, the seven water samples collected on July 31 either had no crypto (one sample); 1 oocyst (two samples); two oocysts (three samples); or three oocysts (one sample).
Yet six of those samples were taken at a point where water from multiple sources had been mixed, so all six of those samples could have contained some Elk Creek water.
In other words, if the Elk Creek water on July 31 had a high level of crypto - that was definitely the case four days later, when the sample with 913 oocysts was collected - those oocysts were considerably diluted by the time water reached the city.
The second series of samples, the nine samples taken on Sunday and including the Elk Creek water, poses the same question.
One of those samples was taken from a water line on College Street in Baker City, a sample that conceivably included water from Elk Creek as well as several other sources.
The College Street water sample contained one oocyst.
It's impossible to say what percentage of the volume of water in that sample was from Elk Creek.
Elk Creek is the farthest south of the series of streams and springs the city uses as water sources.
One potential concern with the Elk Creek area is cattle that graze in the vicinity on Forest Service allotments.
Cattle are a known carrier of crypto.
Owen said Jake Jones, the city's watershed manager, inspected the area around the Elk Creek diversion both on July 31 and again Wednesday morning, after the high crypto concentration had been reported. Neither time did Jones find any evidence that cattle had been in the area, Owen said.
Elk Creek drains a large area, however - thousands of acres - so the source of the crypto could be miles upstream from where the city diverts the creek water into its pipeline.
Elk can also spread the parasite through their feces, and Owen said a herd of about 50 elk is known to roam the upper part of the Elk Creek drainage, also upstream from the city's diversion.
For much of the crypto investigation city and state officials have focused their efforts on the far northern part of the watershed, specifically Goodrich Reservoir.
That's about seven air miles northwest of the Elk Creek diversion.
Officials have said during the past week that a plausible theory of the crypto source is feces from a herd of mountain goats that lives near Goodrich Reservoir.
A water sample taken at Goodrich Reservoir on July 31 contained two crypto oocysts.
A sample taken from the Goodrich water pipeline on Sunday contained no oocysts, Owen said.
The city has not brought any Goodrich water into the distribution system since July 31, and City Manager Mike Kee said the city probably won't use the reservoir for the rest of the summer.
With Elk Creek also shut off, the city's concern about having a sufficient water supply grows, Owen said.
Elk Creek, fortunately, is not a major source of water in late summer because its flow diminishes considerably by August.
Results from other water samples taken Sunday include: Marble Springs, Little Marble Creek, the city's well, Salmon Creek, all 0 oocysts; Little Mill Creek, one oocyst; and Mill Creek, two oocysts.
Most of the water drawn from the city's well actually originated as surface water from streams in the mountains.