Measurements of radioactive waste in Hanford's underground tanks appear to show that the level of waste in 52 of the tanks is rising.
The reason for the readings, taken over decades in some cases, is a mystery.
But the worst-case scenario, that precipitation or snowmelt may be getting into some of the waste tanks, has the state of Washington concerned.
Hanford stores 56 million gallons of high-level radioactive and hazardous chemical waste left from the production of weapons plutonium in underground tanks. That includes 28 newer double-shell tanks and about 141 older single-shell tanks, some of which have leaked in the past.
An assessment of routine measurements of the single-shell tanks concluded that levels may be rising in 52 of the tanks. For some tanks, an average increase of a couple thousands of an inch a year has been measured. But measurements taken in 21 of the tank indicate larger increases of up to an average of about three-quarters of an inch a year.
The measurements have been taken four times a year for decades for some of the tanks.
"At this point, we are not sure what the cause is," said Dennis Washenfelder, engineering manager for tank and pipeline integrity for Washington River Protection Solutions.
Some of the measurements are taken by periodically lowering a plummet on a wire in the tank until it hits the waste. But the surface of the waste can be rough, with peaks and valleys across tanks that can have diameters of 75 feet and capacities of up to 1 million gallons.
It's possible that the plummet, which stays in the tank between measurements, could be hitting different places on the surface of the waste when it is lowered.
In addition, the waste contaminates the plummet, sometimes getting trapped on its end. A small amount of water is flushed down the wire to remove it.
That water could be pooling in a pocket beneath the plummet, Washenfelder said. In that case the measurement would not indicate an increase in the level of waste across the tank.
However, the change in measurements also could be caused by moisture infiltrating the waste, possibly at the below-ground joint between the concrete tank top and metal risers that give access to the underground tank.
Access to the risers usually is through pits, or boxes buried flush with the ground, and water infiltrating the pits also could be getting into the tanks. The pits are topped with interlocking concrete blocks that have a urethane weather cover, plus joints are taped where needed, which would restrict the amount of rain and snow melt that gets into the pits, Washenfelder said.
Washington River Protection Solutions plans to increase monitoring to as often as weekly in some of the tanks.
It also is making plans to inspect the insides of the enclosed tanks using a video camera as the first step to determine the cause of the increasing waste level measurements.
"The visual inspection will provide us the information to confirm whether water intrusion has occurred, and, if it has, what actions should be taken," said Lori Gamache, DOE spokeswoman.
The camera will look at the waste surface in the vicinity of the plummets and at places where a second type of measurement device has been used. The second device relies on a neutron monitor sensitive to water to measure waste levels, and it also has shown an increase in liquid levels in some cases.
In addition, video will be recorded of the inside of the concrete dome of the tank, with the camera searching for damp places that would indicate water is infiltrating it.
"As we begin to inspect the tanks and the causes unfold, we will determine what to do with DOE and Ecology," Washenfelder said.
The Washington State Department of Ecology is following the issue closely, concerned that water may be infiltrating the tanks, said Jeff Lyon, the agency's project manager for tank waste storage.
One of the state's main concerns is the length of time the waste will remain in the old tanks, he said.
Some of the tanks date back to World War II and could contain waste until the last of the single shell tank waste is transferred to double shell tanks about 2040. Because of the limited capacity of the double shell tanks, the Hanford vitrification plant now under construction has to be treating waste from the double shell tanks to make space in them for more of the waste from single shell tanks.
The state's other chief concern is the potential of waste leaking from the tanks into the soil beneath them. Some of the tanks with increasing waste level measurements are known to have leaked in the past, Lyon said.
DOE temporarily stabilized the tanks, completing work in 2004 to remove pumpable liquids in the single shell tanks to regulatory standards. But some of the tanks still have more than 1,000 gallons of free liquid, Lyon said. In addition, more liquid is trapped in the pores of the solid waste that remains in most of the single shell tanks.