WDFW NEWS RELEASE
Washington Department of Fish and Wildlife
600 Capitol Way North, Olympia, WA 98501-1091
http://wdfw.wa.gov/

March 16, 2015
Contact: Cindy LeFleur, (360) 696-6211

Coho fry lost after equipment fails at Kalama Falls Hatchery

VANCOUVER, Wash. – An estimated 200,000 coho salmon fry died last week after a generator and pump systems failed at the Kalama Falls Hatchery in Cowlitz County.

The salmon fry suffocated March 9 after a generator failed, destroying one pump and damaging two others, which reduced the flow of water to the hatchery’s incubation room, said Cindy Le Fleur, regional fish manager for the Washington Department of Fish and Wildlife (WDFW).

Le Fleur estimates that approximately two-thirds of the hatchery’s 2014 late coho fry, measuring about an inch long, were lost.

“We don’t have a firm estimate yet, because some of the fish that survived are in a delicate condition and we don’t want to disturb them to count the survivors,” she said.

Le Fleur said approximately 1.1 million fall chinook and 500,000 spring chinook fry have been moved to state hatcheries on the Lewis River as a safeguard, while WDFW personnel work to repair the damaged pumps.

“We hope to have things back in operation at the Kalama Falls Hatchery by the end of this week,” she said.

Built in 1959, the Kalama Falls Hatchery raises spring and fall chinook, late coho and summer and winter steelhead.

RIP.

Reminds me of the springer smolts lost in Orygun due to the drunk hatchery truck driver, but more of a break down than a crime in this case.

Or two years of lost smolt on the Satsop, because the water intake, plugged up with leaves during a storm and the smolt sufficated.

Wow, this kind of stuff just shouldn't happen these days. Good luck,


SZ

You can have ever fail safe in the world at a fish hatchery. What back up systems do is reduce the frequency of failures. You cannot eliminate them. Just one more reason why total reliance on hatchery systems is a limited choice.

Why do they run pumps on a generator instead of utility power? Is the hatchery located to remote to use the grid? It seems to me the generator would be used for backup in case there was a power outage. If they have to run on a generator they should have had backup that came online when a problem was sensed.

The generator is for when the main power fails. Remember that heavy wind and rain last weekend?

I don't pay much attention to Oregon weather but I see it was stormy down there. If I was running that hatchery I would have a tech standing by whenever the emergency generator came on line. They could have an automated call to get someone there soon after the backup generator came on line just to make sure all is well. An ounce of prevention.

The Kalama Falls hatchery has on-station housing for staff and standby requirements for staff that don't live on-station, so I'm sure this was not a situation of nobody responding to the problem. Additionally, every hatchery that I know of has alarms and emergency procedures to respond to situations just like this. They also test all emergency water supply equipment once a week to detect any problems with the system. However, sometimes stuff just happens that is nobody's fault, no matter how much they prepare. I'm not sure what happened in this case, but since they reported that pumps were damaged, I'm just going to guess that the emergency generator came on like it should, but perhaps was only providing single stage voltage rather than 3-stage voltage required by the pumps. There is generally not much that an operator can do to fix that in a timely enough manner to protect very small fish, unless of course a technician from Cummins NW happens to live next door or happens to be driving down I-5 in the vicinity of the hatchery when this happens.

It's just speculation on what happened if you are not personally involved in the situation. I would think they would take all the precautions in exercising the emergency equipment. In my experience standby generators are often neglected because they are a big expense for rarely used equipment. If personnel was watching the meters and gauges a problem like this would be readily apparent and could be acted upon in time to save the fish,or maybe not. I don't know how long they have before the fish start dieing without the pumps running. Maybe a rental power unit could have been delivered in time to save the fish if that contingency was planed as second level backup ,which it should be. BTW,nitpicking here..but it's single or 3 phase.

Yes, you are correct, it is single phase and 3 phase. I misspoke. And yes, you are nitpicking. In my experience, I have never seen a WDFW hatchery that does not exercise their generator once a week and have service from Cummins once a year. And BTW, you are correct that without being there personally you are just speculating as you did when you speculated that no one was there or standing by when the generator came on and that preventative measures could have been taken to prevent this.

Since we're speculating....

An incubation tray will hold around 6000+ coho eggs/fry. After the eggs are picked and put back down in the tray, a vexilar screen is added to help keep the fry from crowding up on each other if they get spooked by a noise, light, etc. I'm presuming it is a stack system, with 3 to 4 gallons of water per minute going in the top of the stack and draining out the bottom. There are usually 7 or 8 trays per stack. If the water supply shuts off, not all of the water will drain out of the tray, but the remaining water has a finite amount of oxygen and 6000 fry, if stressed, would use up the oxygen pretty quickly. 10 or 15 minutes might use up all the O2. With the generator and pumps out, there probably wasn't much that could be done.

Pump systems for hatcheries do have limitations that a gravity feed system doesn't. The hatcheries I am aware of run their emergency generators on a regular basis so they'll work in an emergency. If the emergency generator goes out, you're screwed. Mechanical systems can fail and usually seem to do so at the worst possible time. It reads like this was one of those situations.

Interesting technicalities. 10 or 15 minutes doesn't give much room for error. I guess it boils down to the risk of loosing all those fish isn't worth the cost of a redundant system to protect them.

A redundent system could be as little as a trash pump

The normal operating system is with power from the grid. The back-up generator is the redundant system. Should there be double redundancy or triple redundancy? Every hatchery person I've ever known would love to have that, but generally the agencies can't afford that. The exception might be where highly endangered populations are taken into a hatchery facility to prevent extinction such as some of the captive rearing programs, but even then systems can fail. The redundancy in those cases generally means spreading the risk across multiple facilities. The only place you routinely see multiple redundant systems is on airplanes.

And for all we know (as we are speculating), the hatchery staff could have been moving any number of temporary (trash) pumps around the station trying to keep all their fish alive, as I speculate that coho fry are not the only fish on-station.

A bunch of Chinook got moved off-site. They could have been in raceways.

It is likely that the coho fry, if they were still in the incubation trays, had just come out of the egg and hadn't buttoned up yet from absorbing the yolk sac. If that was the case, you'd probably lose most if not all, trying to move them to another location. Sounds like a "Perfect Storm situation" for the facility

From my experience with staff at State hatcheries, the last thing they want to do is lose fish. They'll jury-rig something, try something else, as long as they can. Those fish probably died because there simply wasn't anything more to do.

How many here, or in the Leg, would support having two or three backup systems? Say the place can run on one pump. How many backups do you want to pay for? How many generators when you hardly ever actually use one?

It is a cost-benefit. You could add redundancies and take the money from somewhere else. Or, raise license fees to cover additional pumps and generators.

Who else in the Department actually has a day to day relationship with the fish?

In Oregon, hatchery workers live on station for this very reason. There is someone at the facility, 24/7/365

And many facilities are set up as gravity fed, not pumping stations.

sac fry cannot me moved. They will, among other issues, get coagulated yolk and die.

Other problems with back-ups to the back-ups... is you not only add complexity to the system, thus increasing the chances for catastrophic failure, but you also greatly increase worker complacency. "Ehh...if there is a problem, the alarm/back up/etc.. will catch it..."