New Colgate Powerhouse Failure: What We’re Seeing on the Yuba

Disasters like the failure of the penstock water pipe at New Colgate powerhouse are not something anyone hopes for. But in these moments, it is crucial to take a slow, measured response to understand the environmental impacts and let the data inform the response. 

As news of the penstock failure spread the afternoon of Friday February 13th, my first job was to reach out to our partners at Yuba Water Agency to offer the support and assistance of SYRCL, if at all possible. At SYRCL, we are a science-focused, data-driven organization. That approach is one that I deeply value, and, as Executive Director, I believe is critical to our ongoing success as an organization. As the events unfolded Friday afternoon, I knew that collecting data was going to be crucial to helping us understand the near and long-term environmental impacts. I think about events like this on three different time scales: immediate response and engagement, medium-term monitoring, and the long-term implications for the Yuba watershed and water infrastructure. 

But first, let’s describe what happened.

The New Colgate Powerhouse is located approximately eight miles along the river downstream of New Bullards Bar dam. For a hydroelectric powerhouse to generate electricity, it requires hydraulic head – essentially the height of a volume of water above a reference point. To generate the head necessary for the New Colgate Powerhouse, water flows from an outlet point in New Bullards Bar dam horizontally for nearly five miles before emerging at the surface. From there, the water is piped straight downhill through the penstock to the Powerhouse where it spins the turbines generating electricity and is then put back in the river channel. The penstock pipe drops about 800 feet over roughly a mile. 

The failure to the penstock on February 13th occurred during testing of a newly installed section as part of preventative maintenance. This failure occurred at the top of the penstock pipe where it emerges from the tunnel, allowing all the water in the tunnel – about 400 acre-feet — to flow downhill (one acre-foot is an acre area with one foot of water covering it). This caused massive flooding and a mudslide which destroyed much of the Powerhouse and sent a tremendous amount of sediment and debris into the Yuba River upstream of Englebright reservoir. 

In the lower Yuba, downstream of Englebright where we see salmon, this event was characterized by a brief, but rapid drop in river flow: from 1,400 cubic feet per second (cfs) down to 560 cfs where it stayed for about an hour before increasing again to 1,400 cfs. This drop in flows was a result of a loss of power at the operations center at New Colgate Powerhouse. The facilities at Englebright (Narrows 1 and Narrows 2) are controlled in coordination with operations at New Colgate. Under normal circumstances, if there is an issue in the system that could threaten flows in the lower Yuba, YWA has personnel on site to address the issue. However, in this case, those staff were fleeing the ensuing flood and mudslide that the penstock failure caused.  

Immediate response 

Environmental disasters never occur during regular business hours. Watershed Science Director Alecia Weisman and I started planning what we needed to do Friday night and early Saturday morning, discussing where to prioritize sampling and what conditions were highest priority. With the support of Danielle Horvath, our Fisheries Restoration Program Manager, checking conditions in the lower Yuba downstream of Englebright, Alecia and I focused on water quality in the area immediately impacted by the event. The priority for water quality was those metrics necessary for biological organisms to survive: dissolved oxygen, pH, and turbidity. Dissolved oxygen is a measure of the amount of oxygen dissolved in the water. pH is a measure of acidity. Most life needs a pH close to 7 so readings much above or below that could indicate a serious water quality issue. Over the five monitoring days since SYRCL’s response, dissolved oxygen and pH have both held steady within normal ranges and comparable to the control site sampling at Bridgeport. 

Turbidity is a measure of the “cloudiness” of water as measured by how light is scattered by the tiny particles suspended in the water. The standard unit is Nephelometric Turbidity Unit or NTU. While high turbidity was obvious from observations, quantification is important. Salmonids are resilient to short-lived turbidity spikes typically associated with natural disturbances such as storm or landslide events, prolonged turbidity over 20 NTUs is known to negatively impact salmonids, detrimentally affecting feeding behaviors and ultimately resulting in poor health and sometimes death. High turbidity levels also pose a threat to the rest of the aquatic food chain.  

Our long-term river monitoring data has given us a clear picture of what normal conditions are like throughout the watershed. That data is crucial when it comes to understanding deviations from normal conditions such as these. 

At the same time, Danielle’s observations on the lower Yuba confirmed our worst fears. The rapid drop in flows had stranded thousands of juvenile Spring and Fall-run Chinook and steelhead in the cobbles along the banks. As the water dropped rapidly, juveniles that had a clear path to deeper water were able to follow the receding river water to safety. But those fish that either didn’t sense water levels dropping, or were unlucky enough to have their path blocked were left high and dry ultimately dying. Fortunately, SYRCL was on-site to collect the necessary data before the river reclaimed it. 

At that point, we began coordinating our response and plans with YWA and state agencies. 

Storm monitoring 

As winter’s second significant storm moved into the region, it became important to understand how the evolution of the mudslide and debris from New Colgate would continue to impact water quality. We are currently implementing a plan to continue to monitor the evolving situation. We will continue to monitor and sample water quality at Rice’s Crossing as the accessible location most impacted, and Bridgeport as a “control” site so that we can better understand how much the measured variables are fluctuating in a natural system. As the days go on, we will be revisiting monitoring sites so that we can coordinate a response if conditions change. The sustained efforts of our trained river monitoring volunteers make this kind of effort possible. This data is crucial to our ability to help YWA and regulating agencies understand the impacts of this event as it continues to evolve. 

As we continue to collect data during SYRCL’s Wild & Scenic Film Festival weekend, I am more than happy to answer questions or describe the latest results.